Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biogeochemical cycling of copper in the Northeast Pacific Ocean : role of marine heterotrophic bacteria Posacka, Anna 2017

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

Item Metadata

Download

Media
24-ubc_2017_november_posacka_anna.pdf [ 5.88MB ]
Metadata
JSON: 24-1.0355246.json
JSON-LD: 24-1.0355246-ld.json
RDF/XML (Pretty): 24-1.0355246-rdf.xml
RDF/JSON: 24-1.0355246-rdf.json
Turtle: 24-1.0355246-turtle.txt
N-Triples: 24-1.0355246-rdf-ntriples.txt
Original Record: 24-1.0355246-source.json
Full Text
24-1.0355246-fulltext.txt
Citation
24-1.0355246.ris

Full Text

              Biogeochemical cycling of copper in the Northeast Pacific Ocean: role of marine heterotrophic bacteria. by ANNA POSACKA B. Sc. (Hons), Zoology with Marine Zoology, University of Bangor, 2008 M.Sc., Applied Marine Science, University of Plymouth, 2010   A thesis submitted in partial fulfillment of  the requirements for the degree   DOCTOR OF PHILOSOPHY in The Faculty of Graduate and Postdoctoral Studies  (Oceanography)  The University of British Columbia (Vancouver)   August 2017   © Anna Posacka, 2017    ii  Abstract      Copper (Cu) is important in regulating microbial activity in the ocean, as it can act both as a limiting nutrient and a toxic inhibitor depending on its concentration. Yet, our knowledge of its biogeochemical cycle is limited in many oceanic regions including the subarctic Northeast (NE) Pacific, as is our knowledge of Cu nutrition in marine heterotrophic bacteria. To address this, I investigated Cu biogeochemical cycling along a coastal‒oceanic transect, Line P, in the subarctic NE Pacific (Chapter 2). I also explored physiological responses to varying Cu availability (limiting to sufficient) of taxonomically diverse heterotrophic bacteria, which include isolates from surface waters of the Line P transect (Flavobacteriia member: Dokdonia sp. Dokd-P16, and Gammaproteobacteria members Pseudoalteromonas sp. strain PAlt-P2 [coastal] and PAlt-P26 [oceanic]), and a member of the marine Roseobacter clade within class Alphaproteobacteria (Ruegeria pomeroyi DSS-3).  Several important processes were identified to moderate dissolved Cu along Line P. These include fluvial and sedimentary inputs (near the coast), upwelling of deep, Cu-rich waters in the Alaskan gyre (offshore), atmospheric inputs (offshore), as well as scavenging within the intermediate waters of the Oxygen Minimum Zone (OMZ) across the transect. Bacterial responses to changing Cu availability were diverse. Flavobacteriia member Dokd-P16 reduced its growth rate, carbon metabolism, and Cu quota (Cu:P) under Cu limitation, but enhanced its Mn quota. In contrast, both Pseudoalteromonas spp. were mostly unaffected by different Cu levels. Ruegeria pomeroyi maintained constant growth rates but moderated quotas of several metals (under low Cu: decreased Cu and Co, but increased Mn and Fe quotas), and some aspects of its C metabolism. These findings illuminate on the role of Cu in shaping bacterial species composition in the ocean, and the bacterially-mediated cycles of carbon and bioactive metals (i.e. Fe, Zn, Mn, Co). Copper quotas of heterotrophic bacteria are similar to those of cultured marine phytoplankton. Estimates of Cu partitioning between these planktonic groups in the euphotic zone of the NE Pacific revealed that up to 50% of biogenic Cu could be associated with bacterial biomass. Therefore, marine heterotrophic bacteria should not be overlooked in studies of Cu biogeochemical cycling.   iii  Lay Summary        Copper is an important element to the microbial life in the ocean. It acts as an essential nutrient or a toxic inhibitor, depending on its levels. Predicting how copper will influence marine microbes requires a good understating of 1) the processes that regulate its levels in the ocean, and 2) copper nutritional needs and toxicity thresholds of different microbial groups. The goal of my research was to improve the state of knowledge on the copper biogeochemical cycling in the subarctic Northeast Pacific. To this end, I identified the main processes that influence copper levels along a study transect in this region, the Line P. I also investigated copper nutrition in an important microbial group in the ocean, the marine heterotrophic bacteria, for the first time. My results provide insight into how copper could influence their species composition, as well as the processes that these bacteria carry out in the ocean.  iv  Preface Chapter 2 is now published as:   Posacka AM, Semeniuk DM, Whitby H., Orians K, van den Berg S, Cullen J., Maldonado MT. (2017). Dissolved copper (dCu) biogeochemical cycling in the subarctic Northeast Pacific and a call for improving methodologies. Marine Chemistry. doi.org/10.1016/j.marchem.2017.05.007  Dissolved copper samples were collected by Nari Sim & Jason McAlister (UBC). Nutrient analysis of seawater samples was performed by Department of Fisheries and Oceans scientists aboard the CCGS John P. Tully. I performed most of the sample analysis of dCu using FIA-CL method with a subset being performed by Dr. David Semeniuk. The experiments testing the effects of UV oxidation and storage on dCu concentration were designed by me with input from Dr. Kristin Orians. I performed the experiments and analyzed the samples. The dissolved Cu analysis using competitive ligand CSV method was performed by Dr. Hannah Whitby at the University of Liverpool. I generated all figures and wrote the manuscript with Dr. Maria Maldonado and Dr. Kristin Orians and the input from co-authors.  Chapter 3 is in preparation for submission to a peer-reviewed journal.  Posacka AM, Semeniuk DM, Maldonado MT. Effects of Cu availability on growth and carbon metabolism of marine heterotrophic bacteria.  Northeast Pacific bacterial strains were isolated by Dr. Nina Schuback (UBC) and phylogenetically identified by Melanie Scofield from Microbiology and Immunology at the Life Sciences Centre, UBC. The idea of examining Cu nutrition in heterotrophic bacteria was Dr. Dave’s Semeniuk and Dr. Maite’s Maldonado. I designed the experiments with the input from MM and DS. I performed all the experimental work and the majority of sample analysis, wrote scripts to analyze the data and to generate the figures. Carbon and nitrogen samples were run by Maureen Soon at UBC. Dr. Maria Maldonado and I wrote the manuscript with the input from DS.      v  Chapter 4 is in preparation for submission to a peer-reviewed journal.   Posacka AM, Semeniuk DM, Maldonado MT. Bioactive metal composition of marine heterotrophic bacteria and its response to changing Cu availability.   Northeast Pacific bacterial strains were isolated by Dr. Nina Schuback (UBC) and phylogenetically identified by Melanie Scofield from Microbiology and Immunology at the Life Sciences Centre, UBC. The idea of examining metal composition of heterotrophic bacteria as a function of Cu was Dr. Dave’s Semeniuk. I designed the experiments with the input from MM and DS. I performed all the experimental work and the majority of sample analysis, wrote scripts to analyze the data and to generate the figures. Carbon and nitrogen samples were run by Maureen Soon at UBC. Dr. Maria Maldonado and I wrote the manuscript.  vi  Table of Contents  Abstract .......................................................................................................................................... ii Lay Summary ............................................................................................................................... iii Preface ........................................................................................................................................... iv Table of Contents ......................................................................................................................... vi List of Tables .............................................................................................................................. viii List of Figures .................................................................................................................................x List of Abbreviations ................................................................................................................. xiii Acknowledgements .................................................................................................................... xiv Chapter 1: Introduction ............................................................................................................... 1 1.1 Role of copper in prokaryotes ..................................................................................... 3 1.2 Copper biogeochemistry in the ocean ......................................................................... 5 1.3 Marine heterotrophic bacteria ..................................................................................... 8 1.4 Study region: Line P, subarctic NE Pacific ................................................................ 9 1.5 Microbial ecology ..................................................................................................... 10 1.6 Objectives of the thesis ............................................................................................. 12 Chapter 2: Dissolved copper (dCu) biogeochemical cycling in the Northeast Pacific and a call for improving methodologies .................................................................................................... 20 2.1 Summary ................................................................................................................... 20 2.2 Introduction ............................................................................................................... 21 2.3 Methodologies........................................................................................................... 24 2.4 Results ....................................................................................................................... 28 2.5 Discussion ................................................................................................................. 32 2.6 Conclusions ............................................................................................................... 47 Chapter 3: Effects of Cu availability on growth and metabolism of marine heterotrophic bacteria. ..................................................................................................................................... 60 3.1 Summary ................................................................................................................... 60 3.2 Introduction ............................................................................................................... 61 3.3 Methodologies........................................................................................................... 63 vii  3.4 Results ....................................................................................................................... 72 3.5 Discussion ................................................................................................................. 74 Chapter 4: Bioactive metal composition of marine heterotrophic bacteria and its response to changing Cu availability ........................................................................................................... 97 4.1 Summary ................................................................................................................... 97 4.2 Introduction ............................................................................................................... 98 4.3 Methodologies......................................................................................................... 101 4.4 Results ..................................................................................................................... 102 4.5 Discussion ............................................................................................................... 104 Chapter 5: Conclusions ........................................................................................................... 124 5.1 Towards the improvement of dCu measurements in seawater ............................... 124 5.2 Biogeochemical cycling of Cu along Line P in the NE Pacific .............................. 127 5.3 Characterizing Cu interactions with marine heterotrophic bacteria ....................... 128 5.4 On the trace metal stoichiometry of marine heterotrophic bacteria ........................ 131 5.5 Emerging questions on the biological utilization of metals .................................... 133 5.6 Future directions ..................................................................................................... 134 References ...................................................................................................................................139 Appendices ..................................................................................................................................157         Appendix A: Supplementary material to Chapter 2……………………………………….157         Appendix B: Supplementary material to Chapter 3……………………………………….174         Appendix C: Supplementary material to Chapter 4……………………………………….180 viii  List of Tables  Table 1.1: A list of bacteria cuproproteins. .................................................................................. 16  Table 2.1: Comparison of dCu values (nmol kg-1) determined using FIA-CL with the GEOTRACES inter-calibration materials consensus values as of May 2013 (SAFe S D1 and D2) and NASS-6 certified reference material for ocean water. ............................................ 48  Table 3.1: Phylogenetic affiliation of gram-negative bacterial strains isolated from 25 m at various stations along Line P (June 2012 cruise).. ............................................................................ 86 Table 3.2: Total dissolved copper (Cutot), inorganic Cu (Cuʹ) and free Cu ion concentrations (pCu, -log[Cu2+]) in bacterial growth media. .................................................................................. 87 Table 3.3: Specific growth rates (d-1), relative growth rates (μ/μmax) and phosphorus normalized Cu quotas (Cu:P, mmol:mol) of bacterial strains at various levels of Cu in the growth media (nmol L-1). ............................................................................................................................. 88 Table 3.4: One-way ANOVA values for the effect of Cu concentration in the media on various metabolic variables of the bacterial strains ........................................................................... 89 Table 3.5: List of copper containing proteins identified in the proteome of R. pomeroyi DSS-3 obtained from the NCBI database (https://www.ncbi.nlm.nih.gov/). ................................... 90 Table 3.6: Calculations of biogenic Cu concentration (pmol L-1) for marine bacteria and phytoplankton (algae and cyanobacteria) in the seasonal euphotic zone (winter, spring, summer, Sherry et al. 1999)  at station P26 in the subarctic NE Pacific .............................. 91  Table 4.1: Total dissolved metal (Metot), inorganic metals (Meʹ) and free metal ion concentrations (pMe, -log[Me]) in bacterial growth media.. ...................................................................... 114 Table 4.2: Trace metal content of bacteria grown at different levels of Cutot (nmol L-1). Values are mean P-normalized metals (mmol:mol P) ± SE. ................................................................ 115 Table 4.3: One-way ANOVA values for the effect of Cu concentration in media on P-normalized metals (mmol Me:mol P). Statistically significant values are shown in bold. .................... 116 Table 4.4: Comparison of metal quotas (Me:C, μmol:mol) of marine heterotrophic bacteria with those of other heterotrophic bacteria from literature. ......................................................... 117  Table A.1: Comparison of samples from station P26 (collected during Aug 2011 cruise) after ~ 3 months of acidic storage (analysis Jan‒Feb 2012) and ~ 4 years of acidic storage (analysis in July 2015) without UV (storage only) and with UV treatment (storage + 2 hr of UV)...... 157 Table A.2:  Dissolved Cu (nmol kg-1) in samples from OSP at different time points after acidification to pH 2 measured using FIA-CL, with and without UV (2 h). ...................... 158 Table A.3: Dissolved Cu concentrations at OSP for 2010, 2011 and 2012 cruises.. ................. 159 ix  Table A.4: Station locations and methodologies dCu studies in the North Pacific plotted in Fig. 2.5 of Chapter 2 (Storage period was included where reported) ........................................ 161 Table A.5: Comparison of protocols used in seawater dCu determination.  ............................. 164 Table A.6: GEOTRACES reference material consensus values as of May 2013. ..................... 168 Table A.7: Calculations of vertical copper fluxes at Ocean Station Papa (OSP). ..................... 169  Table B.1: Mean (± SE) macronutrient quotas (C, N, P, S, fmol cell-1) and elemental molar stoichiometry (C:N, S:P) of marine heterotrophic bacteria as a function of Cu (nmol L-1) in the culture media. ................................................................................................................ 174 Table B.2: One-way ANOVA values for the effect of Cu concentration in media on major nutrients quota: C, N, P, and S (fmol cell-1) and the stoichiometric ratios of C:N and S:P (mol:mol) in 4 bacterial strains. ............................................................................................................... 175 Table B.3: Metabolic rates of marine heterotrophic bacteria at different levels of Cu (nmol L-1) in growth media.. .................................................................................................................... 176  Table C.1: Cellular trace metal quotas (10-20 mol cell-1) of marine heterotrophic bacteria. Values represent mean metal quotas (± SE) at different levels of Cu (nmol L-1) in culture media. 180 Table C.2: Carbon normalized trace metal quotas (μmol Me:mol C) of marine heterotrophic bacteria. ............................................................................................................................... 181 Table C.3: Comparison of trace metal concentrations of Aquil used in marine phytoplankton studies (Ho et al, 2003, Quigg et al, 2011) and this study. ................................................. 182 Table C.4:  Medians and ranges of metal contents (μmol Me:mol C) of different microorganisms. This data is plotted in Figure 4.5 of Chapter 4. .................................................................. 183   x  List of Figures  Figure 1.1: Comparison of dissolved Cu profiles in the offshore waters in the central Atlantic (Stn 14, 50°E, 27.58° N, Jacquot & Moffett, 2015), central Indian Ocean (Stn ER 10, 72°E, 20°S, Vu & Sohrin, 2013), and Northeast Pacific Ocean (Stn 17, 144.59° W. 32.41°N, Bruland 1980) ..................................................................................................................................... 17 Figure 1.2: Map of the Line P transect in the subarctic NE Pacific. ........................................... 18 Figure 1.3: Microbial transformation of phytoplankton-derived organic matter. Figure used with permission from Nature Reviews (Buchan et al., 2014) ....................................................... 19  Figure 2.1: Map of the Gulf of Alaska showing Line P transect and the 5 sampling stations (P4-P26). Bathymetry was contoured at 1000 m intervals. ......................................................... 49 Figure 2.2: Profiles of dCu (nmol kg-1) at station P26 (Ocean Station Papa). ............................. 50 Figure 2.3: Changes in labile Cu with increasing storage time: (A) 48 h, (B) 2 weeks and (C) 2 months. .................................................................................................................................. 51 Figure 2.4: Comparison of dCu datasets at station P26. .............................................................. 52 Figure 2.5: Comparison o dCu datasets in the North Pacific. ...................................................... 53 Figure 2.6: Potential temperature - salinity plot (θ-S) including all stations along the transect. . 54 Figure 2.7: Top panel (A) shows profiles of dCu (nmol kg-1), PO43- (µmol kg-1) and sigma-t (kg m-3) in the upper 300 m of stations along the Line P transect. Bottom panel (B) shows depth profiles of dCu, PO43- and O2 (µmol kg-1) throughout the entire water column sampled (10-2000 m) with the bottom depths at each station. .................................................................. 55 Figure 2.8: Profiles of dCu (nmol kg-1), PO43- (µmol kg-1) and sigma-t (kg m-3) at station P26 obtained with samples collected in Aug 2010, 2011, and 2012. ........................................... 56 Figure 2.9: Macronutrient relationships: dissolved copper with PO43- (A) and dissolved copper with Si(OH)4 (B) using data from the entire transect. .......................................................... 57 Figure 2.10: Mean mixed layer dCu and dAl trends along the Line P transect in August 2011 (dAl data from Cain, 2013) with an inset showing the spatial surface dCu distribution along the transect. ................................................................................................................................. 58 Figure 2.11: Contour plot of dCu in the upper 600m along the Line P transect in August 2011................................................................................................................................................. 58 Figure 2.12: Area averaged Aerosol Optical Depth (AOD at λ=550 nm, AquaMODIS, 1ᵒ resolution) in the Gulf of Alaska between June-August 2010, 2011 and 2012. ................... 59  Figure 3.1: Bacterial growth rates and Cu content as a function of Cu availability (nmol L-1).  . 92 Figure 3.2:  Cellular carbon (fmol C cell-1), nitrogen (fmol C cell-1), phoshorous (fmol P cell-1), sulfur (fmol S cell-1), molar ratios of C:N and S:P in Dokdonia sp. strain Dokd-P16, coastal xi  Pseudalteromonas sp. strain PAlt-P2, oceanic Pseudoalteromonas sp. strain PAlt-P26 and R .pomeroyi DSS-3 as a function of Cutot (nmol L-1) in the growth media.. ............................ 93 Figure 3.3:  Carbon metabolism of marine heterotrophic bacteria as a function of Cu (nmol L-1) in the culture media.  ................................................................................................................. 94 Figure 3.4: Comparison of Cu:C ratios (μmol: mol) of heterotrophic bacteria with other prokaryotes from literature and eukaryotic phytoplankton ................................................... 95 Figure 3.5: Comparison of Cu quotas (Cu:C, μmol: mol) of cultured marine phytoplankton (A, algae and cyanobacteria) and marine heterotrophic bacteria from this study (B).  .............. 96  Figure 4.1: Growth rates and metal quotas (mmol Me:mol P) of coastal isolate Pseudoalteromonas sp. strain PAlt-P2 and oceanic isolate Pseudoalteromonas sp. strain PAlt-P26, Dokdonia sp. strain Dokd-P16, R. pomeroyi DSS-3 at different Cu levels in culture media (nmol L-1).  119 Figure 4.2: Correlations between Cu:P and growth rate (A), Fe:P and growth rate (B), Cu level (D) in Dokdonia sp. Dokd-P16 (C); and between Co:P and Cu:P in R. pomeroyi DSS-3. . 120 Figure 4.3: Approximate inorganic metal levels [Meʹ] in culture media (A). The metal abundance profile of marine heterotrophic bacteria (B) ....................................................................... 121 Figure 4.4: Comparison of elemental stoichiometry of cultured marine heterotrophic bacteria and algae.. .................................................................................................................................. 122 Figure 4.5: Comparison of metal quotas (Me:C, μmol: mol) of different microbes.. ................ 123  Figure A.1: Comparison of dCu datasets obtained with FIA-CL between Jan-Feb 2011 (after ~ 3 months of storage, no UV) and July‒August 2015 (after ~ 4 years of storage, with 2 hrs of UV) at along the Line P transect. ........................................................................................ 170 Figure A.2: Hovmoller-longitude plot of area averaged Aerosol Optical Thickness (AOD, dark target at 0.55 microns for both Ocean and Land, 1ᵒ spatial resolution monthly means) .... 171 Figure A.3: Comparison of dissolved Cu profiles in the offshore waters in the central Atlantic (Stn 14, 50°E, 27.58° N, Jacquot and Moffett, 2015), and Northeast Pacific Ocean (Stn P26, 140.60° W, 39.60°N, Martin et al. 1989). ........................................................................... 172 Figure A.4: Dissolved Cu [dCu (nmol kg-1)] and silicic acid [Si(OH)4 (μmol kg-1)] profiles at different stations along the Line P transect. The dCu dataset is for Aug 2011 cruise (values represent those measured in UV oxidized samples after ~ 4 years of acidic storage). ....... 173  Figure B.1: Growth rates (d-1) of Pseudoalteromonas sp. strains isolated from different stations along line P (station labels on x axis).. ............................................................................... 177 Figure B.2: Regression of phosphorus values (ppb) determined in same digest samples with ICP-Q-MS and ICP-OES. ........................................................................................................... 177 Figure B.3: Trends in phosphorus (mmol Cu: mol P, top panel), carbon (μmol Cu: mol C, middle panel) and cell number normalized Cu quotas (zmol Cu cell-1, bottom panel) of 4 bacterial strains. ................................................................................................................................. 178 xii  Figure B.4: Effects of oxalate and DTPA washes on the 14C accumulation of Pseudoalteromonas sp. strain PAlt-P26.. ............................................................................................................ 179 Figure B.5: Relationship between growth rate (d-1) and respiration rate (BRcell [fmol O2 cell-1 day-1]) of Dokdonia sp. strain Dokd-P16 measured at different levels of Cu (nmol L-1) represented by different symbols. .......................................................................................................... 179  xiii  List of Abbreviations  AOD Aerosol Optical Density BB bacterial biomass BP bacterial productivity BCD bacterial carbon demand BGE bacterial growth efficiency Cuʹ copper prime, inorganic Cu Cu-L copper bound to ligand (L) CLE-CSV Competitive Ligand Exchange‒Cathodic Stripping Voltammetry dCu dissolved copper dAl dissolved aluminum FIA-CL Flow Injection Analysis‒Chemiluminescence  GOA Gulf of Alaska Me Metal Meʹ inorganic metal Me:C quota as metal to carbon ratio  Me:P quota as metal to phosphorus ratio NESP Northeastern Subarctic Pacific OMZ Oxygen Minimum Zone OSP Ocean Station Papa pCu negative log of Cu2+ concentration SOD superoxide dismutase SOW synthetic ocean water SAFe Sampling and Analysis of Iron reference material SD standard deviation SE standard error UVO Ultraviolet oxidation   xiv  Acknowledgements          I would like to express my deepest gratitude to my supervisor Maria (Maite) Maldonado. Thank you, Maite, for granting me a generous degree of autonomy in my research, while always being available for moral, scientific and hands-on technical support, even during the most inconvenient of times. I also appreciate you providing me with the opportunities of exploring different research directions, and to take time off when I very much needed to.          I am also indebted to my committee member Kristin Orians for her substantial input into my research, as well as Jay Cullen and Andrew Ross for their guidance and constructive comments throughout my project. I would also to acknowledge all my supervisory committee members, especially Maite, for extensive editorial assistance with my thesis. I would also like to thank my former MSc supervisor Maeve Lohan for encouraging me to pursue a Ph.D., and introducing me to the research opportunities at Maite’s Maldonado and Kristin’s Orians labs.         Along my Ph.D. journey, I was extremely fortunate to receive the support (both moral and scientific) of many wonderful people from my research group and others in the EOAS department. I would like to especially thank Dave Semeniuk, whose science and ideas inspired many aspects of my project, including the investigation of Cu nutrition in marine bacteria. Thanks Dave for your mentorship over the years and helping me to become a better scientist. I owe enormous thanks to Jan Finke for his substantial technical support with the flow cytometry analysis of my samples and the Curtis Suttle lab for hosing me to do this work. I would also like to acknowledge the help of Melanie Scoffield and the Steven’s Hallam lab with the identification of my bacterial strains. Thank you, Jason McAlister, Nari Sim and Amy Cain – for collecting my samples and your input on their analysis and interpretation, Anna Hippmann and Nina Schuback – for always being there for me, Robert Izett – for being my office’s Oprah, Tereza Jarnikova and Alysia Herr – for giving me sneaky hugs as I was writing my thesis, Dave Capelle – for contemplating science with me on the lab couch. Also, massive thanks to Lindsay Fenwick, William Burt, Ross McCulloch, Mariko Ikehata, Jade Schiller, Lizzy Asher, Kristina Brown, Chen Zeng, Iselle Flores, Sarah Rosengard, Jingxuan Li, Manuel Colombo, Francois Choi, Kang Wang, Jian Guo, Jeffrey Charters and Julia Gustavsen, for your wonderful company on campus. xv          I wish to thank the following scientists for their insights, advice, inspiration or technical support: Yeala Shaked, Robert Sherrell, Hannah Whitby, Dave Janssen, Christina Schallenberg, Philippe Tortell, Stan van den Berg, Marion Fourquez, Maeve Lohan, Kristen Buck, Jessica Fitsimmons, Travis Mellet, Claire Parker, Rob Middag, Vivian Lai, Marg Amini and Maureen Soon. I would also like to extend my gratitude to the captains and crew of the CCGS JP Tully, and the scientists at the Institute of Ocean Sciences, Sidney for giving me the opportunity to experience work at sea.         I would like to express my sincere gratitude to the University of British Columbia, Marine Technology Society and NSERC for financial assistance.        Thank you, Mom, Dad, Pietras, and Asia for fueling my zest for adventure, and for helping me to travel to Italy, the UK and eventually to Canada, to explore oceanography. I must also thank my wonderful Babcias (Grandmas), for their love and prayers, and for inspiring me to be tenacious, hardworking but also to have fun in life. I am also extremely grateful for having the support of Louise, Ciara, Cecilia, Rosie, Bronwen, Fran, Jen, Mike, Agata, Kamila, and Kasia. Kochani, dziekuje.         Completing this work would have not been possible without my “Canadian family”: Shona, Stuart, Ross, Nicole, Zoe, Sarah, George, Kyle, Anne and our new member Aidan who decided to join us during my defense. I could have not wished for more amazing, supportive friends. Thanks for the many fun and uplifting moments during my project, as well as for looking after me and Jeff.         Finally, there are no words that can fully express my gratitude for the love and support of my soulmate Jeff. Thanks for being there for all three degrees, attempts to help me with my experiments and for cooking me pierogis.    1  Chapter 1:  Introduction         Microbial activity is a critical component of global biogeochemical cycles, including those of the climate-active gases CO2, N2O and CH4 (Arrigo 2005; Glass and Orphan 2012). In addition to the basic requirement for major elements (C, H, N, O, P and S) as building blocks of macromolecules (carbohydrates, lipids and proteins), microbes also rely on a suite of essential trace metals that are extremely scarce in the open ocean (ranging from pico [10-12] to nanomolar [10-9] concentrations). These include iron (Fe), copper (Cu), manganese (Mn), cobalt (Co) and zinc (Zn), the first-row transition metals, which function as cofactors for various enzymes and as centers for stabilizing enzyme and protein structures. While important as nutrients, some of these metals, such as Cu, can also be toxic at moderate concentrations, which can inhibit sensitive microorganisms  (Brand et al. 1986). Among the essential trace metals, iron (Fe) has been studied most intensively due to its importance in limiting phytoplankton growth in ~30‒50% of the world’s oceans (Moore et al. 2004). Although large-scale spatial limitation of biological productivity has not been reported for metals other than Fe, they are recognized to have an important role in structuring microbial community composition (Sunda 2012), which has implications for the cycling of C and  N in marine ecosystems (Twining and Bains, 2013). On the other hand, microbial communities play a role in the oceanic cycles of bioactive metals through a variety of mechanisms. These include metal uptake (phytoplankton and bacteria), regeneration (bacterial decomposition of organic matter), and production of organic chelators, either to sequester essential metals  (e.g siderophores produced by marine heterotrophic bacteria, Granger and Price 1999) or to decrease the bioavailability of those that are potentially toxic (e.g ligands produced by cyanobacteria in response to high Cu stress, Moffett and Brand 1996).          Given the importance of bioactive metals in regulating biological activity in the ocean, there is a need to understand the processes that influence trace metal distribution (sources and sinks), and this ultimately metal availability. Knowledge of the trace metal biogeochemical cycling, including that of Cu, is still profoundly limited for many parts of the world’s oceans. There is also a need to understand the metal nutritional requirements and toxicity thresholds of different 2  members of the microbial community. While significant advances have been made on this front for some groups, particularly the marine phytoplankton (algae and cyanobacteria) (e.g. see reviews of Sunda 2012; Twining and Baines 2013), others such as the non-photosynthetic prokaryotes have so far received limited attention. Rare examples, however, demonstrate that insufficient supply of certain metals limits model organisms from important bacterial groups such as denitrifying bacteria (Cu; Granger and Ward 2003; Moffett et al. 2012), ammonia oxidizing archaea (Cu; Amin et al. 2013) and aerobic marine heterotrophic bacteria (Fe; Tortell et al. 1996; Kirchman et al. 2003a; Fourquez et al. 2014a). It is of great interest to understand metal nutrition of the heterotrophic members of the bacterial community as they account for ~ 50% of the total particulate carbon in the open ocean (e.g. Kirchman et al. 1993; Sherry et al. 1999) and regulate carbon fluxes through remineralization of organic material (Azam et al. 1983). Furthermore, depending on their metal nutritional requirements, these bacteria could play an important role as a biological sink of metals. Indeed, evidence for such a role of heterotrophic bacteria already exists for Fe (Tortell et al. 1996, 1999; Maldonado and Price 1999). However, in contrast to Fe, there is a complete lack of understanding of the nutritional role of other bioactive metals (Cu, Mn, Co, Zn, Ni) in marine heterotrophic bacteria, and similarly the bacterial role in cycling of these metals in the ocean.     In this thesis, I decided to study Cu because of its dual biological role, as either limiting or toxic metal to planktonic communities depending on concentrations, and because Cu distribution and cycling is poorly understood in many oceanic regions. To this end, my goal was to improve the state of knowledge on Cu biogeochemistry in the subarctic northeast (NE) Pacific, where only fragmentary datasets of dissolved Cu are available. My work provides the first water-column profiles of dissolved copper along the coastal‒oceanic transect Line P, which is one of the longest time-series programs in the world (Chapter 2). Another goal of my thesis was to explore, for the first time, the importance of Cu nutrition in marine heterotrophic bacteria. To achieve this, I conducted experiments on strains isolated from the surface waters along Line P, as well as a bacterium from a culture collection (all are gram-negative bacteria and belong to major bacterial clades of the world’s oceans, Chapter 3 and 4). These experiments focused on characterizing bacterial responses in growth, C metabolism and elemental stoichiometry (C, N, S, P, Fe, Zn, Cu, Mn, Co) to variable Cu availability (limiting‒replete). In the following sections, I provide an 3  overview of the biological role of Cu in prokaryotes, Cu biochemical cycling in the ocean, hydrography of the study area, and the ecology of the bacterial strains investigated in my thesis. In the final section of this introduction, I provide the thesis outline.  1.1 Role of copper in prokaryotes     Copper is ideal for the catalysis of electron transfer reactions in the cell, as it can easily undergo inter-conversion between the reduced Cu(I) and oxidized Cu(II) forms. However, while this property is useful for a variety of biological pathways, it is also what makes copper potentially hazardous. This is because of its participation in the Fenton-like reactions when cycling between Cu(I) and Cu(II) states, which generates reactive oxygen species (ROS), and may promote oxidative stress (Rowley and Halliwell 1983; Gunther et al. 1995). Another mechanism by which copper could be harmful to the prokaryotic cell is by damaging essential enzymes. In the reducing environment of bacterial cytoplasm copper is expected to be mainly present as Cu(I), a species that has a strong affinity for binding with thiolates and other sulfides (Dupont et al. 2011; Rensing and McDevitt 2013). As a result, Cu can degrade Fe-S clusters of dehydratases (Macomber and Imlay 2009), and by displacing the Fe atom, it can deactivate these critical enzymes and increase oxidative stress. A challenge for the prokaryotic is to satisfy its physiological requirement for copper-catalyzed processes while at the same time avoiding potential Cu-induced toxicity. Prokaryotes have, therefore, evolved efficient mechanisms to maintain strict homeostatic control over Cu (see below and reviews of Rensing and Grass 2003; Solioz and Stoyanov 2003; Solioz et al. 2010; Bondarczuk and Piotrowska-seget 2013; Porcheron et al. 2013)      There are 11 known prokaryotic Cu-containing proteins (cuproproteins, Table 1.1), and genomic studies suggest that many more may be awaiting identification (e.g. Ridge et al. 2008; Gladyshev and Zhang 2013). Genomic analysis of prokaryotes (450 and 35 of sequenced bacterial and archaeal genomes, respectively) revealed that the heme-Cu respiratory enzyme COX is the most common bacterial cuproprotein (identified in 91% of bacteria, Ridge et al. 2008). Other important enzymes with significant occurrence in prokaryotes include the respiratory enzyme NADH dehydrogenase-2 (34%) and anti-oxidative stress enzyme Cu/Zn superoxide dismutase (21%). Remaining cuproproteins appear to be less common in prokaryotes (nitrosocyanin – 15%, 4  plastocyanin – 15%, Cu-containing nitrite reductase ‒10%, Cu amine oxidase – 6%, pMMO – 2%, CotA – 1%, and tyrosinase ‒1%).     Prokaryotic pathways involved in Cu uptake and nutrition are not well understood, in contrast to those involved in Cu resistance (see reviews of Ma et al. 2010; Argüello et al. 2013; Porcheron et al. 2013). In gram-negative bacteria (having an outer membrane and inner plasma membrane), the passage of Cu through the outer membrane may occur via porins (Balasubramanian et al. 2011; Mermod et al. 2012). Several non-copper specific transporters were identified in different model bacteria as importers of Cu at the next barrier – the plasma membrane. These include zinc transporter ZupT in Escherichia coli (Grass et al. 2002), the MtsABC in Streptococcus pyogenes (Janulczyk et al. 1999), HmtA (the PB1-type ATPase) in Pseudomonas aeruginosa (Lewinson et al. 2009). A unique example of a Cu uptake mechanism is that of the facultative phototroph, R. capsulans. In this bacterium, copper influx through the cytoplasmic membrane is coupled to its immediate efflux (via the MSF family importer (CcoA) and the efflux pump PB1-type ATPase), which appears to be required for the assembly of periplasmic Cu-containing terminal oxidase (cbb3 – COX) (Ekici et al. 2012). How other bacteria acquire copper from their environment is largely unknown at present.    Mechanisms employed by bacteria to reduce Cu toxicity involve restricting the use of Cu in the cytoplasm and Cu efflux. For instance, in the gram-negative bacterium E.coli the respiratory enzyme COX is located in the plasma membrane, with the catalytic sites facing the periplasmic space (Tottey et al. 2005). Copper efflux via Cu+ transmembrane transporters PIB1-type ATPases is the most common Cu detoxification mechanism the bacterial kingdom (Ridge et al. 2008). These are believed to pump Cu at a faster rate than PB1-type ATPase involved in Cu insertion into cytochrome c oxidases (Rensing and McDevitt 2013).    Studies on Cu limitation (as opposed to toxicity) of prokaryotes have been rare. Only a few examples in literature demonstrating how these organisms respond to low Cu. For instance, in denitrifying bacteria Pseudomonas stutzeri and Paracoccus denitrificans, Cu limitation leads to accumulation of nitrous oxide (N2O) gas as a result of the inhibition of the nitrous oxide reductase, a copper containing enzyme necessary to complete denitrification to N2 (Granger and Ward 2003). Copper starvation inhibits growth and oxidation of ammonium (NH3) to nitrite (NO2-) by the archaeon Candidatus Nitrosopumilus maritimus, which requires Cu-dependent ammonia 5  monooxygenase (AMO) for this process (Amin et al. 2013).  Gene regulation caused by low Cu availability has been investigated in the facultative anaerobe gram-negative pathogenic bacterium P. aeruginosa (Frangipani et al. 2008). This bacterium has four Cu-containing terminal oxidases, namely bo3 oxidase (Cyo), the aa3 oxidase (aa3), cbb3 oxidase 1 (cbb3-1), and cbb3 oxidase 2 (cbb3-2), along with cyanide insensitive oxidase (CIO), a terminal oxidase that lacks Cu (Kawakami et al. 2010). When experiencing Cu starvation, P. aeruginosa relies on the CIO for aerobic respiration and depresses expression of genes involved in Fe uptake, which may be related to the reduced Fe requirement for respiratory system under low Cu conditions (Frangipani et al. 2008).   1.2 Copper biogeochemistry in the ocean 1.2.1  Sources of Cu to the ocean    Sources of Cu to the ocean include riverine discharge (Bruland and Franks 1983; Jacquot and Moffett 2015), fluxes from deep-sea marine sediments (Boyle et al. 1977; Klinkhammer 1980; Fischer et al. 1986), inputs from the continental shelves (Heggie et al. 1987) as well as hydrothermal vents (Sander and Koschinsky 2011) and atmospheric deposition (Duce et al. 1991). The latter may be a particularly important for Cu inputs in regions of the ocean that are downwind of major natural and anthropogenic aerosol sources such as the Mediterranean (e.g. Koçak et al. 2004; Jordi et al. 2012), and the East China Sea (Hsu et al. 2010). In such regions, deposition of Cu-rich aerosols of anthropogenic origins, which are high in soluble Cu, has ecotoxicological implications (Paytan et al. 2009; Jordi et al. 2012).    Dissolved copper concentrations in surface waters are lower in regions far removed from aeolian and sedimentary continental sources. For instance, in the central Atlantic Ocean dissolved Cu (dCu) concentrations are ~ 0.5 nmol L-1, compared to ~1‒1.5 nmol L-1 near continental margins (Jacquot and Moffett 2015; Roshan and Wu 2015). In coastal bays and estuaries, especially those that are influenced by anthropogenic activity, dCu concentrations can be as high as 50 nmol L-1  (eg. San Francisco Bay, Buck and Bruland 2005). In surface waters, dCu levels are lower relative to those at depth due to the biological acquisition. The residence time of Cu in the upper ocean is 6  approximately 2 to 9 years (Takano et al. 2014; Semeniuk et al. 2016a), a much longer time than that of other bioactive metals, such as Fe (17 to 424 days, Bergquist and Boyle 2006; Bowie et al. 2015). Estimates of the overall Cu residence time in the ocean vary from 1500 to 6400 years (Boyle et al. 1977; Little et al. 2014; Takano et al. 2014), which is intermediate relative to the average ventilation time of ocean water (∼1000 years for deep ocean water). Deep-water Cu concentrations in the Atlantic Ocean are lower than in the Indian and the Pacific oceans. As young, deep-water is transported from its source of formation (the Atlantic) towards the Pacific, copper is regenerated at depth and consequently it concentrations increase with the age of deep water. The ocean-circulation driven patterns in deep-water dCu concentrations can be seen in Fig. 1.1, which compares its distribution throughout the water column in the Atlantic, Indian and Pacific oceans.  1.2.2  Copper chemistry in seawater    Copper has a complex geochemical behaviour in seawater, as it is both particle reactive and redox sensitive.  In oxic seawater, the major forms of inorganic Cu (Cu´) are CuCl+(aq), CuCO3°, Cu(OH)+ and Cu2+ (Yiğiterhan et al. 2011), with CuCO3° being most abundant (60%, van den Berg, 1984). However, these inorganic complexes are only a minor fraction of the total dCu pool (<1%) which is primarily composed of strong organic ligand complexes, termed Cu-L (Coale et al. 1988; Moffett and Dupont 2007; Buck et al. 2010; Bundy et al. 2012; Jacquot and Moffett 2015; Semeniuk et al. 2016a). Organic complexation plays an important role in modulating the bioavailability of Cu to marine microorganisms. By reducing the levels of the most biologically available form of Cu (Cuʹ), organic complexation may aid in preventing Cu-induced toxicity in sensitive organisms such as cyanobacteria (Moffett and Brand 1996), while potentially limiting those with elevated Cu requirements (Annett et al. 2008; Guo et al. 2012a; Amin et al. 2013). However, there is a growing evidence from both field and lab studies that phytoplankton are also able to acquire organically-bound Cu (Quigg et al. 2006; Annett et al. 2008; Semeniuk et al. 2009, 2015; Guo et al. 2010), raising uncertainty regarding the degree to which organic complexation limits Cu bioavailability (Semeniuk et al. 2015).      In reducing environments Cu occurs mainly in its cuprous form, Cu(I). In this form, Cu acts as a “soft” class B metal with a tendency to react with H2S (> 1µM H2S, Nameroff et al, 2002) and 7  form inert precipitates (Jacobs et al. 1985; Landing and Lewis 1991; Yiğiterhan et al. 2011). In some anoxic regimes with high sulfide levels, such as the deep Black Sea (> 100 μM), Cu is almost completely stripped out of solution, resulting in Cu enriched sediments (Haraldsson and Westerlund 1988). Hence, anoxic environments may play an important role as sinks for dCu.  1.2.3 Copper profile in the water column      The distribution of copper within the water column is largely influenced by biological cycles. In the upper ocean, copper concentrations are low as a result of microbial uptake and increase gradually with depth as the metal is being remineralized from sinking organic particles (Bruland et al. 2014). The influence of biological cycles (uptake and remineralization) is most evident in the northeastern Pacific where linear correlations between the dissolved metal and reactive phosphate have been observed in the oceanic nutricline (Sunda and Huntsman 1995, using dataset of Bruland 1980). The ratio of Cu:P ratio in the NE Pacific nutricline (0.44 mmol Cu:mol P, given by the slope of the linear Cu:P plot) agrees with the average ratios of cultured marine phytoplankton (0.38 mmol Cu:mol P, Ho et al. 2003), highlighting the role of phytoplankton assimilation of Cu in regulating the dCu concentrations in surface waters. However, the positive y-intercept of the Cu:P relationship for the northeastern Pacific indicates that Cu is not completely draw down by the resident phytoplankton community in these waters (Sunda 2012). In contrast to other biologically-controlled trace metals such as Zn, the copper profile lacks a mid-depth maximum, and its concentrations in oceanic deep waters increase in an almost linear fashion towards the ocean floor as shown in Fig 1.1. Mechanisms that have been proposed to explain this feature include flux of Cu from the seafloor and its scavenging by particles in the water column (Boyle et al. 1977) or reversible-scavenging (Little et al. 2013). A recent study found that the isotopic composition of dCu becomes heavier with the age of seawater, likely due to the preferential scavenging of the lighter Cu isotope (63Cu, Takano et al. 2014). This observation supports the idea that copper profile is likely maintained by scavenging of the metal throughout the water column and its flux from seafloor sediments as originally proposed by Boyle et al. (1977). Owing to the combination of processes described, the dCu profile has been classified as a nutrient-scavenged hybrid type (Bruland et al. 2014). 8   1.3 Marine heterotrophic bacteria    Heterotrophic bacteria are the major remineralizers of organic material (OM) in the ocean, and as a result impact the cycling of energy, carbon and essential nutrients (Azam et al. 1983; Azam and Malfatti 2007). They are an important component of the so-called microbial loop, influencing the carbon cycle in two ways: (1) via the release of CO2 through oxidation of OM, and (2) assimilation of the dissolved organic carbon (DOC) into biomass, which facilitates the transfer of carbon to higher trophic levels (Fig 1.3). On average, approximately 50% of the oceanic C fixed by primary producers is subject to bacterial consumption, highlighting their significance in the global C cycle (Azam et al. 1983). In addition, bacterial DOC transformation generates recalcitrant organic carbon, and through this process termed the microbial carbon pump carbon can be sequestered in the deep ocean for thousands of years (Fig. 1.3, Jiao et al. 2010). To understand the impact of marine heterotrophic bacteria on the global C cycle, we need to elucidate variables that regulate bacterial carbon metabolism, including parameters such as bacterial respiration (BR), productivity (BP), carbon demand (BCD) and bacterial growth efficiency (BGE). The latter parameter quantifies the proportions in which bacteria designate carbon towards biomass formation (anabolism) and energy production (catabolism) (del Giorgio and Cole 1998; Rivkin and Legendre 2001), and varies in response to different environmental conditions. For instance, under Fe deficiency, some heterotrophic bacteria lower the amount of C allocated towards biomass (reduced BGE), resulting in a larger amount of C respired as CO2 (Tortell et al. 1996; Kirchman et al. 2003a). Iron availability is critical to the growth and metabolism of marine heterotrophic bacteria (Tortell et al. 1996; Kirchman et al. 2003a; Fourquez et al. 2014a; Obernosterer et al. 2014), which can be explained by a high Fe requirement for the bacterial respiratory chain (Tortell et al. 1996). In response to iron deficiency, heterotrophic bacteria reduce their Fe quota  (Tortell et al. 1996; Granger and Price 1999; Fourquez et al. 2014a), electron transport chain activity (Tortell et al. 1996), and modify carbon metabolic pathways (citric acid cycle, glycolysis and respiratory chain, Fourquez et al. 2014a). Much less is known about the requirements for other essential metals in marine heterotrophic bacteria. Besides iron, the only other trace metals for which bacterial quotas exist to my knowledge are Zn (4 species) and Mn (2 species, Vogel and 9  Fisher 2010), while assessments of physiological effects of varying availability of many essential metals are lacking. This includes copper, even though this metal has the potential to influence how bacteria metabolize and cycle C in aquatic ecosystems given its role in respiration (via COX, NADH-2) and acquisition of complex carbon sources in some bacteria (via laccases, amine oxidases).   1.4 Study region: Line P, subarctic NE Pacific    Line P is a coastal to open ocean transect in the subarctic Northeast Pacific that extends from the continental shelf of Vancouver Island (BC) to Ocean Station Papa (OSP, ~ 1600 km offshore) in the anti-clockwise rotating, upwelling Alaska gyre (Fig. 1.2). Surface water movement in this region includes an eastward flowing North Pacific Current, which bifurcates into the northward Alaskan Stream current and southward California Current near the west coast of North America (Fig 1.2). Near the coast of BC, there is an oscillation between downwelling conditions in the winter (southerly flowing winds) and upwelling conditions in the summer (northerly flowing winds caused by an offshore high-pressure system Whitney et al. 2005). Coastal stations of Line P are influenced by the poleward flowing California Undercurrent (150‒200 m, Thomson, and Krassovski, 2010 σt= 26.5–26.8, Pierce, 2000). The westernmost station of the transect (OSP) is at the edge of the North East Pacific Intermediate (NPIW) water (Talley 1993; Ueno and Yasuda 2003) which forms in the Sea of Okhotsk near the coast of Japan (Shimizu et al. 2004) travelling at 150‒500 m (σt= 26.5–27.0, Talley 1993). In the interior waters of the transect, there is an extensive Oxygen Minimum Zone (OMZ) extending from ~300 down to 2000 m with O2 concentrations ranging between ~7 – 60 µmol kg-1.     Line P is characterized by distinct nutrient regimes: continental shelf waters (NO3 limited in summer), a transition area (which may experience NO3 depletion in summer), and an open ocean region characterized by high macronutrients and low chlorophyll a (HNLC) (Boyd and Harrison 1999), where biological productivity is Fe-limited (Martin and Fitzwater 1988). Bacterial biomass (BB) and production (BP) in the euphotic zone of Line P shows little variability in the winter (~12 µg C L-1 and ~0.5 µg C L-1 d-1, respectively) while in spring and summer BB ranges from ~20 to 49 µg C L-1, and BP from 1 to 6 µg C L-1 d-1 along the transect (Sherry et al. 1999). Growth rates 10  of heterotrophic bacteria measured in the euphotic zone at station Papa were reported to be low (< 0.1 d-1) compared to growth rates of phytoplankton (0.1 to 0.8 d-1)  (Kirchman et al. 1993). The major taxonomic bacterial lineages in surface waters along Line P (10 m) include the members of Alphaproteobacteria cluster SAR11, the cyanobacterial genus Synechococcus, Marine Group A (MGA), Gammaproteobacteria cluster SAR86, and subgroups affiliated with Bacteroidetes (order Flavobacteriales), Alphaproteobacteria (order Rhodobacteraceae among many others), Gammaproteobacteria and Betaproteobacteria (order Methylophilales) (Wright 2013).  1.5 Microbial ecology    The strains used in this thesis belong to major bacterial phyla in the ocean, namely Alphaproteobacteria, Bacteroidetes and Gammaproteobacteria (Yilmaz et al. 2016). Members of the Gammaproteobacteria (order Alteromonadaceae) were isolated from surface waters (~ 25 m) at two distinct stations of Line P and identified as Pseudoalteromonas sp. strain PAlt-P2 (coastal station P2) and PAlt-P26 (oceanic, low Fe station P26). A member of the Bacteroidetes was isolated from midway station (P16) and identified as Dokdonia sp. Dokd-P16 (class Flavobacteriia). The Alphaproteobacteria member Ruegeria pomeroyi DSS-3 of the marine Roseobacteria clade (MRC) was obtained from a culture collection (Chapter 3, methods section 3.2 provides detailed of the identification). Alphaproteobacteria, Bacteroidetes, and Gammaproteobacteria are important components of the prokaryotic communities in surface waters of Line P (Wright, 2015). These microbial phyla are the major players in transformations of algal-derived organic matter (Buchan et al. 2014), and during phytoplankton blooms, they typically display successive patterns in abundance and substrate decomposition (Pinhassi et al. 2004; Teeling et al. 2012).      Members of the Bacteroidetes, formerly known as Cytophagia-Bacteroidetes-Flavobacteria (CFB), account for 10‒70% of the microbial communities in pelagic zones, in both productive coastal and offshore waters  (Glöckner et al. 1999; Eilers et al. 2000; Kirchman et al. 2003b; Abell and Bowman 2005; Gόmez-Pereira et al. 2010; Williams et al. 2013). They are known for their algal associations, responsiveness to algal blooms (Riemann et al. 2000; Pinhassi et al. 2004; Teeling et al. 2012; Williams et al. 2013), and adaptations to decomposition of complex bio-11  polymers  (eg. chitin and protein, Cottrell and Kirchman 2000; algal polysaccarides, Mann et al. 2013), and were also found to play role in biogenic silica remineralization (Bidle et al. 2003). Their populations typically increase during the decay of phytoplankton blooms (Teeling et al. 2012), and it has been proposed that their decomposition of complex algal material during these events provides simple substrates for alpha- and gammaproteobacteria (Teeling et al. 2012; Williams et al. 2013). There is increasing evidence, that Flavobacteriia are the major group of the Bacteroidetes (Cottrell and Kirchman 2003; Alonso et al. 2007), a phylum that dominates the whole genome sequences of proteorhodopsin-containing bacteria (Riedel et al. 2013).  Proteorhodopsins (PR’s) are membrane bound proteins that function as light-harvesting proton pumps (Béjà and Suzuki 2008), considered to supplement bacterial growth under oligotrophic conditions (Kwon et al. 2013). Indeed, PR-mediated phototrophy enhances growth of the flavobacterium Dokdonia sp. MED123, under low availability of labile carbon sources (Gómez-Consarnau et al. 2007). PR encoding genes were identified in other isolates of the genus Dokdonia (strain PRO95, 4H-3-7-5, and Dokdonia donghaensis DSW-1), and these bacteria have served as models for studies on bacterial PR-mediated phototrophy in the ocean (Gómez-Consarnau et al. 2007; Gonzalez et al. 2011; Riedel et al. 2013; Bogachev et al. 2016; Kim et al. 2016). Many Flavobacteriia members, including the Dokdonia sp. Dokd-P16 used in this study, produce a flexiburin-type pigment, which results in their yellow-orange pigmentation  (Reichenbach et al. 1980).    Gammaproteobacteria are widespread in the ocean’s photic zone (Rappé et al. 2000; Venter et al. 2004), accounting for up to 30% of the bacterial community in some regions (Ruiz-González et al. 2012; Signori et al. 2014). Gammaproteobacteria of the of Pseudoalteromonas genus are amongst the most widely known and readily cultivable microorganisms from the marine environments (Giovannoni and Rappé 2000). These bacteria are frequently found in associations with eukaryotic organisms (e.g. algae, fish), and are known for the production of a variety of bioactive compounds with antibacterial, algicidal, antifungal, agarolytic and antiviral properties (Holmström and Kjelleberg 1999). Isolates of Pseudoalteromonas sp. served as useful models for examining physiological role of Fe in marine heterotrophic bacteria (Tortell et al. 1999), and the production of siderophores by these microbes (Sijerčić and Price; Granger and Price 1999; Armstrong et al. 2004)  12           Members of the marine Roseobacter group within the family Rhodobacteraceae comprise 5-20% of coastal and oceanic bacterioplankton (González and Moran 1997; Eilers et al. 2000; Giovannoni and Rappé 2000; Selje et al. 2004). This group is characterized by having diverse metabolic strategies including sulfur metabolism, secondary metabolite production, methylotrophy, and anoxygenic photosynthesis via bacteriochlorophyll a (Buchan et al. 2005, and references herein). Roseobacter genomes feature highly diverse trace metal acquisition pathways, a feature which may be beneficial for exploitation of different trace metal niches (Hogle et al. 2016). A reference strain for the Roseobacter clade is Ruegeria pomeroyi DSS-3 (Cunliffe 2016, formerly known as Silicobacter pomeroyi), whose genome was published in 2004 and represents the first sequence of a bacterium from this clade (Moran et al. 2004). This bacterium can supplement its heterotrophic metabolism with lithoheterotrophy (using inorganic compounds carbon monoxide and sulphide), and perform denitrification (Moran et al. 2004). Ruegeria pomeroyi DSS-3 has been used to study a variety of ecological processes, including organic sulfur degradation of algal osmolite dimethylsulphoniopropionate (DMSP, Howard et al. 2006; Todd et al. 2012; Reisch et al. 2013), the cryptic organic carbon and sulfur degradation via catabolism of algal sulfonate compound 2,3- dihydroxypropane-1-sulfonate (DHPS, Durham et al. 2015), and purine degradation (Cunliffe 2016).  1.6 Objectives of the thesis       The goals of this thesis were 1) to improve our knowledge of Cu biogeochemistry in the subarctic NE Pacific Ocean, and 2) to explore Cu nutrition in marine heterotrophic bacteria and assess their role in Cu cycle. The study area selected for my work is a coastal-oceanic transect located in the Gulf of Alaska called Line P, were water-column Cu concentrations were not measured prior to my study, except for one station (Station Papa, Martin et al. 1989). To investigate the importance of Cu availability in heterotrophic bacteria, I used bacterial strains isolated from surface waters along the Line P transect (Dokdonia sp. Dokd-P16, Pseudoalteromonas sp. coastal strain PAlt-P2 and oceanic strain PAlt-P26) and a model bacterium from culture collection (R. pomeroyi DSS-3). This dissertation includes the three chapters summarized below.  13  Chapter 2: Dissolved copper (dCu) biogeochemical cycling in the subarctic Northeast Pacific and a call for improving methodologies      This chapter is composed of two parts: 1) an evaluation of the analytical methodology used to determine dissolved Cu (dCu) concentrations in collected samples, and 2) the interpretation of this dCu dataset. The first methodological component of Chapter 2) was motivated by a recent recommendation for UV oxidation (UVO) of seawater samples prior to the determination of the total dCu (GEOTRACES Standards & Reference material statement, 2013, http://www.geotraces.org/science/intercalibration/322-standards-and-reference-materials). In this study, I determined dCu concentrations in Line P samples with a flow injection analysis-chemiluminescence method (FIA-CL), following different storage periods under acidic conditions. To my knowledge, this is a first study to investigate the effect of storage time of seawater samples on dCu concentration. In addition, a subset of the Line P samples was analyzed using two distinct methods (FIA-CL and CLE-CSV). I also performed a comprehensive comparison of my dCu dataset with other published datasets in the North Pacific as well as reviewed recent methods for dCu analysis in seawater. The results demonstrate that dCu concentrations are influenced by UV-oxidation, storage period and the analytical method.     In the second component of Chapter 2, I interpreted the distribution of dCu along Line P within the context of regional oceanographic features (water masses, OMZ, upwelling in the Alaskan gyre) and processes (aeolian deposition, continental inputs). In addition, a temporal investigation of dCu at the offshore terminal station of the transect (OSP) was examined over three consecutive years (2010‒2012). The results of this study identify fluvial and sedimentary sources near the coast of BC, and upwelling of deep, dCu rich waters in the Alaska gyre offshore, as the main processes moderating dCu along Line P. Preliminary evidence suggests that aeolian deposition (Cu inputs) and scavenging of Cu within the OMZ (Cu removal) may also play a role in regulating Cu levels in the study region.      14  Chapter 3: Effects of Cu availability on growth and metabolism of marine heterotrophic bacteria      The goal of this chapter was to evaluate the role of Cu nutrition in marine heterotrophic bacteria. I used phylogenetically diverse bacterial strains isolated from different locations along the Line P transect characterized by contrasting ecological regimes; coastal Pseudoalteromonas sp. PAlt-P2 (Stn P2), oceanic Dokdonia sp. Dokd-P16 (Stn P16) and oceanic/low Fe waters Pseudoalteromonas sp. PAlt-P26 (Stn P26), as well as a model heterotrophic bacterium from culture collection R. pomeroyi DSS-3. Bacterial growth, Cu quotas (Cu:P), macronutrient content and stoichiometry (cellular C, N, P, S, and C:N, S:P), and carbon metabolism (respiration, productivity, carbon demand, and growth efficiency) were monitored across a gradient of Cu conditions from limiting to replete. The study revealed diverse physiological adaptations of the bacterial strains to varying Cu availability and a well-controlled regulation of their intracellular Cu. Availability of Cu was most critical to the flavobacterium Dokdonia sp. Dokd-P16, with growth-limiting effects at Cu levels that are on par with those measured in the surface waters of the NE Pacific. My results thus suggest that the activity of this bacterium may be limited in these waters, and raise a question of the importance of Cu in the ecologically significant microbial clade Flavobacteriia. In addition, I carried out a preliminary comparison of Cu quotas of marine heterotrophic bacteria with cultured marine eukaryotic phytoplankton and used the average Cu quotas of both groups to estimate the partitioning of biogenic Cu between these two planktonic groups in the euphotic zone. Copper quotas of both groups were found to be relatively similar, and the estimates of Cu associated with marine heterotrophic biomass bacteria ranged from 4‒50%, indicating that bacterial Cu uptake may play an important role in cycling Cu in surface waters.  Chapter 4: Bioactive metal composition of marine heterotrophic bacteria and is response to changing Cu availability      This chapter provides a further characterization of the effects of Cu on marine heterotrophic bacteria, by assessing the response of the bacterial metal contents (Fe, Zn, Mn, Co, and Cu) to varying Cu conditions. This research also provides a first look at the extended elemental 15  stoichiometry of marine heterotrophic bacteria (C, N, S, P, Fe, Zn, Mn, Cu, Co), and its comparison with the stoichiometry of marine eukaryotic phytoplankton cultured under comparable trace metal availability. Iron, zinc, copper and cobalt contents were similar in all strains, while those of Mn were lower in Pseudoalteromonas sp. strains relative to Dokd-P16 and R. pomeroyi DSS-3. Heterotrophic bacteria and eukaryotic phytoplankton were found to contain similar amounts of Fe and Cu per biomass, while bacteria appeared to be enriched in Zn but depleted in Mn and Co relative to phytoplankton. Furthermore, in this chapter, I demonstrate that Cu availability influences accumulation of metals in marine heterotrophic bacteria, in a way that is unique to the strains examined. In Flavobacteriia member Dokd-P16, low Cu availability is associated with slow growth rates and reduced Cu quotas, but with increased Fe and Mn quotas relative to Cu-replete conditions. In Roseobacter clade member R. pomeroyi DSS-3 growth rates did not vary significantly at different Cu levels, while Cu and Co quotas were reduced, and Fe and Mn were increased as Cu availability declined. Coastal and oceanic Pseudoalteromonas sp. strains (PAlt-P2 and PAlt-P26) were largely unresponsive to changing Cu availability in terms of their growth and metal quotas, except for Zn quotas in PAlt-P26, which were increased when Cu was in low supply.               16   Table 1.1: A list of bacteria cuproproteins. Cuproprotein Physiological role Reference Cytochrome c oxidase (COX) Heme-Cu respiratory family. Terminal enzyme in the respiratory chain located in the cell membrane. Catalyzes four-electron reduction of O2 to H2O, and couples this process to generation of proton gradient across the membrane. Iwata, 1998 (Iwata 1998) NADH dehydrogenase-2 (NDH2) Facilitates electron transfer from NADH to quinone in the respiratory chain. Jaworowski et al., 1981 (Jaworowski et al. 1981) Superoxide dismutase (SOD1) Catalyzes dismutation of superoxide radicals (O2-) into molecular oxygen (O2) or hydrogen peroxide (H2O2); responsible for reducing oxidative stress. Miller, 2012 (Miller 2012) Plastocyanin/azurin Multicopper oxidase (MCO) family. Electron transfer between photosystem II and I. Cavet et al., 2003 (Cavet et al. 2003) Cu amine oxidase MCO family. De-amination of amino groups; may facilitate growth on different carbon or nitrogen sources. Zeng and Spiro, 2013 (Zeng and Spiro 2013) Laccase MCO family. Oxidation of a wide range of substrates. Examples of physiological roles: • cell-division • pigmentation • sporulation • Mn oxidation • Cu efflux   Deckert et al., 1998 Givaudan et al., 1993 Claus and Filip, 1997 Francis and Tebo, 2001 Kim et al., 2001 Tyrosinase MCO family. Involved in early steps of melanin synthesis Sendovski et al., 2011 (Sendovski et al. 2011) Nitrocyanin Unconfirmed function. Isolated from ammonium oxidizing bacterium N. europea. May have a role in a novel nitric oxide dehydrogenase (carrying out the last step in nitrite synthesis), a nitric oxide reductase, or the oxidation of ammonia. Arciero et al, 2002 (Arciero et al. 2002) Nitrite reductase (NiRK) Reduction of NO2- to NO, a step in denitrification Granger and Ward, 2003 Particulate methane monooxygenase (pMMO) Oxidation of methane to methanol, a first step of metabolic pathway of methane oxidizing bacteria Balasubramanian and Rosenzweig, 2007 Ammonia monooxygenase (AMO) Catalyzes oxidation of NH3 to hydroxylamine (NH2OH) Gilch et al., 2010 (Givaudan et al. 1993; Claus and Filip 1997; Deckert et al. 1998; Francis and Tebo 2001; Kim et al. 2001; Balasubramanian and Rosenzweig 2007; Gilch et al. 2010) 17                           Figure 1.1: Comparison of dissolved Cu profiles in the offshore waters in the central Atlantic (Stn 14, 50°E, 27.58° N, Jacquot and Moffett, 2015), central Indian Ocean (Stn ER 10, 72°E, 20°S, Vu and Sohrin, 2013), and Northeast Pacific Ocean (Stn 17, 144.59° W. 32.41°N, Bruland 1980)     18                                                          Figure 1.2: Map of the Line P transect in the subarctic NE Pacific.         19                               Figure 1.3: Microbial transformation of phytoplankton-derived organic matter. Figure used with permission from Nature Reviews (Buchan et al., 2014)  20  Chapter 2:   Dissolved copper (dCu) biogeochemical cycling in the subarctic Northeast Pacific and a call for improving methodologies  2.1 Summary We investigated biogeochemical cycling of dissolved copper (dCu) along the Line P transect, spanning from the coastal waters of British Columbia, Canada, to the High Nutrient, Low Chlorophyll (HNLC) open Ocean Station Papa (OSP or P26), in the subarctic Northeast Pacific. DCu concentrations ranged from 1.4–3.7 nmol kg−1 throughout the water column along the transect, and were elevated in the upper and bottom waters near the continental margin (< 300 m and > 1100 m respectively) as well as in the upper waters offshore (< 300 m). These trends were attributed to the fluvial and sedimentary sources near the coast of BC, and upwelling of deep, dCu rich waters in the Alaska gyre offshore. In addition, we conducted a temporal investigation of dCu at OSP over three consecutive years (2010−2012), which revealed dynamic variability in the top 300 m that was accompanied by elevated sub-surface concentrations, indicating Cu supply from atmospheric deposition. We explore atmospheric inputs in the Gulf of Alaska and suggest that they may play a significant role in moderating dCu distribution in this region. Consistent with previous investigations in the North Pacific, dCu distributions in the nutricline throughout the transect were strongly linked to those of phosphate and silicate. However, within the Northeast Pacific Oxygen Minimum Zone (OMZ) silicate and dCu distributions were noticeably decoupled suggesting a deficit or loss of dCu in these deep, oxygen depleted waters. In this study, we also assessed the requirement for UV oxidation of our samples (pH 2) prior to dCu analysis by flow injection-chemiluminescence (FIA‐CL). We found that UV oxidation leads to a significant increase in labile Cu (up to 40%), however, the variation in this increase is largely dictated by the length of sample storage under acidic conditions. Our results suggest that UV oxidation is essential prior to FIA-CL 21  analysis of young samples (aged for 48 h–2 months), but may not be required if samples have been stored for an extended period at low pH (≥4 years). We found a good agreement between dCu values at station P26 obtained with FIA-CL and those obtained with cathodic stripping voltammetry (CSV). However, comparison of our dataset with all published dCu profiles in the North Pacific revealed some discrepancies in dCu values in this region. Here, we briefly discuss the role of sample storage period, UV oxidation and differences in analytical methodologies as factors causing uncertainties in dCu values           2.2 Introduction   Copper (Cu) is an essential micronutrient for marine organisms and, like other bioactive trace elements, its vertical distribution is governed by its interactions with biota. In laboratory settings, low availability of Cu has been shown to impair growth and metabolism of eukaryotic phytoplankton (Peers et al. 2005; Annett et al. 2008; Guo et al. 2012b), ammonium oxidizing archaeon Nitrosopumilus maritimus (Amin et al. 2013), denitrifying bacteria (Granger and Ward 2003) as well as marine heterotrophic bacterium Dokdonia sp. Dokd-P16 (Posacka et al, unpublished). Furthermore, some iron-stressed phytoplankton in culture have an increased metabolic demand for Cu, through the upregulation of a high-affinity iron (Fe) uptake transport system (HAFeTS) that requires Cu (Maldonado et al. 2006; Annett et al. 2008; Guo et al. 2012b) or/and substitution of Fe-containing enzymes with Cu-containing homologs (Peers and Price 2006). Thus, Cu nutrition is likely to be important in the High Nutrient Low Chlorophyll (HNLC) waters, where low Fe concentrations limit the growth of primary producers. Indeed, recent field work demonstrates such metabolic coupling between Cu and Fe in the HNLC waters of the subarctic Northeast Pacific (Semeniuk et al. 2016b). However, while low Cu may hinder metabolic functions, exposure to Cu levels that exceed cellular requirements causes toxicity (Brand et al. 1986; Gordon et al. 1994; Mann et al. 2002; Paytan et al. 2009; Debelius et al. 2011). Because of a narrow range where Cu supports optimal growth and the distinct Cu requirements of different taxa, Cu bioavailability may influence the taxonomic composition of the marine microbial community (Brand et al. 1986; Mann et al. 2002; Paytan et al. 2009).  22     In the global ocean, vertical distribution of dCu is typically characterized by low surface water concentrations that are accompanied by a vertical gradient of increasing dCu towards the ocean seafloor. Such a behaviour has been classified as hybrid nutrient-like (Bruland and Lohan 2003), and may be attributed to the collective control of internal cycles (biological utilization and regeneration processes), particle scavenging in intermediate waters and fluxes from bottom sediments (Boyle et al. 1977; Bruland 1980; Bruland and Lohan 2003; Takano et al. 2014). Dissolved Cu concentrations in the interior waters of the North Pacific are enriched relative to other oceans, with concentrations being nearly twofold of those measured in the North Atlantic, due to global ocean circulation patterns. Profiles of dCu in different parts of the North Pacific are consistent with its hybrid nutrient-like behaviour (Boyle et al. 1977; Bruland 1980; Martin et al. 1989; Fujishima et al. 2001; Ezoe et al. 2004; Takano et al. 2014; Tanita et al. 2015). Here, strong correlations between dCu and major nutrients (NO3-, PO43-) can be observed in the oceanic nutricline (Bruland 1980; Martin et al. 1989), indicating a tight influence of biological processes on Cu distribution in the North Pacific. However, our understanding of sources and sinks of Cu is still fairly limited in many parts of the basin, including the Northeast subarctic region.    In particular, the role of systems such as the Oxygen Minimum Zone (OMZ) of the subarctic NE Pacific in moderating dCu cycles is not well understood, although this zone has been proposed to affect redox-sensitive elements such as Cd, Zn as well as Cu (Janssen et al. 2014; Janssen and Cullen 2015). NE Pacific OMZ is an intermediate water feature (~ 300‒2000 m) characterized by persistently low, although non-zero, dissolved O2 levels. Such conditions may favour scavenging of Cu via formation of insoluble sulfides precipitates, proposed to be facilitated by sulfidic microenvironments on sinking particles in low O2 zones (Janssen et al., 2014; Janssen and Cullen, 2015). Thus, OMZ may have an important role in reducing the dCu inventory in the NE Pacific.            In addition, control of atmospheric deposition on dCu in the NE subarctic Pacific also remains to be explored. This process may represent an important source of Cu across the subarctic North Pacific, which experiences a seasonal transport of Asian dust from desert and loess region in central Asia (Duce and Tindale 1991) In addition to being the second largest source of dust in the world (Shao et al. 2011), Asia is also a significant emitter of anthropogenic pollutants which may enhance the atmospheric content of elements, such as Cu (Lee et al. 2013). Influence of atmospheric sources on surface Cu distribution is evident in the profiles from the western Pacific 23  (eg. Ezoe et al. 2004; Tanita et al. 2015). In contrast, little is known regarding the influence of atmospheric inputs on the dCu in the remote waters of the NE Pacific, where Asian sources may reach episodically (Boyd et al. 1998).     Our ability to obtain reliable measurements of the dCu is instrumental to the interpretation of Cu cycling in aquatic environments and its impact on microbial communities. Recently, some uncertainties have emerged in dCu analysis due to the observation that UV oxidation of acidified oceanic samples leads to increases in labile Cu (e.g. Biller and Bruland, 2012; Middag et al., 2015; Milne et al., 2010; GEOTRACES Standards & Reference material statement, 2013, http://www.geotraces.org/images/stories/documents/intercalibration/Files/Reference_Samples_May13/SAFe_Ref_Cu_05_13.pdf). Typically, samples for total dCu are acidified (pH 1.7‒2) to dissociate any organically bound Cu (ligands or colloids) prior to analysis. However, recent evidence suggests that some organic Cu complexes may persist under acidic conditions even after prolonged storage, thus requiring UV oxidation (Milne et al. 2010; Biller and Bruland 2012; Middag et al. 2015). Lack of UV oxidation was used to explain the underestimation of dCu values in the GEOTRACES reference material SAFe (D2) (e.g. Boye et al., 2012; Milne et al., 2010). However, effects of UV on labile Cu have been variable in recent investigations, and it remains unclear how best to obtain accurate dCu measurements.    In the present study, we focus on characterizing Cu biogeochemistry along Line P, a coastal-open ocean zonal transect in the subarctic Northeast Pacific spanning from the continental shelf of Vancouver Island (BC) to Ocean Station Papa (OSP, ~ 1600 km offshore) in the Alaskan gyre, where primary productivity is Fe-limited (Fig. 2.1). Previous work there examined the role of Cu nutrition and its availability to resident plankton communities, the interactive control of Cu and Fe availability on Fe-limited phytoplankton, as well as the distribution of dCu within the mixed layer (Semeniuk et al. 2009, 2015, 2016a; b). However, there exists a dearth of dCu measurements at depth across the transect nor are there systematic investigations of its temporal variability. Here, we report high-resolution dCu profiles down to 2000 m at five major stations of Line P, and examine Cu behaviour in the context of oceanographic features encountered along this transect including: a) the Northeast Pacific Oxygen Minimum Zone (OMZ, 300‒2000 m); b) the coastal California Undercurrent (CUC) intruding waters at stations P4 and P12; c) the North Pacific Intermediate Water (NPIW) at station P26 (McAlister 2015), and d) the offshore upwelling  24  Alaskan gyre. In addition, temporal variability of dCu was examined over 3 sampling cruises (in 2010, 2011 and 2012) at the offshore station P26 (OSP). In this study, we also investigated the requirement for UV oxidation of samples (GEOTRACES Standards & Reference material statement, 2013, http://www.geotraces.org/images/stories/documents/intercalibration/Files/Reference_Samples_May13/SAFe_Ref_Cu_05_13.pdf) prior to our analysis of dCu by flow injection-chemiluminescence (FIA-CL). Furthermore, we compare our results to published dCu profiles in the North Pacific and discuss variability inherent in the dataset. We propose that methodological factors explored in this study (acidified sample storage, UV oxidation and analytical methods) likely contribute to this variability, and warrant more detailed investigation into how to improve current dCu methodologies.  2.3 Methodologies 2.3.1  Sampling and study transect        Samples were collected at five stations along the Line P (P4-P26) during the August 17- 26th cruise on board the C.C.G.S. John P. Tully (Cruise, 2011-27) (Fig. 2.1). For a temporal study of dCu at OSP, we used samples from three cruises: Aug 17th-Sept 3rd, 2010 (cruise 2010-14), August 17- 26th 2011 (Cruise, 2011-27) and Aug 13-30th, 2012 (cruise 2012-13). We obtained 18-22 samples at each station within the depth range of 10-2000 m of the water column using a 12 bottle powder coated trace metal clean rosette system equipped with 12 L Teflon- coated GO-FLO bottles (General Oceanics, FL, USA)  (Measures et al. 2008) Filtered seawater (0.22 µm AcroPak, Pall Corporation) was collected into trace metal clean 500 mL BellArt bottles (according to GEOTRACES protocols, http://www.geotraces.org/libraries/documents/Intercalibration/Cookbook.pdf). Samples for total dissolved analysis were acidified at sea to pH 2 using 1 mL of ultrapure 12N HCl per litre of sample (Seastar Chemicals, Sidney BC). Another set of dCu samples was collected during the 2011 and 2012 Line P cruises as described above, but were immediately frozen after filtration at ambient pH and stored at -20ᵒC until analysis. Macronutrients Si(OH)4, NO3-, PO43- were analyzed 25  according to Barwell-Clarke and Whitney (1996). Macronutrient and CTD data are courtesy of Institute of Ocean Sciences, Department of Ocean & Fisheries and are publicly available (https://www.waterproperties.ca/linep/index.php).  2.3.2 Dissolved copper analysis          Samples from the Aug 2011 cruise were initially analyzed for dCu by flow injection-chemiluminescence (FIA-CL), without prior UV oxidation (~ 3 months of sample storage at pH 2). We found a good agreement between these data and a historical profile of dCu at OSP (P26, Fig. 2.2, Martin et al. 1989). Furthermore, our values of GEOTRACES inter-calibration standard SAFe D2 and a reference material NASS-6 were in good agreement with the consensus values and certified values, respectively (Table 1). Following a recent recommendation of UV oxidation of samples prior to dCu analysis by the GEOTRACES community, all the transect samples were re-analyzed between July‒August 2015 (~ 4 years of storage at pH 2) after 2 hours of UV oxidation (twice in 60 min bursts). UV apparatus used in this study consisted of 30 mL silica sample tubes with PTFE capping units surrounding a high -pressure mercury lamp (125 W) (van den Berg 2017). Once UV oxidized, samples were left for at least 2 hours or overnight before analysis to remove the effect of hypochlorite or hypobromite, which can be produced by UV oxidation (Achterberg & van den Berg, 1994).     There was a large difference between our new dCu dataset and the initial analysis at all stations across the transect (6.5‒43% more dCu), with no apparent trend (e.g. onshore-offshore, Appendix A, Fig. A.1). Originally, we attributed this increase to the effect of UV oxidation. However, we also found higher dCu values in a small subset of non-UV oxidized samples in this most recent re-analysis. Nevertheless, UV oxidation of those aged samples was associated with a minor increase in dCu (7 out of 9 samples showed an increase in dCu after UV ranging 3.4‒18 %, Appendix A, Table A.1). This led us to hypothesize that both long sample storage at low pH and UV oxidation can lead to increases in FIA-CL labile dCu. Subsequently, we sought to investigate how the length of sample storage and UV oxidation affect Cu lability in FIA-CL analysis. For these experiments, we used August 2011 samples, from different depths at station P26, which were not acidified, but instead were stored at – 20°C until 2015. After acidification to pH 2, 26  corresponding measurements of dCu in UV and non-UV oxidized sample aliquots were performed at different time points (48 h, 2 weeks and 2 months) and compared to the 2015 data. Next, we compared FIA-CL measurements of dCu at P26 (2015 re-analysis, 4 years of storage and 2 h of UV oxidation) with values obtained by CSV (no storage, 45 min of UV oxidation) and previous dCu measurements by Semeniuk et al. (2016a). Lastly, we compared our results with other published dCu datasets in the North Pacific.  2.3.2.1 Flow-injection analysis chemiluminescence (FIA-CL)          For measurements of the total dissolved Cu, we used a flow injection analysis, chemiluminescence (FIA-CL) method of Zamzow et al (1998). This method is based on the oxidation of a complex formed between Cu and 1,10-phenanthroline in the presence of hydrogen peroxide. A working standard of Cu was prepared in MilliQ from 1000 ppm Cu ICP-MS standard and was used to quantify dCu in the samples by the standard additions method. A five-standard calibration curve was acquired every 8‒10 samples. Each day several samples previously analyzed (≥ 5) were run to determine the precision of the method. Precision based on percent relative standard deviation (% RSD) of n = 36 replicate sample was 2.93 % RSD, median = 1.36 % RSD, and range = 0.14 – 10.2 % RSD. The accuracy of the method was assessed by analyses of SAFe D1 & D2 (GEOTRACES reference samples, http://www.geotraces.org/science/intercalibration/322-standards-and-reference-materials) and NASS-6 (National Research Council Canada). Additionally, GEOTRACES inter-calibration sample GSC 318 (for which there is no current consensus) was also analyzed (Table 1). We run the GEOTRACES inter-calibration standards with and without UV oxidation. All reference materials were in good agreement with the consensus values and we found no statistical difference between UV oxidized and non-UV oxidized SAFe samples (t-test, p=0.582). System blanks for FIA-CL were 0.12 ± 0.022 nmol kg-1 (n=14), yielding a detection limit (3 × blank standard deviation) of 0.064 nmol kg-1. Data was plotted using packages “m_map” (Pawlowicz 2015) in commercial software MATLAB (The Mathworks, Inc, Natick, Massachusetts, United States), Sigmaplot (Systat Software, San Jose, CA),  and “ggplot2” (Wickham 2009)  in open source programming language R (R Core Team 2016). The dCu dataset along with the ancillary data, as 27  well as scripts used to produce the plots in this study can be downloaded from a public repository: https://github.com/AnnaMagdalena/DCu_LineP-Subarctic-Pacific. 2.3.2.2 Cathodic Stripping Voltammetry (CSV)       Cathodic Stripping Voltammetry (CSV) was performed on non-acidified samples collected during the August 2012 cruise. Samples were defrosted overnight at 4C, swirled gently to re-dissolve any particulate matter formed during freezing, and poured into conditioned sterilin containers to warm up to room temperature (20C) in the dark. Total dissolved copper was measured by CSV at natural pH. In addition, surface (10 m) and deep (1200 m and 1400 m) samples were analyzed by anodic stripping voltammetry (ASV) at pH 1.9, allowing inter-comparison with reference seawater of the same pH (NASS-6). UV digestion at natural pH avoids the formation of hypochlorite which interferes with CSV Cu determinations (Campos and van den Berg 1994). For CSV measurements, the sample was poured into conditioned silica 40 mL UV tubes or into a conditioned silica cell. UV oxidation was performed for 45 minutes using the same UV apparatus as for the FIA-CL method, with the lamp either surrounded by four 40 mL sample tubes with PTFE caps or positioned horizontally above a sample aliquot in the voltammetric cell.  Dissolved copper was then measured in the cell by CSV in the presence of 20 μM salicylaldoxime (SA) and 0.01 M borate/ammonia pH buffer (pH 8.15 on NBS scale) (Campos and van den Berg 1994) The borate/ammonia pH buffer (1 M boric acid/0.3 M ammonia) was UV oxidized to remove organic matter and contaminating metals were removed by equilibration with 100 μM manganese dioxide (MnO2) followed by filtration (van den Berg 1982). Sample was purged with N2 for 5 minutes to remove dissolved oxygen prior to analysis. The voltammetric equipment was a μ-Autolab III potentiostat (Ecochemie, Netherlands) connected to a 663 VA stand (Metrohm) with hanging mercury drop electrode (HMDE). The system used a glassy carbon counter electrode, an Ag/AgCl reference electrode with a 3 M KCl salt bridge, with a rotating polytetrafluoroethylene (PTFE) rod for stirring solutions. The software was modified to discard 2, instead of the usual 4, drops of Hg between scans, to minimize Hg usage. CSV measurements were in the differential pulse mode, at a deposition potential of -0.15 V, a deposition time of 30 s, and a 1s potential jump to -1.2 V to desorb any residual organic matter. For ASV measurements, acid-cleaned quarts UV tubes were not conditioned, and prior to UV digestion, the sample was acidified to pH 1.9 with 28  trace metal grade HCl. Voltammetric scans for ASV used the square-wave mode with a 5-minute deposition time, with no reagents added. Comparative measurements between CSV and ASV were found to give the same result within the standard deviation of three repeat measurements, and ASV measurements on NASS-6 reference material gave results within 5% of the certified value (3.94 ± 0.32 nmol L-1 versus 3.90 nmol L-1 for NASS-6).  2.4 Results 2.4.1 Methodological considerations of dCu analysis 2.4.1.1 Effect of storage time and UV oxidation    Time dependent analysis of dCu in non-UV oxidized acidified samples demonstrated that labile Cu increases progressively with the length of storage period (48 h‒2 months, Fig 2.3, dataset in Appendix A, Table A.2). Subjecting those same samples to UV oxidation generated an additional increase in labile Cu at each time point, which was greatest for the ‘youngest’ samples (~ 40 % offset between UV and non-UV oxidized samples at 48 h).  However, the amount of Cu released by UV oxidation decreased when measurements were performed after a longer storage, either 2 weeks or 2 months (~ 6‒30%).  After 2 months of acidified sample storage, the profile of dCu after UV oxidation is on par with the more recent analyses done in 2015 (~ 4 years of storage with 2 h of UV oxidation), although upper water column samples are still somewhat lower (50 m sample ~24 % lower; 75 m sample ~11% lower than samples for these depths).   2.4.1.2 Comparison of datasets obtained by FIA-CL, CSV, and the historical data at Ocean Station Papa (OSP)       Measurements of dCu at P26 performed with distinct sample pre-treatment, hardware, and modes of detection (FIA-CL and CSV) were in good agreement, particularly at depths below 250 m (Fig. 2.4A, Appendix A, Table A.3). However, there is some disparity between the two profiles 29  in surface waters (0‒250 m), likely due to the seasonal variation in dCu, as samples analyzed by CSV were collected in Aug 2012, while those analyzed by FIA-CL were collected in Aug 2011. Our dataset is also in good agreement with the surface dCu measured along the transect (P4‒P26) by Semeniuk et al (2016a) for the same cruise and analytical method. The various measurements are linearly correlated (p =<0.001 Fig. 2.4B). In contrast, dCu profiles at P26 from our most recent analysis (2015) are noticeably offset from the historical data from Martin et al (1989) (Fig. 2.2). The differences between the two profiles are up to 40 % in the surface and approximately 14 % at depths below 250 m, likely reflecting differences in methodological approaches.  2.4.1.3 Comparison of published dCu profiles in the North Pacific      To compare published vertical profiles of dCu in the North Pacific we grouped data into oceanographically similar regions (Region 1‒7, Fig. 2.5). Within these regions, we see substantial disagreement between studies with dCu differences of ~10-100 %, even for profiles collected at the same sampling location. For instance, at Ocean Station Papa (Region 3) our dCu values are 15‒40 % and ~ 60 % higher than those of Martin et al (1989) and Coale and Bruland (1990), respectively. Interestingly, the results for these two published studies disagree even though samples were obtained during the same cruise (Vertex VII) and analyzed with a similar method, although with differences in filtration (0.45 and 0.3 µm for Martin et al (1989) and Coale and Bruland (1990), respectively). However, we also note some agreement between studies when comparing stations farther away (note that not all vertical dCu profiles are plotted). The profiles from Tanita et al. (2015) (station 22, profile not plotted; station 5) and Moffett and Dupont (2007) (stations 6 and 4) agree with each other and are more similar to the dCu values reported in this study. On the other hand, the results of Fujishima et al (2001) (station 19, profile not plotted; station 17), Takano et al., (2014)  and Ezoe et al (2004)  (stations BO01‒07 stations, profiles not plotted) agree with one another, but are much lower than our dCu values. Finally, at the GEOTRACES inter-calibration station SAFe (region 6) there is an excellent agreement between the analyses of Bruland (1980), Coale and Bruland (1990), and Biller & Bruland (2012).         While in the near surface waters these differences likely reflect temporal variability in dCu, this is unlikely to be the cause for the variations in the deep-water values of nearby stations. Thus, 30  methodological differences such as the analytical approaches used, sample storage time at low pH, filtration and whether a UV step was utilized or not, may help explain these differences (methodological details of each study are listed in Appendix A, Table A.4). For instance, the dCu profile at OSP from Martin et al (1989) was obtained using non-UV oxidized samples and is similar to our profile also obtained without UV oxidation (2012 analysis, ~ 3 months of sample storage).  In contrast, the difference between UV and non-UV treated samples is not observed when the datasets of Bruland (1980) at station 17 and Biller and Bruland (2012) at SAFe are compared (APDC-DDDC chloroform organic extraction without UV and NOBIAS P1A resin extraction with UV, respectively). Thus, UV oxidation alone is not the sole cause of these disagreements.    2.4.2 Copper biogeochemistry in the subarctic Northeast Pacific 2.4.2.1 Line P hydrography      Surface waters along Line P are characterized by low salinity and a strong density stratification (Fig. 2.6), with a mixed layer of ~ 20‒30 m (Fig. 2.7). In the eastern reaches of the transect (P4 and P12), the permanent pycnocline is centered on the σt= 26.5 kg m-3, and is shifted to the σt= 26.8 kg m-3 in the water column of station P26. Coastal stations (P4‒P12) are influenced by the California Under Current (CUC) along σt = 26.5 – 26.8 kg m-3 (Pierce et al. 2000) characterized as warm and salty (McAlister 2015). At the western terminus of the transect, waters along the σt = 26.8 kg m-3 (P26) represent the fresher and cooler waters of the North Pacific Intermediate Water (NPIW) (Talley 1993; McAlister 2015). In this system, an oxygen minimum zone (OMZ) extends from ~ 300‒2000 m with O2 concentrations between ~7 – 60 µmol kg-1, with the lowest O2 occurring at depths ~ 1000 m.  2.4.2.2 General trends in dCu along Line P transect      Vertical profiles of dissolved Cu along the transect were typical of a nutrient-like element with surface depletions relative to deep-water and increasing concentrations with depth (Fig. 2.7A 31  & B), consistent with dCu behaviour in other ocean provinces  (Martin et al. 1989; Boye et al. 2012; Vu and Sohrin 2013; Heller and Croot 2015; Jacquot and Moffett 2015). Surface dCu concentrations were highest at station P4 (2.16 nmol kg-1), gradually declined towards P16 (1.49 nmol kg-1), and increased west of P16 towards stations P20 and P26 (1.78 and 1.83 nmol kg-1, respectively). This surface dCu trend is consistent with a previous investigation along Line P during the same cruise (Semeniuk et al. 2016a). In contrast, the onshore-offshore trend in deep water dCu (> 300 m) was characterized by increasing concentrations from P4 to P12 followed by a decline at P16, and minor dCu increases at P20 and P26 (Fig. 2.7B). In the upper 300 m, patterns of dCu distribution matched well with those of PO43- at the offshore stations P16, P20 and P26. In contrast, at the two stations closer to the coast (P4 and P12), dCu levels were in excess of PO43- at the shallowest depths (0‒40 m). Vertical distributions of dCu and PO43- followed the density trend at P26 with a summer (0‒29 m) and winter mixed layers (50‒130 m) derived from sigma-t profiles (Fig. 2.7A). As such, the presence of two ‘cuproclines’ at this station can be observed; one between 20‒75 m and another between 135‒200 m.    Deep dCu profile values at P12 (400‒2000 m) were higher (2.99‒3.72 nmol kg-1) than those at P16, P20 & P26 (2.74‒3.48 nmol kg-1) (Fig. 2.7B). Across the transect, dCu levels within the OMZ (~ 300–2000 m with O2 concentrations ranging between ~7 – 60 µmol kg-1) were consistent. However, sharp increases in dCu can be seen below 1000 m at P4, likely reflecting sedimentary sources.   3.2.3 Annual variability in dCu values in upper waters at Ocean Station Papa (OSP, P26).       Dissolved Cu distribution at the offshore station P26 during three consecutive August Line P cruises: 2010‒2012 were characterized by dynamic shifts in dCu in the upper 300 m (Fig. 2.8), while below this depth dCu patterns and concentrations remain relatively homogenous (Appendix A, Table A.3). Between 40 and 200 m, dCu levels were typically lower in 2010 (≤ 2 nmol kg-1), than in 2011 and 2012 (> 2 nmol kg-1). Surface enrichments in dCu at P26 were observed in 2010 and 2012 data, with values of 2.3‒2.4 nmol kg-1 (5‒10 m) relative to 1.83 nmol kg-1 in 2011 (10 32  m) (Fig. 2.8, Appendix A, Table A.3). This could be an indication of an atmospheric deposition event in 2010 and 2012. The shape of the dCu profile in 2011 can be partially explained by the density structure and agrees somewhat with the distribution of phosphate, while this is not observed in 2010 and 2012. However, lower sampling resolution for the latter years limit our ability to relate water column structure and dCu distribution.  2.4.2.3 Dissolved Cu and macronutrient relationships across the transect     Metal-macronutrient relationships are useful tools for examining the importance of both biological and abiotic processes in influencing the distribution of hybrid type metals such as Cu. As such, strong metal-macronutrient correlations are suggestive of trace metal distribution being largely driven by biological processes (uptake and remineralization), while deviations from those correlations imply abiotic effects on metal distribution (e.g. particle scavenging). Here, we observed a strong correlation of dCu with PO43- and Si(OH)4 in the upper 400 m of the transect (P4‒P26, [dCu] = 0.53[PO43-] + 1.27, r2 =0.86; [dCu] = 0.015[Si(OH)4] + 1.68, r2=0.73., Fig. 2.9A and B, respectively).  However, dCu distribution in deeper waters was uncoupled from that of both macronutrients. In the Cu-PO43- relationship, dCu values below 400 m (~2.7 µmol kg-1 PO43-) begin to increase abruptly above the nutricline trend, suggesting an excess of dCu concentrations relative to phosphate (Fig. 2.9A). On the other hand, decoupling between dCu and Si begins at a shallower depth (300 m) with Cu values falling well below the nutricline trend (Fig. 2.9B). A noticeable plateau can be seen in copper-silicate relationship between ~ 75 µmol kg-1 to ~ 150 µmol kg-1 Si(OH)4 (~ 300‒1600 m), suggesting scavenging of dCu in the deeper waters.  2.5 Discussion    This discussion is divided into 2 distinct sections. First, we discuss methodological aspects accounting for variability among dCu values from different laboratories and analytical methods 33  (section 2.5.1), followed by a section focusing on the biogeochemistry of dCu in the NE Pacific Ocean (section 2.5.2).  2.5.1 Methodological uncertainties of dCu analysis     The variability in the North Pacific dCu datasets we highlight here, indicate that there is some uncertainty regarding the concentration and distribution of Cu in this basin, and perhaps in other oceans. There have been other reports on the variability in dCu analysis in recent literature, and this has so far been attributed primarily to UV oxidation (e.g. Milne et al., 2010; Biller and Bruland 2012; Boye et al., 2012; GEOTRACES Standards & Reference material statement, 2013, http://www.geotraces.org/images/stories/documents/intercalibration/Files/Reference_Samples_May13/SAFe_Ref_Cu_05_13.pdf). However, as we explored here, dCu results can also be largely influenced by sample storage time and the choice of analytical method. Below, we discuss the interplay between these three factors and make a call for an improvement of current dCu analyses.  2.5.1.1  UV oxidation    Recent studies, examining the importance of a UV oxidation step to obtain quantitative detection and accurate dCu, report variable recovery of Cu from acidified seawater samples. For instance, a study by Milne et al., (2010) found ~ 10% increase in labile Cu in the SAFe D2, Biller and Bruland (2012) report ~ 30 % increase in an offshore California Current sample, while we found between 6.4 to 43 % increase in P26 samples following UV oxidation, depending on acidic storage period. In a recent inter-comparison study at Bermuda Atlantic Time Series station, dCu measurements throughout the water column obtained with and without UV oxidation were compared by two separate labs (Middag et al. 2015). Increases in labile Cu due to UV oxidation were on average 12 % and 16 % for US and Netherlands GEOTRACES labs, respectively (Middag et al. 2015). This variability could be partially explained by differences in sample origins (and hence the amount of organic compounds), and UV oxidation conditions.     In UV oxidation, the interaction between UV light with dissolved oxygen (DO) and water leads to production of ozone, superoxide radicals (• O2- ) and hydroxyl radicals (• OH) (Achterberg and van den Berg 1994). The superoxide will ultimately disproportionate to hydrogen peroxide 34  (H2O2) unless it reacts with redox-active species. The energetic UV radiation combined with the superoxide acts to break down dissolved organic matter, ultimately to CO2 and water. A side-reaction of the superoxide with chloride (abundant in seawater) leads to formation of hypochlorite (ClO-) (Haag and Hoigne 1983). Hypochlorite formation and the radiative heating of the water lead to a rapid decrease in the residual concentration of DO, which is the end of the oxidative destruction of organic matter, typically after 30 to 45 minutes of irradiation (Achterberg and van den Berg 1994). Once the DO in the sample is exhausted H2O2 may be added to continue the process. The efficacy of UV oxidation is influenced by several factors such as the intensity of the UV lamp, the distance between the UV lamp and the sample, as well as the sample volume irradiated. For instance, using on-line UV oxidation methods where surface-to-volume ratio is high requires much shorter UV exposure time to destroy DOM than batch techniques (minutes vs hours, respectively; e.g. Achterberg and van den Berg (1994) and Achterberg et al.,(2001)). Thus, there is a need to establish consistent UV oxidation protocols for future studies.      In our experiments, dCu values obtained after UV oxidation of samples stored in acidic conditions for 48 h and 2 weeks were substantially lower relative to the values obtained after ~ 4 years of acidic storage with UV oxidation. A possible explanation for this could be that early in the sample storage period when Cu complexation by organics is expected to be highest, our UV oxidation process was not effective. This may explain why we did not observe any significant increases in labile Cu following UV oxidation using longer exposure time (4 h on samples from 50 m and 75 m stored for 2 months, Appendix A, Table A.2). Further investigation is required to determine whether adding H2O2 or O2 to these ‘young’ samples would cause greater increase in labile Cu after UV oxidation. However, given the low DOC content of waters along Line P (80‒100 µmol kg-1, Wong et al., (2000)) addition of H2O2 should not be required (Achterberg and van den Berg 1994). We considered that the differences between dCu values of short-term and long-term stored acidic samples (48 h‒2 months vs 4 years) could be caused by Cu leaching from sampling bottles during storage. While we were not able to directly test for Cu leakage from sampling bottles, we eliminated this possibility on the basis that we found an agreement between the dataset obtained in 2015 for station P26 with two different methods: FIA-CL (stored acidified for 4 years) and CSV (non-acidified and frozen).  35  2.5.1.2  Sample storage period     As we demonstrated here (using FIA-CL analytical method for dCu), changes in labile Cu in acidified samples are highly dependent on their storage time. Long storage at low pH promotes the release of Cu from organic complexes in samples, which introduces a potential for variable results depending on the time from collection to analysis. This has implications for whether a UV oxidation of samples is necessary prior to analysis. SAFe reference materials were collected and acidified in 2004, thus were much older than our Line P samples (stored in acidic conditions for 48 h‒4 years). We found that the consensus value for dCu in the SAFe D2 agreed with the value from our initial analysis in 2011 without UV oxidation (at this point SAFe D2 sample age was ~ 7 years), and with our values in the 2015 analyses, with and without UV oxidation. Similarly, Tanita et al., (2015; using CSV analytical method for dCu) and Butler et al. (2013; using solid-phase extraction with Toyopearl AF chelate 650 M, followed by ICPMS analytical method for dCu) without UV oxidation, obtained values on par with the consensus value (Supplementary Material, Table S4). In contrast, Milne et al. (2010; using pre-concentration step with Toyopearl AF chelate 650 M resin, followed by isotope dilution/ICPMS) did find a small increase (~ 10 %) in labile Cu in UV irradiated SAFe D2 sample (collected from 1000 m), despite its long storage. Taken together, these studies suggest that long storage period potentially leads to degradation of organic complexes binding Cu in the SAFe sample, and values for dCu in agreement with each other, though some differences may emerge due to specific analytical methodologies (discussed in 4.1.3). The hypothesis that aging of samples can destroy inert Cu is further supported by our observations of minor (if none) increases in labile dCu in samples from station P26, stored for ~ 4 years and UV oxidized relative to non-UV oxidized (Supplementary Material, Table S1). Our study suggests that UV oxidation is critical for the determination of dCu by FIA-CL in acidified samples stored for ≤ 3 months, but may not be required for samples stored for more than ~ 4 years.       Middag et al. (2015) compared dCu values at the Bermuda Atlantic Time Series station, with and without UV oxidation, obtained by two research groups (US GEOTRACES and Netherlands GEOTRACES), and analyzed with the same method (solid-phase extraction (Nobias-chelate PA1) with ICP-MS, Biller and Bruland, 2012). As in our study, UV oxidation enhanced labile Cu in acidified samples. However, the percent enhancement was not significantly different between the 36  two datasets (US group: mean = 11.7 ± 10.1 %; Netherlands group, mean = 16.2 ± 10.5 %, two sample t-test, p-value=0.15), even though their samples were stored for different time (1 and 2 years, respectively). This suggests that 2 years of acidic storage may not be sufficient to destroy all the inert dissolved Cu.      Interestingly, for several samples in their study (Middag et al., 2015), dCu values with and without UV oxidation were within analytical uncertainty. Similarly, we found no significant change in dCu values for a small number of samples after UV oxidation, coupled with ~ 4 years’ acidic storage (Supplementary Material, Fig. S1). Finally, Milne et al. 2010 found that UV treatment had no effect on the dCu concentration in SAFe D1 sample (also collected from 1000 m as D2 sample). These inconsistent results with the idea that UV oxidation is needed for accurate dCu determinations, especially in young samples, may be exceptions and may reflect differences in analytical methods or simply nature of the samples.  2.5.1.3  Role of methodologies in current dCu uncertainties         Methodologies employed in dCu analysis include: solid phase extraction with quantification via either standard additions or external calibration (Lohan et al. 2005; Sohrin et al. 2008; e.g. Biller and Bruland 2012; O’Sullivan et al. 2013) or isotope dilution (Milne et al. 2010; e.g. Lee et al. 2011; Lagerström et al. 2013), electrochemical methods (Campos and van den Berg 1994; e.g. Buck and Bruland 2005), and flow injection analysis methods, chemiluminescence methods (Zamzow et al. 1998; e.g. Achterberg et al. 2001). These methodologies and their variations differ in terms of their recovery of Cu (and its species) in the seawater sample. For instance, analysis by cathodic stripping voltammetry (CSV) using a ligand with lower Cu-complex stability, such as tropolene, can cause greater underestimation of the total dCu, than the use of stronger ligands such as oxine in non-UV treated seawater (Achterberg and van den Berg, 1994). For solid phase extraction methods, underestimation of dCu may result from the need to buffer the sample to a higher pH, possibly allowing re-complexation of Cu with organic ligands (Ndung’u et al. 2003); although, it has been suggested this may be prevented with the use of in-line buffering (O’Sullivan et al. 2013). Thus, in addition to storage time of acidified samples, analytical methodologies may introduce another source of variability in dCu values. 37     Indeed, our comparison of two distinct methodologies in terms of sample preparation and mode of detection for dCu analysis supports this. Here, the values for dCu obtained with FIA-CL and CSV methods agreed only when samples for FIA-CL analysis were stored for ≥ 2 months and UV oxidized. This suggest these two methods have distinct efficiencies of Cu recovery from the sample. In the FIA-CL system, mostly labile and some weakly associated Cu species are measured, due to the short reaction time (fraction of a second) with the synthetic ligand 1,10 phenanthroline (48 µM)  before detection (Zamzow et al. 1998). On the other hand, in CSV system there is a longer reaction time (4‒5 min) between Cu and the competing ligand (SA, 20 µM), which may result in greater Cu acquisition from the organic Cu pool. The higher efficiency of CSV is further supported by the similarity between the values for dCu in a non-UV treated sample (from 100 m at P26) analyzed by CSV (1.6 ± 0.1 nmol kg-1) and that for same sample, aged for ~ 3 months (non-UV oxidized), and analyzed by FIA-CL (1.5 ± 0.05 nmol kg-1). Overall, our results indicate that CSV is far more efficient at accessing the strongly complexed organic Cu than the FIA-CL and that for accurate measurements of dCu, in addition to UV oxidation aging of samples is required for the latter method.      In recent studies, underestimation dCu values were mainly explained in terms of UV oxidation (Milne et al. 2010; Boye et al. 2012; Biller and Bruland 2012). However, we surveyed recent dCu studies that employ the GEOTRACES inter-calibration standards and observed variability that cannot be fully explained by the presence or absence of a UV oxidation step (Supplementary Material, Table S5). For instance, some methods obtain reference material values within the current GEOTRACES consensus value without UV oxidation (Lee et al. 2011; Butler et al. 2013; Jacquot et al. 2013; Lagerström et al. 2013; Vu and Sohrin 2013; Jacquot and Moffett 2015), while others underestimate the reference value if UV oxidation is omitted (Sohrin et al. 2008; Milne et al. 2010; Boye et al. 2012).  Methods incorporating isotope dilution (ID) may underestimate dCu in non-UV oxidized samples (e.g.  Milne et al., 2010, Boye et al., 2012), despite the recommendation that this analytical method does not require UV oxidation (GEOTRACES Reference & Standards Statement, 2013, http://www.geotraces.org/science/intercalibration/322-standards-and-reference-materials). In isotope dilution methods, unlike those based solely on solid phase partitioning, extraction efficiency of metals is not crucial for metal quantification (Milne et al., 2010). However, in principle metals must be in an isotopically exchangeable form before 38  equilibration with the isotope spike for ID methods. Otherwise, the isotopic ratio measured in the sample will not reflect the ratio of the spiked sample, and consequently, the original sample concentration will be underestimated. Thus, attention should be paid to dCu sample pre-treatment (storage and UV oxidation) prior to analysis with ID methods.      Given the methodological issues explored here and the observed variability in dCu values in the literature, we believe that there is an urgent need to establish an optimal and consistent treatment of dCu samples. Future studies should: 1) compare dCu values with different analytical methods using acidified samples of the same age; 2) establish standard UV conditions to be used in future studies (intensity, time, use/no use of H2O2), and which methods may require this treatment; 3) address the identity of the inert Cu species in acidified samples (e.g. colloidal vs dissolved Cu-L species).   2.5.2  Copper cycling in the subarctic Northeast Pacific     Although exploration of Cu cycling along Line P is relatively new, a number of processes are known to moderate the distribution of nutrients and other trace metals (e.g. Cd, Fe, Zn, Ag, Mn, Ga, Pb) in this region, and are likely to influence Cu as well. These include: coastal upwelling, river discharge (Whitney et al. 2005), atmospheric deposition (Boyd et al. 1998; Hamme et al. 2010), onshore-offshore shelf transport (Lam et al. 2006; Cullen et al. 2009), and mesoscale eddies (Johnson et al. 2005; Xiu et al. 2011; Brown et al. 2012). Trace metals could also be sourced to this region via water masses penetrating the western and eastern reaches of the Line P transect (CUC (P4‒P12) and NPIW (P26), respectively, McAlister, 2015). In addition, recent work has shown that the extensive Oxygen Minimum Zone (300‒2000 m) of the subarctic NE Pacific plays an important role in the cycling and speciation of redox-sensitive metals such as Cd, Zn, Ag and Fe (Kramer et al. 2011; Janssen and Cullen 2015; Schallenberg et al. 2015), and may also affect Cu (Janssen et al. 2014).   39  2.5.2.1 Controls of dCu concentrations in the upper waters (<600m). 2.5.2.1.1 Water masses and continental shelf     The continental shelf of BC (~ 150‒200 m) is an important source of both dissolved and particulate Fe that can be transported long distances offshore (Lam et al. 2006; Cullen et al. 2009; Schallenberg et al. 2015). In addition, coastal stations of Line P are influenced by the intrusion of warm and salty waters of the California Undercurrent (CUC, 150‒200 m), which acts as a source of metals such as Ga to stations P4 and P12 (McAlister, 2015). In this study, however, we did not observe dCu enrichments along isopycnals surfaces associated with either the CUC or the continental shelf (~ 200 m, σt = 26.5‒26.8, Fig.2.7A). While Cu is known to have strong sedimentary source, it also has a strong affinity for organic matter (Fischer et al. 1986).  Inputs from the CUC or the continental shelf could be reduced by scavenging of Cu near the coast of BC where productivity in the summer is high (Peña and Bograd 2007). Instead, remineralization processes appeared to be the dominant source of dCu in waters between 40‒400 m as maxima were mostly associated with the bottom of the nutricline. This depth also coincides with the upper boundary of the Oxygen Minimum Zone (~ 400 m) in this region (Fig.7B). Thus, our data suggests that neither the continental margin nor advected water masses were major sources of dCu to Line P. 2.5.2.1.2 Sources of dCu to the mixed layer along the transect         The mixed layer distribution of dCu along Line P was examined in relation to that of dissolved aluminum (dAl) (Cain 2014, Fig. 2.10A & B), a proxy for atmospheric deposition and continental sources. The gradual decline of dCu from the coastal station P4 to the offshore station P16 (2.19 nmol kg-1 and 1.6 nmol kg-1, respectively) agree with the dAl trend. This suggests that North American continental sources, such as mineral aerosol deposition or fluvial runoff, may partially explain the elevated dCu at our coastal stations. A recent study suggests that fluvial inputs, rather than atmospheric deposition, from North America are more significant sources of Pb to stations P4 and P12 (McAlister, 2015). Furthermore, Semeniuk et al., (2016a) found a decreasing gradient of surface dCu (7‒10 m) with increasing salinity between coastal station P3 and offshore 40  station P16, suggesting that indeed fluvial sources are most likely responsible for the dCu trends observed near the coast.      Westward of station P16, the latitudinal trends of dCu and dAl in the mixed layer diverge, with dAl continuing its east-west decline (> 3-fold between P4 and P26), while dCu starts to increase towards P26 (OSP) (~ 0.4 nmol kg-1 between station P16 to P26).  Low dAl concentrations at OSP indicate the relatively small atmospheric inputs to this station. Thus, the disparity in the dAl and dCu trends at the offshore stations (P20 and P26) suggests that atmospheric deposition alone is unlikely to account for the increasing dCu towards OSP. Instead, upwelling of dCu rich waters in the Gulf of Alaska (GoA) is likely driving this trend (Fig. 2.11). Upwelling would bring high dCu concentrations to the surface while lowering dAl, as deep waters are typically depleted in dAl, due to its short residence time (Orians and Bruland 1985).      Transport of copper by mesoscale eddies is also a possible mechanism for increased dCu levels offshore, however satellite altimetry anomalies did not indicate the presence of an eddy near P26 during our study (Semeniuk et al. 2016a). Hence, we propose that the trend of increasing dCu levels towards the offshore stations is largely controlled by the upwelling, which is in agreement with previous explanations for the high surface water dCu in the Gulf of Alaska (Vertex VII, Martin et al., 1989). Using a simple-one dimensional model, we estimate that upwelling is responsible for supplying 0.58‒1.95 nmol L-1 y-1 dCu to the euphotic zone at P26, assuming the summer and winter mixed layer of 30 and 100 m, respectively (Appendix A, Table A.3).  2.5.2.1.3 Temporal variability of dCu at Ocean Station Papa (P26)     We showed that dCu concentrations in upper waters at OSP are temporally variable and identified sub-surface dCu maxima in Aug 2010 and 2012, that are suggestive of atmospheric inputs (Fig. 2.8). To explore the possibility of atmospheric aerosol deposition of Cu at OSP, we used Aerosol Optical Depth measurements (AOD, λ=550 nm) by Moderate Optical Imaging Spectroradiometer (MODIS) on NASA's Aqua satellite, a proxy for aerosol concentration in the atmosphere. The area averaged AOD plots between July‒August for all three years indicate the presence of elevated atmospheric aerosol concentrations in the western region of GoA near OSP and diminishing levels towards North America (Fig. 2.12). The potential deposition of these 41  atmospheric aerosols could explain high dCu in the subsurface waters at OSP in 2010 and 2012, but would not corroborate with the lower values measured in 2011. It is possible that atmospheric Cu inputs occurred in 2011 as well, but that we did not observe any effects on dCu surface distribution that year due to either Cu consumption by the resident community or the low solubility of Cu in deposited material.      The identification of the aerosol plumes we observed in the GoA requires further study. However we take an opportunity to examine their potential origins by assessing the atmospheric sources that are known to influence this region including : glacial flour from coastal Alaska (Crusius et al. 2011), volcanic ash from the Aleutian Islands (Hamme et al. 2010) and Asian dust (Boyd et al. 1998; Bishop et al. 2002). Deposition events associated with these sources may enhance local primary productivity, likely by inducing Fe-fertilization (Bishop et al., 2002, Hamme et al., 2010). Yet, due to the episodic character of such events, as well as their dependence on prevailing wind patterns (Takeda 2011) and local hydrology (Crusius et al., 2011), deposition of these sources to surface waters in the GoA is highly unpredictable, while their metal content remains unknown. It is unlikely that the high AOD we observed in this study was associated with the glacial flour from river valleys in Alaska as its transport tends to occur in autumn when coastal river levels are low and riverbed sediments are exposed (Crusius et al. 2011). While major eruptions from Aleutian Island volcanos can provide a spectacular amount of air-borne ash into the GoA as seen during the Kasatochi volcano eruption in August 2008 (Hamme et al. 2010), such events did not occur during our study. We did note, however, that some eruptive activity associated with small ash cloud formation for Mt. Cleveland volcano were reported in June 2010 and July 2012 (Herrick et al. 2014; Neal et al. 2014), but the role of this atmospheric source requires further exploration. A possible source of atmospheric inputs prior to our sampling may be dust from Asia, typically reaching the remote waters of the subarctic NE Pacific in early spring and summer (Uematsu et al. 1983). Asian dust originates from arid and semi-arid desert area in Northern China, the second largest dust source in the world (Shao et al. 2011) contributing  ~ 400 Tg dust to the North Pacific and beyond (Zhang et al. 1997). Indeed, the deposition of Asian sources in the GoA is supported by the distinct signature of lead isotopes measured in the mixed layer at OSP (McAlister, 2015). 42      Our observations suggest that surface dCu at OSP may be moderated via atmospheric deposition events in the Gulf of Alaska. There is a need to quantify dCu inputs from such events, which are likely to vary in time and space along the Line P transect. Indeed, the Hovmoller-longitude time series of monthly averaged AOD levels over the area covering Line P indicate annual, seasonal as well as spatial variability in the atmospheric aerosols across the transect (Appendix A, Fig. A.2). Generally, aerosol plumes tend to be associated with spring-summer periods and are confined to the western section of the transect. In contrast, the easternmost locations are characterised by persistently low aerosol levels all year round.   2.5.2.1.4 Biological implications of potential atmospheric inputs of Cu at OSP     In certain oceanographic settings atmospheric inputs can represent an important source of nutrients like nitrogen (Duce et al. 2008; Mackey et al. 2010), phosphorus (Markaki et al. 2003; Hsu et al. 2014), Fe (Duce and Tindale 1991; Jickells 1995, 2005; Boyd et al. 1998; Hamme et al. 2010), Co, Mn and Ni (Mackey et al. 2012) to the surface ocean. While aerosol deposition can be beneficial to phytoplankton by increasing availability of essential elements, it may also inhibit microbial growth if the concentration of potentially toxic metals, such as Cu, are too high (Paytan et al. 2009; Jordi et al. 2012).     Although future work is needed to assess the role of atmospheric deposition in dCu cycling at OSP and elucidate its origins, we explore the potential biological effects associated with inputs of Asian sources at this station, given the recent evidence for their deposition there (McAlister 2015). Highly urbanized and industrialized areas of east Asia are a source of harmful pollutants that may mix with mineral dust from deserts on the trajectory of its long-range transport across the North Pacific (Hoell et al. 1996; Hsu et al. 2010; Li et al. 2012). For instance, dust samples collected over urbanized areas such as Daejeon, Korea during Asian dust (AD) events (when strong monsoons carry dust from deserts) contain Cu levels that are on average 23‒57 times higher than uncontaminated Chinese desert soil (12.9 mg kg-1 versus 291‒740 mg kg-1, Lee et al., 2013). Furthermore, anthropogenically modified aerosols are often characterized by a high fractional solubility of Cu, hence have a greater potential to be solubilized in surface waters in deposition 43  areas than mineral aerosols. Indeed, as much as half of Cu in aerosols deposited over the East China Sea during the AD period is soluble (Hsu et al. 2010), compared to 1‒7% solubility of mineral aerosols from Saharan deserts (Sholkovitz et al. 2010). Thus, enhanced Cu content and high fractional solubility of contaminated aerosols that may be advected from Asia could have inhibitory effects on phytoplankton communities in deposition areas across the North Pacific, including potentially the GoA. A number of factors are likely to control such effects including Cu dissolution time from aerosols (Mackey et al. 2015), the buffering capacity of the in situ ligand pool, and taxonomic composition of the resident community. Abundance of smaller sized groups such as cyanobacteria and dinoflagellates, which are very sensitive to Cu toxicity may be diminished relative to more resilient diatoms (Brand et al. 1986). In our study region (station P16), a reduction in Cu bio-availability using an artificial, strong organic ligand cyclam, increased the abundance of cyanobacteria, suggesting that these organisms may be experiencing Cu stress (Semeniuk 2014). Thus, cyanobacteria may be vulnerable to any potential increases in ambient Cu that could result from atmospheric deposition events in the GoA. Our current understanding of the composition, deposition rates and biological effects of atmospheric sources deposited in this region is, however, still profoundly limited. As such, future aerosol studies would be highly beneficial from an ecological standpoint as well as to better understand the cycling of Cu and other bioactive elements in the GoA.  2.5.2.1.5 Correlations of dCu with macronutrients     As in previous investigations in the North Pacific, distribution of dCu was strongly coupled to that of the macronutrients phosphate and silicate in the oceanic nutricline (e.g. Martin et al.,1984, Bruland, 1980). The Cu:P nutricline ratio (0.53 mmol:mol, using data from all stations, depths 0‒400 m) in our study is in the range of ratios in this region and in others (e.g. 0.3‒0.8 mmol:mol for Central and North Pacific, Boyle et al., 1977, Martin et al, 1989, Bruland, 1980; 0.41‒0.47 mmol:mol for North Atlantic, Bruland and Franks, 1983; and 0.33‒0.54 mmol:mol for Indian Ocean, Morley et al., 1993). In contrast to our findings, recent studies in the Atlantic (Jacquot and Moffett, 2015, Roshan and Wu, 2015) and Southern Oceans (Heller and Croot, 2015) found poor correlations between Cu and PO34-, whereas its distribution showed a better correlation 44  with Si(OH)4 throughout the water column. In our study, the Cu:Si ratio in the nutricline (0.015 mmol:mol, all stations, depths 0‒300 m) is similar to that reported for the Atlantic sector of the Southern Ocean (0.012-0.020 mmol:mol, depths 10-5300 m, Heller and Croot, 2015) and for the Indian Ocean (0.02 mmol:mol, depths 0-5000 m, Vu and Sohrin, 2013), yet lower than the ratio in the Atlantic Ocean (0.035 mmol:mol, depths 0-5000 m, Roshan and Wu, 2015).      The correlations between Cu and macronutrients observed here suggest that the distribution of Cu along Line P is strongly coupled to biological assimilation and regeneration cycles. Indeed, converting the Cu-PO34- for the entire transect using the Redfield ratio of 106C:1P yields a Cu:C ratio of 5.1 µmol:mol, which is in the range of assimilation ratios determined in cultured phytoplankton (Annett et al. 2008; Guo et al. 2012b). Interestingly, the Cu:C of diatoms cells in field samples (4.2 µmol:mol) is also on par with the assimilation ratio determined here, while those of flagellates were substantially elevated (25-33 µmol:mol, Twining et al. 2015). This raises some interesting questions regarding the potential role of diatoms in influencing the Cu-PO34- ratios in the oceanic nutricline in the North Pacific.  2.5.2.2 Processes affecting dCu in deep waters (> 600 m) 2.5.2.2.1 Sedimentary sources of dCu     Sediments are considered to represent an important source of Cu to the overlying waters, as supported by the benthic flux estimates using water column and sediment pore-water Cu profiles (Boyle et al. 1977; Klinkhammer 1980; Klinkhammer et al. 1982; Heggie et al. 1987). Copper may be liberated from sedimentary material via aerobic respiration of organic matter (oxic sediments: Klinkhammer et al. 1982; Shaw et al. 1990) or via microbially mediated reductive dissolution of Mn/Fe oxides to which Cu is bound (suboxic sediments: Murray 1975; Hines et al. 1984). In contrast, anoxic sediments may to trap Cu due to the formation of highly insoluble precipitates with sulfide minerals, which tends to occur under reducing conditions (Morse and Luther 1999).            Given the large distance from the deepest sample and the seafloor at most of our stations (~ 1000‒2000 m, P12‒P26), we only examine the role of benthic sources at the shallowest station, P4 (deepest sample 200 m above the seafloor). The bottom enrichment of dCu at this station (1100 and 1200 m, Fig. 2.7B) is consistent with sedimentary input sources. Bottom sediments at station 45  P4 may be reducing as enhanced Fe (II) levels have been measured in deep waters at this station (1000 and 1200 m, Schallenberg et al. 2015), although during a different cruise than those in our study (June 2012 and Aug 2013). Reducing conditions in these sediments would likely favor formation of insoluble copper sulfides, thus diminishing the dCu inputs from bottom sediments. However, resuspension of bottom sediments and oxidation of CuS in overlaying waters may help explain the enrichment in bottom dCu concentrations at P4. We did not find evidence of sedimentary inputs of Cu from the continental shelf (~ 200 m), in contrast with other regions of the North Pacific where shelf source of Cu was suggested (Biller and Bruland 2013). It is possible that scavenging of dCu by sinking particles, which is likely to be intense at P4 given its high productivity in the summer (Whitney et al. 2005), reduced the signature of dCu input from the continental shelf. Alternatively, Cu may be scavenged from the dissolved phase upon sediment re-suspension (Fischer et al. 1986) on the shelf, or scavenged by sulfide minerals if shelf sediments are sufficiently reducing, both of which would limit Cu inputs into water column at P4.  2.5.2.2.2 dCu in the Oxygen Minimum Zone (OMZ)    Previous studies strongly suggest that the OMZ of the subarctic NE Pacific acts as a sink for elements such as Ag, Zn, Cd, and potentially Cu (Kramer et al. 2011; Janssen et al. 2014; Janssen and Cullen 2015). Both thermodynamic considerations and field observations suggest that these class B metals are depleted under anoxic conditions via formation of insoluble sulfide solid phases (Jacobs and Emerson 1982; Jacobs et al. 1985; Landing and Lewis 1991). Since waters within the OMZ of the subarctic NE Pacific still contain traces of O2 (lowest in this study O2 = 7 µmol kg-1), the presence of free sulfide in the water column is unlikely as it would become rapidly oxidized by O2. However, sulfidic conditions may develop within microenvironments on sinking particles in the OMZ’s (Janssen & Cullen, 2015). Formation of insoluble ZnS within these particles may explain the decoupling between Zn and Si distributions within the OMZ waters of subarctic NE Pacific (Janssen & Cullen, 2015).      Similarly, to the recent Zn study, we also observed a noticeable break in the relationship between Cu and Si that coincided with the Oxygen Minimum Zone (300‒2000 m) (Fig. 2.9B). Analogous to the Zn:Si, decoupling between Cu and Si within these O2 deficient waters might be 46  indicative of the scavenging process described by Janssen and Cullen (2015). However, unlike Zn & Si, the water column behavior of Cu and Si are not as tightly coupled in the world's oceans, creating some uncertainties in such an interpretation. Furthermore, behavior of Cu in low O2 settings is not straightforward because processes that act to release and remove dCu may be occurring simultaneously.  For instance, dissolution of Mn/Fe-oxyhydroxide minerals (under hypoxic and suboxic conditions), as well as POM remineralization, would enhance dCu in the water column of OMZ. Both processes may partially explain the accumulation of dCu we observed in the upper reaches of the OMZ (400 m) where low O2 levels might be favoring reductive dissolution of Cu, and where remineralization of POM is highest.  In contrast, formation of CuS within microenvironments on sinking particles would act as a sink of dCu in the OMZ. However, it is currently uncertain how effective this mechanism might be in the presence of strong organic ligands binding most of the dissolved Cu, even in these deep waters (Moffett and Dupont, 2007). Furthermore, Cu redox chemistry, which is poorly understood under such conditions, adds another level of uncertainty because Cu in +1 state is a definite class B acceptor (soft metal forming stable complexes with donor atoms such as S), while Cu in +2 state can behave as both soft and hard metal (forming stable complexes with donor atoms such as N, O, and F; Ahrland et al. 1958). However, along Line P, the dCu concentrations at depths corresponding to the OMZ were generally uniform. Similar behavior of dCu can be seen in a previous study at station P26 by Martin et al. (1989) (Appendix A, Fig. A.3). In contrast, it is typical for dCu concentrations to gradually increase with depth (e.g. Jacquot and Moffett, 2015, Fig. A.3), which can be explained by the flux of Cu from bottom sediments and particle scavenging throughout the water column (Boyle et al. 1977; Takano et al. 2014). The behavior of dCu within the OMZ of the NE Pacific suggests that it may be scavenged more intensely in these low O2 waters by the process described by Janssen and Cullen (2015).      Cu distributions within the OMZ waters near Mauritania (Jacquot and Moffett, 2015, Roshan and Wu,2015) and eastern subtropical North Pacific (Nameroff et al. 2002) did not show any anomalies associated with low O2 conditions. In contrast, accumulation of dCu was observed within the secondary nitrite maximum of the eastern tropical South Pacific OMZ, although shelf sources were likely responsible for this trend rather than low O2 conditions (Jacquot et al, 2013). Interpreting Cu behavior within OMZ’s in continentally influenced study sites, as is the case of 47  these previous investigations, is challenging because of strong dCu sources there that may be masking potential removal processes. While further research is required to confirm this, our dataset may support the recent hypothesis for the scavenging of Cu within the OMZ of the Northeast Pacific (Janssen and Cullen, 2015). Using the Cu:Si relationship for each station along Line P, we estimated that approximately 0.01‒1.3 nmol kg-1 of dCu is ‘missing’ within the OMZ, suggesting these intermediate low O2 waters may potentially have significant role in reducing the dissolved Cu inventory in the subarctic NE Pacific.  2.6 Conclusions     In the present study, we identified major biogeochemical and physical processes that drive the spatial and temporal patterns of dCu along Line P. Our dataset also adds to the growing body of dCu measurements across the North Pacific, and aims to improve our understanding of Cu behaviour throughout this basin. In the eastern sections of our transect, fluvial rather than atmospheric inputs from North America were responsible for enhanced dCu levels in surface waters. Remote, Fe-deplete stations along Line P were also found to be enriched in dCu, a feature we propose to be largely driven by upwelling in the Gulf of Alaska. Atmospheric deposition may represent an additional source of Cu to these remote waters as indicated by the sub-surface dCu enrichments at station Papa, which coincide with elevated atmospheric aerosol levels. Furthermore, we explored the possible sources of atmospheric Cu to this region and discuss the potential ecological impacts of these inputs in surface waters at OSP.  Across our transect, Cu was strongly correlated with both phosphate and silicate in the nutricline, reflecting the influence of biological processes (uptake and remineralization) on the Cu distribution in this region. We also provided some evidence that dCu may be sensitive to scavenging within the OMZ, behaving similarly to other soft metals such as Cd and Zn in these O2 deficient waters (Janssen et al. 2014, Janssen & Cullen, 2015). Finally, our study addressed some of the recent methodological uncertainties regarding dCu analysis. We observed that the length of acidified sample storage largely impacts the outcome of dCu analysis by FIA-CL. With shorter storage periods (up to 2 months) a UV oxidation pre-treatment of the samples is required for accurate measurements of dCu, while this treatment was found to be less important if samples were stored for an extended 48  length of time (several years). In addition, we also compared different analytical methods to measure dCu and discussed how various methodologies may also contribute to current uncertainties in dCu values. We proposed that the interplay between the three factors briefly explored in our study (sample storage, UV oxidation, analytical methodologies) may explain the inconsistencies in dCu concentrations of GEOTRACES reference materials and in dCu profiles in the North Pacific.    Table 2.1: Comparison of dCu values (nmol kg-1) determined using FIA-CL with the GEOTRACES inter-calibration materials consensus values as of May 2013 (SAFe S, D1, and D2) and NASS-6 certified reference material (gamma-irradiated) for ocean water (National Research Council Canada, http://www.nrc-cnrc.gc.ca/eng/solutions/advisory/crm/certificates/nass_6.html). Also included are values for the new GEOTRACES standard, GSC 318 for which there is no consensus.                  Reference material  +UV (2 h) No UV Consensus/ Certified SAFe D1 (2015) 2.24 ± 0.19 (n=2) 2.31 ± 0.11 (n=3) 2.28 ± 0.15 SAFE D2 (2012) nd 2.34 ± 0.16 (n=3) 2.27 ± 0.11 SAFe D2 (2015) 2.33 ± 0.12 (n=4) 2.27 ± 0.15 (n=3) 2.27 ± 0.11 GSC 318 (2015) 1.46 ± 0.11 (n=5) 1.38 ± 0.02 (n=2) No consensus NASS-6 (2012) nd 3.88 ± 0.15 (n=4) 3.81 ± 0.03 49                            Figure 2.1: Map of the Gulf of Alaska showing Line P transect and the 5 sampling stations (P4-P26). Bathymetry was contoured at 1000 m intervals.                                    50                                      Figure 2.2: Profiles of dCu (nmol kg-1) at station P26 (Ocean Station Papa). Samples from August 2011 cruise analyzed with FIA-CL without UV oxidation between Jan‒Feb 2012 (open circles) and analyzed again between June‒August 2015 with 2 h of UV oxidation (closed circles). Also plotted is the dCu dataset reported by Martin et al (1989) using APDC/DDDC-chloroform organic extraction without UV treatment (triangles).     51                                        Figure 2.3: Changes in labile Cu with increasing storage time of (A) 48 h, (B) 2 weeks and (C) 2 months. At each time point, acidified samples were analyzed without (closed circles) and with 2 hr of UV treatment (open circles) and analyzed by FIA-CL. DCu profile obtained in 2015 (~ 4 years of storage) is also shown for reference (triangles).                        52                              Figure 2.4: Comparison of dCu datasets at station P26 analyzed by FIA-CL with 2 h of UV oxidation and sample storage at pH 2 of ~ 4 y (Aug 2011 cruise, circles) and CSV with UV oxidation and no acidified storage (Aug 2012 cruise, gray triangles) (A). Linear regressions of data obtained in this study by FIA-CL (AP), versus CSV (HW, gray triangles), and versus surface transect data from Semeniuk et al (2016a) (DS, open squares) (B). The regression fits are:  y=1.058x-0.26, r2=0.78, between datasets of AP & HW (using all data points), and y= 0.959x-0.104, r2=0.7, between datasets of AP & DS, excluding the outlier data point at 2 nmol kg-1 dCu. 53                           Figure 2.5: Comparison o dCu datasets in the North Pacific. Upper and bottom panel shows plots of selected dCu profiles from different regions (Region 1 [R1]-Region 7 [R7]) as indicated in the map. The details of methodologies used in each study are provided in Appendix A, Table A.2). 54           Figure 2.6: Potential temperature - salinity plot (θ-S) including all stations along the transect (P4 - ◊, P12- □; P16 - x; P20 - ☆; P26 - ∆) with dissolved oxygen (µmol kg-1) as symbol colors.                        55                       Figure 2.7: Top panel (A) shows profiles of dCu (nmol kg-1), PO43- (µmol kg-1) and sigma-t (kg m-3) in the upper 300 m of stations along the Line P transect. Bottom panel (B) shows depth profiles of dCu, PO43- and O2 (µmol kg-1) throughout the entire water column sampled (10-2000 m) with the bottom depths at each station.   56                               Figure 2.8: Profiles of dCu (nmol kg-1), PO43- (µmol kg-1) and sigma-t (kg m-3) at station P26 obtained with samples collected in Aug 2010, 2011, and 2012. Samples from 2010 & 2011 were analyzed with FIA-CL (2 h of UV), and samples from 2012 were analyzed by CSV (45 min of UV). Density and PO43-data are courtesy of the Line P program, Institute of Ocean Sciences, Department of Oceans & Fisheries, Canada. This data is available in Appendix A, Table A.3    57                             Figure 2.9: Macronutrient relationships: dissolved copper with PO43- (A) and dissolved copper with Si(OH)4 (B) using data from the entire transect. Linear regressions (solid lines) were performed using data between 50-400m for phosphate ([DCu]=0.53[PO43-] + 1.27, r2 =0.86) and 0‒300m for silicic acid ([dCu]= 0.015[Si(OH)4] + 1.68, r2=0.73). Horizontal dashed lines indicate the extension of linear regressions assuming dCu data was to follow the nutricline trends. Vertical dashed lines represent data points used for the linear regressions. 58          Figure 2.10: Mean mixed layer dCu and dAl trends along the Line P transect in August 2011 (dAl data from Cain, 2013) with an inset showing the spatial surface dCu distribution along the transect.                        Figure 2.11: Contour plot of dCu in the upper 600m along the Line P transect in August 2011. White dashed lines with labels represent dissolved PO43- (µmol kg-1). Shoaling of high dCu and phosphate can be seen at the offshore stations (P16‒P26) being suggestive of the upwelling conditions.59                         Figure 2.12: Area averaged Aerosol Optical Depth (AOD at λ=550 nm, AquaMODIS, 1ᵒ resolution) in the Gulf of Alaska between June‒August 2010, 2011 and 2012. AOD data was downloaded from NASA Giovanni visualization and analysis online data system developed and maintained by the NASA GES DISC (http://giovanni.gsfc.nasa.gov/giovanni/) and plotted using MATLAB. The script used to produce the plot can be found in a public repository: https://github.com/AnnaMagdalena/DCu_LineP-Subarctic-Pacific.60  Chapter 3:  Effects of Cu availability on growth and metabolism of marine heterotrophic bacteria  3.1 Summary         We explored the effects of copper (Cu) availability on the physiology and metabolism of oceanic and coastal marine heterotrophic bacteria from ecologically significant microbial clades (Flavobacteriia within phylum Bacteroidetes, marine Roseobacter clade within class Alphaproteobacteria and Alteromonadales within class Gammaproteobacteria). Bacterial growth, Cu quotas (Cu:P), macronutrient content and stoichiometry (cellular C, N, P, S and C:N, S:P), as well as carbon metabolism (respiration, productivity, carbon demand, growth efficiency) were monitored across a gradient of Cu conditions characteristic of coastal and open-ocean surface waters. The effects of changing Cu availability were most striking in a Flavobacteriia member Dokdonia sp. strain Dokd-P16 for which we observed significant impairment in all metabolic rates and reductions in Cu quotas when Cu was limiting. Other strains did not significantly reduce their growth rate, but adjusted their Cu content and some C metabolic rates (Ruegeria pomeroyi DSS-3, Roseobacter clade) or were largely unaffected (Pseudoalteromonas sp. strain PAlt-P2, Alteromonadales clade). These diverse responses were accompanied by constant cellular composition of major elements and stoichiometric ratios by all strains. Changes in bacterial Cu quotas occurred within a relatively narrow range (~2-fold range) despite a 50-fold variation in total Cu. The range of Cu quotas in our study was modest (0.03‒0.15 mmol Cu:mol P) despite the contrasting phylogenies and ecological provenance of our strains. This may reflect a well-controlled Cu homeostasis in marine heterotrophic bacteria. We conducted a preliminary comparison of Cu quotas of our marine heterotrophic bacteria with other prokaryotes from literature as well as marine phytoplankton, which suggested that Cu requirements may not be markedly different among these groups. Using Cu quotas obtained in our study, we examined the potential contribution of bacterial Cu to the biogenic Cu pool in the oceanic euphotic zone in the 61  NE Pacific. These estimates attribute a significant portion of biogenic Cu to marine heterotrophic bacteria (~ 4 to 50%) and implicate these prokaryotes as important players in the Cu biogeochemical cycle. Our study sheds light on the interactions between Cu and marine heterotrophic bacteria, which has implications for prokaryotic ecology and oceanic Cu cycling.  3.2 Introduction      Planktonic bacteria dominate the living biomass in the world's oceans (Whitman et al. 1998). In surface waters, prokaryotic assemblages are mainly dominated by autotrophic and heterotrophic bacteria, while archaea are more abundant in the ocean interior (Karner et al. 2001). Heterotrophic members of the bacterioplankton community play a critical role as the recyclers of organic matter, thus influencing fluxes of energy and essential elements in the oceanic ecosystem. Their activity is constrained by the availability of organic substrates as well as inorganic nutrients. This includes the trace metal, iron (Fe), which has been shown to regulate growth and metabolism of heterotrophic microbes in the lab (Tortell et al. 1996; Kirchman et al. 2003a; Fourquez et al. 2014a) and field (Pakulski et al. 1996; Bertrand et al. 2011). In the Northeast Pacific heterotrophic bacteria accounted for ~ 50 % of biogenic Fe and 20‒70% of community Fe uptake rates (Tortell et al. 1996; Maldonado and Price 1999), indicating their significant influence on the biogeochemical cycling of Fe (Tortell et al. 1999). Interestingly, in a recent study in the Southern Ocean (Fourquez et al. 2014b) the contribution of the heterotrophic bacteria to the community Fe uptake was found to be low (1‒2%), suggesting that the bacterial impact on the Fe cycle may be regionally variable. In contrast, the interaction between bioactive trace elements and oceanic heterotrophic bacteria remains largely unexplored.     Among these micronutrients, Cu is unique because even though it is required for growth, it can be extremely toxic even at minute levels. It has the potential to induce oxidative stress, due to its ability to participate in Fenton-like reactions that generate harmful hydroxyl radicals (Rowley and Halliwell 1983). Another important Cu-induced toxicity mechanism is the destruction of Fe-S clusters in enzymes by direct complexation of Cu(I) to the coordinating sulfur atom (Macomber and Imlay 2009). Hence, prokaryotes must handle their cellular Cu carefully to avoid toxicity, while at the same time satisfying their Cu nutritional requirements. Prokaryotic Cu homeostasis 62  has been the subject of extensive research in model bacteria (see reviews of Rensing and Grass 2003; Solioz and Stoyanov 2003; Solioz et al. 2010; Argüello et al. 2013; Bondarczuk and Piotrowska-seget 2013), with a focus on elucidating mechanisms for Cu detoxification. In this context, the interactions between Cu and aquatic bacteria have also been explored (Gordon et al. 1993, 1994, 2000; Moffett and Brand 1996; Mann et al. 2002). In contrast, little is known on how prokaryotes cope with low Cu availability.     To our knowledge, the effects of Cu deficiency have only been investigated in two groups of marine microbes; heterotrophic denitrifying bacteria and autotrophic ammonia-oxidizing archaea (Granger and Ward 2003; Moffett et al. 2012; Amin et al. 2013); whereas, aerobic heterotrophic bacteria have been largely overlooked. Copper serves as a catalyst in ten known prokaryotic cuproenzymes, of which the respiratory enzyme, cytochrome c oxidase (COX), has the most prominent use in aerobic heterotrophic bacteria (Ridge et al. 2008). Hence, Cu starvation may potentially affect C metabolism in these organisms, impacting a cell’s energetic status and growth. Moreover, there is a lack of measurements of Cu contents of marine heterotrophic bacteria, thus, their role in the marine biogeochemistry of Cu remains largely unknown.      The goal of this investigation was to provide an integrated characterization of the physiological responses of marine heterotrophic bacteria across a range of environmentally relevant Cu conditions. We examined bacteria isolated from eutrophic and oligotrophic surface waters of the Northeast Pacific Ocean, as well as a model bacterium Ruegeria pomeroyi DSS-3. Our organisms belong to ecologically significant microbial groups within the world’s ocean prokaryotic populations, namely Flavobacteriia class within the phylum of Bacteroidetes, Roseobacter clade within the Alphaproteobacteria class (Proteobacteria), and the order Alteromonadales within the Gammaproteobacteria class (Proteobacteria). Members of Flavobacteriia and Roseobacter clades are abundant in both coastal and oceanic surface waters in various habitats (Roseobacteria: 15‒25%, e.g. González and Moran, 1997, Eilers et al. 2000, Giovannoni and Rappé, 2000, Selje et al. 2004;  Flavobacteria: 10‒70%, eg. Glöckner et al. 1999, Eilers et al., 2000, Abell & Bowman 2005), while Alteromonadales are less abundant (Eilers et al. 2000), but can increase substantially during natural phytoplankton blooms (Lucas et al. 2015). Collectively, bacteria affiliated with Gammaproteobacteria, Roseobacter, and Flavobacteriia are key players in biogeochemical nutrient cycling, regulating phytoplankton bloom-events and 63  transforming phytoplankton-derived organic matter (as reviewed in Buchan et al. 2014). Furthermore, bacterial clades such as Flavobacteriia and Roseobacter containing family Rhodobacteraceae belong to the dominant subgroups of to the prokaryotic communities in surface waters of the NE Pacific (Wright 2013). The responses of our strains Dokd-P16 and R. pomeroyi DSS-3 may provide some insight into the prokaryotic response to Cu availability in these waters, although we acknowledge and discuss the uncertainties of extrapolating these responses to field populations. Our investigation represents one of the very few examinations of adaptive strategies to Cu starvation in prokaryotes.   3.3 Methodologies 3.3.1 Study organisms    The heterotrophic bacteria used in this study were isolated from different locations along Line P, a transect extending from the coast of BC into the open-ocean waters of the Gulf of Alaska, 1500 km offshore. Near-surface seawater (~ 25 m) was collected at four different stations along the transect (P2, P4, P16 and P26) during the June cruise in 2012 (Line P Program, cruise 2012-12, http://linep.waterproperties.ca/2012-12/index.php). The seawater was then plated onto marine agar supplemented with either 0.5 g L-1 (for seawater from the coastal stations P2 and P4) or 0.05 g L-1 (for seawater from oceanic stations P16 and P26) of organic substrate (bactopeptone and casein hydrolysate in a 1:1 ratio). The agar plates were prepared using filtered seawater (0.22 μm) collected from ~ 40 m at station P26. The colonies which grew on the plates were purified by sequential plating on marine agar. Pure cultures were preserved and stored in 15% glycerol stock at -80oC and plated as needed for experiments.      Isolated strains were identified using 16S ribosomal DNA PCR amplification. PCR was performed on Line P isolates using liquid cultures from colonies originally grown on agar plates. Liquid cultures were grown by adding a single colony to 2 mL of sterile liquid media made with station P26 seawater (from 10 m) supplemented with 0.5g DOC [0.25 g L-1 bactopeptone + 0.25 g L-1 casein hydrolysate].  Test tubes were placed in a shaker for 24 hours at room temperature. Following culture growth, the full-length 16S rRNA gene was PCR amplified with primers 27F (5′-AGAGTTTGATCMTGGCTCAG -3′) and 1492R (5′- GGTTACCTTGTTACGACTT-64  3′).  PCR amplification was performed in a 25 µL reaction volume with 16.7 µL of nuclease free water, 2.5 µL of buffer, 1.5 µL MgCl2, 2 µL dNTPs, 0.5 µL forward and reverse primers, 0.3 µL of Taq polymerase and 1µL of liquid culture. The reaction mixture was placed in the Thermocycler and run under the following conditions 95°C for 10 minutes, followed by 25 cycles of 95°C for 30 seconds, 55°C for 30 seconds, 72°C for 90 seconds, followed by a final extension for 10 minutes at 72°C.  The amplicon size was verified with an agarose gel, and the amplicons were purified with the NucleoSpin Gel and PCR Clean-up kit (Macherey-Nagel).  Samples were diluted to ~ 4 ng µl-1 and sent to the GeneWiz sequencing center (GeneWiz, Inc.) for sequencing using three primers, 27F, 1492R, and a center primer, 519R (5’-GNTTTACCGCGGCKGCTG-3’).  The three sequences for each clone were aligned using Sequencher 5.1 and a consensus sequence for each clone was obtained.  The sequences were compared to the SILVA database version 115 (http://www.arb-silva.de/) and isolates were taxonomically identified to the genus level (Table 3.1). In addition to the Line P isolates, we studied the model marine bacterium Ruegeria pomeroyi DSS-3, DMS 15171 (isolated from coastal Georgia), which was obtained from the American Type Culture Collection culture collection (ATCC, Manassas, VA, USA). The bacterial strains examined here are diverse in terms of their provenance and phylogeny (Table 3.1), and thus are likely to have distinct Cu requirements and adaptations to changing Cu availability. In our preliminary studies, we found a similar growth response to low and high Cu availability in different Pseudoalteromonas sp. isolates (Appendix B, Fig. B1). Ultimately, we selected one coastal (station P2, strain PAlt-P2) and one oceanic isolate (station P26, strain PAlt-P26) of Pseudoalteromonas sp. for more detailed investigations.  3.3.2 Experimental design and culture conditions      Bacterial physiology and metabolism were examined at five dissolved Cu levels (dCu) from limiting to replete, with a 50-fold variability in dCu (Table 3.2). At each Cu concentration, the following measurements were obtained: 1) growth rates; 2) Cu content 3) major elemental composition [carbon (C), nitrogen (N), phosphorus (P) and sulfur (S)]; and 4) bacterial O2 consumption rates. Bacteria were grown at 19 ± 1°C as semi-continuous batch cultures in the chemically well-defined artificial seawater medium Aquil (Price et al. 1988) modified for culturing 65  marine heterotrophic bacteria (Granger and Price 1999). Organic substrates were added to a final concentration of 0.5 g L-1 and consisted of casein hydrolysate and bactopeptone (1:1 ratio, Sigma-Aldrich). Organic substrate stocks were purified separately with Chelex 100 resin (25g per 100 mL of organic substrate, Bio-Rad) prepared following the protocol of Price et al. (1988/1989). These purified stocks were then used to amend 250 mL of artificial seawater (SOW) in trace metal-clean polycarbonate bottles and the media were microwave-sterilized for approximately 7 minutes. Once cooled, SOW with organic substrates was amended with filter-sterilized additions of vitamins and trace metal stocks. Trace metals were buffered with 100μM ethylenediaminetetraacetic acid (EDTA), and except for Cu, their total concentrations in the media were identical to those in Maldonado et al.  (2006). Copper was either omitted or added separately as Cu-EDTA complex (1:1.1) to create 5 different Cu treatments (Table 3.2). All plastics used in the growth experiments were sterilized and rigorously cleaned prior to use by storage in Extran and 10% HCl for at least 24 hrs. All manipulations of bacterial cultures were done under sterile and trace metal clean conditions using a laminar flow hood.     In Aquil medium, EDTA and trace metals are manipulated to buffer the concentrations of the free metals—typically the most bioavailable metal species—at environmentally relevant concentrations. Because of the unknown Cu affinity of the organic substrate added to the bacteria growth media, Cu speciation in these media cannot be precisely estimated. For this reason, Cu treatments are primarily reported here in terms of the total dissolved Cu (Cutot). However, we estimated the concentrations of free Cu (pCu = - log [Cu2+]) and of inorganic Cu (Cuʹ) in the growth media (ignoring the organic substrate additions) using the chemical equilibrium model Visual MINTEQ  (version 3.1, Gustafsson 2016, Table 3.2). These values are only approximate but are included here for reference, as previously done for other studies of marine heterotrophic bacteria involving organically-amended Aquil (Granger and Price 1999; Fourquez et al. 2014a). Granger and Price (1999) found that the concentration of kinetically labile Fe was higher in Aquil modified for heterotrophic bacteria compared to the traditional Aquil (with the same amount of Fe and EDTA), and suggested that the additions of organic substrates for bacterial growth may have enhanced Fe lability. There is some evidence suggesting that the presence of weaker ligands (e.g. cysteine and histidine) in seawater culture containing a stronger ligand (e.g. EDTA) can also increase Cu and Zn availability (e.g. Aristilde et al. 2012; Kim et al. 2015; 66  Semeniuk et al. 2015) due to a shift in the equilibrium of the metal-strong ligand complex towards dissociation (Aristilde et al. 2012). Unfortunately, the labile concentrations of Cu, Co, Mn, and Zn in organics-amended Aquil have not been evaluated, but it is possible that, as with Fe, their levels are higher than in the traditional Aquil media. The approximate levels of inorganic Cu [Cuʹ] in our experiments encompass a range of concentrations that may be experienced by bacteria in both coastal and open ocean environments (Moffett 1995; Jacquot et al. 2014; Semeniuk et al. 2016a).     3.3.3 Growth rate measurements     Bacterial growth rates were determined as previously described (Granger and Price, 1999). Briefly, cells acclimated to different Cu treatments (for 8 generations) were inoculated into trace metal clean, sterile polystyrene cuvettes with caps and a stir bar (all sterilized by microwaving). Cultures were incubated in the spectrophotometer (Cary 1E UV-Vis, Varian) at 19°C and their optical density (A=600 nm) was measured at 5 min interval.  Absolute growth rates were determined during the exponential phase from log-linear regressions of absorbance versus time.  3.3.4 Enumeration of bacteria     Bacterial cell numbers were estimated using flow cytometry analysis as described in Brussaard (2004). Cultures were diluted (x 5) with pre-filtered SOW (0.22 μm PC filters) before the addition of glutaraldehyde (Electron Microscopy Grade, Sigma-Aldrich) to a final concentration of 0.5 % (i.e. 980 μL of culture sample containing 20 μL of 25% glutaraldehyde). Samples were taken in duplicate, flash frozen with liquid nitrogen and stored at -80° C until analysis. For cytometry analysis with FACS Calibur (Becton-Dickinson, Franklin Lakes, New Jersey, USA) samples were diluted up to 1:1000 in sterile 0.1 μm filtered 1 x TE buffer (10 mM Tris-HCL and 1 mM EDTA, pH 8), stained with SYBR Green I (Invitrogen, Waltham, MA, USA) at a final concentration of 0.5 x 10-4 of commercial stock. Samples were incubated for 15 min in the dark before the processing run for 1 min at a high rate. Event rates were kept between 100‒1000 per second and green fluorescence and side scatter detectors were used. Data were processed and gated using Cell-Quest software (Becton-Dickson) 67   3.3.5 Intracellular carbon and nitrogen analysis      Samples in mid- to late-exponential growth phase (10 to 20 mL) were filtered by gravity onto 25-mm pre-combusted Whatman GF/F filters (> 4 hr, 450°C). Cells were rinsed with 2 mL of 0.6M NaCl to remove extracellular organics and the filters were preserved at -20°C until analysis. To normalize the C and N data, samples for cell number enumeration by flow cytometry were taken prior to filtration. Samples for flow cytometry analysis were collected from bacterial cultures before filtration onto GF/F filters, as well as from the filtrate. This allowed us to correct the cell numbers for any cell losses during filtration. However, on average, more than 99 % of the bacteria were retained on the filters, similarly to a previous study using gravity filtration to collect cultured heterotrophic bacteria for C analysis (Pedler et al. 2014). Procedural blanks consisted of filters that were rinsed with media containing no bacterial cells. These blanks were typically < 5 % of the signal of the sample and the precision of the analysis ranged from 1‒6 % (range in RSD of 20 duplicate samples). Filters were dried overnight at 50°C, fumed with HCl, and dried again at 50°C before analysis on Perkin Elmer analyzer.   3.3.6 Intracellular P, S, and trace metal analysis     Samples for trace metal analysis were collected in mid-exponential growth phase. From each biological replicate, 50 mL of culture were collected and centrifuged at 9,000 x g for 10 min. The supernatant was decanted and cells were rinsed with cold (4°C) 20 mL of oxalate wash (pH 7.0, Tang & Morel, 2006) for 15 min at 4°C. Preliminary experiments showed that washing bacterial cells with the oxalate solution did not cause cell leakage, as described in Methods section 3.3.6. Cells were then centrifuged for 10 min (9,000 x g, 4°C) and rinsed with 20 mL of chelexed 0.6M NaCl for 5 mins at 4°C, and centrifuged again. The pellet was then re-dissolved in 1 mL of chelexed 0.6M NaCl and transferred to a clean microcentrifuge tube. An aliquot (5 μL) was taken for cell enumeration by flow cytometry. Subsequently, the sample was centrifuged for 10 min (10,000 x g, 4°C), the supernatant was removed and pellets were immediately stored at -20°C until analysis. The metal content of bacterial pellets was determined via inductively-coupled plasma-68  mass spectrometry (ICP-MS) using an instrument equipped with a hyperbolic quadrupole mass analyzer and octupole collision cell (Agilent, 7700). Bacterial pellets were digested in 5 mL Teflon digestion vials, which were cleaned following these steps: 1) detergent Extran for 24 hr (50°C), 2) 6N HCl for one-week (50°C), and 3) 1N HNO3 for one week (50°C). Vials were rinsed in between the cleaning steps with MilliQ water (x 6 rinses) and were leached in MilliQ for a week at 50°C before use. Pellets were digested in 3 mL of ultrapure HNO3 at 120°C for 7 hours, cooled down, and evaporated to dryness overnight at 120°C.  Subsequently, 1 mL of ultrapure H2O2 was added to each vial and evaporated to dryness at 120°C for ~ 4 h. Procedural blanks consisted of vials with no pellets treated in the same way as the samples. For ICP-MS analysis, vials were filled with a matrix, consisting of 2% HNO3 spiked with 10 ppb In as an internal standard, and allowed to stand for 10 minutes before being transferred to 15 mL trace metal clean polyethylene tube. The final volume of samples was subsequently adjusted to ~10 mL.     The instrument was calibrated using multi-element standard (High-Purity Standards, Charleston, SC, USA) prepared in the same matrix as the samples. Two isotopes for all elements with multi-isotopes (Fe 56/57, Cu 63/65, Zn 66/68) were measured to ensure that background and interference were acceptably low. Determination of phosphorus (31P) by quadrupole ICP-MS is challenging due to high background and interferences from polyatomic ions such as 15N16O+; 14N17O+; 14N16O1H+ and 30Si1H+. We assessed P data obtained using ICP-MS by comparison with measurements of P in the same samples by ICP-Optical Emission Spectroscopy (OES), which is widely used in P analysis of aqueous samples. There was an excellent agreement between the P measurements made by these two different instruments (Appendix B, Fig B.2), thus confirming the accuracy of the P values obtained with ICP-MS.  Sulfur data were obtained using ICP-OES. The limit of detection calculated as 3 x SD of the blank was 0.07 nmol L-1 for Cu, 1.5 nmol L-1 for Fe, 0.3 nmol L-1 Zn, 0.02 nmol L-1 Mn, 0.04 nmol L-1 Co. Samples measured on average > 40 to 100 times higher than the detection limit for all metals except Co (10 times higher). Trace metal data were normalized using phosphorus, cell numbers or cellular carbon data. We focus on P normalized quotas because this element was measured simultaneously with Cu during ICP-MS analysis, thus eliminating errors due to different analytical techniques. P-normalization allows for comparison of metal contents independently of bacterial cell volume, which was not measured during this study. While bacterial P content, can be plastic depending on environmental 69  conditions (e.g. organic substrate quality, Godwin and Cotner 2015a), we found that it not to change significantly in response to Cu (see section 3.4.2). Indeed, P-normalized Cu quotas followed the same trends as those of the cell- and C-normalized Cu quotas (Appendix B, Fig. B.3), suggesting it was an appropriate biomass proxy.  3.3.7 Assessment of cell leakage due to washes          Determination of intracellular elements requires a washing step to remove any elements adsorbed to the outside of the cells. There are various washing protocols, depending on the element of interest (e.g. Fe vs. Cu) and the method used for quota determination (e.g. radioisotope techniques vs. ICP-MS). However, washing solutions could affect cell membrane integrity and lead to a loss of intracellular elements, resulting in underestimations of elemental quotas. In this study, we examined whether two commonly used washes, oxalate-EDTA  (Tang and Morel 2006) and DTPA  (Croot et al. 2003), induce cell leakage in bacteria using 14C radiolabeled glucose. For these experiments, the oceanic Pseudoalteromonas sp. strain PAlt-P26 was selected as it grew successfully in modified media containing glucose as a C source (μ ~ 3 d-1), unlike Dokdonia sp, strain Dokd-P16, which we were not able to culture in this medium. The modified medium contained 35 mM glucose, 6 mM ammonium and 0.1 mM phosphate in filtered seawater (Weaver et al. 2003). We used water from Ocean Station Papa (OSP, station P26, 10 m) amended with vitamins and trace metals, as for the Aquil media recipe. A small inoculum of exponentially growing PAlt-P26 acclimated to the modified media was transferred into 250 mL medium containing 20 μCi 14C and cultured overnight at room temperature.  Wash test experiments involved filtering two aliquots (2.5 mL) of exponentially growing bacterial culture (0.22 μm polycarbonate filters) and incubating one aliquot with a wash (2.5 mL, oxalate or DTPA solutions) and the other aliquot with sterile SOW (2.5 mL) for 15 min in the dark at room temperature. This procedure was performed in triplicate. The oxalate and the DTPA incubated aliquots of cell culture were rinsed with sterile SOW before final filtration. Filter blanks (n = 3) were collected by subjecting filters to the same washing procedure but without bacterial cells and were used to correct the 14C activity of sample filters. Cell leakage was determined by comparing the 14C activity of wash-treated (oxalate or DTPA) and SOW only-treated cells.  A paired student’s t-test was used 70  to determine whether the differences between the washes (oxalate or DTPA) and SOW were statistically significant. There was no significant difference between wash-treated (oxalate, T-value=3.85; p=0.061; DTPA, T-value=1.6; p=0.24) and SOW-treated cells indicating that the washing solutions do not induce a cellular content loss (Appendix B, Fig. B.4). We thus assumed that just like with 14C-glucose, the loss of Cu assimilated by the cells would be minimal. We do acknowledge, however, that the C and Cu pools in bacteria may not be the same. In gram-negative bacteria, Cu-containing proteins are believed to be mostly located in the inner membrane and/or the periplasmic space (Tottey et al. 2005). We speculate that a damage to the cell’s membranes, in addition to a loss of cytoplasmic C constituents, would also lead to a loss of membrane-bound Cu and Cu ions in the periplasm. Ideally, to assess Cu leakage, cellular Cu pools in the cells would have been labelled with a radioactive Cu tracer (e.g. 64Cu or 67Cu). Unfortunately, these tracers are not commercially available. Ultimately, oxalate solution was selected for cell washing during the collection of samples for ICP-MS analysis (as described in Section 3.3.5) due to its previous use in the removal of multiple biogenic elements (e.g. Wilhelm et al. 2013), as opposed to the DTPA wash, which was developed specifically for extracellular Cu removal.   3.3.8 Oxygen consumption rates       Bacterial respiration was measured using a S1 Clark type oxygen electrode (Hansatech Instruments, Norfolk, England). Cultures were sampled in mid-exponential phase as 1 mL aliquots that were transferred to a clean O2 electrode chamber connected to a 19°C water bath. Oxygen consumption rates were derived as the slope of O2 consumption over time (10 min) in the chamber containing the bacterial culture. Prior to these measurements, the electrode was calibrated by filling the electrode chamber with O2-saturated synthetic seawater (SOW, bubbled with O2 for 1 h prior to experiment in a water bath at 19°C). Subsequently, the SOW in the electrode chamber was gently bubbled with N2 to record the O2-minimum signal. The electrode was controlled by a CB1D O2 electrode control box (Hansatech) and the data were collected and processed with the Labjackoxy software (Hansatech). For each biological replicate, two or more technical replicates were collected during exponential phase. The O2 consumption rates were normalized to bacterial cell abundance. Technical replicates were averaged and the error associated with the O2 71  measurements was propagated in all calculations of the cell normalized bacterial respiration (BRcell) and carbon normalized bacterial respiration (BRcarbon).  3.3.9 Bacterial carbon metabolism     Growth and respiration rates, as well as C content data, were used to derive different estimates of heterotrophic metabolism as previously described (del Giorgio and Cole 1998):                BP = Growth rate (d-1) x Cellular C (fmol C cell-1)                               (1)         BCD = BP (fmol C cell d-1) + BRcell (fmol C cell d-1)                           (2)         BGE = BP/BCD                                                                                     (3) where BP stands for bacterial productivity (fmol C cell-1 d-1), BCD for bacterial carbon demand (fmol C cell-1 d-1), BR for bacterial respiration (fmol C cell-1 d-1), and BGE for bacterial growth efficiency (ratio, unitless).      These metabolic rates were derived for Dokdonia sp. strain Dokd-P16, oceanic Pseudoalteromonas sp. strain PAlt-P26 and R. pomeroyi DSS-3 using measurements of three biological replicates. For these estimates, bacterial respiration rates were converted from the amount of oxygen consumed (fmol O2 cell-1 h-1)  to the amount of carbon respired (fmol C cell-1 h-1) using a respiratory quotient (RQ - ratio of O2 consumed to CO2 produced) of 1, as has been assumed in other studies (del Giorgio and Cole 1998; Reinthaler and Herndl 2005; Hörtnagl et al. 2011; Pradeep Ram et al. 2016).   3.3.10 Statistical analysis      The statistical analyses were conducted using an open source programming language R (R Core Team 2016). Data processing was done using package “dplyr” (Wickham 2016), while all figures were generated using “ggplot” (Wickham 2009) and “cowplot” (Wilke and Wickham 2016) packages in R. All the data presented were calculated as means ± standard error (S.E.). One-way Analysis of Variance (ANOVA) was used to determine the significance among different Cu treatments for individual bacterial strains with a significance cutoff of 0.05. Post-hoc analyses 72  were performed using a pairwise t-test with Bonferroni correction of α (at 0.05) when the variance was homogeneous using the R package “stats”. For data that displayed heterogeneity of variance, we applied Tukey’s honest significance difference, using “multcomp” package in R (version 1.4-6; Herberich et al., 2010).   3.4 Results 3.4.1 Growth rates and Cu content in response to changing Cu availability      We found diverse responses in bacterial growth rates and Cu quotas to changing Cu availability (Fig. 3.1). Both coastal and oceanic strains of Pseudoalteromonas sp. (PAlt-P2 and PAlt-P26, respectively), had the fastest growth rates, and minor reductions in growth rates in lowest Cu treatments (Cutot = 0.6 nmol L-1, μ/μmax of 0.72 and 0.84 for Pseudoalteromonas sp. PAlt-P2, and PAlt-P26, respectively, Table 3.3). These strains also had similar Cu quotas (Cu:P, mmol:mol) across all Cu treatments (Fig. 3.1G and H). Statistical analysis confirmed that neither their growth rates nor the Cu:P ratio were significantly affected by Cu concentration (Table 3.4). In contrast, while the growth rate of the Roseobacter member R. pomeroyi DSS-3 was not affected by changes in Cu availability (F= 2.07, p= 0.158, Table 3.4, Fig. 3.1B), its Cu quota (Cu:P, mmol: mol) varied significantly (F =20.25, p <0.001, Table 3.4, Fig 3.1 F), with a 2-fold reduction from the lowest Cu treatment to the highest Cu treatment (Cu:P =0.05 ± 0.003 vs 0.15±0.01, mmol: mol, Table 3.3). The growth rate of the Flavobacteriia member Dokdonia sp. strain Dokd-P16 was significantly affected by Cu availability (F= 78.47, p <0.001, Table 3.4), with reductions of up to ~ 80% from its maximal growth rate (0.192 μ/μmax) at the lowest Cu level (0.6 nmol Cu L-1, Table 3.3). In this strain, reductions in growth rates were accompanied by ~ 50% decrease in Cu quotas at the two lowest Cu levels (average 0.7 mmol Cu:mol P) compared to the two highest Cu replete treatments (average 0.13 mmol Cu:mol P, Fig 3.1 E, Table 3.3). Interestingly, similar decreases in Cu quota were observed at moderate Cu levels in the media (10 nmol Cu L-1) even though growth rate was unaffected (Fig 3.1 E, Table 3.3).  73  3.4.2 Macronutrient cellular composition in response to Cu       The major elemental compositions of bacterial strains at varying Cu levels were assessed as cell-normalized carbon (C), nitrogen (N), phosphorus (P) and sulfur (S) (Fig. 3.2; Appendix B, Table B.1). In general, cellular macronutrients did not vary significantly in response to Cu treatments (Fig. 3.2, ANOVA results, Appendix B, Table B.2). When averaged across all treatments, the oceanic Pseudoalteromonas sp. strain PAlt-P26 had the highest macronutrient quotas (cellular fmolar ratio of 27C: 6N: 0.5P: 0.2S), followed by the coastal Pseudoalteromonas sp. strain PAlt-P2 (22C: 6N: 0.5P: 0.2S), R. pomeroyi DSS-3 (16C: 4N: 0.2P: 0.15S) and Dokdonia sp. strain Dokd-P16 (10C: 2N: 0.2 P: 0.12S). The larger cellular macronutrient content of the Pseudoalteromonas sp. strains may reflect their larger cell size compared to R. pomeroyi DSS-3 and Dokd-P16. To further assess the effects of Cu on the elemental composition of bacteria, we determined the stoichiometric ratios of major elements (C:N & S:P, Fig. 3.2). We consider ratios only for those elements that were analyzed simultaneously (C and N, P and S) to minimize the error due to differences in analytical techniques. In all bacterial strains, the molar ratios of both C:N and S:P were unaffected by changing Cu availability (ANOVA results, Appendix B, Table B.2). The elemental ratios determined in this study are on par with those of native and cultured marine and freshwater bacteria (Fig. 3.2, Fagerbakke et al. 1996).  3.4.3 Regulation of bacterial carbon metabolism by Cu availability    Using growth rate, cellular C, and bacterial respiration data, we estimated various aspects of heterotrophic metabolism and examined its sensitivity to changing Cu availability in three strains: Dokd-P16, oceanic PAlt-P26, and R. pomeroyi DSS-3 (Fig. 3.3). Carbon metabolism of bacteria can be distinguished into anabolic (bacterial productivity, BP) and catabolic reactions (bacterial respiration, BR), which can be used to derive bacterial carbon demand (BCD; the carbon biomass that is required to sustain net bacterial metabolic needs) and bacterial growth efficiency (BGE; carbon assimilation efficiency into biomass) (del Giorgio and Cole 1998). Of all strains, only Dokd-P16 was found to significantly reduce metabolic rates (BR, BP, and BCD) and BGE in response to low Cu availability (One-way ANOVA, Table 3.4). Copper limitation (0.6 & 2 nmol L-1 Cu treatments) was associated with ~20% reduction in cellular respiration (BRcell), ~ 40% 74  reduction in productivity, and ~ 40 to 50% reduction in both BCD and BGE, compared to Cu replete conditions (25 & 50 nmol L-1 Cu, Fig. 3.3, Appendix B, Table B.3). Furthermore, BR and BCD were significantly reduced at moderate Cu levels (10 nmol Cu L-1) compared to the two highest Cu treatments (Fig. 3.3), yet, BGE estimates were similar across these treatments. In R. pomeroyi DSS-3, BR and BCD were also significantly affected by Cu availability (Table 3.4), but in contrast to Dokd-P16, these rates were substantially elevated under Cu limitation (~ 50% higher at 0.6 than 50 nmol L-1, p <0.001, Fig. 3.3, Appendix B, Table B.3). Interestingly, none of the carbon metabolism parameters we assessed for PAlt-P26 varied significantly in response to changing Cu availability (Table 3.4, Fig. 3.3). The BGE estimates in this study varied from 0.3 to 0.6 and are in the range of those reported for Vibrio harveyi  (<0.01-0.5 in natural seawater with glucose, Kirchman et al., 2003), and are typical of bacterial communities in eutrophic systems (as reviewed in del Giorgio and Cole 1998). Respiration rates obtained for PAlt-P26 (344 to 449 fmol O2 cell-1 d-1 range from all treatments, Appendix B, Table B.3) are higher than those of coastal and oceanic Alteromonas macleodii, also of the order Alteromonadales cultured in Aquil modified media with replete Fe (226 fmol O2 cell-1 d-1, Fourquez et al. 2014a).  3.5 Discussion 3.5.1 Bacterial growth and Cu content in response to Cu     Bacterial responses to Cu availability were taxonomically-distinct and may offer some insight into the ecology of prokaryotic groups represented by our strains. Copper appeared to be critical for maximal growth of Flavobacteriia member Dokdonia sp. Dokd-P16 but the mechanisms underlying this response have yet to be unraveled. When Cu conditions are considered in terms of the free divalent Cu (Cu2+), substantial growth reduction in Dokd-P16 (33% of μmax at 2 nmol L-1 Cu; 1.38 x 10-15 mol L-1 Cu2+) is induced at Cu levels that also limit several coastal phytoplankton species, including the diatoms Thalassiosira weissflogii and Chaeteoceros decipens, the coccolithophore Emiliana huxleyi, and the prymnesiophyte Phaeocystis cordata (1 x 10-15 mol L-1, Annett et al., 2008, Guo et al., 2012). Investigations of Cu limitation in non-photosynthetic marine prokaryotes have so far been limited to two groups occupying highly specific ecological niches, denitrifying bacteria and ammonia oxidizing archaea (Matsubara et al. 75  1982; Granger and Ward 2003; Moffett et al. 2012; Amin et al. 2013). The levels of Cu2+ that limit the growth of Dokd-P16 are one order of magnitude higher than those found to reduce the growth of some marine denitrifiers (<10-16 mol L-1, Moffett et al., 2012), but are two orders of magnitude lower than levels that limit the ammonia oxidizing archaeon (AOA) Nitrosopumilus maritimus SCM1 (10-13 mol L-1, Amin et al., 2013). These results suggest that the Flavobacteriia member Dokd-P16 may have Cu requirements that are higher than for denitrifying bacteria but lower than for AOA.      Furthermore, the levels of Cu2+ reported for surface waters at station P16 in the NE Pacific (Cu2+ = 2.5 x 10-15 mol L-1 (10 and 35 m), Semeniuk et al., 2016a), from where Dokd-P16 was isolated, fall within the range of levels in our study that resulted in growth rates of 74% and 33% of μmax in this strain (Cutot treatments of 10 nmol Cu L-1, Cu2+= 5.6 x 10-15 nmol L-1; and 2 nmol Cu L-1, Cu2+= 1.38 x 10-15 nmol L-1, respectively). Is it possible that Cu conditions in surface waters at station P16 may be limiting the activity of Dokdonia sp. strain Dokd-P16? Bacterioplankton growth rates in the subarctic NE Pacific are > 10 times lower (< 0.1 d-1, Kirchman et al. 1993) than the growth rates measured for Dokd-P16 in our study (~1‒9 d-1). This raises the question of whether our observations could be extrapolated to the field. However, community growth rates are an average of many taxa growing at various rates (Kirchman 2016) and are unlikely to reflect those of individual taxa (Lankiewicz et al. 2015). Furthermore, there is uncertainty about the growth rates of indigenous marine heterotrophic bacteria (see the discussion in Kirchman 2016) and growth rate measurements of individual taxa or phyla in situ are rare. However, some studies, find that fast-growing, copiotrophic (nutrient-loving) coastal bacteria affiliated with Alteromonadaceae (Ferrera et al. 2011) and Flavobacteriia (Yokokawa et al. 2004) can grow at rates as fast as ~ 5 d-1. Given our isolation method, we expect that in nature the growth rate of Dokd-P16 in nature would be fast, as that of a typical copiotroph, suggesting that it is possible that Cu levels at station P16 limit this bacterium.  Yet, what is the ecological significance of this? Copiotrophs are generally characterized by low numbers in nature (rare) but can peak in abundance during episodic events such as algal blooms. For example, rare Flavobacteriia taxa (Ulvivbacter and Formosa) dominated the early stages of an algal bloom in the North Sea (Teeling et al 2012). There is also increasing evidence that rare strains can have a significant impact on the ecosystem dynamics, which may be related to their high activity when conditions favor growth. For instance, a large, 76  fast-growing single strain of Alteromonas sp. AltSIO (Alteromonadaceae) was shown to consume the entire pool of dissolved organic carbon in a coastal ecosystem (Pedler et al. 2014). At present, little is known about the abundance and activity of the Dokdonia genus. However, this taxon could have a potential role in proteorhodopsin (PR)-mediated bacterial phototrophy, which is a significant process potentially impacting fluxes of C and energy in the ocean (Béjà et al. 2001), as many Dokdonia isolates can carry out this process (e.g. Gómez-Consarnau et al. 2007; Gonzalez et al. 2011; Kimura et al. 2011; Riedel et al. 2013; Bogachev et al. 2016; Kim et al. 2016). The abundance of bacteria affiliated with Dokdonia sp. was found to be low in a recent survey of Flavobacteriia populations in the North Atlantic Ocean (Gόmez-Pereira et al. 2010). However, to our knowledge, this is the only reporting on the distribution of Dokdonia sp. in ocean environment. Future metagenomic investigation in surface waters of the Line P transect will help elucidate the abundance and role of this taxon.     The growth response of Dokd-P16 contrasts that of the coastal and oceanic strains of Pseudoalteromonas sp. strain PAlt-P2 and PAlt-P26, and R.  pomeroyi DSS-3 which were not affected by Cu availability. This may indicate lower Cu requirements for these organisms. Indeed, in both Pseudoalteromonas sp. strains Cu:P ratios were ~ 50% lower than those of Cu replete Dokd-P16 suggesting a relatively lower Cu utilization (Fig. 3.1, Table 3.3). The ability of both Pseudoalteromonas sp. to maintain relatively invariant cellular Cu levels, despite up to 50-fold variation in external Cu concentrations, indicates that these organisms have excellent Cu homeostasis mechanisms.    In contrast, the similarity in Cu:P ratios in R. pomeroyi DSS-3 and Dokd-P16 across the range of Cu conditions does not support the hypothesis that R. pomeroyi DSS-3 has lower cellular Cu requirements, allowing it to maintain optimal growth under low Cu. In addition, the proteome of R.  pomeroyi DSS-3 features several Cu-containing enzymes (Table 3.5), suggesting a diverse use of Cu, although some of these proteins may be used only under certain conditions. The growth rate of this strain was not affected even when deprived of Cu, but its quota dropped drastically (by 60%). This suggests that R. pomeroyi DSS-3 reduces its Cu use under Cu-limitation and this strategy is sufficient to allow it to maintain maximum growth rates, as opposed to Dokd-P16.    One possibility for the lack of Cu limitation of Pseudoalteromonas spp. and R. pomeroyi DSS-3 could be their ability to acquire the organically-bound Cu in culture media. It is unclear if the 77  inorganic metals are the sole substrates for uptake in our bacteria, or if they could also access metals bound to EDTA. The work of Granger & Price (1998) suggests that inorganic Fe [Feʹ] was likely the primary bioavailable form of Fe in the EDTA-buffered Aquil media designed for marine heterotrophic bacteria. However, to our knowledge, bioavailability of other metals has not been evaluated in their study, nor in those examining metal composition of other bacteria (e.g. E. coli by Outten and Halloran 2001, thermophilic archaea by Cameron et al 2012, P. aeruginosa by Cunrath et al. 2016). In these studies, however, EDTA was not used to buffer metals. We expect that under Cu-replete conditions (10-50 nmol L-1) our strains would not need to access organically bound Cu and that Cuʹ would be the main uptake substrate, especially given its potential to induce toxicity as Cu concentrations increase. However, we cannot rule out that organically-bound Cu may be acquired when our bacteria are Cu-limited. Perhaps, this may be the reason why strains such as R. pomeroyi DSS-3 and Pseudoalteromonas spp. were not growth limited when Cu was in low supply, but a further study is required to confirm this hypothesis.      One of the major findings of this study was that, although there was some strain-specific variability in bacterial Cu quotas, the range of Cu:P ratios was narrow (0.03 to 0.15 mmol Cu:mol P), despite contrasting bacterial phylogenies and ecological provenance of our strains. Much of the understanding of bacterial Cu homeostasis derives from studies with model bacteria such as E. coli. These studies suggest that bacteria are highly sensitive to Cu toxicity and that their Cu trafficking and cellular inventories are under strict management via a variety of homeostasis mechanisms (Rensing and Grass 2003; Solioz et al. 2010; Dupont et al. 2011; reviewed in Bondarczuk and Piotrowska-seget 2013). Indeed, the cytoplasm of the gram-negative bacterium E. coli is almost completely devoid of Cu, and it is believed that Cu-containing proteins are confined to the periplasm or the cytoplasmic membrane (Tottey et al. 2005). The Cu-regulatory protein of E. coli that oversees Cu efflux (CueR) is activated at zeptomolar (10-21) Cu2+ levels (< one atom per cell), suggesting that this bacterium is extremely sensitive to free Cu2+, and operates the mechanisms involved in Cu detoxification even under Cu deprivation (Changela et al. 2003). Hence, there is a need to keep bacterial Cu at low level and within a narrow range to avoid Cu toxicity, which can cause oxidative stress and destruction of Fe-S clusters in proteins (reviewed in Dupont et al. 2011). Thus, it appears that the responses of marine heterotrophic bacteria to changes 78  in Cu availability we observed are consistent with the model of Cu regulation in gram-negative bacteria.     3.5.2 Comparison of prokaryotic and eukaryotic Cu quotas- insights into Cu utilization       Trace metal content in microorganisms reflects present trace metal availability (varying in time and space), their physiological requirements, as well as trace metal accessibility during their evolution (Quigg et al. 2003; Twining and Baines 2013). Therefore, metal content comparisons among different organisms can yield valuable insights into the trends in biological metal usage. To this end, we compared the Cu quotas of marine heterotrophic bacteria with those published for other prokaryotes, as well as eukaryotic phytoplankton (Fig. 3.4).     Cu quotas of marine heterotrophic bacteria agree well with those estimated for the model heterotroph E. coli (Outten et al. 2001; Cameron et al. 2012), despite differences in culturing conditions. The range of Cu quotas of our strains and E. coli (0.6‒5 μmol Cu: mol C) may, therefore, provide a good approximation of Cu requirements in aerobic heterotrophic prokaryotes. Although the quotas of cyanobacteria and eukaryotic phytoplankton vary more broadly, the overall trends across different microbial groups suggest that Cu content does not change markedly. The larger range of quota values in autotrophs may reflect a greater taxonomic variability in Cu requirements, wider range of Cu tolerances in the cell or differences in Cu availability in culture studies. To explore these factors, we selected datasets from studies using the EDTA-buffered growth medium Aquil and compared Cu quotas of marine heterotrophic bacteria, cyanobacteria and eukaryotic phytoplankton across a comparable range of bioavailable Cu (inorganic Cu [Cuʹ], Fig. 3.5). In general, heterotrophic bacteria maintain a well-controlled inventory of Cu across varying levels of Cuʹ, and their average Cu content agrees particularly well with that of eukaryotic phytoplankton and cyanobacteria under conditions of low Cu availability (-log[Cuʹ] < 13, Fig. 3.5). The similarities of Cu quotas under limiting conditions suggest that the minimum Cu requirements of these marine microbial groups may be comparable. In contrast, as Cu availability increases there is a larger variation in Cu quotas in eukaryotic phytoplankton and cyanobacteria compared to heterotrophs. The larger variation in phytoplankton (both eukaryotes and prokaryotes) may reflect more variability in their Cu requirements, as well as their ability to detoxify Cu 79  intracellularly. In eukaryotic algae, excess metals in the cell are sequestered by phytochelatins and metallothioneins (cysteine-rich peptides binding metals) and delivered to vacuoles, where these metals can be safely stored (Cobbett and Goldsbrough 2002). In contrast, bacteria detoxify metals mainly by efflux (see reviews of Nies 1999; Rensing and McDevitt 2013), suggesting a tight control over a range of cellular metal content, such as that of Cu.    Nevertheless, these preliminary observations challenge our present understanding of metal utilization in prokaryotes and eukaryotes. It has been hypothesized that trace metal availability during the evolution of various organisms influenced metal selection for biological usage (Williams and Fraústo da Silva 2003). Prokaryotes evolved in an anoxic ocean (Archean, ~ 2750 Myr) where Cu chemistry, as indicated by speciation models, was dominated by complexation with sulfide, rendering Cu less bioavailable (Saito et al. 2003). This contrasts with the evolutionary environment of eukaryotic organisms (Paleozoic ~1000 Myr to modern ocean ~ 540 to 120 Myr), which was characterized by higher Cu availability as a result of Earth’s oxygenation (Saito et al. 2003). These considerations have been used to explain the greater sensitivity of modern autotrophic prokaryotes to Cu toxicity (Brand et al. 1986; Saito et al. 2003). On this basis, we may expect Cu to be less important in prokaryotes than eukaryotes, but our preliminary analysis does not support this hypothesis given the similarity of Cu contents in both groups that we observed.   3.5.3 Composition of major elements in response to Cu     Microbial biosynthesis requires a balanced stoichiometry of carbon (C), nitrogen (N), and phosphorus (P) (Herbert 1976). Since bacterial biomass represents a major sink for these elements in the aquatic environment, understanding factors that influence bacterial stoichiometry is critical to predicting the fate of C, N, and P in the ocean. Sulfur is not naturally limiting in the ocean (seawater SO4 concentrations are ∼28 mmol L−1, Emerson and Hedges 2008), and rarely considered in stoichiometric studies. However, bacterial S content is of interest due to the involvement of S-containing compounds (e.g. glutathione) in Cu binding and trafficking in the cell (Helbig et al. 2008; Solioz et al. 2010). The elemental stoichiometry of bacteria is thought to be regulated within a relatively narrow range (Makino et al. 2003), with a C:N:P molar composition of ~ 50:10:1 (Goldman et al. 1987; Fagerbakke et al. 1996). However, observations of flexible 80  bacterial stoichiometries have challenged this paradigm (Scott et al. 2012; Godwin and Cotner 2015a). Bacterial stoichiometry may be influenced by growth rate and its phase (Fagerbakke et al. 1996; Makino et al. 2003), resource composition (Scott et al. 2012) and temperature (Cotner et al. 2006). The low bacterial C:P ratio observed under some conditions can be related to rapid growth rates, which elevates the requirement for P-rich ribosomal RNA (rRNA, representing up to 30% of microbial biomass) (Neidhardt et al. 1992; Makino et al. 2003; Chrzanowski and Grover 2008). Since reduced rRNA requirements are expected under conditions of slower growth (according to “growth rate hypothesis”, Elser et al. 2000), bacterial P content should scale proportionately, although microbial storage of P may cause deviations (eg. Elser et al. 2003). Our results do not support this hypothesis since, in our experiments with Dokdonia sp. strain Dokd-P16, cellular P content was not significantly affected despite a 4.4-fold variation in growth rate (Table 3.3, Appendix B, Table B.2).          In the present study, bacterial C:N and S:P ratios were essentially unchanged across various Cu growth conditions. Molar ratios of all elements normalized to P (C:N:S:P) and averaged across all Cu levels were 60:13:0.8:1 for Dokd-P16, 53:14:0.5:1 for the oceanic PAlt-P16, 41:16:0.6:1 for the coastal PAlt-P2, and 85:21:0.8:1 for R. pomeroyi DSS-3. These stoichiometric ratios are consistent with those published for a variety of native and cultured bacteria, as well as R. pomeroyi DSS-3 (Goldman et al. 1987; Fagerbakke et al. 1996; Chan et al. 2012). In agreement with our findings, constancy in C:N ratios was reported for marine heterotrophic bacteria grown under different Fe regimes (Tortell et al. 1996; Fourquez et al. 2014a). These invariant bacterial C:N ratios indicate that assimilatory pathways of C and N in bacteria remain highly interconnected under changing availability of bioactive metals, such as Fe and Cu.  3.5.4 Cu regulation of microbial carbon metabolism     Microbial growth and metabolism is dependent on the consumption of organic carbon, which at a cellular level is partitioned between biosynthetic (anabolic – bacterial production) and energy-yielding (catabolic – bacterial respiration) processes (del Giorgio and Cole 1998; Carlson et al. 2007). This partitioning is flexible (Teixeira de Mattos and Neijssel 1997), a strategy that allows bacteria to maximize growth depending on environmental conditions (Tempest and Neijssel 1992; 81  Russell and Cook 1995). One way to parameterize the relationship between catabolism and anabolism in a bacterial cell is to estimate bacterial growth efficiency (BGE- sometimes termed carbon use efficiency, CUE), which represents the ratio of C allocated towards bacterial biomass production relative to the total amount of C consumed to support the observed bacterial production [BGE = BP/(BP+BR)] (del Giorgio and Cole 1998; Rivkin and Legendre 2001). Bacterial growth efficiency, a well-known proxy for bacterial physiological state, can also be used to infer the ecological function of bacterial communities in aquatic ecosystems (del Giorgio and Cole 1998). In oceanic and coastal systems BGE ranges from 0.01 to 0.4 (being higher in eutrophic ecosystems, del Giorgio and Cole 1998), indicating that between 60 and 99% of the assimilated carbon is used for respiration. Higher BGE values are suggestive of more efficient C utilization (i.e. allocation towards bacterial growth) and relatively low amounts of C respired (as CO2). In contrast, a low BGE points to relatively large losses of C (as CO2) and inefficient energy metabolism, which has been linked to imbalanced substrate stoichiometry (Keiblinger et al. 2010), viral infection (Pradeep Ram et al. 2016) and high UV exposure (Hörtnagl et al. 2011). Iron deficiency has also been shown to reduce BGE in some marine heterotrophic bacteria (Tortell et al. 1996; Kirchman et al. 2003a) due to a less efficient Fe-rich respiratory systems (Tortell et al. 1996). In contrast, the role of other bioactive trace metals, such as Cu, in controlling BGE remains unexplored.    In our study, the largest Cu-limiting effects on bacterial C metabolism were found in Flavobacteriia member Dokdonia sp. strain Dokd-P16. In this bacterium, Cu starvation caused a severely reduced growth (82% and 67% of μmax at 0.6 and 2 nmol L-1 Cu levels, respectively, Table 3.3) as well as BCD, and slower BR and BP rates (Fig 3.3, Appendix B Table B.3). Slower BR rates may be partly related to the impairment of the respiratory system, which in aerobic bacteria is the major repository of cellular Cu (in the form of COX, Ridge et al. 2008).  On average, the reduction in BP (~ 40%) was greater than that in BR (~ 20%), indicating a higher flux of C for energy production than for biomass in Cu-limited Dokd-P16, which was confirmed by a reduction in BGE (by 50%). We also found a poor correlation between its growth and respiration rates (R2= 0.12, Appendix B, Fig. B.5), which has been similarly noted for bacteria growing under varying Fe availability, although with a higher statistical strength (R2= 0.45, Fourquez et al. 2014a). ATP formation (mainly via respiration) and growth may be poorly correlated because bacteria allocate energy to multiple processes that are independent of cell biomass formation, notably, overflow 82  metabolism, futile cycles and maintenance metabolism (reviewed by Russell and Cook 1995). The energetic requirements for processes such as cell maintenance (e.g. activation of nutrient uptake system) may become particularly important under constrained growth (Russell and Cook 1995), which may explain the trends we observed in BR and BP, and ultimately reduced efficiency of C use in Cu-limited Dokd-P16. Iron limitation led to a similar response, lowering BGE in 3 out of 5 marine heterotrophic bacteria (up to 70%, Tortell et al. 1996). Our results suggest that low Cu availability may potentially have implications for the fate of organic C transformed by Dokdonia sp. strain Dokd-P16 in marine waters.      In contrast to the flavobacterium, the growth rates of R. pomeroyi DSS-3 and PAlt-P26 were not constrained by Cu limitation (Table 3.3), but their BR and BCD increased substantially (R.  pomeroyi DSS-3) or remained relatively invariant (PAlt-P26, Fig. 3.3, Appendix B, Table B.3). Given that BCD provides an estimate of the C needed to support bacterial production (Carlson et al. 2007), its increase in R. pomeroyi DSS-3 under Cu limitation suggest that the energetic demand to support BP under these conditions is higher than when Cu is replete. Perhaps this reflects the energetic investment in mechanisms allowing R. pomeroyi DSS-3 to maintain constant BP even when Cu is scarce (Fig. 3.3). In this regard, our data suggested that this strain is lowering its Cu use under Cu-limitation (Fig. 3.1). In contrast, strain PAlt-P26 was not affected in either growth or Cu quota under varying Cu availability, which may explain why we did not observe any major modifications to the C metabolism in this strain.   3.5.5 Oceanographic implications     While microbial Cu toxicity has been investigated extensively, rarely has Cu limitation been explored in bacteria. Here, we characterized the effects of low Cu availability on the physiology and biochemistry of a diverse group of marine heterotrophic bacteria. We investigated strains from environmentally relevant microbial groups, including Flavobacteriia, marine Roseobacter clades, and Alteromonadales. Our study indicates that Cu is a critical micronutrient for optimal growth of our Flavobacteriia member Dokdonia sp. Dokd-P16. Indeed, the Cu levels in the medium ([Cutot] 2 nmol Cu L-1; [Cuʹ] = 19.8 fmol L-1) where the growth of this bacterium was substantially impaired, approximate those in surface waters of many oceanic regions, including the western 83  section of the Line P transect in the NE Pacific (station P26; [Cuʹ] = 18.7 fmol L-1 at 10 m; Semeniuk et al. 2016a). Therefore, the activity of Dokd-P16 may be limited in this oceanic area. Our results also indicate that Cu availability could act as a regulator of bacterial species composition. When Cu is low, bacteria such as R. pomeroyi DSS-3 (Roseobacteria) and Pseudoalteromonas sp. strains PAlt-P2 and PAlt-P26 (Alteromonadales) may outcompete those relying heavily on Cu for growth, such as Dokdonia sp. strain Dokd-P16 (Flavobacteriia).      There is a large uncertainty when extrapolating laboratory results of readily culturable strains to natural populations as they tend to be rare (which tend to be rare in nature, e.g. Eilers et al. 2000), although significant fractions of culturable bacteria have been observed in indigenous populations (e.g. Lebaron et al. 2001). How taxonomically representative is the response of our strains remains to be confirmed in future studies. However, it is promising that the members of Flavobacteriia responded preferentially to an increase in Cu availability in a recent field study in the Southern Ocean (Flavobacteria-Cytophagia cluster, Ramaiah et al., 2015), supporting our hypothesis of an important role of Cu and Flavobacteriia. Given the high abundance of Flavobacteriia in the surface waters in various regions (10 to 70 %, Glöckner et al. 1999; Eilers et al. 2000; Kirchman et al. 2003b; Abell and Bowman 2005; Gόmez-Pereira et al. 2010; Williams et al. 2013) as well as their significance in organic matter transformation (particularly the complex, high molecular weight compounds), evaluating the role of Cu nutrition in this group is of interest.     Considering the genus level, there is growing evidence of an important ecological role of the taxa that our strains are affiliated with. For instance, many Dokdonia sp. isolates carry out rhodopsin based bacterial phototrophy (Gómez-Consarnau et al. 2007; Kimura et al. 2011; Riedel et al. 2013; Bogachev et al. 2016), suggesting that this taxon may plan a role in cycling of carbon and energy in the ocean.  Ruegeria pomeroyi DSS-3 has a range of adaptations that have been implicated in various biogeochemical cycles (Green et al. 2012; e.g. Cunliffe 2013, 2016; Durham et al. 2015). Of particular interest is its ability to degrade the algal osmolite dimethylsulphoniopropionate (DMSP, Howard et al. 2006; Todd et al. 2012; Reisch et al. 2013), an important process in the marine sulfur cycle (see review of Reisch et al. 2011). This strain has been used as a model for examining the ecological role of the marine Roseobacter clade (Cunliffe 2016). Finally, transient peaks in the abundance of Pseudoalteromonas sp. were recently reported during a phytoplankton bloom in the North Sea (Lucas et al. 2015), suggesting that these bacteria 84  can play role in processing algal material during bloom events. Pseudoalteromonas spp. are also well known for the production of Fe-binding siderophores (Sijerčić and Price; Granger and Price 1999; Armstrong et al. 2004), hence may be influencing its speciation and availability in the ocean.        The cellular Cu content of heterotrophic bacteria determined in this study may offer some insights into the potential role of this important planktonic group in the biogeochemical cycling of Cu. Bacteria dominate living biomass in the ocean (Whitman et al. 1998), and may equate or even exceed phytoplankton biomass in euphotic zones, particularly in oligotrophic regions (Cho and Azam 1990; Ducklow 1999). Hence, the Cu associated with these microbes might represent a significant fraction of the biogenic Cu pool in the upper ocean. To explore this, we calculated biogenic Cu associated with bacterial and phytoplankton biomass, using mean Cu:C ratios from culture studies of both groups (Table 3.6). For these estimates, we employed euphotic zone integrated bacterial and phytoplankton biomass measurements made at station P26 in the NE Pacific in different seasons (Table 3.6). As exemplified using this model system, heterotrophic bacteria may account for ~ 4 to 50% of the total biogenic Cu (associated with planktonic organisms), suggesting that bacterial biomass can act as biological sink of Cu in surface waters.       Bacterially assimilated Cu is unlikely to be exported into the deep ocean as bacteria are not dense enough to sink on their own. Therefore, bacterial Cu uptake may contribute to metal retention in surface waters. This retained Cu could be recycled back into the dissolved phase via viral lysis of bacterial cells or be transferred to protozoans that graze upon them. However, we recognize that there are large uncertainties when extrapolating our limited dataset to the natural system, and our preliminary estimates should be taken with caution. These include an assumption that bacterial Cu:C ratios in our study are representative of the bacterioplankton community at station P26, which is likely composed of highly diverse bacteria with different Cu requirements. In addition, the Cu:C ratios of our copiotrophic bacteria may not be representative of those of bacterioplankton with oligotrophic life strategies (nutrient-poor adapted), such as SAR11, a significant component of prokaryotic community in the NE Pacific surface waters (Wright 2013). However, copiotrophic taxa, owing to their fast-growth rates, could have at times a disproportional effect on the amount of the dissolved Cu assimilated into bacterial biomass. For instance, if we use a community growth rate of 0.1 d-1 that is typical of oligotrophic bacteria (Kirchman, 2016) and an average estimate of Cu:C in bacterial biomass for the summer (Table 3), bacteria can assimilate 85  0.3 pmol Cu L-1 d-1. In contrast, using rates of 1‒5 d-1 as reported for some copiotrophic taxa (Yokokawa et al. 2004; Ferrera et al. 2011; Kirchman 2016) (Kirchman, 2016) yields an estimate of 2.9‒15 pmol Cu L-1 d-1, whereas phytoplankton account for an uptake of 3 pmol Cu L-1 d-1(assuming the typical phytoplankton community growth rate of 1 d-1). While some uncertainties exist, our observations hint that marine heterotrophic bacteria, just like phytoplankton, could play a role in the oceanic biogeochemical Cu cycle.   86  Table 3.1: Phylogenetic affiliation of gram-negative bacterial strains isolated from 40 m at various stations along Line P (June 2012 cruise). Information on the isolation location and the phylogeny of the model organism Ruegeria pomeroyi DSS-3 (isolated in June 1996; ATCC culture collection) is also included. Dissolved Cu (dCu), pCu and Cuʹ, represent ranges of values in samples from 10-40 m at each station determined by Semeniuk et al. (2016a).  Isolated from  Co-ordinates  Habitat   [dCu] (nmol L-1)  pCu (-log[Cu2+])  [Cuʹ] (fmol L-1)  Phylum  Class  Order  Family  Relative in Silva  Shared sequence homology (%)  Coastal Georgia  31.98 N 81.022 W  Coastal  nd  nd  nd  Proteobacteria  α  Rhodobacterales  Roseobacteraceae         Ruegeria pomeroyi DSS-3 (ATCC collection)  n/a NE Pacific  Stn P2  48.6° N 126° W  Coastal  nd  nd  nd  Proteobacteria  γ  Alteromonadales  Pseudoalteromondaceae   Pseudoalteromonas sp. BSw20080,  99.79 NE Pacific  Stn P4  48.66° N 126.66° W  Coastal  1.8-2.4  14.7-18.8  44-66  Proteobacteria  γ  Alteromonadales  Pseudoalteromondaceae   Pseudoalteromonas sp. (148Z-14, 146Z1-2, BSw10087)  99.65 NE Pacific  Stn P16  49.28° N 134.66° W  Oceanic  1.7-1.9  14.6  59-60  Proteobacteria  γ  Alteromonadales  Pseudoalteromondaceae   Pseudoalteromonas 148Z-14  99.38 NE Pacific  Stn P16  49.28° N 134.66° W  Oceanic  1.7-1.9  14.6  59-60  Bacteroidetes  n/a  Flavobacteriia  Flavobacteriaceae  Dokdonia sp. 4H-3-7-5 Dokdonia diaphoros   99.29 99.28 NE Pacific  Stn P26  49.99° N 144.99° W  Oceanic  2.1-2.2  14.9-15  19-60  Proteobacteria  γ  Alteromonadales  Pseudoalteromondaceae   Pseudoalteromonas sp.   (148Z-14, 146Z1-2, BSw10002)  99.72 nd (not determined), n/a (not applicable) 87         Table 3.2: Total dissolved copper (Cutot), inorganic Cu (Cuʹ) and free Cu ion concentrations (pCu, -log[Cu2+]) in bacterial growth media.  Calculations were performed with Visual MINTEQ version 3.1 (software is available from https://vminteq.lwr.kth.se/) excluding the organic substrate additions from calculations. The dissolved Cu reported in the lowest Cu treatment (without Cu addition) is that measured in the chelexed SOW by FIA-CL (Semeniuk, 2014), and indicates media background Cu contamination.                                                                                                                                                      1 from background Cu contamination         Treatment [Cutot] (nmol L-1) [Cuʹ] (fmol L-1) [pCu] (-log[Cu2+])  Low (no addition)  0.61  4.6  15.49 Low 1.92 19.8 14.86 Replete 10 81.4 14.25 High 25 204.4 13.85 High 50 388.9 13.57     88     Table 3.3: Specific growth rates (d-1), relative growth rates (μ/μmax) and phosphorus normalized Cu quotas (Cu:P, mmol:mol) of bacterial strains at various levels of Cu in the growth media (nmol L-1).  Strain Cutot (nmol L-1) n Growth rate (d-1)   μ/μmax n Cu:P (mmol:mol) Dokdonia sp.  0.6 6 1.61 ± 0.3 0.18 4 0.06 ± 0.02 strain Dokd-P16 2 11 2.83 ± 0.1 0.33 6 0.08 ± 0.01  10 7 6.48 ± 0.3 0.74 7 0.08 ± 0.01  25 9 6.94 ± 0.2 0.79 6 0.14 ± 0.01  50 10 8.72 ± 0.8 1 3 0.12 ± 0.003 Pseudoalteromonas sp. 0.6 4 9.00 ± 0.5 0.72 4 0.03 ± 0.01 strain PAlt-P2 (coastal) 2 nd nd nd  nd  10 5 12.4 ± 0.9 1 3 0.04 ± 0.01  25 7 11.3 ± 0.4 0.92 3 0.03 ± 0.01  50 nd nd nd  nd Pseudoalteromonas sp. 0.6 20 15.2 ± 0.4 0.84 4 0.06 ± 0.01 strain PAlt-P26 (oceanic) 2 6 18.2 ± 0.7 1 4 0.05 ± 0.01  10 33 18.0 ± 0.4 1 3 0.05 ± 0.02  25 3 17.1 ± 0.4 0.95  nd  50 6 15.2 ± 0.9 0.84 3 0.03 ± 0.01 R. pomeroyi DSS-3 0.6 11 4.68 ± 0.4 0.85 3 0.05 ± 0.003  2 4 5.47 ± 0.3 1 4 0.12 ± 0.01  10 9 4.42 ± 0.3 0.80 3 0.15 ± 0.01  25 3 3.94 ± 0.3 0.72 nd nd  50 9 5.47 ± 0.3 1 3 0.15 ± 0.01         nd - not determined  89   Table 3.4: One-way ANOVA values for the effect of Cu concentration in the media on various metabolic variables of the bacterial strains. Cell normalized bacterial respiration (BRcell), carbon normalized respiration (BRcarb), bacterial productivity (BP), bacterial carbon demand (BCD), bacterial growth efficiency (BGE). Statistically significant effects are shown in bold. Cellular C, N, P, S as well as C:N and S:P were tested with no significance (results of ANOVA can be found in Appendix B, Table B.2).                                                                          ANOVA results  Dokd-P16 PAlt-P2 PAlt-P26 R. pomeroyi DSS-3 Variable F statistic p-value F statistic p-value F statistic p-value F statistic p-value Growth rate  79 >0.001 2.08 0.170 0.25 0.612 2.1 0.158 Cu quota (Cu:P) 7.1 >0.001 0.84 0.467 0.74 0.698 20 >0.001 BRcell 14 >0.001 nd nd 0.97 0.452 7.7 0.022 BRcarb 4.1 0.032 nd nd 0.74 0.554 5.6 0.043 BP 17 >0.001 nd nd 0.32 0.805 0.3 0.756 BCD 46 >0.001 nd nd 0.25 0.857 11.9 0.008 BGE 9.0 0.002 nd nd 1.11 0.399 1.98 0.218     nd- not determined90   Table 3.5: List of copper containing proteins identified in the proteome of R. pomeroyi DSS-3 obtained from the NCBI database (https://www.ncbi.nlm.nih.gov/).  Protein description Accession no Functional role Heme/copper-type cytochrome/quinol oxidase  Subunit I  Subunit I  Subunit II  Subunit III   WP_011241964  WP_011241963 WP_011048769 WP_044029533  Transmembrane protein in the respiratory chain responsible for dioxygen reduction Nitric oxide reductase WP_011241995            Transmembrane protein carrying out nitric oxide (NO) reduction in denitrification Plastocyanin/azurin family  copper-binding protein  WP_011046018  Proteins in this family are involved in electron shuttling during deamination processes as well as denitrification Polyphenol oxidase WP_011046655  Periplasmic protein from the family of multi-copper oxidases, involved in phenol and diamine oxidation  Copper oxidase WP_011242145    Uncharacterized Cu protein with electron carrier activity  Multicopper-oxidases (MCO)   WP_044029625 WP_011047099 WP_011046443 Unknown functions. Proteins in the MCO family facilitate oxidation of a wide range of substrates.    91   Table 3.6: Calculations of biogenic Cu concentration (pmol L-1) for marine bacteria and phytoplankton (algae and cyanobacteria) in the seasonal euphotic zone (winter, spring, summer, Sherry et al. 1999)  at station P26 in the subarctic NE Pacific. Estimates were performed using average Cu:C quotas of cultured phytoplankton from literature and the heterotrophic bacteria in this study (Fig. 3.4). Bacterial biomass to phytoplankton biomass ratio (BB:PB) and bacterial biomass were obtained from Sherry et al. 1999, with values for two different years (summer: 1995 & 1996, winter and spring: 1996 & 1997).    Heterotrophic bacteria Phytoplankton      (1.36 μmol Cu:mol C)  (2.4 μmol Cu:mol C)     Season BB:PB Biomass (μmol C L-1) Biogenic Cu (pmol L-1) Biomass (μmol C L-1) Biogenic Cu (pmol L-1) Total (pmol L-1) Bacteria % of total Phytoplankton % of total Winter 0.08 1.1 1.5 14 23 34 4 96  0.64 0.8 1.0 1.2 2.0 3.8 27 73 Spring 0.87 2.1 2.8 1.1 2.0 5.6 51 49  1.82 1.8 2.4 2.0 3.5 7.2 33 67 Summer 1.29 2.3 3.1 1.7 3.0 7.2 42 58  1.49           2.0 2.7 1.3 2.3 5.9 46 54 92   Figure 3.1: Bacterial growth rates and Cu content at different levels of Cu (nmol L-1). Specific growth rates (d-1) and Cu:P ratios (mmol:mol) of Dokdonia sp. strain Dokd-P16 (A, E), Ruegeria pomeroyi DSS-3 (B, F), coastal Pseudoalteromonas sp. strain PAlt-P2 (C, G), oceanic Pseudoalteromonas sp. PAlt-P26 (D,H).  Data are the mean (± SE) and gray circles superimposed over the bars are the data points. Within a single graph, values with the same letter above were not statistically significant (pairwise t.test with Bonferroni correction, p < 0.05). All data are presented in Table 3.3 and the ANOVA results are presented in Table 3.4. Note x-axis is not linear.93            Figure 3.2: Cellular carbon (fmol C cell-1), nitrogen (fmol C cell-1), phosphorus (fmol P cell-1), sulfur (fmol S cell-1), molar ratios of C:N and S:P in Dokdonia sp. strain Dokd-P16, coastal Pseudoalteromonas sp. strain PAlt-P2, oceanic Pseudoalteromonas sp, strain PAlt-P26 and R. pomeroyi DSS-3 at different levels of Cu (nmol L-1) in the growth media. Non-filled circles represent means ± SE and the superimposed gray circles are the data. The horizontal dashed lines indicate the average C:N and S:P for native and cultured bacteria reported in a study by Fagerbakke et al., 1996. The data used to produce the figures can be found in Appendix B, Table B.1, and the ANOVA results for the effect of Cu on major elemental composition in the Appendix B, Table B.2. Note x-axis is not linear. 94        Figure 3.3: Carbon metabolism of marine heterotrophic bacteria at different levels of Cu (nmol L-1) in the culture media. Cell normalized respiration (BRcell, fmol O2 cell-1), bacterial Productivity (BP, fmol C d-1), bacterial carbon demand (BCD, fmol C d-1) and bacterial growth efficiency (BGE, unitless) of Dokdonia sp. strain Dokd-P16, Ruegeria pomeroyi DSS-3, Pseudoalteromonas sp. strain PAlt-P26. Points with error bars are mean (± SE) for BRcarb and BP, and propagated errors for BCD and BGE (as described in Methods Section 3.3.9); open circles represent the individual data. Significant effects (p <0.05) are shown with different letters for Dokd-P16 and R. pomeroyi DSS-3 (pairwise t.test with Bonferroni correction of α at 0.05). The data used to produce the figure can be found in the Appendix B, Table B.3. Note x-axis is not linear.  95                                                    Figure 3.4: Comparison of Cu:C ratios (μmol: mol) of heterotrophic bacteria from this study (data from all Cu treatments) with literature values for other prokaryotes and phytoplankton. E. coli  (Outten and Halloran 2001; Cameron et al. 2012) and Archaea values (Cameron et al. 2012) were estimated assuming 50% C by dry mass. Cyanobacteria ratios include data from include diazotrophic and non-diazotrophic organisms from culture studies and single cells from field (Quigg et al. 2010; Fe-replete treatments in Guo et al. 2012b; Nuester et al. 2012; Walve et al. 2014; Fe-replete treatments in Cunningham and John 2017). Phytoplankton ratios (Sunda and Huntsman 1995; Ho et al. 2003; Fe-replete treatments in Annett et al. 2008; Fe-replete treatments in Guo et al. 2012b) reported quotas as Cu:P, we used C:P ratios reported in those studies to estimate Cu:C. Note that Cu conditions under which Cu contents were obtained are variable. For clarity, errors were not included and an outlier value of 28 μmol Cu:mol C in the phytoplankton dataset is not shown in the plot.                                                      96                   Figure 3.5: Comparison of Cu quotas (Cu:C, μmol:mol) of cultured marine phytoplankton (A, algae and cyanobacteria) and marine heterotrophic bacteria from this study (B). Phytoplankton quotas are from for cultures grown in Aquil media (Sunda and Huntsman 1995; Ho et al. 2003; Annett et al. 2008; Quigg et al. 2010; Guo et al. 2012b). Data from Ho et al., 2003 and Quigg et al., 2010 were converted from Cu:P to Cu:C using the C:P ratios reported for each species in both studies. Colors in all figures represent different taxa. Dashed lines in both plots show the mean quota averaged across all Cu levels, and dotted line in phytoplankton plot shows the median (for phytoplankton). The shaded region represents the range of Cu:C remineralization ratios obtained by converting Cu:P ratios (regressions for upper waters, < 800 m) reported for different oceans assuming the 106C:1P stoichiometry (Fe-replete waters in Annett et al. 2008; reviewed by Twining and Baines 2013). The log [Cuʹ] is based on data of Semeniuk et al. (2016a) and Mofett (1995). For clarity, errors are not shown and the figure with phytoplankton data does not display an outlier quota of 28 μmol Cu:mol C (Glaucocytophyceae member, Quigg et al., 2010), which was also excluded from the calculation of mean and median Cu:C ratios.  97  Chapter 4:  Bioactive metal composition of marine heterotrophic bacteria and its response to changing Cu availability  4.1 Summary    The trace metal requirements of marine heterotrophic bacteria are largely unknown. Here, we present data on the Fe, Zn, Mn, Cu and Co content of four taxonomically distinct marine heterotrophic bacteria (marine Roseobacter clade member Ruegeria pomeroyi DSS-3, Flavobacteriia member Dokdonia sp. Dokd-P16, Alteromonadales members Pseudoalteromonas sp. coastal strain PAlt-P2 and oceanic strain PAlt-P26). Additionally, we examined the variation in bacterial metal content as a function of changes in non-toxic Cu concentrations. In Dokd-P16 and R. pomeroyi DSS-3, metal contents decreased in the order of Fe>Zn>Mn>Cu>Co, while in the Pseudoalteromonas spp., we found a pattern of Fe>Zn>Mn≈Cu>Co. All strains had similar Fe, Zn, Cu and Co content, but both Pseudoalteromonas sp. were depleted in Mn relative to other strains. Metal abundance profiles data and Cu-induced physiological responses indicate that Pseudoalteromonas spp. have low Mn and Cu requirements. We also performed a preliminary comparative assessment of the elemental stoichiometry of marine heterotrophic bacteria [(P0.018N4.2S0.01C1)1000Fe42Zn15Mn5.3Cu1.3Co0.28] and eukaryotic phytoplankton [(P0.010N8.8S0.01C1)1000Fe78Zn6.4Mn29Cu1.72Co1.2] grown under similar conditions of trace metal availability. Average elemental signatures and the variation in metal contents suggests that both groups contain relatively similar amounts of Fe and Cu, while bacteria appear to be richer in Zn but depleted in Mn and Co compared to phytoplankton. These trends do not fit with the evolutionary context of metal use by these two groups (i.e. higher Fe, Mn, Co and lower Zn and Cu utilization in prokaryotes relative to eukaryotes). Here, we also demonstrate that Cu availability can affect bacterial accumulation of metals. In both Pseudoalteromonas sp., growth rates and metal contents were largely unaffected by Cu, except for the increased Zn quota of the oceanic strain PAlt-P26 under low Cu. In contrast, the reduced Cu quotas in R. pomeroyi DSS-3 and Dokd-P16 98  under Cu limitation were associated with elevated Fe and Mn quotas, while growth was only limited in the former strain. We speculate that these responses may reflect an increased requirement for Fe- and Mn-containing metalloproteins under Cu limitation, or non-specific metal uptake. Furthermore, we found a reduction in Co quota in R. pomeroyi DSS-3 in response to reduced Cu availability, which was correlated to changes in Cu quota indicating a potential link between Cu and Co homeostasis in this strain. Collectively, our results provide insights into trace metal requirements of marine heterotrophic bacteria, and thus their role in trace metal biogeochemical cycling, and potential ecological interactions between major microbial groups in the ocean  4.2 Introduction     Bioactive metals such as Fe, Zn, Mn, Cu, and Co are indispensable to many biological pathways in the cell. Their functions fall into the following three categories: enzymatic catalysis, the transfer of electrons, and the structural stabilization of molecules. It is believed that metals are found in approximately 30% of the cell’s proteins, highlighting their biological significance (Tainer et al. 1992). Metals are employed in different ways by the prokaryotic cell. Iron is needed to mediate tricarboxylic acid (TCA) cycle, electron transfer in oxidative phosphorylation, nitrogen fixation, methanogenesis, and DNA synthesis (Andrews et al. 2003). Zinc is employed in hydrolytic enzymes responsible for cleavage of amino acids and phosphate groups from compounds (peptidases and phosphatases, respectively), and in polymerases involved in DNA and RNA binding and synthesis  (Lipscomb and Sträter 1996). Manganese enzymes play a wide range of roles, including detoxification of superoxide radicals and hydrogen peroxide, dephosphorylation of cellular proteins, glucogenesis and regulation of RNA synthesis (Kehres and Maguire 2003). Copper functions primarily in electron transfer reactions of the respiratory chain, but also has roles in oxidative stress reduction, amine metabolism, sporulation and pigmentation (Ridge et al. 2008; Argüello et al. 2013 and references herein). Cobalt forms the cobalamin cofactor of the essential vitamin B12, which in prokaryotes is required for enzymes involved in fermentative pathways, DNA synthesis and repair (adenosylcobalamin-dependent isomerases), amino acid metabolism, one carbon metabolism (methylcobalamin-dependent methyltransferases) and organohalide 99  respiration (B12-dependent reductive dehalogenase, Banerjee and Ragsdale 2003 and references herein).     There is a variation in the relative usage of bioactive metals in biology, which gives rise to an elemental blueprint known as the metallome. As originally proposed by Williams (2001),  the metallome contains valuable information about an organism, just like its genome (DNA) is the blueprint of all molecules required to carry out cellular processes and the proteome is the repertoire of individual proteins expressed by an organism at a given time. Much can be learned from metallome studies about metal utilization, which can be approached by analyzing metal concentrations of the whole cell or its cellular compartments, and the content and structure of metalloproteins (Mounicou et al. 2009). The latter approach has been invaluable to our understanding of the general principles of metal use across the three domains of life, and have helped unravel relationships between evolutionary environment and metal selection (Andreini et al. 2006; Dupont et al. 2006; Ridge et al. 2008; Zhang et al. 2009; Zhang and Gladyshev 2010; Decaria et al. 2011). Characterization of metal bio-signatures of marine phytoplankton has aided in our understanding of the links between ocean biogeochemistry and their trace metal physiology (Ho et al. 2003; Twining and Baines 2013; Twining et al. 2015). Variables influencing phytoplankton metal contents have also been explored, with phylogeny (Ho et al. 2003; Quigg et al. 2003), Fe availability (Wilhelm et al. 2013; Cunningham and John 2017) and irradiance (Finkel et al. 2006), as some examples. Furthermore, analyses of metal contents have also provided insights into the unique metal requirements of microbial groups such as hypothermophilic archaea (e.g. for Ni, Co, W in M. jannaschii and P. furiosis, Cameron et al. 2012) and marine diazotrophic cyanobacteria (e.g. for Fe, Mo and Ni in Trichodesmium, Nuester et al. 2012).     Marine bacteria, archaea, and phytoplankton collectively shape the oceanic cycles of elements and affect the fluxes of climate-active gases such as CO2 and N2O (Arrigo 2005; Kirchman 2008), with many of these processes being metal-dependent (e.g. photosynthesis, respiration, denitrification, N2 fixation). However, in contrast to the efforts to establish the elemental signature of cultured and field marine phytoplankton, the trace metal stoichiometry of marine prokaryotes remains poorly characterized, particularly that of heterotrophic bacteria. Although some data are available on the Fe content in cultured heterotrophic bacteria (Tortell et al. 1996; Granger and Price 1999; Fourquez et al. 2014a), there is limited information on the 100  cellular quotas of other essential metals. Vogel and Fisher (2010) examined adsorption and accumulation of bioactive metals Fe, Zn and Mn (and non-essential metals Cd, Cs, and Am) in various marine bacterioplankton in coastal seawater. However, essential metals were not measured consistently in all strains examined, and a complete set of elements (Fe, Zn, Mn) is only available for one strain (Roseobacter litoralis). Therefore, there are uncertainties regarding how these metals vary in marine heterotrophic bacteria, while the contents of other elements such as Cu and Co (that we examined here) are largely unknown. In addition, there are no data on the elemental stoichiometry of marine heterotrophic bacteria cultured under similar trace metal availabilities as marine phytoplankton, which limits our ability to assess the potential differences in their trace metal requirements. Such information would, however, aid in the understanding of the ecological interactions between these two important planktonic groups. For instance, evidence suggests that some marine heterotrophic prokaryotes may enhance availability of trace elements to phytoplankton, such as Co-containing vitamin B12 (Durham et al. 2015) or Fe (Amin et al. 2009), or may directly compete with phytoplankton for poorly available Fe (Tortell et al. 1999; Fourquez et al. 2014b). Bacteria-phytoplankton interactions involving other metals, besides Co and Fe, are likely to exist; however, our ability to predict those is constrained by the limited information on the relative metal utilization by these two groups.      Studies with phytoplankton and cyanobacteria provide some examples of how changes in the availability of one essential metal may influence the stoichiometry of the other metals. For instance, Fe starvation was found to elevate Cu, Zn and Mn quotas in the coral endosymbiont Symbiodinium kawaguii (Rodriguez et al. 2016), and Cd, Co and Mn quotas in cyanobacteria Synechoccocus and Prochlorococcus (Cunningham and John 2017). In both cases, it was hypothesized that these changes may reflect an increased physiological requirement for these elements to compensate for low Fe, or non-specific metal uptake via divalent metal transporters. Fe-limitation also elevates Cu quotas in some diatoms, reflecting an increased demand for Cu by the high-affinity Fe uptake system (Annett et al. 2008; Guo et al. 2012b). Macronutrient availability can also influence phytoplankton metal stoichiometry, as has been observed in two species of freshwater algae under phosphorus (P) limitation (Ji and Sherrell 2008). In Chlorella sp. UTCC522, for example, low P availability leads to an enrichment in Co, Zn, and Cd in the cell. This was linked to an increased requirement for alkaline phosphatase, which facilitates acquisition 101  of organic phosphorous and can use these metals as cofactors (Ji and Sherrell 2008). While much focus has been given to phototrophs, less is known of how limitation by essential elements, such as Cu, may influence metal stoichiometry in non-photosynthetic prokaryotes, particularly heterotrophic bacteria.     The goal of this study was to examine the metal stoichiometry of marine heterotrophic bacteria. For this analysis, we used three bacterial strains isolated from coastal and open ocean surface waters in the Northeast Pacific, as well as a model bacterium from culture collection Ruegeria pomeroyi DSS-3 (ATCC, strain DMS 15171). Our isolates were phylogenetically characterized in a previous study (Chapter 3), and include Dokdonia sp. strain Dokd-P16 (oceanic) a member of the Flavobacteriia class within the phylum of Bacteroidetes, and Pseudoalteromonas sp. strain PAlt-P2 (coastal) and PAlt-P26 (oceanic) of the order Alteromonadales in the Gammaproteobacteria class. In addition, R. pomeroyi DSS-3 has been used as a reference strain for the Marine Roseobacter Clade (e.g. Green et al. 2012; Cunliffe 2016). We hypothesized that our strains would display distinct metal signatures due to their unique phylogeny and/or ecology. In addition, we sought to understand how the environmentally relevant variations in Cu availability affect bacterial metal content.  4.3 Methodologies     Bacterial trace metal content data reported here (Fe, Zn, Mn, Cu, and Co) were collected during experiments described in Chapter 3. The culture conditions, experimental design, and instrumental analysis are identical to those described in Sections 3.3 of that chapter. The composition of metals in culture media can be found in Table 4.1. Here, metals normalized to cell number or biomass proxies’ C and P (major cellular constituents) are used as appropriate. Metal normalization to P (Me:P, mmol:mol) is preferred in plankton stoichiometry studies because P values are obtained simultaneously with trace metal values during ICP-MS analysis (eg. Ho et al. 2003; Finkel et al. 2006). However, marine microorganisms may vary their P content in response to changing environmental conditions (Elser et al. 2003), thus influencing the metal to phosphorus stoichiometry. Although varying stoichiometry of the organic substrate (C:P ratio) is known to affect bacterial phosphorus content (Godwin and Cotner 2015a), much less is known about the 102  effects of trace metal availability in this context.  In Chapter 3, we found that intracellular P did not vary significantly in a bacterial strain under conditions of varying Cu availability. Hence, we will focus on P-normalized metals (Me:P) to examine the effects of changing Cu availability on metal contents of in a bacterial strain. Phosphorus content can, however, vary between organisms, and here we found that R. pomeroyi DSS-3 was deplete in P compared to other strains (molar C:P of R. pomeroyi DSS-3 was 85C:1P; 60C:1P for Dokd-P16; 53C:1P for oceanic PAlt-P16; and 41:1 for the coastal PAlt-P2). We, therefore, used C-normalized metals (Me:C, μmol: mol) to examine the metal signatures of different strains, and for a comparative assessment of metal content in our bacteria with respect to values for microorganisms found in the literature. We discuss the use of P versus C-normalization of metal concentrations in more detail in section 4.3. The carbon normalized and cell-normalized metal contents of the four bacterial strains investigated are available in the appendix (Appendix C, Table C.1, and C.2, respectively). Carbon-normalized values were obtained by dividing cellular metal concentration (metals normalized to cell abundance in the bacterial pellet used for ICP-MS analysis) by cellular carbon concentration obtained during the same experiment (measured by CNH analysis, Chapter 3, Section 3.3.5).  4.4 Results       Bacterial metal quotas (Cu:P, Fe:P, Mn:P, Zn:P, Co:P, [mmol Me:mol P]) were examined at total copper (Cutot) concentrations in the growth media, ranging from 0.6 to 50 nmol L-1 (Fig. 4.1). The growth rate and Cu content of the coastal and oceanic Pseudoalteromonas sp. strains (PAlt-P2 and PAlt-P26, respectively) were unaffected by changing Cu availability (as discussed in more detail in Chapter 3), as were the quotas of other metals, which remained mostly constant in these strains (Fig. 4.1, one-way ANOVA, Table 4.3). An exception was the Zn quota of the oceanic PAlt-P26, which was significantly lower (0.4 ± 0.1 mmol Zn: mol P) under high Cu availability (50 nmol L-1) compared to lower Cu treatments (1 to 2.1 mmol Zn:mol P at 0.6 to 10 nmol L-1 Cu, Table 4.2). Post-hoc analysis confirmed that at this highest Cu treatment (50 nmol L-1), the Zn quota of PAlt-P26 was significantly lower than at 2 and 10 nmol L-1 Cu (Fig. 4.1).      The Roseobacteria member, R. pomeroyi DSS-3 maintained constant growth rates across a range of Cu availabilities but showed some changes in C metabolic rates and Cu content (Chapter 103  3).  In addition to these responses, we found that the Mn, Fe and Co quotas of R. pomeroyi DSS-3 were also affected by Cu availability (Fig. 4.1, one-way ANOVA, Table 4.3). Manganese and Fe quotas were both significantly elevated at a moderately low Cu treatment of 2 nmol L-1, relative to Cu-replete treatments of 10 and 50 nmol L-1 (Fig 4.1), although the increase in Fe was larger than that of Mn (on average a factor of 2.2 and 1.3, respectively, Table 4.2). Further decreases in Cu availability (to 0.6 nmol L-1) caused both Mn and Fe quotas to drop, although this was most pronounced for Fe (Fig. 4.1). Interestingly, Co quotas of R. pomeroyi DSS-3 decreased significantly as Cu levels in the growth media were reduced from 10 to 0.6 nmol L-1 (Fig. 4.1). The overall trend in Co quota with decreasing Cu concentrations mimicked that of Cu quotas (Fig. 4.1), and the quotas of these two metals were found to covary significantly in R. pomeroyi DSS-3 (R2=0.67, p=0.0003, Fig. 4.2C).      In contrast to the other strains, the Flavobacteriia member Dokdonia sp. strain Dokd-P16 is highly dependent on Cu for its growth and metabolism (Chapter 3). In response to low Cu, Dokd-P16 reduced its Cu quota, and associated with this response were changes in Mn and Fe contents (Fig. 4.1, One-Way ANOVA, Table 4.3). We found that the Mn quota was significantly higher (by 22 to 39%) at lower Cu treatments (2 and 10 nmol L-1) compared to the highest Cu treatment (50 nmol L-1, Fig. 4.1, One-Way ANOVA, Table 4.3). On average, the iron quota of Dokd-P16 doubled on going from 50 nmol L-1 to 0.6 nmol L-1 Cu in growth media, and the effect of Cu availability on its Fe quota was at the significance cutoff (p=0.051, One-Way ANOVA, Table 4.3). On further analysis, we found a significant inverse correlation between the growth rate of Dokd-P16 and its Fe quota (R2=0.24, p=0.006; Fig. 4.2B), suggesting an elevated requirement for Fe when the growth of this strain is limited under low Cu. While a correlation was observed for Fe quotas and Cu levels in the growth media, we did not perform regression analysis on these data because it displayed heterogenous variance. The Fe quota-growth rate relationship was opposite of the positive correlation between Cu quota and growth rate in Dokd-P16 (R2=0.21, p=0.009; Fig. 4.2A).     Interestingly, assessing metals quotas that varied significantly across different Cu concentrations of Dokd-P16 and R. pomeroyi DSS-3 suggests that these strains regulate their metal content within a similar range (by a factor < 3). In contrast, content of most metals remained 104  constant in Pseudoalteromonas spp. strains as Cu availability changed, whereas Zn content was found to vary by a factor of up to 5 in PAlt-P26.  4.5 Discussion     This study aimed to gain some insight into the poorly characterized trace metal composition of marine heterotrophic bacteria. In this discussion, we first focus on examining the general trends in metal signatures of the four bacterial strains investigated. Next, we compare the trace metal stoichiometry of cultured marine heterotrophic bacteria and other prokaryotes, as well as eukaryotic phytoplankton to understand the relative importance of metals in the physiology of different microbial groups. Finally, we discuss the effects of Cu availability on cellular Fe, Zn, Mn and Co content in our heterotrophic bacteria, and discuss potential mechanisms responsible for the interaction between these metals.   4.5.1 Trends in metal composition of marine heterotrophic bacteria      We used C-normalized quotas from all Cu treatments to provide a summary of the metal signature for each strain (Fig. 4.3B). Metal profiles of our bacteria can be explained by the cellular biochemical usage of each metal in the cell. The most abundant metals in our bacteria were typically Fe and Zn, their content being similar among the strains. In contrast, metals with relatively fewer known biological functions, such as Mn, Cu, and Co, were required at much lower levels (by 1 to 2 orders of magnitude), with Mn showing the most pronounced relative variation among the strains. The hierarchy of metal abundance in each strain can be summarized as Fe>Zn>Mn>Cu>Co for Dokdonia sp. strain Dokd-P16, and R. pomeroyi DSS-3, and Fe>Zn>Cu≈Mn>Co for both Pseudoalteromonas sp. strains (PAlt-P2 and PAlt-P26). In general, these relative trends are consistent with those of model pathogenic bacteria, such as E. coli (Fe>Zn> Mn≈Cu > Co, Cameron et al., 2002; Fe≈Zn>Mn≈Cu>Co, Outten and O’Halloran 2001) and Pseudomonas aeruginosa (Fe≈Zn>Mn≈Cu>Co, Cunrath et al. 2016). Furthermore, we find that for many metal quotas of our marine heterotrophic bacteria are comparable with those of E. coli (Outten and Halloran 2001; Cameron et al. 2012, Table 4.4). This agreement, especially for metals such as Zn, is striking given the ~ 170-fold higher Zn concentration in the metal-replete 105  culture media used for E. coli (Outten and O’Halloran, 2001) relative to our study, and that the bacterial Zn quota varies by ~ 3-fold. When normalized to carbon, our Fe quotas (26 to 44 μmol Fe:mol C) are in the range of those reported in Fourquez et al. 2014a (16.1 and 141 μmol Fe:mol C) and those we estimated using data from Granger and Price (1999) (1.5 to 20 μmol Fe:mol C) for oceanic and coastal heterotrophic bacteria cultured in Fe-replete Aquil-modified media (amended with bactopeptone and casein hydrolysate as in this study, Table 4.4). In contrast, our Fe:C ratios are much lower than those we estimated from values reported for three marine bacterioplankton species in Vogel and Fisher (2010) (i.e. 270 to 2242 μmol Fe:mol C). This discrepancy may be related to a potential methodological artifact in the study of Vogel and Fisher (2010), since the 55Fe spike added to the coastal seawater for bacterial metal uptake was not pre-complexed to a ligand (e.g. EDTA), creating the possibility of some precipitation of the added Fe. However, we find an agreement of our bacterial Mn:C (0.1 to 14 μmol Mn:mol C) and Zn:C (11 to 47 μmol Zn:mol C) with certain species in study of Vogel & Fisher (2010) (Zn:C of 37 and 14.5 in P. atlantica & R. litoralis, respectively, and Mn:C of 7.5 in P. atlantica, Table 4.4).     There has been some debate as to whether the measured metal quotas in cultures reflect the cellular physiology of an organism or the metal composition of the growth medium (eg. Ho et al. 2003; Cameron et al. 2012). In our experiments, the approximate concentration of bioavailable Fe ([inorganic Fe or Feʹ] = 0.27 nmol L-1) in the bacterial culture medium was an order of magnitude higher than that of Zn ([Znʹ] = 0.027 nmol L-1, Table 4.1, Fig 4.3A), yet the Zn and Fe quotas in all the strains were similar. Likewise, despite bioavailable Mn ([Mnʹ] = 7.6 nmol L-1) being ~ 2 orders of magnitude higher than zinc ([Znʹ] = 0.027 nmol L-1), its content in bacterial cells was either on par with that of Zn (R. pomeroyi DSS-3) or lower (Dokd-P16, PAlt-P26, PAlt-P2, Fig. 4.3). Bacterial copper content was consistently higher than that of Co, although bioavailable Cu, even at our highest Cu treatment (50 nmol L-1), was approximately one order of magnitude lower than Co (0.0003 versus 0.008 nmol L-1 for [Cuʹ] and [Coʹ], respectively, Table 4.1, Fig 4.3A)., observations may make a case for a highly-regulated accumulation of metals in our strains. However, it is unclear if the inorganic metals are the sole substrates for uptake in our bacteria, and if they could also access metals bound to EDTA (as discussed in Chapter 3, section 3.5.1). We expect that, given the replete levels of Fe, Mn, Zn and Co in the culture media, our strains would not need to access the organically bound metal species, and that the Meʹ would be the main uptake 106  substrate. We speculate that the same is likely for Cu under replete conditions (10-50 nmol L-1), especially given the potential to induce toxicity as Cu concentrations increase. However, we cannot rule out that organically-bound Cu may be acquired when our bacteria are Cu-limited.  The idea that metal content of bacteria is well regulated is further reinforced by the fact that C-normalized metal content of our bacteria fall in the range of metal content in the model heterotrophic bacterium E. coli, though it was cultured under very different conditions. Hence, we believe that the ranges of metal quotas reported here serve as a good approximation of those of natural bacterioplankton.     Our results may provide insights into the trace metal requirements of ecologically bacterial important taxa to which our strains belong (see Chapter 3, section for more detailed discussion on strain relevance), and their potential influence on metal cycles. Our findings show that Fe, Zn, Cu, and Co are similarly abundant in our bacterial strains, and we hypothesize similar nutritional requirements for these metals in Roseobacter sp. (marine Roseobacter clade), Dokdonia sp. (Flavobacteriia) and Pseudoalteromonas sp. (Alteromonadales within the Gammaproteobacteria). In contrast, the Mn content of Pseudoalteromonas sp. PAlt-P2 and PAlt-P26 was extremely low Mn compared to other strains. Therefore, we hypothesize that Mn assimilation by bacteria belonging to Pseudoalteromonas genus, which can be abundant during phytoplankton blooms (e.g. Lucas et al. 2015), would be small compared to those with potentially higher requirements such as Dokdonia or Roseobacter clades. Metal requirements of our strains require further confirmation in future studies, where physiological responses under sufficient and limiting trace metal conditions could be compared, as done for Cu in Chapter 3. In that chapter, based on growth and physiological data, we speculated that Cu availability could be important to Dokdonia sp. and possibly Flavobacteriia (based on their response to Cu in the field), but is unlikely to affect members of Roseobacteria and Pseudoalteromonas. As discussed in Chapter 3, there are uncertainties when extrapolating observations from our strains to their entire taxa. Indeed, there is a need for larger study surveying metal requirements in a broad range of species from these taxa, as well as evaluating metal requirements in field populations.    107  4.5.2 Comparison of trace metal stoichiometry in marine heterotrophic bacteria and phytoplankton       The extended elemental stoichiometries (C, N, P, S, Fe, Zn, Mn, Cu, Co) of marine heterotrophic bacteria and photosynthetic eukaryotes have not been compared previously; however, some differences may be expected based on the distinct evolutionary histories, functionalities, and metabolisms of these microbial groups. Changes in trace metal availabilities during evolutionary periods have been hypothesized to influence metal selection for biological usage (Williams and Fraústo da Silva 2003). This concept has been explored via comparative analyses of eukaryotic and prokaryotic metalloproteomes (inferred from genome sequences), which linked the higher abundance of Fe-, Mn-, and Co-binding proteins of prokaryotes to a greater availability of these metals in their evolutionary environment (Dupont et al. 2006). In contrast, Zn-binding proteins are more abundant in eukaryotic proteomes (Andreini et al. 2006; Dupont et al. 2006), consistent with their proliferation after the availability of Zn in the environment increased because of Earth’s oxygenation (Dupont et al, 2006). In addition, Cu availability was low in the evolutionary environment of prokaryotes (Archaean), and it has been hypothesized that this may explain their greater sensitivity to Cu toxicity compared to eukaryotic phytoplankton (Brand et al. 1986; Saito et al. 2003). Taken together, these observations suggest that Fe, Mn and Co requirements may be relatively higher in prokaryotes than in eukaryotes, while the opposite could be inferred for Zn and Cu.      Here, to explore these hypotheses we compare the metal contents of cultured marine heterotrophic bacteria with that of phytoplankton. As in previous studies of phytoplankton metal physiology, we used Aquil media. In Aquil, inorganic trace metals are buffered with a chelate EDTA to mimic bioavailable (labile) metal concentrations in surface waters (Price et al., 1988/1989). The exception is Fe and Mn, for which the inorganic levels are typically raised above those found in the open ocean to facilitate phytoplankton growth in culture. In our study, we used very similar total metals and EDTA concentrations to those in algal Aquil studies used for comparisons with bacterial quota data (Ho et al. 2003; Quigg et al. 2010), except for the Fetot, which was ~17 times higher in our study than in algal studies (Appendix C, Table C.3). The required addition of organic substrates to the bacterial media (according to Granger and Price, 108  1999), creates some uncertainties regarding the concentration of labile metals because of the unknown capacity of these substrates for metals (as discussed in Chapter 3, section 3.3). However, according to Granger and Price (1999) kinetically labile Fe was higher in Aquil amended with organics than in traditional Aquil (with the same amount of Fe and EDTA). Unfortunately, the labile concentrations of other metals in organic substrate amended Aquil have not been determined; however, it is possible that, as with Fe, their levels are higher than in the traditional Aquil media (as discussed in Chapter 3, section 3.3). Our results, nevertheless, suggest that bacterial accumulation of metals is highly regulated. Therefore, we believe that the comparison of bacteria and phytoplankton performed here can provide meaningful insights.     The metal composition of bacteria and phytoplankton is visualized using both P- and C-normalized metal content (Fig. 4.4). Typically trace metals determined by ICP-MS analysis or synchrotron x-ray fluorescence (SXRF), are normalized to P in marine plankton studies; however, this approach may not be ideal when comparing different groups of organisms. For instance, phosphorus to carbon ratio of marine heterotrophic bacteria in our study and others (molar ratio close to 0.02 P:1C, Goldman et al. 1987; Fagerbakke et al. 1996; and the nutrient-replete bacteria in Vrede et al. 2002) is twice that of the eukaryotic phytoplankton (0.009 P: 1C, Redfield, 1963), suggesting that bacterial biomass is P-rich. It is a common paradigm that aquatic bacteria have low and invariant biomass C:P, and a high P-content (Godwin and Cotner 2015b), although it has been shown that bacterial elemental stoichiometry can be flexible (Scott et al. 2012; Godwin and Cotner 2015a). However, only a few studies have investigated the C:N:P of natural bacterial populations in the marine environments. A study of single cell bacteria from coastal and freshwater systems reports and average C:P of 50:1 (thus molar ration of P:C of 0.02, Fagerbakke et al. 1996), which is closer to the average ratio we found in our bacteria. In contrast, individual bacteria collected from an oceanic site (Bermuda Atlantic Time-Series Study [BATS]) were reported to have ratios closer to the Redfield ratio (C:P range of 58-143:1 Gundersen et al. 2002).      The low C:P ratio of bacteria has been linked to a higher abundance of P-rich ribosomes in these organisms relative to that in phytoplankton (Goldman et al. 1987), for which requirements are elevated under rapid growth (Makino et al. 2003; Vrede et al. 2004). Indeed, our cultured strains grow at rates that are up to 18 times faster (growth rates from all experiments range from 1‒18 d-1) than the typical growth rates of phytoplankton in the culture and in the field (e.g. 0.2‒109  0.8 d-1 reported by Ho et al. 2003, and ~ 1 d-1 for phytoplankton in the field, Kirchman, 2016). Furthermore, Godwin and Cotner (2015) showed that aquatic bacteria isolated using P-rich media (as in our study) have a high cellular P content, as opposed to those isolated using P-poor media. Therefore, it is possible that the macronutrient-replete conditions in culture could play a role in the high bacterial P:C of our bacterial strains. High nutrient availability has also been hypothesized to explain the non-Redfield ratios of bacteria from coastal sites, as opposed to those from an oceanic site at BATS (Gundersen et al. 2002). An alternative hypothesis is that bacterial cells are C-poor rather than P-rich compared to phytoplankton, owning to the absence of C-rich structural moieties in bacteria. Unfortunately, we were not able to quantitatively assess the differences in the cellular P and C between phytoplankton and bacteria in our dataset in Fig. 4.4, as these data are not available in the literature data for many species. Surprisingly, C:N:P stoichiometry of marine bacteria versus phytoplankton have received very little attention, and there is a need for further study to elucidate the trends in the element stoichiometry of these two groups.     Here, we illustrate how bacterial trace metal stoichiometry relative to that of eukaryotic phytoplankton is slightly different depending on whether metals are normalized to P or C (Fig. 4.4). In analogy to the extended Redfield ratio for phytoplankton, the overall elemental composition of heterotrophic bacteria using P (averages of metal contents for all strains and all Cu treatments) yields the formula of (C61N15S0.6P1)1000 Fe2.3Zn1.1Mn0.4Cu0.1Co0.02, whereas that for eukaryotic phytoplankton is: (C124N16S1P1)1000 Fe13Zn0.9 Mn2.5 Cu0.3 Co0.09 (using data from Ho et al. 2003 and Quigg et al. 2011, and median values for trace metals to account for the skewness of the phytoplankton data). This suggest that, except for Zn, the average content of metals of heterotrophic bacteria is lower than that of eukaryotic phytoplankton. However, re-calculating the stoichiometry using C-normalized metals instead of P results in: (P0.018N4.2S0.01C1)1000Fe42Zn15Mn5.3Cu1.3Co0.28 for heterotrophic bacteria and (P0.010N8.8S0.01C1)1000Fe78Zn6.4Mn29Cu1.72Co1.2 for eukaryotic phytoplankton (Me:C ratios calculated for each species using C: P ratio reported in Ho et al. 2003 and Quigg et al. 2011, the coefficients representing the medians reported in Appendix C, Table C.4, calculations having been performed on the data shown in Fig 4.4). Considering these C-normalized metal coefficients and the variation in trace metals for each group (Fig 4.4, Appendix C, Table C.4) we suggest that heterotrophic bacteria and phytoplankton contain similar Fe and Cu, while bacteria appear to be 110  relatively richer in Zn but depleted in Mn and Co. Given the issues with potential P enrichment of bacterial biomass in our study, and the fact that P content can be flexible in microorganisms under different trace metal regimes (e.g. P content increases under Fe-limitation in the diatom T. weissflogii, Price 2005), we suggest that C-normalized stoichiometries are most appropriate for the comparison of heterotrophic bacterial and phytoplankton. Furthermore, C is by far the dominant element in biomass (except for O and H from water), making it an ideal normalizing factor for cellular elements.     Furthermore, we compare the metal profiles of bacteria and eukaryotic phytoplankton (Fig. 4.5) with other microorganisms (cyanobacteria, E. coli, archaea, and field single cell Trichodesmium). The order of metal abundance of prokaryotes (including cyanobacteria) can be summarized as Fe>Zn>Mn>Cu>Co, while in eukaryotes as Fe>Mn>Zn>Cu>Co. Hence, these organisms differ in terms of the abundance of manganese and zinc in the cell. This difference has been noted previously in comparisons of marine phytoplankton metal content with that of E. coli (Ho et al. 2003). The authors partly related this difference to the requirement for Mn for the oxygen-evolving complex of PSII in phytoplankton. However, they also questioned whether the high and low concentrations of Mn and Zn, respectively, in phytoplankton media in contrast to those of E. coli could also account for these differences. The total Mn and Zn concentrations in our media are comparable with the study of Ho et al. 2003 (Appendix C, Table C.3), yet Mn quotas of eukaryotic phytoplankton are indeed markedly higher than those of our heterotrophic bacteria and from cyanobacteria in other studies (Fig. 4.5). Furthermore, our estimated median of carbon-normalized phytoplankton Mn quotas is 29 versus 3-15 μmol Mn: mol C for field samples (the estimate for natural phytoplankton assemblages used for comparison in the study of Ho et al. 2003, Fig. 4.5, Appendix C, Table 4.C). Recent measurements of single cells in the Atlantic by synchrotron X-ray fluorescence (1.5 to 5.5 μmol Mn:mol C, Twining et al. 2015) are also lower than those of cultured phytoplankton (in Ho et al. 2003 and Quigg et al. 2011) (Fig. 4.5). We, therefore, hypothesize that the higher Mn content in photosynthetic eukaryotes is more likely caused by the high Mn in the medium than physiological requirements. On the other hand, the relatively lower bacterial Mn (Fig. 4.5) may reflect a well-controlled Mn homeostasis in cultured prokaryotes relative to eukaryotic phytoplankton. In our preliminary comparison, bacterial Zn quotas are in the range of those of phytoplankton, although at the higher end (Fig. 4.5). This does 111  not support the hypothesis that prokaryotes may have lower Zn quotas relative to eukaryotes due to the lower abundance of Zn finger proteins in the former (Twining and Baines 2013). The Zn quotas of bacteria may reflect their demands for substrate acquisition since hydrolytic enzymes (zinc-binding peptidases) appear to dominate the Zn proteome of prokaryotic organisms (Andreini et al. 2006).      Overall, our observations do not fit into the evolutionary context of metal use inferred from metalloproteome studies (i.e. higher Fe, Mn, Co and lower Zn and Cu utilization in prokaryotes relative to eukaryotes). However, metal contents may not be directly linked to biological demand because metals can be accumulated in excess of their physiological requirements (e.g. Fe as ferritin in some pennate diatoms, Marchetti et al. 2009), although we believe this to be less of an issue in prokaryotes given our observations of their well-controlled metal contents in our bacteria. Future studies comparing the minimum metal quotas and growth responses to their varying availability in marine bacteria and eukaryotic phytoplankton will help in elucidating their relative requirements for trace metals.   4.5.3 Effects of Cu availability on the content of metals in bacteria          In this study, we found that Pseudoalteromonas sp. strains PAlt-P2 and PAlt-P26 behaved very similarly when grown under conditions of varying Cu concentrations, maintaining relatively constant growth and Cu quotas. Furthermore, the quotas of other metals in these bacteria were similar and did not vary significantly with respect to Cu concentration in the growth media, even though the two strains were isolated from contrasting locations with different metal availabilities (PAlt-P2 from coastal waters of BC, and PAlt-P26 from oceanic, Fe-limited waters). An exception was the Zn quota of the oceanic strain PAlt-P26, which increased in response to a reduction in Cu availability. Metals can be accumulated unintentionally, through metal transporters that bind a variety of metals (Nevo and Nelson 2006; Ma et al. 2010). In some organisms, overexpression of divalent metal transporters under trace metal limitation (e.g. overexpression of ZIP and NRAMP transporters Cu-limited T. pseudonana, Guo et al. 2015) may lead to such non-specific uptake of some metals, as hypothesized for increased Cd content in Fe-limited phytoplankton (Lane et al. 112  2009). Perhaps this mechanism may explain the increase in the Zn quota of PAlt-P26 when Cu concentration in the growth media is lowered. Since this strain has a low requirement for Cu (Chapter 3), we would not expect an overexpression of transporters to enhance Cu uptake, nor would we expect an increased demand for Zn, in response to Cu limitation. However, Cu uptake systems in gram-negative bacteria remain poorly understood (Argüello et al. 2013), so it is uncertain how Cu and Zn transporters might interact.     Although the growth of R. pomeroyi DSS-3 was relatively unresponsive to changes in Cu availability, the quotas of all metals except for Zn did vary in this strain. Our results suggest that R. pomeroyi DSS-3 can maintain maximum growth regardless of Cu availability by adjusting its Cu use and regulating some aspects of C metabolism (Chapter 3). Whereas Cu content of R. pomeroyi DSS-3 is reduced as Cu availability is lowered, we find an increase in the content of Mn and Fe, although both metals are mainly affected at moderately low Cu (2 nmol L-1), relative to other Cu treatments (Fig. 4.1). These two metals, along with the Cu/Zn pair, have complementary functions in the cell because all can serve as co-factors in the anti-oxidative enzyme superoxide dismutase (SOD) (Miller 2012). Their increase under moderately limiting Cu concentrations (2 nmol L-1) may imply an increased production of Fe- and Mn-SOD, resulting either from increased oxidative stress or the need to compensate for a possible reduction in Cu/Zn SOD under conditions of low Cu availability. It is also possible that the trends of Mn and Fe content could be related to non-specific uptake caused by overexpression of metal transporters to enhance Cu uptake when Cu concentrations are low.     Changes in the cobalt content in response to changing Cu availability in R. pomeroyi DSS-3 are intriguing (Fig. 4.1). The main role of cobalt in these organisms is as the central atom in the corrin ring core of vitamin B12 (cobalamin, Kobayashi and Shimizu 1999). It may be hypothesized that the requirement for vitamin B12 may be reduced when this strain experiences Cu starvation, although it is difficult to explain the physiological benefit of such a response. The fact that Cu and Co quotas are correlated with Cu availability (Fig. 4.2C) suggests that their homeostasis may be in some ways connected. Current homeostasis models for these two metals in bacteria (Ma et al. 2010; Porcheron et al. 2013) do not indicate a clear connection although, as already noted, Cu transport and homeostasis in prokaryotes are not completely understood (Argüello et al. 2013; 113  Gladyshev and Zhang 2013). To our knowledge, this type of interaction between Cu and Co has not been observed before, and we believe it offers an interesting topic for further study.     The Flavobacteriia member Dokd-P16 was the only strain in our study to be severely impaired by low Cu availability in terms of its growth rate and C metabolism (Chapter 3). In addition to these responses, we found increases in Mn quotas as Cu availability decreased, although for Mn this was not observed at the lowest Cu concentrations (Fig. 4.1). We also found a negative correlation between Fe quota and growth rates and a positive correlation between Cu quota and growth rate in Dokd-P16 (Fig. 4.2B and Fig. 4.2A, respectively), which suggests that Fe requirements increase in Cu-limited Dokd-P16 to compensate for the shortage of Cu. As discussed above for R. pomeroyi DSS-3 this effect could be due to an increased requirement for Mn- or Fe-superoxide dismutase (SOD). In addition, iron is heavily involved in bacterial C metabolic pathways (Andrews et al. 2003), and as the C metabolism of Dokd-P16 was significantly impacted under low Cu conditions (Chapter 3). Hence, an increased Fe quota in slow growing Dokd-P16 may reflect modifications to C metabolic pathways.      We also considered whether the changes in bacterial metal quotas in our study could be explained by a growth rate dilution effect. This effect describes an increase in metal contents of microorganism when their growth rate declines, because of the inverse relationship between cellular metal quotas and the specific growth rate at steady state (Q = Vss/μ, where Q is metal quota, Vss is steady-state metal uptake, and μ is growth rate, Sunda and Huntsman, 1998). However, given that this effect is a function of growth, it is unlikely to explain our results for R. pomeroyi DSS-3 and Pseudoalteromonas spp. which did not significantly reduce their growth under low Cu. In addition, the growth rate dilution does not readily explain the Mn quotas of Dokd-P16 despite the significantly variable growth rates. This is because the Mn contents were elevated only in treatments with moderate Cu levels, and not in the most Cu deplete treatments where the growth rate of this bacterium was most restricted (Fig. 4.2). (2015)(Sunda and Huntsman 1998)114     Table 4.1: Total dissolved metal (Metot), inorganic metals (Meʹ) and free metal ion concentrations (pMe, -log[Me]) in bacterial growth media.  Calculations were performed with Visual MINTEQ version 3.1 (software is available from https://vminteq.lwr.kth.se/) excluding the organic substrate additions from calculations. The dissolved Cu reported in the lowest Cu treatment (without Cu addition) is that measured in the chelexed SOW by FIA-CL (Semeniuk 2014) and indicates media background Cu contamination. Varying Cu levels had not effect of the inorganic and free concentrations of other metals.                                                                                                                                  Metal [Metot] (nmol L-1) [Meʹ] (mol L-1) [pMe] (-log[Me])  Fe  1370  2.8 x 10-10  18.88 Zn 70 2.7 x 10-11 10.89 Mn 110            7.6 x 10-9 8.30 Cu 0.6‒50 4.6 x 10-15 ‒ 3.9 x 10-13 15.47‒13.55 Co 44.2 8.3 x 10-12 11.21                Mo 100 N/A N/A                Se 10 N/A N/A 115  Table 4.2: Trace metal content of bacteria grown at different levels of Cutot (nmol L-1). Values are mean P-normalized metals (mmol:mol P) ± SE.   nd – not determined, bd-below detection of method   Strain [Cutot] (nmol L-1) n Growth rate (d-1) n Fe:P Zn:P Mn:P Cu:P Co:P Dokdonia sp.  0.6 6 1.61 ± 0.3 4 4.8 ± 0.8 1.1 ± 0.3 0.33 ± 0.03 0.06 ± 0.02 0.017 ± 0.002 strain Dokd-P16 2 11 2.83 ± 0.1 6 3.6 ± 0.2 1.0 ± 0.1 0.51 ± 0.06 0.08 ± 0.01 0.019 ± 0.001  10 7 6.48 ± 0.3 7 3.0 ± 0.4 0.6 ± 0.2 0.40 ± 0.02 0.08 ± 0.01 0.018 ± 0.001  25 9 6.94 ± 0.2 6 2.8 ± 0.7 2.4 ± 0.9 0.39 ± 0.03 0.14 ± 0.01 0.017 ± 0.010  50 10 8.72 ± 0.8 3 2.3 ± 0.1 1.0 ± 0.5 0.31 ± 0.02 0.12 ± 0.003 0.015 ± 0.005 Pseudoalteromonas sp.  0.6 4 9.00 ± 0.5 4 1.7 ± 0.1 1.3 ± 0.4 0.07 ± 0.03 0.03 ± 0.01 0.012 ± 0.001 strain PAlt-P2 (coastal) 2 nd nd nd nd nd nd nd nd  10 5 12.4 ± 0.9 3 2.0 ± 0.1 1.1 ± 0.1 0.02 ± 0.003 0.04 ± 0.001 0.013 ± 0.002  25 7 11.3 ± 0.4 4 1.5 ± 0.2 1.5 ± 0.2 0.12 ± 0.05 0.03 ± 0.01 0.014 ± 0.002  50 nd nd nd nd nd nd nd nd Pseudoalteromonas sp. 0.6 20 15.2 ± 0.4 4 2.4 ± 0.5 1.6 ± 0.3 0.01 ± 0.001 0.06 ± 0.01 0.03 ± 0.010 PAlt-P26 (oceanic) 2 6 18.2 ± 0.7 4 2.1 ± 0.1 2.1 ± 0.4 bd  0.05 ± 0.01 0.008 ± 0.002  10 33 18.0 ± 0.4 3 2.4 ± 0.2 1.0 ± 0.1 bd 0.06 ± 0.02 0.009 ± 0.001  25 3 17.1 ± 0.4 nd nd nd nd nd nd  50 6 15.2 ± 0.9 3 2.0 ± 0.2 0.4 ± 0.1 bd 0.04 ± 0.004 0.008 ± 0.002 R. pomeroyi DSS-3 0.6 11 4.68 ± 0.4 3 2.2 ± 0.1 1.4 ± 1.0 0.98 ± 0.05 0.05 ± 0.003 0.016 ± 0.003  2 4 5.47 ± 0.3 4 5.1 ± 0.6 1.2 ± 0.3 1.10 ± 0.03 0.11 ± 0.012 0.023 ± 0.002  10 9 4.42 ± 0.3 3 2.2 ± 0.1 0.5 ± 0.1 0.90 ± 0.02 0.15 ± 0.009 0.037 ± 0.002  25 3 3.94 ± 0.3 nd nd nd nd nd nd  50 9 5.47 ± 0.3 3 2.3 ± 0.2 0.5 ± 0.2 0.84 ± 0.01 0.15 ± 0.006 0.030 ± 0.002 116      Table 4.3: One-way ANOVA values for the effect of Cu concentration in media on P-normalized metals (mmol Me:mol P). Statistically significant values are shown in bold.                                                                               ANOVA results  Dokd-P16 PAlt-P2 PAlt-P26 R. pomeroyi DSS-3 Variable F statistic p-value F statistic p-value F statistic p-value F statistic p-value Fe:P  2.81 0.051 3.17 0.097 0.42 0.741 24.6 >0.001 Zn:P 2.07 0.12 0.29 0.755 4.61 0.028 0.93 0.467 Mn:P 4.07 0.014 2.21 0.172 bd bd 10.4 0.003 Cu:P 7.05 <0.001 0.83 0.467 0.49 0.691 20.5 0.002 Co:P 2.6 0.064 0.71 0.521 2.29 0.140 16.8 >0.001 117  Table 4.4: Comparison of metal quotas (Me:C, μmol:mol) of marine heterotrophic bacteria with those of other heterotrophic bacteria from literature. The range of metal quotas of marine heterotrophic bacteria represents the variability of quotas for all strains and across all Cu treatments (the data for the mean Me:C of different bacteria are shown in Appendix C, Table C.2). The metal data shown here represent mean and errors were not included for clarity.  Reference Organism Media Metal [Metot] (mol L-1) Me:C (μmol:mol) Outten & O'Halloran, 2001a E. coli LB Fe 5.5 x 10-6 23    Zn 1.2 x 10-5 15    Mn 1.6 x 10-7 1.8    Cu 1.5 x 10-7 3.3   Minimal Fe below detection 12    Zn 1.8 x 10-7 14    Mn 2.2 x 10-8 0.2    Cu 6.7 x 10-8 0.5       Cameron et al., 2012a E. coli (aerobic) Purified LB Fe 4.7 x 10-6 153    Zn 7.5 x 10-6 14    Cu 9.7 x 10-8 1.7    Co                6.0 x 10-8 0.01  E. coli (anaerobic)  Fe 4.6 x 10-6 154    Zn 6.2 x 10-6 12    Cu 8.3 x 10-9 4.7    Co 5.7 x 10-8 0.1       This study Marine heterotrophic Aquil-modifiedb Fe 1.4 x 10-6 24-100  bacteria isolates  Zn 7.0 x 10-8 11- 47    Mn 1.1 x 10-7 0.1-14    Cu 0.6-50 x 10-9 0.5-2.7    Co 4.4 x 10-8 0.1-0.4 118    a Metal contents of E. coli from both studies were converted using a conversion factor of 50:100 carbon: dry weight. Metals were determined by ICPMS analysis. Values for whole cell digests were used   from Cameron et al., 2011. b Traditional Aquil media for culturing marine phytoplankton (Price et al. 1988/89) amended with bacterial growth substrates (bactopeptone and casein hydrolysate) as described in Granger & Price, 1999. c Metal content of bacteria was determined using radiotracer technique (55Fe, 54Mn,65Zn) after metal accumulation reached steady-state (48-92 h). We estimated carbon content of bacteria in this study    using cell volumes (μm3) reported for each strain and bacterial carbon versus volume relationship determined by Fagerbakke et al. 1996 [ln(C) = (1.2±0.3) x ln(volume) + (4.28±0.04)]. This carbon     content was used to convert the cellular metals reported by Vogel & Fisher, 2010 to metals per carbon (Me:C, μmol: mol). For these conversions, we used values for cells rinsed with oxalate wash for Fe (Tovar-Sanchez et al. 2003) and EDTA rinse for Zn and Mn (Hassler et al. 2004) except for Zn values for H. aquamarina and Vibrio sp. (only seawater rinsed). d Iron was measured using 55Fe technique in titanium(III)-EDTA-citrate solution rinsed cells (Hudson and Morel 1989) and converted to Fe:C using cellular carbon determined by CNH analysis. e Iron was measured using 55Fe technique in titanium(III)-EDTA-citrate solution rinsed cells (Hudson & Morel, 1989). We converted the cellular metals reported in this study using estimates of C content    derived from bacterial volumes and carbon: volume conversion of Fagerbakke et al. 1996.Reference Organism Media Metal [Metot] (nmol L-1) Me:C (μmol:mol) Vogel and Fisher, 2010c V.natrigens Coastal Fe 1.4 x 10-8 270  H. aquamarina seawater Fe 1.4 x 10-8 903  R. litoralis in situ  Fe 1.4 x 10-8 2242  P. atlantica organic Mn 5.0 x 10-9 7.5  R. litoralis substrates Mn 5.0 x 10-9 62  P. atlantica  Zn 1.3 x 10-7 37  R. litoralis  Zn 1.3 x 10-7 14.5  H. aquamarina  Zn 1.3 x 10-7 182  Vibrio sp.  Zn 1.3 x 10-7 67       Fourquez et al. 2014ad A. macleodii (coastal) Aquil-modifiedb Fe 5.4 x 10-9/5.4 x 10-6 0.56 / 141  A. macleodii (oceanic)  Fe 5.4 x 10-9/5.4 x 10-6 0.43 / 16.1       Granger and Price 1999e Neptune Aquil-modifiedb Fe 1.2 x 10-10/ 8.4 x 10-6 0.1 / 13.9  P20pac  Fe 1.2 x 10-10/ 8.4 x 10-6 0.9 / 20  Pseudomonas sp. (Isol5)  Fe 1.2 x 10-10/ 8.4 x 10-6 0.5 / 1.5  Jul_88  Fe 1.2 x 10-10/ 8.4 x 10-6 0.3 / 8.5  Lmg1  Fe 1.2 x 10-10/ 8.4 x 10-6 1.9 / 20  Pwf3  Fe 1.2 x 10-10/ 8.4 x 10-6 0.2 / 12.3  V.natrigens  Fe 1.2 x 10-10/ 8.4 x 10-6 0.7 / 17.2 119   Figure 4.1: Growth rates and metal quotas (mmol Me:mol P) of coastal isolate Pseudoalteromonas sp. strain PAlt-P2 and oceanic isolate Pseudoalteromonas sp. strain PAlt-P26, Dokdonia sp. strain Dokd-P16, R. pomeroyi DSS-3 at different Cu levels in culture media (nmol L-1). Significant differences among different Cu treatments (p <0.05) are shown with different letters (Tukey’s honest significant difference test or pairwise t.test with Bonferroni correction for datasets with heterogenous and homogenous variances, respectively). Metal quota data in plots with no letters where not statistically significant as indicated by one-way ANOVA (Table 4.2) Open circles and error bars are mean ± SE and the gray filled circles are the data points. For Co quota of PAlt-P26, an outlier point (> 0.06 mmol:mol) at 0.6 nmol L-1 is not shown. The dataset is available in Table 4.2. Note, Mn quotas of PAlt-P26 at Cu treatments of 0.6, 10 and 50 nmol L-1 were below detection limit of the method.  Note that the x-axis is non-linear   120                           Figure 4.2: Correlations between Cu:P and growth rate (A), Fe:P and growth rate (B), Cu level (D) in Dokdonia sp. Dokd-P16 (C); and between Co:P and Cu:P in R. pomeroyi DSS-3.   121               Figure 4.3: Approximate inorganic metal levels [Meʹ] in culture media (A). The metal profile of marine heterotrophic bacteria (B). Data points in the bottom panel are mean C-normalized metals (μmol Me:mol C) from all Cu treatments for each strain (Me:C data can be found in Appendix C, Table C.2, and metal concentrations in growth media can be found in Table 4.1) Error bars were removed for clarity. For Cu, the inorganic concentration is that of the highest Cu treatment (Cutot=50 nmol L-1, [Cuʹ] = 3.9 x 10-13)122            Figure 4.4: Comparison of elemental stoichiometry of cultured marine heterotrophic bacteria, cyanobacteria and algae. The graphs show major elements (C, N, and S) normalized to phosphorus (top panel), and minor elements (Fe, Zn, Mn, Cu and Co) normalized to phosphorus (middle panel) and carbon (bottom panel). Note the changes in bacterial metal content relative to algae when data are normalized to C instead of P, which can be explained by the differences in their C:P ratio (top panel). Values for marine heterotrophic bacteria are from this study (mean quotas of all Cu treatments) and Fe:C of bacteria from Fourquez et al, 2014a. Algae data is from Ho et al. 2003, Annett et al. 2008 (Cu:C only, Fe-replete cultures), Quigg et al. 2010, Guo et al. 2012b (Cu:C only, Fe-replete cultures). Phytoplankton data is from Ho et al. 2003 (algae) and Quigg et al. 2010 (algae and cyanobacteria) was converted from Me:P quotas to Me:C using the C:P ratios reported for each organism in both studies. Filled points are mean metal values reported in each study and the black vertical line is the median estimated for each planktonic group. 123                     Figure 4.5: Comparison of metal quotas (Me:C, μmol:mol) of different microbes. Data points are mean metal quotas and errors were omitted for clarity. The estimated medians of Me:C and the ranges for each group are presented in Appendix C, Table C.4. Archaea data is from Cameron et al. 2012 (whole cell digests, metal contents were estimated assuming 50% C by dry mass). Heterotrophic bacteria data includes bacteria in our study (all Cu treatments), the literature values for E. coli (Fe, Zn, Cu and Co in Cameron et al. 2011; Fe, Zn, Cu, Mn, Co in Outten and O’Halloran 2001, estimated assuming 50% C by dry mass) and marine heterotrophic bacteria (Fe quotas determined in Fe-replete cultures by Granger and Price, 1999, Fourquez et al. 2014a). Data for photosynthetic bacteria includes diazotrophic and non-diazotrophic species from Nuester et al. (2012), Quigg et al. (2010), Guo et al. (2012b, Cu:C from Fe-replete treatments). Phytoplankton values include those reported in Ho et al. (2003), and Quigg et al. (2010; Fe, Zn, Mn, Cu, and Co, converted using C:P ratio for each species reported in these studies), Sunda and Huntsman (1995; Cu:C, range of [Cuʹ] in media: 18-759 fmol L-1), Annett et al. (2010), and Guo et al. (2012b, Cu:C determined for Fe-replete cultures growing at [Cuʹ] of 23.9 and 239 fmol L-1). Field data for natural phytoplankton assemblages estimated by Ho et al. 2003 (gray triangles) and the single cell measurements by synchrotron X-ray fluorescence (SXRF) of phytoplankton cells (diatoms, autotrophic dinoflagellates, and picoplankton) from the Atlantic by Twining et al. (2015) (cyan triangles). (Redfield 1958; Emerson and Hedges 2008; Gόmez-Pereira et al. 2010; Ramaiah et al. 2015) 124  Chapter 5:  General discussion     The research presented in this thesis couples field observations with laboratory studies to gain insights into the processes that control the biogeochemical cycling of Cu, particularly in the subarctic Northeast Pacific Ocean. I examined the distribution of dissolved Cu (dCu) along the well-studied Line P transect in this region (Chapter 2) and characterized the local sources and sinks of dCu. Furthermore, in light of recent methodological issues associated with dCu determinations, I evaluated a series of techniques to measure dCu, using samples collected from Line P. The findings of the methodological component of Chapter 2 identify factors causing uncertainties in total dCu analyses, and can be used to establish international protocols for dCu analysis and intercalibrations among laboratories. In Chapter 3, I focused on characterizing the interaction between Cu and marine heterotrophic bacteria using laboratory experiments with strains isolated from surface waters along Line P, and a model bacterium from a culture collection. The results of these laboratory studies were interpreted within the context of Cu biogeochemistry, as well as bacterial physiology and ecology. The effects of Cu on marine heterotrophic bacteria were further characterized in Chapter 4 where the response of bacterial metal content (Fe, Zn, Mn, and Co) to changing Cu availability was assessed. To examine how different metals control microbial ecology in the sea, I also compared the content of trace metals in marine heterotrophic bacteria and in other well-studied microorganisms (e.g. E. coli, cyanobacteria, phytoplankton) in Chapter 3 and 4. In the following sections, I provide a synthesis of the major findings of my thesis, highlighting specific contributions to the field of trace metal biogeochemistry, discussing emerging questions from this work, as well as limitations of my research.     5.1 Towards the improvement of dCu measurements in seawater    Our ability to obtain reliable measurements of metals is key to the interpretation of their cycles and impacts on microbial processes in the ocean. Low concentrations of metals in the open ocean, combined with ease of contamination during sampling imposed significant challenges on early 125  trace metal analyses. Overcoming these challenges required development of rigorous protocols for trace metal-clean sampling, evaluation of metal contamination from sample containers, and optimization of analytical methods to allow low-level metal detection. The efforts to refine oceanographic sampling protocols for trace metals and analytical techniques are ongoing. An important aspect of this effort has been the establishment of an inter-calibration programs, such as the ‘Sampling and Analysis of Iron’ (SAFe) program adopted by the GEOTRACES (http://www.geotraces.org/sic/about-s-i), which compares a suite of analytical techniques and trace-metal clean sampling methodologies from different laboratories. In the most recent SAFe intercalibration for dCu it emerged that, to obtain accurate dCu values, UV oxidation of samples prior to analysis was required for some methods (SAFe inter-calibration document, 2013). Acidifying samples to low pH (1.7 to 2) prior to analysis, combined with some storage period, has been the standard procedure for dissociating the organically-bound Cu (which accounts for > 99.9% of the total dCu pool in seawater) to the free form and hence, ensuring determination of the total dCu in the samples. However, for some analytical methods, Cu concentration was found to be higher in SAFe samples that have been UV oxidized (UVO) prior to analysis compared to non-UVO samples, even after long storage periods (Milne et al. 2010). This suggests that some organically-bound Cu may persist in acidified seawater samples. However, the effects of UVO on dCu analysis have been inconsistent, and it is still uncertain which methodologies require this treatment.   Initial analysis of samples for the biogeochemical study of Cu along Line P was carried out without prior UVO using the FIA-CL method, following storage at low pH for ~ 3 months. An accompanying analysis of the reference sample SAFe D2, also without UVO, provided values on a par with the consensus values, thus indicating that there is consistency between our dCu analysis by FIA-CL and other methods, even though UVO was not used. However, we sought to re-evaluate initial analysis (Line P samples and SAFe reference materials), based on the UVO recommendation by the SAFe community. This work was performed using the same Line P samples, but approximately 4 years later, with UVO pre-treatment. For our field samples, large offsets (6.5-43%) in dCu concentrations relative to initial analysis were found, but there was no difference between UVO and non-UVO SAFe samples (D1 and D2), which were much older than the Line P samples (>7 years). These results suggested to us that 1) sample storage, 2) UVO and 126  3) analytical methodologies affect the total dCu analysis. To better understand the interaction of these factors, we designed a series of experiments (dCu measurements with and without UVO at different time points during acidic storage), compared dCu measurements using different methodologies (FIA-CL and CSV-CLE), evaluated published dCu profiles in the North Pacific and reviewed recent literature on dCu analysis. Based on this work, several important observations regarding sample storage time, UVO and analytical methods for dCu determination were made. Our work provides valuable information for the trace metal community and brings uncertainties in dCu values into the spotlight. The principal findings from the methodological component of Chapter 2 can be summarized as follows: (1) Storage time influences the labile Cu in acidified samples. Long-term storage at low pH promotes the release of Cu from organic complexes, which introduces a potential for variable results depending on the time from collection to analysis. This has implications for whether UVO of samples is necessary prior to analysis. (2) UVO is essential prior to FIA-CL analysis of young samples (aged for 48 h to 2 months), but may not be required if samples have been stored for an extended period at low pH (≥ 4 years). (3) In addition to the storage time of acidified samples, analytical methodologies may introduce another source of variability in dCu values. For instance, FIA-CL analysis of samples from P26 required at least 2 months of storage with UVO pre-treatment to provide values in agreement with those obtained by CLE-CSV analysis of non-stored samples, also with UVO. (4) There is a substantial variation in dCu values in the NE Pacific profiles. It is highly unlikely that oceanographic processes can account for these variations. A combination of different methodologies and sample storage times may help to explain these variations. (5) Based on our review of published SAFe values, there is no consistency in terms of which methods (UVO or no UVO) provides the more accurate measurements of total dCu. For example, some methods utilizing isotope dilution, which has been proposed not to require UVO, reported dCu values that were lower than the SAFe consensus values.    There are some additional questions that need to be addressed in future studies. For instance, if organic complexation is responsible for the analytical problems discussed in this section, we would expect larger offsets between non-UVO and UVO samples near the coast, where organic 127  matter is likely to be higher. Our data does not demonstrate such a trend. Also, why is UVO necessary for dCu analysis but not for the analysis of dFe, which is also known to be complexed by very strong organic ligands? One could speculate that the nature of Cu-binding and Fe binding ligands is very different; however, uncertainty remains.   5.2 Biogeochemical cycling of Cu along Line P in the NE Pacific    Cu measurements in the ocean have been made for over 40 years, yet its biogeochemical cycling remains to be fully characterized at both regional and global scales. One geographical area for which our understanding of the processes governing the Cu cycle is limited is the subarctic Northeast Pacific. Previous work in this region has provided valuable insights into the mechanisms shaping the dCu depth profile, such as biological uptake and remineralization, mid-depth scavenging, and Cu inputs from sediments (Bruland 1980; Martin et al. 1989; Fujishima et al. 2001; Takano et al. 2014). However, studies integrating dCu measurements along zonal sections have been rare, and this has hampered our ability to assess the effects of spatial-scale processes (e.g. water mass transport of dCu, fluxes from the continents, etc.) on Cu biogeochemistry in this area. Likewise, little is known regarding the temporal variation in dCu in this region. To address this, I sought to identify and characterize some of the processes influencing Cu biogeochemistry along the coastal-open ocean Line P transect in the subarctic NE Pacific. Previous dCu studies along this transect examined the spatial distribution of dCu concentration and its speciation in the mixed layer, and Cu interactions with the resident microbial communities (Semeniuk et al, 2016). Work described in Chapter 2 provides a next step towards a better understanding of Cu cycling along Line P by examining high-resolution dCu profiles down to 2000 m at five major stations along the transect. The data were interpreted within the context of local oceanographic features, such as water masses (NPIW and CUC), oxygen minimum zone, and the offshore upwelling Alaskan gyre. We gained additional information on the controls of Cu biogeochemistry in this region by assessing the temporal variation of dCu at the terminal offshore station (P26) and from satellite observations. Our results are summarized in Fig. 5.1 and provide an overview of the major players in the Cu cycle along Line P. Mechanisms responsible for delivering Cu to surface waters along the transect include continental inputs near the coast of BC (eastern side of transect in Fig 128  5.1) and upwelling of dCu rich waters in the Alaskan gyre (western side of transect in Fig 5.1). Satellite observations of the aerosol optical thickness (AOD) for the Gulf of Alaska, and surface trends of dCu at station P26 suggest that atmospheric deposition may act as a potential episodic supply of dCu. We must await future studies to characterize the nature and magnitude of this supply. However, as highlighted in Chapter 2, these atmospheric inputs may be anthropogenically influenced Asian sources and could have adverse ecological effects in this area. We were constrained in our ability to examine bottom sources of Cu due to a lack of measurements at depths greater than 2000 m at most stations. The exception was the coastal station P4, where we observed that deep water samples were enriched in dCu, suggesting a sedimentary supply of dCu (bottom box Fig. 5.1). The copper-macronutrient relationships revealed that dCu is tightly coupled to biological uptake in surface waters and remineralization at depth (dCu:PO3-4) and that it may be scavenged within the intermediate depths of the OMZ (dCu:Si). As highlighted in the schematic representation of the Cu cycle along Line P (Fig 5.1), some of the biochemical processes examined in Chapter 2 require further study. In addition, there is a component of the Cu biogeochemical cycle that has not been well-explored in this region (nor in the global ocean), which is Cu assimilation by marine heterotrophic bacteria. When I began my Ph.D. research, estimates of Cu contents in marine heterotrophic bacteria were not available, nor were there studies investigating the role of Cu in their physiology and metabolism. This provided a context for the experimental work in Chapter 3 and 4, which are summarized below.   5.3 Characterizing Cu interactions with marine heterotrophic bacteria     In Chapter 3 I examined the effects of low Cu on the physiology of marine heterotrophic bacteria, by surveying various bacterial responses (Cu quotas, growth, carbon metabolism, elemental stoichiometry) across a range of non-toxic Cu concentrations, mimicking those of coastal and open ocean surface waters. The first estimates of marine heterotrophic bacteria Cu quotas generated in this thesis allowed me to infer the importance of bacterial accumulation of Cu in surface waters. Similarly, metabolic responses of different strains to Cu limitation allowed me 129  to discuss how Cu may modulate ecological interactions among different microbial groups. The results presented in Chapter 3 have broader implications for oceanography, as well as for prokaryotic Cu nutrition in general. This is because, although there is a great deal of information on how bacteria protect themselves against Cu toxicity, little is known about their Cu requirements, uptake mechanisms and metabolism under low Cu availability.    Bacterial strains used for the experiments in Chapter 3 are affiliated with ecologically significant microbial clades in the ocean; marine Roseobacteria, Flavobacteriia, and Gammaproteobacteria.  These bacteria were isolated from various locations along Line P: Dokdonia sp. strain Dokd-P16 (Flavobacteriia), Pseudoalteromonas sp. strain PAlt-P2, Pseudoalteromonas sp. strain PAlt-P26 (Alteromonadales within Gammaproteobacteria), while R. pomeroyi DDS-3 was obtained from culture collection (isolated from Georgia coastal waters). Bacterial responses to changing Cu availability were diverse. In the Flavobacteriia member Dokd-P16, the only strain to be Cu-limited, C metabolic rates, and Cu quotas were reduced as Cu concentrations in culture media were lowered. These negative effects were observed in treatments with inorganic Cu levels ([Cuʹ] = 19.8 fmol L-1) that are similar to those in surface waters of many oceanic regions, including the western section of the Line P transect in the NE Pacific Ocean (e.g. from 10 m at station P26, [Cuʹ] =18.7 fmol L-1, Semeniuk et al., 2016a). Hence, the activity of the Dokd-P16 strain and potentially, other Flavobacteriia, may be impaired in those waters. In contrast, Cu limitation did not affect the growth of the roseobacterium R. pomeroyi DSS-3 and Pseudoalteromonas spp. strains (PAlt-P2 and PAlt-P26), indicating a potential for these bacteria to out-compete Dokd-P16 when Cu is low. To avoid growth limitation under conditions of Cu availability, Pseudoalteromonas spp. strains maintain low Cu requirements regardless of Cu availability, while R. pomeroyi DSS-3 reduces its Cu use.    Results presented in Chapter 3 suggest that intracellular Cu in marine bacteria may be tightly regulated, given that Cu quotas of bacterial strains examined either remained constant (Pseudoalteromonas sp. strains) or varied by a factor of three (Dokd-P16 and R. pomeroyi DSS-3), relative to the 50-fold variation in [Cutot] in the growth media. These observations are consistent with the model of Cu homeostasis in reference organisms such as E. coli, which indicates that intracellular Cu is highly controlled by prokaryotic cells (see reviews of Rensing and Grass 2003; Solioz et al. 2010; Dupont et al. 2011; Bondarczuk and Piotrowska-seget 2013). Given the dangers 130  associated with free Cu (including oxidative stress and Fe-S cluster destruction), its use in biology has been limited in comparison to other metals such as Zn. For example, bacterial copper proteins represent less than 1% of the total proteome (Andreini et al. 2008), compared to 4.9% for Zn proteins (Andreini et al. 2006). In aerobic bacteria, the biological requirement for Cu is primarily associated with the cytochrome c oxidase (COX) (Ridge et al., 2008). However, there are some additional, although less common uses of Cu, such as polyphenol oxidation by copper-containing laccases (Ridge et al., 2008). These alternative Cu uses may allow some bacteria to exploit different ecological niches. Bacterial quota data from Cu-limiting experiments suggest that the minimum Cu requirements are quite similar in all strains [Cu:C ratios (μmol:  mol) were 1.3 ± 0.3 for Dokd-P16, 0.8 ± 0.3 for PAlt-P2, 1.4 ± 0.4 for PAlt-P26 and 0.6 ± 0.1 for R. pomeroyi DSS-3, for Cu:C comparisons refer to Fig. B.3 in the Appendix B]. Perhaps these values reflect the basic demand for the main cuproprotein-COX in aerobic bacteria. The fact that in Dok-P16 this Cu quota was not sufficient to allow maximal growth, compared to the other strains, may suggest that there is an additional Cu-dependent metabolic pathway in this strain.      Several ecological implications can be derived from research presented in Chapter 3). Firstly, my results indicate that Cu has the potential to limit the Flavobacteriia member Dokdonia sp. Dokd-P16 in situ. Bacteria of the genus Dokdonia may play an important role in proteorhodopsin-mediated bacterial phototrophy (Gómez-Consarnau et al. 2007; Gonzalez et al. 2011; Kimura et al. 2011; Riedel et al. 2013; Kim et al. 2016), a significant process in the ocean potentially impacting fluxes of C and energy (Béjà et al. 2001). Furthermore, these bacteria are typically characterized by their ability to degrade high-molecular-weight compounds (Gonzalez et al. 2011; Klippel et al. 2011), consistent with the adaptations of its ecologically significant phylum Bacteroidetes (Fernandez-Gomez et al. 2013), which may be dominated by Flavobacteriia (Cottrell and Kirchman 2000; Alonso et al. 2007). Despite the significance of Flavobacteriia such as Dokdonia spp. in the ocean, the role of trace metal availability in regulating their activity is not well understood. This thesis suggests that Cu is likely essential to the metabolism of Dokdonia spp. and its availability may impact the C transformations mediated by this bacterial group. Is high Cu requirement for growth and metabolism also a trait of other Flavobacteriia? A recent field study in the Southern Ocean found that increasing Cu availability had a positive impact on Flavobacteriia populations (Flavobacteria-Cytophagia cluster, Ramaiah et al., 2015), suggesting 131  that Cu may indeed be important to this microbial group. Further research is required to identify which critical processes are limited by Cu in Dokdonia sp. Dok-P16, which will help elucidate the interactions between Cu and Flavobacteriia.      In addition, my results highlight a potential a role of Cu in regulating the bacterial species composition, as well as the importance of heterotrophic bacteria in the Cu biochemical cycling. Heterotrophic bacteria that rely on high Cu availability for growth, such as Dokdonia sp. strain Dokd-P16 (Flavobacteriia), may be replaced by those with efficient strategies to maintain maximal growth when Cu is scare, such as R. pomeroyi DSS-3 (Roseobacteria) and Pseudoalteromonas sp. strains PAlt-P2 and PAlt-P26 (Alteromonadales). Furthermore, my estimates of the relative proportions of Cu associated with bacterial and phytoplankton biomass in surface waters at station P26 (NE Pacific) revealed that bacterial Cu could represent a significant fraction of the biogenic Cu in surface waters (~ 4 to 50%). The biomass of heterotrophic bacteria is not likely to be exported to greater depths; hence, bacterial accumulation of Cu (and likely other metals) could be most important for retaining Cu in surface waters. Copper accumulated into bacterial biomass could be released back into the dissolved pool during viral lysis of bacterial cells, or be transferred to organisms that graze on bacteria.    5.4 On the trace metal stoichiometry of marine heterotrophic bacteria    In Chapter 4 I further examined the effects of Cu on heterotrophic bacteria by assessing the changes in bioactive metal quotas (Fe, Zn, Mn, Co, and Cu) under conditions of varying Cu availability. The research presented in this chapter has implications for bacterial metal regulation and role of different metals in microbial ecology.    Findings presented here make the case for a well-controlled metal homeostasis in heterotrophic prokaryotes as observed for Cu in Chapter 3. On average, metals that varied significantly in response to different Cu treatments did so by a factor of < 3 in different strains (except for Zn in PAlt-P26, factor of 5), and patterns of metal abundance in all bacteria were independent of those in growth media. Metal variations observed in Chapter 4 may define the limits of metal contents in natural bacterioplankton. However, given the taxonomic limitation of 132  my dataset, our understanding of metal regulation in marine prokaryotes is still in its infancy, and advancing this field would require a broad survey of metal content in diverse prokaryotes. In contrast to marine heterotrophic bacteria, marine phytoplankton display a substantially broader variation in metal composition (by a factor of ~20 for most metals, Ho et al., 2003). This, in part, likely reflects a broader diversity of phytoplankton metal requirements, but may also indicate the greater flexibility of metal regulation in eukaryotic versus prokaryotic cells. The overall order of abundance for metals of interest in heterotrophic bacteria (Fe>Zn>Mn>Cu>Co for R. pomeroyi DSS-3 and Dok-P16, Fe>Zn>Mn≈Cu>Co in Pseudoalteromonas sp. strains) is consistent with model prokaryotes such as E. coli  (Outten and Halloran 2001; Cameron et al. 2012), and reflect the relative importance of these elements in bacterial physiology. However, I found that my bacteria differed in terms of their Mn content, with Pseudoalteromonas sp. strains being Mn depleted relative to the other strains. This likely reflects their low Mn requirements, which may be advantageous under conditions of low Mn availability.   The metal stoichiometry of marine heterotrophic bacteria presented in this thesis provides a baseline for future studies on the trace metal physiology of this important microbial group and will help in understanding their impacts on metal biogeochemical cycles. While other assessments of metal content in marine bacterioplankton have been carried out (Fe by Tortell et al., 1999; Maldonado and Price, 1999; Granger et al, 1999; Fourquez et al., 2014a; Fe, Zn, Mn, Cs, and Am by Vogel and Fisher, 2010), this study is the first to consider the elemental stoichiometry including major elements (C, N, S, P), and provides first measurements of Cu and Co. In addition, the effect of changing the availability of specific trace metals such as Cu on this extended elemental stoichiometry have not been examined before. In Chapter 3 I demonstrated that the composition of major nutrients in marine heterotrophic bacteria is essentially unchanged in response to Cu variability, which contrasts with the variable trace metal compositions observed in Chapter 4. Low Cu availability was associated with elevated quotas of specific metals, including Fe and Mn in Dokd-P16; Fe, Mn and in R. pomeroyi DSS-3 and Zn in the oceanic PAlt-P26. Interestingly, Co content in R. pomeroyi DSS-3 showed the opposite trend and was tightly linked to the variation in its Cu content. These variations may reflect changing physiological requirements under conditions of varying Cu availability or uptake via non-specific transporters. While the mechanisms responsible for bacterial metal quota responses need to be elucidated in a future study, my research 133  suggest that by dictating variable accumulation of metals in marine heterotrophic bacteria, Cu availability may influence the cycling of other metals, such as Fe, Mn, Zn, and Co.   5.5 Emerging questions on the biological utilization of metals     There have been conflicting ideas regarding the trends in metal composition between different microbial groups. For instance, it has been suggested that the metal content of algae and bacteria are similar (Barton et al. 2007). In contrast, other researchers have hypothesized that algal and bacterial metal requirements are different based on the content and structure of their metalloproteomes, and consideration of the contrasting trace metal availabilities in their evolutionary environments. For example, it has been hypothesized that Zn quotas may be lower in prokaryotes relative to eukaryotes, due to the lower abundance of Zn finger proteins in the former group (Twining and Bains, 2015). The higher abundance in Fe-, Mn-, and Co-binding proteins in prokaryotes relative to eukaryotes (Dupont et al., 2006) may suggest a higher requirement for these metals in prokaryotes. Furthermore, given the high sensitivity of modern autotrophic prokaryotes to Cu, which has been linked to low Cu availability during bacterial evolution (Brand et al. 1986; Saito et al. 2003), low Cu requirements may be expected in bacteria relative to eukaryotes which have evolved under conditions of higher Cu availability.     In Chapter 3 and 4, the elemental stoichiometry of marine heterotrophic bacteria was compared with that of eukaryotic phytoplankton cultured under similar conditions of trace metal availability. These preliminary comparisons indicate that while marine heterotrophic bacteria and eukaryotic phytoplankton contain similar amounts of Fe and Cu, bacteria appear to be relatively rich in Zn but depleted in Mn and Co compared to phytoplankton. This contradicts the hypothesis of the similar metal composition between those two microbial groups (Barton et al. 2007), and does not fit well with the trends in metal use inferred from metalloproteome studies and evolutionary considerations of metal availability (higher Fe, Mn, Co and lower Zn and Cu uses in prokaryotes relative to eukaryotes). There are some uncertainties in linking metal content obtained under metal-replete conditions with physiological requirements, as some organisms may accumulate metals in excess of their physiological needs (e.g. Fe in phytoplankton, Marchetti et al. 2009). Comparison of the growth responses and metal content of bacteria and phytoplankton 134  under conditions of minimum trace metal availability will help to quantify the relative importance of specific metals to those two groups.  5.6 Future directions     In this section, I highlight the key questions prompted by the research presented in this thesis and provide recommendations for future studies.  5.6.1  Cu sources and sinks along Line P in the subarctic NE Pacific    To constrain the processes controlling Cu cycling along Line P (Chapter 2) future research should focus on assessing dCu inputs from aeolian deposition and bottom sediments, as well as the dCu behavior in the intermediate waters of the OMZ. Analysis of satellite AOD data between 2003 and 2015 along Line P reveal annual, seasonal, and spatial variability in the atmospheric aerosols across this transect (Appendix A, Fig. A.2). Generally, the aerosol plumes tend to be associated with spring‒summer periods and are confined to the western section of the transect. Collection of aerosol samples along the transect during the July and August Line P cruises over several years could, therefore, provide valuable insight into the spatial and temporal trends in the potential aeolian deposition of Cu. Confirming Cu inputs from this source would require determining the Cu content of aerosols as well as its fractional solubility, coupled with dCu measurements in surface waters. In addition, analysis of air mass trajectories together with the aerosol content of Al (a tracer of crustal sources) and isotopic composition of stable isotopes of Pb (tracer of anthropogenic sources; eg. Ho et al. 2015; McAlister 2015), would aid in identifying the origin of aerosol derived Cu to Line P. It should be noted, that this work could be easily expanded to include other essential elements (e.g. Fe, Zn, Mn, Co, Mo, Ni), which may provide insights into the biochemical cycling and biological effects these other elements along Line P.    One of the limitations of this study was the lack of dCu measurements below 2000 m at most stations. These measurements are necessary to assess sources of Cu from bottom sediments along the transect (seafloor at 3000‒4000 m) and would also help with the interpretation of dCu behavior in the OMZ. Recent evidence suggests that within the OMZ, metals such as Cu may be sensitive to scavenging by sulfide within sinking particles (Janssen et al. 2014, Janssen & Cullen, 2015). 135  Dissolved Cu concentrations were uniform within the OMZ waters of Line P, in contrast to the linear slope of dCu towards the bottom typically observed in other regions. In addition, Cu:Si relationship was strongly decoupled in the OMZ, thus supporting the hypothesis of Cu loss. However, there are uncertainties in the interpretation of dCu behavior in low O2 waters in the NE Pacific OMZ, and to elucidate the process described by Janssen et al., 2014, measurements of the particulate Cu and its stable isotopic composition are required.     Research presented in this thesis suggests that assimilation of Cu by marine heterotrophic bacteria could represent a potentially significant component of the Cu biogeochemical cycle (Chapter 3). To substantiate these findings, future studies should include field assessments of Cu partitioning between biogenic (living) pools in surface waters. An elegant approach, first used by Tortell et al. (1996), to measure Fe:C ratios of phytoplankton and bacteria in the NE Pacific could be applied. This approach involved incubation of surface waters with 55Fe and H14CO3 over 6 days, allowing the labelling of phytoplankton-derived organic carbon incorporated into marine heterotrophic bacteria with 14C (bacterial C quotas). Similar experiments could be performed using the radioisotope 67Cu (half-life = 2.58 days) in surface waters along Line P.  5.6.2 What is the role of Cu in Dokdonia sp. Dokd-P16, and other Flavobacteriia members?    Elucidating the causes of Cu limitation in Dokdonia sp. strain Dokd-P16 is of interest given the ecological significance of Flavobacteriia, including those of the genus Dokdonia. Genome analysis could be performed to identify the genes encoding Cu-dependent pathways in this strain, while proteomic analysis of Cu sufficient and deficient Dokd-P16 under conditions of sufficiency and deficiency would help in understanding how the response to Cu availability is regulated on a molecular level. It would be interesting to assess if the effects of Cu on Dokd-P16 are related to its unique properties such as pigmentation, given the involvement of multi-copper oxidases in pigment biosynthesis in some bacteria (Givaudan et al. 1993). Carbon metabolic rates in Dokd-P16 were severely impaired under conditions of low Cu availability, and examining the expression of proteins in the C metabolic pathways is likely to offer some insights into the role of Cu in this strain (as has been recently done for Fe, Fourquez et al. 2014a). Once the processes underpinning 136  Cu-dependence in Dokd-P16 are identified, the genomes of other Flavobacteriia (e.g. Dokdonia sp. strain 4H-3-7-5, Klippel et al., 2010; Dokdonia sp. strain MED 132, Riedel et al. 2013) could be screened to evaluate if Cu-dependence is a feature of bacteria belonging to this ecologically significant microbial group.  5.6.3 Variations in metals content in marine heterotrophic bacteria    To advance our understanding of the trace metal requirements of marine heterotrophic bacteria, as presented in Chapters 3 and 4, there is a need for a broad survey of metal content in taxonomically diverse species. As the findings of my research indicate, metal content of cells may be maintained within a narrow range in heterotrophic bacteria (compared to phytoplankton); however, these observations are based on only 4 strains. To assess the diversity of metal content in heterotrophic bacteria, future work should use isolates with contrasting phylogenies, ecologies (coastal & oceanic) and life strategies. Bacteria used in this thesis are copiotrophic (prefer nutrient rich environments), as are most bacteria that can be successfully cultured. It is quite likely that their metal requirements may differ from those adapted to life under oligotrophic (low nutrient) conditions. Such differences are seen in the trace metal acquisition strategies of copiotrophic (Roseobacter clade) and oligotrophic (SAR11 clade) bacteria (Hogle et al. 2016), but the differences in their metal content remain unknown. Future studies should, therefore, consider including oligotrophic strains such as the cultivated member of the SAR11 clade Candidatus ‘Pelagibacter ubique’, in the studies of the trace metal requirements of marine bacterioplankton.     5.6.4 Identification of mechanisms involved in metal accumulation by marine heterotrophic bacteria in response to Cu     In Chapter 4 several metals were found to vary with changes in Cu availability in heterotrophic bacteria depending on the bacterial strain. In general, for metals other than Cu, decreasing Cu availability in the media was accompanied by increases in quotas; Fe and Mn in Dokdonia sp. strain Dokd-P16, Fe and Mn in R. pomeroyi DSS-3 and Zn in oceanic 137  Pseudoalteromonas sp. strain PAlt-P26. Cobalt was the only element, whose quota was reduced under low Cu availability (R. pomeroyi DSS-3), mimicking the trend of Cu quotas. There is uncertainty regarding the specific mechanisms responsible for these metal quota variations, but they may be driven by physiological requirements for metalloproteins in response to Cu limitation or non-specific metal uptake. These mechanisms could be elucidated with a comparative proteomic study of bacterial strains under Cu sufficiency and deficiency, targeting the expression of Fe, Mn, and Zn metalloproteins as well as their metal transporters (to confirm whether these metals are acquired via specific transport systems). Currently, little is known about how Cu is acquired in different heterotrophic bacteria. The observations of Cu and Co covariation in R. pomeroyi DSS-3 is exciting, as it suggests that the homeostasis of both metals may be connected. It would be interesting to examine potential Cu acquisition through Co uptake systems in this strain in a future study.                   138                         Figure 5.1 Schematic representation of the Cu biogeochemical cycling along the Line P transect in the subarctic NE Pacific. Major sources of dCu to surface waters along Line P are upwelling within the Alaska gyre to the east and coastal runoff to the west. Coastal margins may also serve as deep dCu source at Station P4. Atmospheric deposition during summer months may play role in supplying Cu to the eastern most sections of Line P. In upper waters, Cu is subject to removal by active biological uptake by the resident microbial community. The role of heterotrophic bacteria in Cu acquisition and remineralization is largely unknown, but bacterially accumulated Cu is likely to remain in the surface waters and serve as a recycled Cu source. At intermediate depths, Cu is remineralized by bacterial decomposition of biogenic material, with a Cu:P signature close to the average ratio of cultured phytoplankton (0.35 mmol: mol, Ho et al., 2003). DCu inventory in the subarctic NE Pacific could be reduced by scavenging processes within the Oxygen Minimum Zone (OMZ). The processes that affect Cu cycling along Line P that remain to be elucidated (indicated with question marks) in a future study are atmospheric deposition, dCu inputs from marine sediments, and dCu retention in surface waters by marine heterotrophic bacteria.      139  References  Abell, G. C. J., and J. P. Bowman. 2005. Ecological and biogeographic relationships of class Flavobacteria in the Southern Ocean. FEMS Microbiol. Ecol. 51: 265–277. doi:10.1016/j.femsec.2004.09.001 Achterberg, E. P., and C. M. G. van den Berg. 1994. In-line ultraviolet-digestion of natural water samples for trace metal determination using an automated voltammetric system. Anal. Chim. Acta 291: 213–232. doi:10.1016/0003-2670(94)80017-0 Achterberg, E. P., C. B. Braungardt, R. C. Sandford, and P. J. Worsfold. 2001. UV digestion of seawater samples prior to the determination of copper using flow injection with chemiluminescence detection. Anal. Chim. Acta 440: 27–36n. doi:10.1016/S0003-2670(01)00824-8 Ahrland,  a, J. Chatt, and N. R. Davies. 1958. The relative affinities of ligand atoms for acceptors. Quart. Rev. Chem. Soc 12: 265. doi:10.1039/QR9581200265 Alonso, C., F. Warnecke, R. Amann, and J. Pernthaler. 2007. High local and global diversity of Flavobacteria in marine plankton. Environ. Microbiol. 9: 1253–1266. doi:10.1111/j.1462-2920.2007.01244.x Amin, S. A., D. H. Green, M. C. Hart, F. C. Küpper, W. G. Sunda, and C. J. Carrano. 2009. Photolysis of iron-siderophore chelates promotes bacterial-algal mutualism. Proc. Natl. Acad. Sci. U. S. A. 106: 17071–6. doi:10.1073/pnas.0905512106 Amin, S. A., J. W. Moffett, W. Martens-Habbena, and others. 2013. Copper requirements of the ammonia-oxidizing archaeon Nitrosopumilus maritimus SCM1 and implications for nitrification in the marine environment. Limnol. Oceanogr. 58: 2037–2045. doi:10.4319/lo.2013.58.6.2037 Andreini, C., L. Banci, I. Bertini, and A. Rosato. 2006. Zinc through the three domains of life. J. Proteome Res. 5: 3173–3178. doi:10.1021/pr0603699 Andreini, C., L. Banci, I. Bertini, and A. Rosato. 2008. Occurrence of Copper Proteins through the Three Domains of Life : A Bioinformatic Approach research articles. J. Proteome Res. 7: 209–216. Andrews, S. C., A. K. Robinson, and F. Rodríguez-Quiñones. 2003. Bacterial iron homeostasis. FEMS Microbiol. Rev. 27: 215–237. doi:10.1016/S0168-6445(03)00055-X Annett, A. L., S. Lapi, T. J. Ruth, and M. T. Maldonado. 2008. The effects of Cu and Fe availability on the growth and Cu:C ratios of marine diatoms. Limnol. Oceanogr. 53: 2451–2461. doi:10.4319/lo.2008.53.6.2451 Arciero, D. M., B. S. Pierce, M. P. Hendrich, and A. B. Hooper. 2002. Nitrosocyanin, a red cupredoxin-like protein from Nitrosomonas europaea. Biochemistry 41: 1703–1709. doi:10.1021/bi015908w Argüello, J. M., D. Raimunda, T. Padilla-benavides, and G. Grass. 2013. Mechanisms of copper homeostasis in bacteria. Front. Cell. Infect. Microbiol. 3: 1–14. doi:10.3389/fcimb.2013.00073 Aristilde, L., Y. Xu, and F. M. M. Morel. 2012. Weak Organic Ligands Enhance Zinc Uptake in Marine Phytoplankton. Environ. Sci. Technol. 5438–5445. doi:10.1021/es300335u Armstrong, E., J. Granger, E. L. Mann, and N. M. Price. 2004. Outer-membrane siderophore receptors of heterotrophic oceanic bacteria. Limnol. Oceanogr. 49: 579–587. doi:10.4319/lo.2004.49.2.0579 Arrigo, K. R. 2005. Marine microorganisms and global nutrient cycles. Nature 437: 2192–2205. doi:10.1038/nature0415 Azam, F., T. Fenchel, J. G. Field, J. S. Gray, L. A. Meyer-Reil, and F. Thingstad. 1983. The Ecological Role of Water-Column Microbes in the Sea. Mar. Ecol. Prog. Ser. 10: 257–263. Azam, F., and F. Malfatti. 2007. Microbial structuring of marine ecosystems. Nat. Rev. Microbiol. 5: 782–91. doi:10.1038/nrmicro1747 de Baar, H. J. W., J. T. M. de Jong, D. C. E. Bakker, B. M. Löscher, C. Veth, U. V. Bathmann, and V. Smetacek. 1995. Importance of iron for plankton blooms and carbon dioxide drawdown in the 140  Southern Ocean. Nature 373: 412–415. doi:10.1038/373412a0 Balasubramanian, R., G. E. Kenney, and A. C. Rosenzweig. 2011. Dual pathways for copper uptake by methanotrophic bacteria. J. Biol. Chem. 286: 37313–37319. doi:10.1074/jbc.M111.284984 Balasubramanian, R., and  a. C. Rosenzweig. 2007. Structural and mechanistic insights into methane oxidation by particulate methane monooxygenase. Acc Chem Res 40: 573–80. doi:10.1021/ar700004s Banerjee, R., and S. W. Ragsdale. 2003. The Many Faces of Vitamin B 12 : Catalysis by Cobalamin-Dependent Enzymes1. Annu. Rev. Biochem. 72: 209–247. doi:10.1146/annurev.biochem.72.121801.161828 Barton, L. L., F. Goulhen, M. Bruschi, N. A. Woodards, R. M. Plunkett, and F. J. M. Rietmeijer. 2007. The bacterial metallome: Composition and stability with specific reference to the anaerobic bacterium Desulfovibrio desulfuricans. BioMetals 20: 291–302. doi:10.1007/s10534-006-9059-2 Barwell-Clarke, J., and F. Whitney. 1996. Institute of Ocean Sciences Nutrient Methods and Analysis. Canadian Technical Report of Hydrography and Ocean Sciences, No. 182, 43 pp. Béjà, O., E. N. Spudich, J. L. Spudich, M. Leclerc, and E. F. DeLong. 2001. Proteorhodopsin phototrophy in the ocean. Nature 411: 786–789. doi:10.1038/35081051 Béjà, O., and M. T. Suzuki. 2008. Photoheterotrophic Marine Prokaryotes. Microb. Ecol. Ocean. Second Ed. 131–157. doi:10.1002/9780470281840.ch5 van den Berg, C. M. . 1982. Determination of copper complexation with natural organic ligands in seawater by equilibration with MnO2 II. Experimental procedures and application to surface seawater. Mar. Chem. 11: 323–342. doi:10.1016/0304-4203(82)90029-9 van den Berg, C. M. G. G. 2017. UV digestion apparatus. http://pcwww.liv.ac.uk/~sn35/Site/UV_digestion_apparatus.html. Bergquist, B. A., and E. A. Boyle. 2006. Dissolved iron in the tropical and subtropical Atlantic Ocean. Global Biogeochem. Cycles 20: 1–14. doi:10.1029/2005GB002505 Bertrand, E. M., M. a Saito, P. a Lee, R. B. Dunbar, P. N. Sedwick, and G. R. Ditullio. 2011. Iron limitation of a springtime bacterial and phytoplankton community in the ross sea: implications for vitamin b(12) nutrition. Front. Microbiol. 2: 160. doi:10.3389/fmicb.2011.00160 Bidle, K. D., M. a. Brzezinski, R. a. Long, J. L. Jones, and F. Azam. 2003. Diminished efficiency in the oceanic silica pump caused by bacterially-mediated silica dissolution. Limnol. Oceanogr. 48: 1855–1868. doi:10.4319/lo.2003.48.5.1855 Biller, D. V., and K. W. Bruland. 2012. Analysis of Mn, Fe, Co, Ni, Cu, Zn, Cd, and Pb in seawater using the Nobias-chelate PA1 resin and magnetic sector inductively coupled plasma mass spectrometry (ICP-MS). Mar. Chem. 130–131: 12–20. doi:10.1016/j.marchem.2011.12.001 Biller, D. V., and K. W. Bruland. 2013. Sources and distributions of Mn, Fe, Co, Ni, Cu, Zn, and Cd relative to macronutrients along the central California coast during the spring and summer upwelling season. Mar. Chem. 155: 50–70. doi:10.1016/j.marchem.2013.06.003 Bishop, J. K. B., R. E. Davis, and J. T. Sherman. 2002. Robotic Observations of Dust Storm Enhancement of Carbon Biomass in the North Pacific. Science (80-. ). 298: 817–821. doi:DOI: 10.1126/science.1074961 Bogachev, A. V, Y. V Bertsova, M. L. Verkhovskaya, M. D. Mamedov, and V. P. Skulachev. 2016. Real-time kinetics of electrogenic Na(+) transport by rhodopsin from the marine flavobacterium Dokdonia sp. PRO95. Sci. Rep. 6: 21397. doi:10.1038/srep21397 Bondarczuk, K., and Z. Piotrowska-seget. 2013. Molecular basis of active copper resistance mechanisms in Gram-negative bacteria. Cell Biol. Toxicol. 29: 397–405. doi:10.1007/s10565-013-9262-1 Bowie, A. R., P. Van Der Merwe, F. Quéroué, and others. 2015. Iron budgets for three distinct biogeochemical sites around the Kerguelen Archipelago (Southern Ocean) during the natural fertilisation study, KEOPS-2. Biogeosciences 12: 4421–4445. doi:10.5194/bg-12-4421-2015 Boyd, P., and P. J. Harrison. 1999. Phytoplankton dynamics in the NE subarctic Pacific. Deep. Res. Part II Top. Stud. Oceanogr. 46: 2405–2432. doi:10.1016/S0967-0645(99)00069-7 141  Boyd, P. W., C. S. Wong, J. Merrill, F. Whitney, J. Snow, P. J. Harrison, and B. E. T. Al. 1998. Atmospheric iron supply and enhanced vertical carbon flux in the NE subarctic Pacific: Is there a connection? Global Biogeochem. Cycles 12: 429–441. Boye, B. D. Wake, P. Lopez Garcia, and others. 2012. Distributions of dissolved trace metals (Cd, Cu, Mn, Pb, Ag) in the southeastern Atlantic and the Southern Ocean. Biogeosciences 9: 3231–3246. doi:10.5194/bg-9-3231-2012 Boyle, E. ., F. R. Sclater, and J. M. Edmond. 1977. The distribution of dissolved copper in the Pacific. Earth Planet. Sci. Lett. 37: 38–54. Brand, L. E., W. G. Sunda, and R. R. L. Guillard. 1986. Reduction of marine phytoplankton reproduction rates by copper and cadmium. J. Exp. Mar. Bio. Ecol. 96: 225–250. doi:10.1016/0022-0981(86)90205-4 Brown, M. T., S. M. Lippiatt, M. C. Lohan, and K. W. Bruland. 2012. Trace metal distributions within a Sitka eddy in the northern Gulf of Alaska. Limnol. Oceanogr. 57: 503–518. doi:10.4319/lo.2012.57.2.0503 Bruland, K., and R. Franks. 1983. Mn, Ni, Cu, Zn and Cd in the western North Atlantic, p. 395–414. In C.S. Wong, E. Boyle, K.W. Bruland, J.D. Burton, and E.D. Goldberg [eds.], Trace metals in sea water. Springer US. Bruland, K. K. W. 1980. Oceanographic distributions of cadmium, zinc, nickel, and copper in the North Pacific. Earth Planet. Sci. Lett. 47: 176–198. doi:10.1016/0012-821X(80)90035-7 Bruland, K. W., and M. C. Lohan. 2003. Controls of Trace Metals in Seawater, p. 23–47. In Treatise on Geochemistry,Volume 6. Editor: Henry Elderfield. Executive Editors: Heinrich D. Holland and Karl K. Turekian. pp. 625. ISBN 0-08-043751-6. Elsevier. Bruland, K. W., R. Middag, and M. C. Lohan. 2014. Controls of Trace Metals in Seawater, 2nd ed. Elsevier Ltd. Brussaard, C. P. D. 2004. Optimization of Procedures for Counting Viruses by Flow Cytometry. Appl. Environmantal Microbiol. 70: 1506–1513. doi:10.1128/AEM.70.3.1506 Buchan, A., J. M. González, and M. A. Moran. 2005. Overview of the marine Roseobacter lineage. Appl. Environ. Microbiol. 71: 5665–5677. doi:10.1128/AEM.71.10.5665 Buchan, A., G. R. LeCleir, C. A. Gulvik, and J. M. González. 2014. Master recyclers: features and functions of bacteria associated with phytoplankton blooms. Nat. Rev. Microbiol. 12: 686–698. Buck, K. K. N., K. K. E. Selph, and K. a. Barbeau. 2010. Iron-binding ligand production and copper speciation in an incubation experiment of Antarctic Peninsula shelf waters from the Bransfield Strait, Southern Ocean. Mar. Chem. 122: 148–159. doi:10.1016/j.marchem.2010.06.002 Buck, K. N., and K. W. Bruland. 2005. Copper speciation in San Francisco Bay : A novel approach using multiple analytical windows. Mar. Chem. 96: 185–198. doi:10.1016/j.marchem.2005.01.001 Bundy, R. M., K. a. Barbeau, and K. N. Buck. 2012. Sources of strong copper-binding ligands in Antarctic Peninsula surface waters. Deep Sea Res. Part II Top. Stud. Oceanogr. 1–13. doi:10.1016/j.dsr2.2012.07.023 Butler, E. C. V, J. E. O’Sullivan, R. J. Watson, A. R. Bowie, T. A. Remenyi, and D. Lannuzel. 2013. Trace metals Cd, Co, Cu, Ni, and Zn in waters of the subantarctic and Polar Frontal Zones south of Tasmania during the “SAZ-Sense” project. Mar. Chem. 148: 63–76. doi:10.1016/j.marchem.2012.10.005 Cain, A. N. 2014. Dissolved Aluminum in the Northeast Pacific and the Western Arctic Ocean. MSc Thesis, University of British Columbia, pp 95-100. Cameron, V., C. H. House, and S. L. Brantley. 2012. A first analysis of metallome biosignatures of hyperthermophilic archaea. Archaea 2012. doi:10.1155/2012/789278 Campos, L. . A. M., and C. M. G. van den Berg. 1994. Determination of copper complexation in sea water by cathodic stripping voltammetry and ligand competition with salicylaldoxime. Anal. Chim. Acta 284: 481–496. doi:10.1016/0003-2670(94)85055-0 Carlson, C., P. del Giorgio, and G. Herndl. 2007. Microbes and the Dissipation of Energy and Respiration: 142  From Cells to Ecosystems. Oceanography 20: 89–100. doi:10.5670/oceanog.2007.52 Cavet, J. S., G. P. M. Borrelly, and N. J. Robinson. 2003. Zn, Cu and Co in cyanobacteria: Selective control of metal availability. FEMS Microbiol. Rev. 27: 165–181. doi:10.1016/S0168-6445(03)00050-0 Chan, L.-K., R. J. Newton, S. Sharma, C. B. Smith, P. Raypati, A. J. Limardo, C. Meile, and M. A. Moran. 2012. Transcriptional changes underlying elemental stoichiometry shifts in a marine heterotrophic bacterium. Front. Microbiol. 3: 1–24. doi:10.3389/fmicb.2012.00159 Changela, A., K. Chen, Y. Xue, J. Holschen, C. E. Outten, and T. V. O. Halloran. 2003. Molecular Basis of Metal-Ion Selectivity and Zeptomolar Sensitivity by CueR. Science (80-. ). 301: 1383–1386. Cho, B., and F. Azam. 1990. Biogeochemical significance of bacterial biomass in the ocean’s euphotic zone . Mar. Ecol. Prog. Ser. 63: 253–259. doi:10.3354/meps063253 Chrzanowski, T. H., and J. P. Grover. 2008. Element content of Pseudomonas fluorescens varies with growth rate and temperature : A replicated chemostat study addressing ecological stoichiometry. Limnol. Oceanogr. 53: 1242–1251. Claus, H., and Z. Filip. 1997. The evidence of laccase-like enzyme activity in Bacillus sphaericus strain. Microbiol. Res. 152: 209–216. doi:10.1016/S0944-5013(97)80014-6 Coale, K. H., and K. W. Bruland. 1990. Spatial and temporal variability in copper complexation in the North Pacific. Deep Sea Res. Part A. Oceanogr. Res. Pap. 37: 317–336. doi:10.1016/0198-0149(90)90130-N Coale, K. H., K. W. Bruland, and W. B. Kenneth. 1988. Copper complexation in the Northeast Pacific. Limnol. Oceanogr. 33: 1084–1101. doi:10.4319/lo.1988.33.5.1084 Cobbett, C., and P. Goldsbrough. 2002. Phytochelatins and metallothioneins : Roles in heavy metal detoxification and homeostasis. Annu. Rev. Plant Biol. 53: 159–182. Cotner, J. B., W. Makino, and B. A. Biddanda. 2006. Temperature Affects Stoichiometry and Biochemical Composition of Escherichia coli. Microb. Ecol. 52: 26–33. doi:10.1007/S00248-006-9040-1 Cottrell, M. T., and D. L. Kirchman. 2000. Natural assemblages of marine proteobacteria adn members of the Cytophaga-Flavobacter clustler consuming low- and high- molecular-weight dissolved organic matter. Appl. Environ. Microbiol. 66: 1692–1697. doi:10.1128/AEM.66.4.1692-1697.2000.Updated Cottrell, M. T., and D. L. Kirchman. 2003. Contribution of major bacterial groups to bacterial biomass production (thymidine and leucine incorporation) in the Delaware estuary. Limnol. Oceanogr. 48: 168–178. doi:10.4319/lo.2003.48.1.0168 Croot, P. L., B. Karlson, J. T. ven Elteren, and J. J. Kroon. 2003. Uptake and efflux of 64 Cu by the marine cyanobacterium Synechococcus ( WH7803 ). Limnol. Oceanogr. 48: 179–188. Crusius, J., A. W. Schroth, S. Gass, C. M. Moy, R. C. Levy, and M. Gatica. 2011. Glacial flour dust storms in the Gulf of Alaska: Hydrologic and meteorological controls and their importance as a source of bioavailable iron. Geophys. Res. Lett. 38: 1–5. doi:10.1029/2010GL046573 Cullen, J. T., M. Chong, and D. Ianson. 2009. British columbian continental shelf as a source of dissolved iron to the subarctic northeast Pacific Ocean. Global Biogeochem. Cycles 23: 1–12. doi:10.1029/2008GB003326 Cunliffe, M. 2013. Physiological and metabolic effects of carbon monoxide oxidation in the model marine bacterioplankton Ruegeria pomeroyi DSS-3. Appl. Environ. Microbiol. 79: 738–740. doi:10.1128/AEM.02466-12 Cunliffe, M. 2016. Purine catabolic pathway revealed by transcriptomics in the model marine bacterium Ruegeria pomeroyi DSS-3. FEMS Microbiol. Ecol. 92: 1–6. doi:10.1093/femsec/fiv150 Cunningham, B. R., and S. G. John. 2017. The effect of iron limitation on cyanobacteria major nutrient and trace element stoichiometry. Limnol. Oceanogr. 846–858. doi:10.1002/lno.10484 Debelius, B., J. M. Forja, and L. M. Lubián. 2011. Toxicity of copper, nickel and zinc to Synechococcus populations from the Strait of Gibraltar. J. Mar. Syst. 88: 113–119. doi:10.1016/j.jmarsys.2011.02.009 Decaria, L., I. Bertini, and R. J. P. Williams. 2011. Zinc proteomes, phylogenetics and evolution. Metallomics 3: 56–60. doi:10.1039/C0MT00045K 143  Deckert, G., P. V Warren, T. Gaasterland, and others. 1998. The complete genome of the hyperthermophilic bacterium Aquifex aeolicus. Nature 392: 353–358. doi:10.1038/32831 Duce, R. A., J. LaRoche, K. Altieri, and others. 2008. Impacts of Atmospheric Anthropogenic Nitrogen on the Open Ocean. Science (80-. ). 320: 893–897. doi:10.1126/science.1150369 Duce, R. A., P. S. Liss, J. T. Merrill, and others. 1991. The atmospheric input of trace species to the world ocean. Global Biogeochem. Cycles 5: 193–259. doi:10.1029/91GB01778 Duce, R. a, and N. W. Tindale. 1991. Atmospheric transport of iron and its deposition in the ocean. Limnol. Oceanogr. 36: 1715–1726. doi:10.4319/lo.1991.36.8.1715 Ducklow, H. W. 1999. Minireview: The bacterial content of the oceanic euphotic zone. FEMS Microbiol. 30: 1–10. doi:10.1016/S0168-6496(99)00031-8 Dupont, C. L., G. Grass, and C. Rensing. 2011. Copper toxicity and the origin of bacterial resistance — new insights and applications. Metallomics 3: 1109–1118. doi:10.1039/c1mt00107h Dupont, C. L., S. Yang, B. Palenik, and P. E. Bourne. 2006. Modern proteomes contain putative imprints of ancient shifts in trace metal geochemistry. Proc. Natl. Acad. Sci. U. S. A. 103: 17822–7. doi:10.1073/pnas.0605798103 Durham, B. P., S. Sharma, H. Luo, and others. 2015. Cryptic carbon and sulfur cycling between surface ocean plankton. Proc. Natl. Acad. Sci. 112: 453–457. doi:10.1073/pnas.1413137112 Eilers, H., J. Pernthaler, F. Glöckner, and R. Amann. 2000. Culturability and in situ abundance of pelagic bacteria from the North Sea. Appl Env. Microbiol 66: 3044–3051. doi:10.1128/AEM.66.7.3044-3051.2000 Ekici, S., H. Yang, H. G. Koch, and F. Daldal. 2012. Novel transporter required for biogenesis of cbb3-type cytochrome c oxidase in Rhodobacter capsulatus. MBio 3: 1–11. doi:10.1128/mBio.00293-11 Elser, J. J., K. Acharya, M. Kyle, and others. 2003. Growth rate – stoichiometry couplings in diverse biota. Ecol. Lett. 6: 936–943. doi:10.1046/j.1461-0248.2003.00518.x Elser, J. J., R. W. Sterner, E. Gorokhova, and others. 2000. Biological stoichiometry from genes to ecosystems. Ecol. Lett. 3: 540–550. Emerson, S., and J. Hedges. 2008. Chemical Oceanography and the Marine Carbon Cycle, 1st ed. Cambridge University Press, pp 13. Ezoe, M., T. Ishita, M. Kinugasa, X. Lai, K. Norisuye, and Y. Sohrin. 2004. Distributions of dissolved and acid-dissolvable bioactive trace metals in the North Pacific. Geochem. J. 38: 535–550. doi:10.2343/geochemj.38.535 Fagerbakke, K. M., M. Heldal, and S. Norland. 1996. Content of carbon, nitrogen, oxygen, sulfur and phosphorus in native aquatic and cultured bacteria. Aquat. Microb. Ecol. 10: 15–27. doi:10.3354/ame010015 Fernandez-Gomez, B., M. Richter, M. Schuler, J. Pinhassi, S. G. Acinas, J. M. Gonzalez, and C. Pedros-Alio. 2013. Ecology of marine Bacteroidetes: a comparative genomics approach. ISME J. 7: 1026–1037. doi:10.1038/ismej.2012.169 Ferrera, I., J. M. Gasol, M. Sebastia, E. Hojerova, and M. Koblíz. 2011. Comparison of Growth Rates of Aerobic Anoxygenic Phototrophic Bacteria and Other Bacterioplankton Groups in Coastal Mediterranean Waters ᰔ. Appl. Environmantal Microbiol. 77: 7451–7458. doi:10.1128/AEM.00208-11 Finkel, Z. V., A. Quigg, J. a. Raven, J. R. Reinfelder, O. E. Schofield, and P. G. Falkowski. 2006. Irradiance and the elemental stoichiometry of marine phytoplankton. Limnol. Ocean. 51: 2690–2701. doi:10.4319/lo.2006.51.6.2690 Fischer, K., J. Dymond, M. Lyle, A. Soutar, and S. Rau. 1986. The benthic cycle of copper : Evidence from sediment trap experiments in the eastern tropical North Pacific Ocean. Geochim. Cosmochim. Acta 50: 1535–1543. Fourquez, M., A. Devez, A. Schaumann, A. Guéneuguès, T. Jouenne, I. Obernosterer, and S. Blain. 2014a. Effects of iron limitation on growth and carbon metabolism in oceanic and coastal heterotrophic 144  bacteria. Limnol. Oceanogr. 59: 349–360. doi:10.4319/lo.2014.59.2.0349 Fourquez, M., I. Obernosterer, D. M. Davies, T. W. Trull, and S. Blain. 2014b. Microbial iron uptake in the naturally fertilized waters in the vicinity of Kerguelen Islands: phytoplankton–bacteria interactions. Biogeosciences Discuss. 11: 15053–15086. doi:10.5194/bgd-11-15053-2014 Francis, C. A., and B. M. Tebo. 2001. cumA Multicopper Oxidase Genes from Diverse Mn ( II ) -Oxidizing and cumA Multicopper Oxidase Genes from Diverse Mn ( II ) -Oxidizing and Non-Mn ( II ) -Oxidizing Pseudomonas Strains. Appl. Environmantal Microbiol. 67: 4272–4278. doi:10.1128/AEM.67.9.4272 Frangipani, E., V. I. Slaveykova, C. Reimmann, and D. Haas. 2008. Adaptation of aerobically growing Pseudomonas aeruginosa to copper starvation. J. Bacteriol. 190: 6706–17. doi:10.1128/JB.00450-08 Fujishima, Y., K. Ueda, M. Mauro, E. Nakayama, C. Tokutome, H. Hasegawa, M. Matsui, and Y. Sohrin. 2001. Distribution of Trace Bioelements in the Subarctic North Pacific Ocean and the Bering Sea (the R/V Hakuho Maru Cruise KH-97-2). J. Oceanogr. 57: 261–273. Gilch, S., O. Meyer, and I. Schmidt. 2010. Electron paramagnetic studies of the copper and iron containing soluble ammonia monooxygenase from Nitrosomonas europaea. BioMetals 23: 613–622. doi:10.1007/s10534-010-9308-2 del Giorgio, P. A., and J. J. Cole. 1998. Bacterial Growth Efficiency in Natural Aquatic Systems. Annu. Rev. Ecol. Syst. 29: 503–541. doi:10.1146/annurev.ecolsys.29.1.503 Giovannoni, S., and M. Rappé. 2000. Evolution, Diversity, and Molecular Ecology of Marine Prokaryotes, p. 47–84. In D.L. Kirchman [ed.], Microbial Ecology of the Oceans. Wiley-Liss. Givaudan, A., A. Effosse, D. Faure, P. Potier, M.-L. Bouillant, and R. Bally. 1993. Polyphenol oxidase in Azospirillum lipoferum isolated from rice rhizosphere: Evidence for laccase activity in non-motile strains of Azospirillum lipoferum. FEMS Microbiol. Lett. 108: 205–210. doi:Doi 10.1016/0378-1097(93)90586-Q Gladyshev, V. N., and Y. Zhang. 2013. Comparative genomics analysis of the metallomes, p. 529–574. In L. Banci [ed.], Metallomics and the cell. Springer Netherlands. Glass, J. B., and V. J. Orphan. 2012. Trace metal requirements for microbial enzymes involved in the production and consumption of methane and nitrous oxide. Front. Microbiol. 3: 61. doi:10.3389/fmicb.2012.00061 Glöckner, F. O., B. M. Fuchs, and R. Amann. 1999. Bacterioplankton Compositions of Lakes and Oceans : a First Comparison Based on Fluorescence In Situ Hybridization. Appl. Environ. Microbiol. 65: 3721–3726. Godwin, C. M., and J. B. Cotner. 2015a. Stoichiometric flexibility in diverse aquatic heterotrophic bacteria is coupled to differences in cellular phosphorus quotas. Front. Microbiol. 6: 1–15. doi:10.3389/fmicb.2015.00159 Godwin, C. M., and J. B. Cotner. 2015b. Aquatic heterotrophic bacteria have highly flexible phosphorus content and biomass stoichiometry. ISME J. 9: 1–4. doi:10.1038/ismej.2015.34 Goldman, J. C., D. A. Caron, and M. R. Dennett. 1987. Regulation of Gross Growth Efficiency and Ammonium Regeneration in Bacteria by Substrate C : N Ratio. Limnol. Oceanogr. 32: 1239–1252. Gómez-Consarnau, L., J. M. González, M. Coll-Lladó, P. Gourdon, T. Pascher, R. Neutze, C. Pedrós-Alió, and J. Pinhassi. 2007. Light stimulates growth of proteorhodopsin-containing marine Flavobacteria. Nature 445: 210–3. doi:10.1038/nature05381 González, J. M., and M. A. Moran. 1997. Numerical dominance of a group of marine bacteria in the α-subclass of the class Proteobacteria in coastal seawater. Appl. Environ. Microbiol. 63: 4237–4242. Gonzalez, J. M., J. Pinhassi, B. Fernadnez-Gomez, and others. 2011. Genomics of the Proteorhodopsin-Containing Marine Flavobacterium. Appl. Environ. Microbiol. 77: 8676–8686. doi:10.1128/AEM.06152-11 Gordon, A. S., J. R. Donat, R. a Kango, B. J. Dyer, and L. M. Stuart. 2000. Dissolved copper-complexing ligands in cultures of marine bacteria and estuarine water. Mar. Chem. 70: 149–160. 145  doi:10.1016/S0304-4203(00)00019-0 Gordon,  a S., V. J. Harwood, and S. Sayyar. 1993. Growth, copper-tolerant cells, and extracellular protein production in copper-stressed chemostat cultures of Vibrio alginolyticus. Appl. Environ. Microbiol. 59: 60–6. Gordon,  a S., L. D. Howell, and V. Harwood. 1994. Responses of diverse heterotrophic bacteria to elevated copper concentrations. Can. J. Microbiol. 40: 408–11. Granger, J., and N. M. Price. 1999. The importance in iron nutrition of siderophores of heterotrophic marine bacteria. Limnol. Oceanogr. 44: 541–555. Granger, J., and B. Ward. 2003. Accumulation of nitrogen oxides in copper-limited cultures of denitrifying bacteria. Limnol. Oceanogr. 48: 313–318. Grass, G., M. D. Wong, B. P. Rosen, L. Ron, C. Rensing, and R. L. Smith. 2002. ZupT Is a Zn ( II ) Uptake System in Escherichia coli. J. Bacteriol. 184: 864–866. doi:10.1128/JB.184.3.864 Green, R. T., J. D. Todd, and A. W. B. Johnston. 2012. Manganese uptake in marine bacteria ; the novel MntX transporter is widespread in Roseobacters , Vibrios , Alteromonadales and the SAR11 and SAR116 clades. ISME J. 7: 581–591. doi:10.1038/ismej.2012.140 Gundersen, K., M. Heldal, S. Norland, D. A. Purdie, and A. H. Knap. 2002. Elemental C , N , and P Cell Content of Individual Bacteria Collected at the Bermuda Atlantic Time-Series Study (BATS) Site. Limnol. Oceanogr. 47: 1525–1530. Gunther, M., P. Hanna, R. Mason, and M. Cohen. 1995. Hydroxyl radical formation from cuprous ion and hydrogen peroxide: A spin trapping study. Arch. Biochem. Biophys. 316: 515–522. Guo, C., J. Yu, T. Y. Ho, L. Wang, S. Song, L. Kong, and H. Liu. 2012a. Dynamics of phytoplankton community structure in the South China Sea in response to the East Asian aerosol input. Biogeosciences 9: 1519–1536. doi:10.5194/bg-9-1519-2012 Guo, J., A. L. Annett, R. L. Taylor, S. Lapi, T. J. Ruth, and M. T. Maldonado. 2010. Copper-Uptake Kinetics of Coastal and Oceanic Diatoms1. J. Phycol. 46: 1218–1228. doi:10.1111/j.1529-8817.2010.00911.x Guo, J., B. R. Green, and M. T. Maldonado. 2015. Sequence analysis and gene expression of potential components of copper transport and homeostasis in thalassiosira pseudonana. Protist 166: 58–77. doi:10.1016/j.protis.2014.11.006 Guo, J., S. Lapi, T. J. Ruth, and M. T. Maldonado. 2012b. The effects of iron and copper availability on the copper stoichiometry of marine phytoplankton. J. Phycol. 48: 312–325. doi:10.1111/j.1529-8817.2012.01133.x Gustafsson, J. P. 2016. Visual MINTEQ (v3.0). A Windows version of MINTEQA2. http://vminteq.lwr.kth.se/. Gόmez-Pereira, P. R., B. M. Fuchs, C. Alonso, M. J. Oliver, J. E. van Beusekom, and R. Amann. 2010. Distinct flavobacterial communities in contrasting water masses of the North Atlantic Ocean. ISME J. 4: 472–487. doi:10.1038/ismej.2009.142 Haag, W. R., and J. Hoigne. 1983. Ozonation of Water Containing Chlorine or Chloramines. Water Res. 17: 1397–1983. doi:doi:10.1016/0043-1354(83)90270-1 Hamme, R. C., P. W. Webley, W. R. Crawford, and others. 2010. Volcanic ash fuels anomalous plankton bloom in subarctic northeast Pacific. Geophys. Res. Lett. 37: 1–5. doi:10.1029/2010GL044629 Haraldsson, C., and S. Westerlund. 1988. Trace metals in the water columns of the Black Sea and Framvaren Fjord. Mar. Chem. 23: 417–424. doi:10.1016/0304-4203(88)90108-9 Hassler, C. S., V. I. Slaveykova, and K. J. Wilkinson. 2004. Discriminating between intra- and extracellular metals using chemical extractions. Limnol. Oceanogr. Methods 2: 237–247. doi:10.4319/lom.2004.2.237 Heggie, D., G. Klinkhammer, and D. Cullen. 1987. Manganese and copper fluxes from continental margin sediments. Geochim. Cosmochim. Acta 51: 1059–1070. doi:10.1016/0016-7037(87)90200-6 Helbig, K., C. Bleuel, G. J. Krauss, and D. H. Nies. 2008. Glutathione and Transition-Metal Homeostasis in Escherichia coliᰔ. J. Bacteriol. 190: 5431–5438. doi:10.1128/JB.00271-08 146  Heller, M. I., and P. L. Croot. 2015. Copper speciation and distribution in the Atlantic sector of the Southern Ocean. Mar. Chem. 173: 253–268. doi:10.1016/j.marchem.2014.09.017 Herberich, E., J. Sikorski, and T. Hothorn. 2010. A robust procedure for comparing multiple means under heteroscedasticity in unbalanced designs. PLoS One 5: 1–8. doi:10.1371/journal.pone.0009788 Herbert, D. 1976. Stoichiometric aspects of microbial growth, p. 1–30. In A. C. R. Dean et al. [eds.], Continuous culture: 6 applications and new fields. Ellis Horwood. Herrick, J. A., C. A. Neal, C. E. Cameron, J. P. Dixon, R. G. McGimsey, and V. H. Program. 2014. 2012 Volcanic Activity in Alaska : Summary of Events and Response of the Alaska Volcano Observatory Scientific Investigations Report 2014 – 5160. 46–52. Hines, M. E., W. Berry Lyons, P. B. Armstrong, W. H. Orem, M. J. Spencer, H. E. Gaudette, and G. E. Jones. 1984. Seasonal metal remobilization in the sediments of Great Bay, New Hampshire. Mar. Chem. 15: 173–187. doi:10.1016/0304-4203(84)90014-8 Ho, P. L., Y. Jo, and M. K. S. Kim. 2015. Seasonal variation in trace element concentrations and Pb isotopic composition of airborne particulates during Asian dust and non- Asian dust periods in Daejeon , Korea. Environ. Earth Sci. 3613–3628. doi:10.1007/s12665-015-4425-4 Ho, T.-Y., A. Quigg, Z. V. Finkel, A. J. Milligan, K. Wyman, P. G. Falkowski, and F. M. M. Morel. 2003. The elemental composition of some marine phytoplankton. J. Phycol. 39: 1145–1159. doi:10.1111/j.0022-3646.2003.03-090.x Hoell, J. M., D. D. D. Davis, S. C. C. Liu, and others. 1996. Pacific Exploratory Mission-West A (PEM-West A): September-October 1991. J. Geophys. Res. 101: 1641–1653. doi:10.1029/95JD00622 Hogle, S. L., J. Cameron Thrash, C. L. Dupont, and K. A. Barbeau. 2016. Trace metal acquisition by marine heterotrophic bacterioplankton with contrasting trophic strategies. Appl. Environ. Microbiol. 82: AEM.03128-15. doi:10.1128/AEM.03128-15 Holmström, C., and S. Kjelleberg. 1999. Marine Pseudoalteromonas species are associated with higter organisms and produce biologically active extracellular agents. FEMS Microb. Ecol. 30: 285–293. doi:10.1111/j.1574-6941.1999.tb00656.x Hörtnagl, P., M. T. Pérez, and R. Sommaruga. 2011. Contrasting effects of ultraviolet radiation on the growth efficiency of freshwater bacteria. Aquat. Ecol. 45: 125–136. doi:10.1007/s10452-010-9341-9 Howard, E. C., J. R. Henriksen, A. Buchan, and others. 2006. Bacterial Taxa That Limit Sulfur Flux from the Ocean. Science (80-. ). 649: 649–652. doi:10.1126/science.1130657 Hsu, S.-C., G.-C. Gong, F.-K. Shiah, and others. 2014. Sources, solubility, and acid processing of aerosol iron and phosphorous over the South China Sea: East Asian dust and pollution outflows vs. Southeast Asian biomass burning. Atmos. Chem. Phys. Discuss. 14: 21433–21472. doi:10.5194/acpd-14-21433-2014 Hsu, S. C., G. T. F. Wong, G. C. Gong, and others. 2010. Sources, solubility, and dry deposition of aerosol trace elements over the East China Sea. Mar. Chem. 120: 116–127. doi:10.1016/j.marchem.2008.10.003 Hudson, R. J. M., and M. M. Morel. 1989. Distinguishing Between Extra- and Intracellular Iron in Marine Phytoplankton. Limnol. Oceanogr. 34: 1113–1120. Iwata, S. 1998. Structure and function of bacterial cytochrome c oxidase. J Biochem Tokyo 123: 369–375. Jacobs, L., and S. Emerson. 1982. Trace metal solubility in an anoxic fjord. Earth Planet. Sci. Lett. 60: 237–252. doi:10.1016/0012-821X(82)90006-1 Jacobs, L., S. Emerson, and J. Skei. 1985. Partitioning and transport of metals across the O2 H2S interface in a permanently anoxic basin: Framvaren Fjord, Norway. Geochim. Cosmochim. Acta 49: 1433–1444. doi:10.1016/0016-7037(85)90293-5 Jacquot, J. E., R. E. a. Horak, S. a. Amin, A. H. Devol, A. E. Ingalls, E. V. Armbrust, D. a. Stahl, and J. W. Moffett. 2014. Assessment of the potential for copper limitation of ammonia oxidation by Archaea in a dynamic estuary. Mar. Chem. 162: 37–49. doi:10.1016/j.marchem.2014.02.002 Jacquot, J. E., Y. Kondo, A. N. Knapp, and J. W. Moffett. 2013. The speciation of copper across active 147  gradients in nitrogen-cycle processes in the eastern tropical South Pacific. Limnol. Oceanogr. 58: 1387–1394. doi:10.4319/lo.2013.58.4.1387 Jacquot, J. E., and J. W. Moffett. 2015. Copper distribution and speciation across the International GEOTRACES Section GA03. Deep. Res. Part II Top. Stud. Oceanogr. 116: 187–207. doi:10.1016/j.dsr2.2014.11.013 Janssen, D., T. Conway, and S. John. 2014. Undocumented water column sink for cadmium in open ocean oxygen-deficient zones. Proc. Natl. Acad. Sci. 111: :6888-93. doi:10.1073/pnas.1402388111 Janssen, D. J., and J. T. Cullen. 2015. Decoupling of zinc and silicic acid in the subarctic northeast Pacific interior. Mar. Chem. 177: 124–123. doi:10.1016/j.marchem.2015.03.014 Janulczyk, R., J. Pallon, and L. Björck. 1999. Identification and characterization of a Streptococcus pyogenes ABC transporter with multiple specificity for metal cations. Mol. Microbiol. 34: 596–606. Jaworowski, A., G. Mayo, D. C. Shaw, H. D. Campbell, and I. G. Young. 1981. Characterization of the Respiratory NADH Dehydrogenase of Escherichia coli and Reconstruction of NADH Oxidase in ndh Mutant Membrane Vesicles. Biochemistry 3621–3628. Ji, Y., and R. M. Sherrell. 2008. Differential effects of phosphorus limitation on cellular metals in Chlorella and Microcystis. Limnol. Oceanogr. 53: 1790–1804. Jiao, N., G. J. Herndl, D. a Hansell, and others. 2010. Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nat. Rev. Microbiol. 8: 593–9. doi:10.1038/nrmicro2386 Jickells, T. 1995. Atmospheric inputs of metals and nutrients to the oceans: their magnitude and effects. Mar. Chem. 48: 199–214. doi:10.1016/0304-4203(95)92784-P Jickells, T. D. 2005. Global Iron Connections Between Desert Dust, Ocean Biogeochemistry, and Climate. Science (80-. ). 308: 67–71. doi:10.1126/science.1105959 Johnson, W. K., L. A. Miller, N. E. Sutherland, and C. S. Wong. 2005. Iron transport by mesoscale Haida eddies in the Gulf of Alaska. Deep. Res. Part II Top. Stud. Oceanogr. 52: 933–953. doi:10.1016/j.dsr2.2004.08.017 Jordi, A., G. Basterretxea, A. Tovar-Sánchez, A. Alastuey, and X. Querol. 2012. Copper aerosols inhibit phytoplankton growth in the Mediterranean Sea. Proc. Natl. Acad. Sci. U. S. A. 109: 21246–9. doi:10.1073/pnas.1207567110 Karner, M. B., E. F. DeLong, and D. M. Karl. 2001. Archaeal dominance in the mesopelagic zone of the Pacific Ocean. Nature 409: 507–510. doi:10.1038/35054051 Kawakami, T., M. Kuroki, M. Ishii, Y. Igarashi, and H. Arai. 2010. Differential expression of multiple terminal oxidases for aerobic respiration in Pseudomonas aeruginosa. Environ. Microbiol. 12: 1399–1412. doi:10.1111/j.1462-2920.2009.02109.x Kehres, D. G., and M. E. Maguire. 2003. Emerging themes in manganese transport, biochemistry and pathogenesis in bacteria. FEMS Microbiol. Rev. 27: 263–290. doi:10.1016/S0168-6445(03)00052-4 Keiblinger, K. M., E. K. Hall, W. Wanek, and others. 2010. The effect of resource quantity and resource stoichiometry on microbial carbon-use-efficiency. FEMS Microbiol. Ecol. 73: 430–40. doi:10.1111/j.1574-6941.2010.00912.x Kim, C., W. W. Lorenz, J. T. Hoopes, and J. F. D. Dean. 2001. Oxidation of phenolate siderophores by the multicopper oxidase encoded by the Escherichia coli yacK gene. J. Bacteriol. 183: 4866–4875. doi:10.1128/JB.183.16.4866-4875.2001 Kim, J.-M., O. Baars, and M. M. Morel. 2015. Bioavailability and Electroreactivity of Zinc Complexed to Strong and Weak Organic Ligands. Environ. Sci. Technol. 49: 10894–10902. doi:10.1021/acs.est.5b02098 Kim, K., S. Kwon, J. Yoon, and J. F. Kim. 2016. Complete Genome Sequence of the Proteorhodopsin-Containing Marine Flavobacterium. Am. Soc. Microbiol. 4: 9–10. doi:10.1128/genomeA.00804-16.Copyright Kimura, H., C. R. Young, A. Martinez, and E. F. Delong. 2011. Light-induced transcriptional responses 148  associated with proteorhodopsin-enhanced growth in a marine flavobacterium. ISME J. 5: 1641–51. doi:10.1038/ismej.2011.36 Kirchman, D. L. 2008. Introduction and Overview, p. 1–13. In D.L. Kirchman [ed.], Microbial Ecology of the Oceans. Whiley-Blackwell. Kirchman, D. L. 2016. Growth Rates of Microbes in the Oceans. Ann. Rev. Mar. Sci. 8: 285–309. doi:10.1146/annurev-marine-122414-033938 Kirchman, D. L., K. A. Hoffman, R. Weaver, and D. A. Hutchins. 2003a. Regulation of growth and energetics of a marine bacterium by nitrogen source and iron availability. Mar. Ecol. Prog. Ser. 250: 291–296. Kirchman, D. L., R. G. Keil, and M. Simon. 1993. Biomass and production of heterotrophic bacterioplankton in the oceanic subarctic Pacific. Deep. Res. Part 1. Oceanogr. Res. Pap. 40: 967–988. Kirchman, D. L., L. Yu, and M. T. Cottrell. 2003b. Diversity and Abundance of Uncultured Cytophaga -Like Bacteria in the Delaware Estuary. Appl. Environ. Microbiol. 69: 6587–6596. doi:10.1128/AEM.69.11.6587 Klinkhammer, G., D. T. Heggie, and D. W. Graham. 1982. Metal diagenesis in oxic marine sediments. Earth Planet. Sci. Lett. 61: 211–219. doi:10.1016/0012-821X(82)90054-1 Klinkhammer, G. P. 1980. Early diagenesis in sediments from the eastern equatorial Pacific, II. Pore water metal results. Earth Planet. Sci. Lett. 49: 81–101. doi:10.1016/0012-821X(80)90151-X Klippel, B., A. Lochner, D. C. Bruce, and others. 2011. Complete Genome Sequences of Krokinobacter sp . Strain 4H-3-7-5 and. J. Bacteriol. 193: 4545–4546. doi:10.1128/JB.05518-11 Kobayashi, M., and S. Shimizu. 1999. Cobalt proteins. Eur. J. Biochem. 261: 1–9. doi:10.1046/j.1432-1327.1999.00186.x Koçak, M., M. Nimmo, N. Kubilay, and B. Herut. 2004. Spatio-temporal aerosol trace metal concentrations and sources in the Levantine Basin of the Eastern Mediterranean. Atmos. Environ. 38: 2133–2144. doi:10.1016/j.atmosenv.2004.01.020 Kramer, D., J. T. Cullen, J. R. Christian, W. K. Johnson, and T. F. Pedersen. 2011. Silver in the subarctic northeast Pacific Ocean : Explaining the basin scale distribution of silver. Mar. Chem. 123: 133–142. doi:10.1016/j.marchem.2010.11.002 Kwon, S., B. K. Kim, J. Y. Song, M. Kwak, C. H. Lee, J. Yoon, T. K. Oh, and J. F. Kim. 2013. Genomic Makeup of the Marine Flavobacterium Nonlabens of Rhodopsins. Genome Biol. Evol. 5: 187–199. doi:10.1093/gbe/evs134 Lagerström, M. E., M. P. Field, M. Séguret, L. Fischer, S. Hann, and R. M. Sherrell. 2013. Automated on-line flow-injection ICP-MS determination of trace metals ( Mn , Fe , Co , Ni , Cu and Zn ) in open ocean seawater : Application to the GEOTRACES program. Mar. Chem. 155: 71–80. doi:10.1016/j.marchem.2013.06.001 Lam, P. J., J. K. B. Bishop, C. C. Henning, M. A. Marcus, G. A. Waychunas, and I. Y. Fung. 2006. Wintertime phytoplankton bloom in the subarctic Pacific supported by continental margin iron. Global Biogeochem. Cycles 20: 1–12. doi:10.1029/2005GB002557 Landing, W., and B. Lewis. 1991. Thermodynamic Modeling of Trace Metal Speciation in the Black Sea. Black Sea Oceanogr. 351: 125–160. doi:10.1007/978-94-011-2608-3_8 Lane, E. S., D. M. Semeniuk, R. F. Strzepek, J. T. Cullen, and M. T. Maldonado. 2009. Effects of iron limitation on intracellular cadmium of cultured phytoplankton: Implications for surface dissolved cadmium to phosphate ratios. Mar. Chem. 115: 155–162. doi:10.1016/j.marchem.2009.07.008 Lankiewicz, T. S., M. T. Cottrell, and D. L. Kirchman. 2015. Growth rates and rRNA content of four marine bacteria in pure cultures and in the Delaware estuary. ISME J. 10: 823–832. doi:10.1038/ismej.2015.156 Lebaron, P., P. Servais, M. Troussellier, and others. 2001. Microbial community dynamics in Mediterranean nutrient-enriched seawater mesocosms: changes in the genetic diversity of bacterial 149  populations. FEMS Microbiol Ecol 34: 243–253. Lee, J.-M., E. a Boyle, Y. Echegoyen-Sanz, J. N. Fitzsimmons, R. Zhang, and R. a Kayser. 2011. Analysis of trace metals (Cu, Cd, Pb, and Fe) in seawater using single batch nitrilotriacetate resin extraction and isotope dilution inductively coupled plasma mass spectrometry. Anal. Chim. Acta 686: 93–101. doi:10.1016/j.aca.2010.11.052 Lee, P. K., S. J. Youm, and H. Y. Jo. 2013. Heavy metal concentrations and contamination levels from Asian dust and identification of sources: A case-study. Chemosphere 91: 1018–1025. doi:10.1016/j.chemosphere.2013.01.074 Lewinson, O., A. T. Lee, and D. C. Rees. 2009. A P-type ATPase importer that discriminates between essential and toxic transition metals. Proc. Natl. Acad. Sci. U. S. A. 106: 4677–82. doi:10.1073/pnas.0900666106 Li, J., Z. Wang, G. Zhuang, G. Luo, Y. Sun, and Q. Wang. 2012. Mixing of Asian mineral dust with anthropogenic pollutants over East Asia: A model case study of a super-duststorm in March 2010. Atmos. Chem. Phys. 12: 7591–7607. doi:10.5194/acp-12-7591-2012 Lipscomb, W. N., and N. Sträter. 1996. Recent advances in zinc enzymology. Chem. Rev. 96: 2375–2434. doi:10.1021/cr950042j Little, S. H., D. Vance, M. Siddall, and E. Gasson. 2013. A modeling assessment of the role of reversible scavenging in controlling oceanic dissolved Cu and Zn distributions. Global Biogeochem. Cycles 27: 780–791. doi:10.1002/gbc.20073 Little, S. H., D. Vance, C. Walker-Brown, and W. M. Landing. 2014. The oceanic mass balance of copper and zinc isotopes, investigated by analysis of their inputs, and outputs to ferromanganese oxide sediments. Geochim. Cosmochim. Acta 125: 673–693. Lohan, M. C. M., A. A. M. Aguilar-Islas, R. P. Franks, and K. W. Bruland. 2005. Determination of iron and copper in seawater at pH 1.7 with a new commercially available chelating resin, NTA Superflow. Anal. Chim. Acta 530: 121–129. doi:10.1016/j.aca.2004.09.005 Lucas, J., A. Wichels, H. Teeling, M. Chafee, M. Scharfe, and G. Gerdts. 2015. Annual dynamics of North Sea bacterioplankton: Seasonal variability superimposes short-term variation. FEMS Microbiol. Ecol. 91: 1–11. doi:10.1093/femsec/fiv099 Ma, Z., F. E. Jacobsen, and D. P. Giedroc. 2010. Metal transporters and metal sensors: How Coordination Chemistry Controls Bacterial Metal Homeostasis. Chem. Rev. 109: 4644–4681. doi:10.1021/cr900077w.Metal Mackey, K. R. M., K. N. Buck, J. R. Casey, A. Cid, M. W. Lomas, Y. Sohrin, and A. Paytan. 2012. Phytoplankton responses to atmospheric metal deposition in the coastal and open-ocean Sargasso Sea. Front. Microbiol. 3: 1–15. doi:10.3389/fmicb.2012.00359 Mackey, K. R. M., C. Te Chien, A. F. Post, M. A. Saito, and A. Paytan. 2015. Rapid and gradual modes of aerosol trace metal dissolution in seawater. Front. Microbiol. 6: 1–11. doi:10.3389/fmicb.2014.00794 Mackey, K. R. M., G. L. Van Dijken, S. Mazloom, A. M. Erhardt, J. Ryan, K. R. Arrigo, and A. Paytan. 2010. Influence of atmospheric nutrients on primary productivity in a coastal upwelling region. Global Biogeochem. Cycles 24. doi:10.1029/2009GB003737 Macomber, L., and J. A. Imlay. 2009. The iron-sulfur clusters of dehydratases are primary intracellular targets of copper toxicity. Proc. Natl. Acad. Sci. U. S. A. 106: 8344–9. doi:10.1073/pnas.0812808106 Makino, A. W., J. B. Cotner, R. W. Sterner, J. J. Elser, W. Makino, J. B. Cotner, R. W. Sterner, and J. J. Elser. 2003. Are Bacteria More Like Plants or Animals ? Growth Rate and Resource Dependence of Bacterial C : N : P Stoichiometry. Funct. Ecol. 17: 121–130. 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. doi:10.4319/lo.2006.51.4.1729 Maldonado, M. T., and N. M. Price. 1999. Utilization of Fe bound to strong organic ligands by phytoplankton communities in the subarctic Pacific Ocean. Deep. Res. II 46: 2447–2473. 150  doi:10.1016/S0967-0645(99)00071-5 Mann, A. J., R. L. Hahnke, S. Huang, and others. 2013. The Genome of the Alga-Associated Marine Flavobacterium Formosa agariphila KMM 3901 T Reveals a Broad Potential for Degradation of. Appl. Environ. Microbiol. 79: 6813–6822. doi:10.1128/AEM.01937-13 Mann, E. L., N. Ahlgren, J. W. Moffett, S. W. Chisholm, and W. Hole. 2002. Copper toxicity and cyanobacteria ecology in the Sargasso Sea. Limnol. Oceanogr. 47: 976–988. doi:10.4319/lo.2002.47.4.0976 Marchetti, A., M. S. Parker, L. P. Moccia, and others. 2009. Ferritin is used for iron storage in bloom-forming marine pennate diatoms. Nature 457: 467–470. Markaki, Z., K. Oikonomou, M. Kocak, G. Kouvarakis, A. Chaniotaki, N. Kubilay, and N. Mihalopoulos. 2003. Atmospheric deposition of inorganic phosphorus in the Levantine Basin, eastern Mediterranean: Spatial and temporal variability and its role in seawater productivity. Limnol. Oceanogr. 48: 1557–1568. doi:10.4319/lo.2003.48.4.1557 Martin, J. H., and S. . Fitzwater. 1988. Iron deficiency limits phytoplankton growth in the north-east Pacific subarctic. Nature 331: 341–343. Martin, J. H., R. M. Gordon, S. Fitzwater, and W. William. 1989. VERTEX : phytoplankton / iron studies in the Gulf of Alaska. Deep Sea Res. 36: 649–680. Matsubara, T., K. Frunzke, and W. G. Zumft. 1982. Modulation by copper of the products of nitrite respiration in Pseudomonas Modulation by Copper of the Products of Nitrite Respiration in Pseudomonas perfectomarinlus. J. Bacteriol. 149: 816–823. McAlister, J. A. 2015. Biogeochemistry of dissolved gallium and lead isotopes in the northeast Pacific and western Arctic Oceans. PhD Thesis, University of British Columbia, pp 53-112. Measures, C. I., W. M. Landing, M. T. Brown, and C. S. Buck. 2008. A commercially available rosette system for trace metal-clean sampling. Limnol. Oceanogr. Methods 6: 384–394. doi:10.4319/lom.2008.6.384 Mermod, M., D. Magnani, M. Solioz, and J. V. Stoyanov. 2012. The copper-inducible ComR (YcfQ) repressor regulates expression of ComC (YcfR), which affects copper permeability of the outer membrane of Escherichia coli. BioMetals 25: 33–43. doi:10.1007/s10534-011-9510-x Middag, R., R. Séférian, T. M. Conway, S. G. John, K. W. Bruland, and H. J. W. de Baar. 2015. Intercomparison of dissolved trace elements at the Bermuda Atlantic Time Series station. Mar. Chem. 177: 476–489. doi:10.1016/j.marchem.2015.06.014 Miller, A. F. 2012. Superoxide dismutases: Ancient enzymes and new insights. FEBS Lett. 586: 585–595. doi:10.1016/j.febslet.2011.10.048 Milne, A., W. Landing, M. Bizimis, and P. Morton. 2010. Determination of Mn, Fe, Co, Ni, Cu, Zn, Cd and Pb in seawater using high resolution magnetic sector inductively coupled mass spectrometry (HR-ICP-MS). Anal. Chim. Acta 665: 200–207. doi:10.1016/j.aca.2010.03.027 Moffett, J. W. 1995. Temporal and spatial variability of copper complexation by strong chelators in the Sargasso Sea. Deep Sea Res. Part I Oceanogr. Res. Pap. 42: 1273–1295. doi:10.1016/0967-0637(95)00060-J Moffett, J. W., and L. E. Brand. 1996. Production of strong, extracellular Cu chelators by marine cyanobacteria in response to Cu stress. Limnol. Oceanogr. 41: 388–395. doi:10.4319/lo.1996.41.3.0388 Moffett, J. W., and C. Dupont. 2007. Cu complexation by organic ligands in the sub-arctic NW Pacific and Bering Sea. Deep Sea Res. Part I Oceanogr. Res. Pap. 54: 586–595. doi:10.1016/j.dsr.2006.12.013 Moffett, J. W. J., C. B. C. Tuit, and B. B. Ward. 2012. Chelator-induced inhibition of copper metalloenzymes in denitrifying bacteria. Limnol. Oceanogr. 57: 272–280. doi:10.4319/lo.2012.57.1.0272 Moore, J. K., S. C. Doney, and K. Lindsay. 2004. Upper ocean ecosystem dynamics and iron cycling in a global three-dimensional model. Global Biogeochem. Cycles 18: 1–21. doi:10.1029/2004GB002220 151  Moran, M. A., A. Buchan, J. M. González, and others. 2004. Genome sequence of Silicibacter pomeroyi reveals adaptations to the marine environment. Nature 432: 910–3. doi:10.1038/nature03170 Morse, J. W., and G. W. Luther. 1999. Chemical influences on trace metal-sulfide interactions in anoxic sediments. Geochim. Cosmochim. Acta 63: 3373–3378. doi:10.1016/S0016-7037(99)00258-6 Mounicou, S., J. Szpunar, and R. Lobinski. 2009. Metallomics: the concept and methodology. Chem. Soc. Rev. 38: 1119–1138. doi:10.1039/b713633c Murray, J. W. 1975. Interaction of Metal-Ions At Manganese Dioxide Solution Interface. Geochim. Cosmochim. Acta 39: 505–519. Nameroff, T. J., L. S. Balistrieri, and J. W. Murray. 2002. Suboxic trace metal geochemistry in the eastern tropical North Pacific. Geochim. Cosmochim. Acta 66: 1139–1158. doi:10.1016/S0016-7037(01)00843-2 Ndung’u, K., R. P. Franks, K. W. Bruland, and A. R. Flegal. 2003. Organic complexation and total dissolved trace metal analysis in estuarine waters: Comparison of solvent-extraction graphite furnace atomic absorption spectrometric and chelating resin flow injection inductively coupled plasma-mass spectrometric analysis. Anal. Chim. Acta 481: 127–138. doi:10.1016/S0003-2670(03)00063-1 Neal, C. A., J. Herrick, O. A. Girina, M. Chibisova, A. Rybin, R. G. McGimsey, and J. Dixon. 2014. 2010 Volcanic activity in Alaska, Kamchatka, and the Kurile Islands—Summary of events and response of the Alaska Volcano Observatory. Neidhardt, F. C., J. L. Ingraham, and M. Schaechter. 1992. Physiology of the bacterial cell. A molecular approach.Sinauer associates, Sunderland,. Nevo, Y., and N. Nelson. 2006. The NRAMP family of metal-ion transporters. Biochim. Biophys. Acta - Mol. Cell Res. 1763: 609–620. doi:10.1016/j.bbamcr.2006.05.007 Nies, D. H. 1999. Microbial heavy-metal resistance. Appl. Microbiol. Biotechnol. 51: 730–751. Nuester, J., S. Vogt, M. Newville, A. B. Kustka, and B. S. Twining. 2012. The unique biogeochemical signature of the marine diazotroph Trichodesmium. Front. Microbiol. 3: 1–15. doi:10.3389/fmicb.2012.00150 O’Sullivan, J. E., R. J. Watson, and E. C. V Butler. 2013. An ICP-MS procedure to determine Cd, Co, Cu, Ni, Pb and Zn in oceanic waters using in-line flow-injection with solid-phase extraction for preconcentration. Talanta 115: 999–1010. doi:10.1016/j.talanta.2013.06.054 Obernosterer, I., M. Fourquez, and S. Blain. 2014. Fe and C co-limitation of heterotrophic bacteria in the naturally fertilized region off Kerguelen Islands. Biogeosciences Discuss. 11: 15733–15752. doi:10.5194/bgd-11-15733-2014 Orians, K. J., and K. W. Bruland. 1985. Dissolved aluminium in the central North Pacific. Nature 316: 427–429. doi:10.1038/314731a0 Outten, C. E., and T. V. O. Halloran. 2001. Femtomolar Sensitivity of Metalloregulatory Proteins Controlling Zinc Homeostasis. Sci. Rep. 292: 2488–2491. Outten, F. W., D. L. Huffman, J. A. Hale, and T. V. O’Halloran. 2001. The Independent cue and cus Systems Confer Copper Tolerance during Aerobic and Anaerobic Growth in Escherichia coli. J. Biol. Chem. 276: 30670–30677. doi:10.1074/jbc.M104122200 Pakulski, J. D., R. B. Coffin, C. a. Kelley, S. L. Holder, R. Downer, P. Aas, M. M. Lyons, and W. H. Jeffrey. 1996. Iron stimulation of Antarctic bacteria. Nature 383: 133–134. doi:10.1038/383133b0 Pawlowicz, R. 2015. M_map: a mapping package for Matlab. Available at https://www.eoas.ubc.ca/~rich/map.html. Paytan, A., K. R. M. Mackey, Y. Chen, I. D. Lima, S. C. Doney, N. Mahowald, R. Labiosa, and A. F. Post. 2009. Toxicity of atmospheric aerosols on marine phytoplankton. Proc. Natl. Acad. Sci. U. S. A. 106: 4601–4605. doi:10.1073/pnas.0811486106 Pedler, B. E., L. I. Aluwihare, and F. Azam. 2014. Single bacterial strain capable of significant contribution to carbon cycling in the surface ocean. Proc. Natl. Acad. Sci. U. S. A. 111: 7202–7. doi:10.1073/pnas.1401887111 152  Peers, G., and N. M. N. Price. 2006. Copper-containing plastocyanin used for electron transport by an oceanic diatom. Nature 441: 341–4. doi:10.1038/nature04630 Peers, G., S. Quesnel, and N. Price. 2005. Copper requirements for iron acquisition and growth of coastal and oceanic diatoms. Limnol. Oceanogr. 50: 1149–1158. Peña, M. A., and S. J. Bograd. 2007. Time series of the northeast Pacific. Prog. Oceanogr. 75: 115–119. doi:10.1016/j.pocean.2007.08.008 Pierce, S. D., R. L. Smith, P. M. Kosro, J. A. Barth, and C. D. Wilson. 2000. Continuity of the poleward undercurrent along the eastern boundary of the mid-latitude north Pacific. Deep Sea Res. Part II 47: 811–829. doi:10.1016/S0967-0645(99)00128-9 Pinhassi, J., M. Montserrat Sala, H. Havskum, F. Peters, O. Guadayol, A. Malits, and C. Marrase. 2004. Changes in Bacterioplankton Composition under Different Phytoplankton Regimens †. Appl. Environ. Microbiol. 70: 6753–6766. doi:10.1128/AEM.70.11.6753 Porcheron, G., A. Garénaux, J. Proulx, M. Sabri, and C. M. Dozois. 2013. Iron, copper, zinc, and manganese transport and regulation in pathogenic Enterobacteria: correlations between strains, site of infection and the relative importance of the different metal transport systems for virulence. Front. Cell. Infect. Microbiol. 3: 24 p. doi:10.3389/fcimb.2013.00090 Pradeep Ram, A. S. P., J. Colombet, F. Perriere, T. Thouvenot, and T. Sime-Ngando. 2016. Viral Regulation of Prokaryotic Carbon Metabolism in a Hypereutrophic Freshwater Reservoir Ecosystem ( Villerest , France ). Froniters Microbiol. 7: 1–13. doi:10.3389/fmicb.2016.00081 Price, N. M., G. I. Harrison, J. Hering, R. J. Hudson, M. V. N. Pascale, B. Palenik, and F. Morel. 1988. Preparation and Chemistry of the Artificial Algal Culture Medium Aquil. Biol. Oceanogr. 6: 443–461. doi:10.1080/01965581.1988.10749544 Quigg, A., Z. V Finkel, A. J. Irwin, and others. 2003. Evolutionary inheritance of elemental stoichiometry in phytoplankton. Lett. to Nat. 425: 291–293. doi:10.1098/rspb.2010.1356 Quigg, A., A. J. Irwin, and Z. V Finkel. 2010. Evolutionary inheritance of elemental stoichiometry in phytoplankton. Proc. Biol. Sci. 526–534. doi:10.1098/rspb.2010.1356 Quigg, A., J. R. Reinfelder, and N. S. Fisher. 2006. Copper uptake kinetics in diverse marine phytoplankton. Limnol. Oceanogr. 51: 893–899. doi:10.4319/lo.2006.51.2.0893 R Core Team. 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/. Ramaiah, N., A. Jain, R. M. Meena, and others. 2015. Response of bacteria and phytoplankton from a subtropical front location Southern Ocean to micronutrient amendments ex-situ. Deep Sea Res. Part II Top. Stud. Oceanogr. 118: 209–220. doi:10.1016/j.dsr2.2015.03.013 Rappé, M. S., K. Vergin, and S. J. Giovannoni. 2000. Phylogenetic comparisons of a coastal bacterioplankton community with its counterparts in open ocean and freshwater systems. FEMS Microbiol. Ecol. 33: 219–232. doi:10.1016/S0168-6496(00)00064-7 Redfield. 1958. The biological control of chemical factors in the environment. Am. Sci. 46: 205–221. Reichenbach, H., W. Kohl, A. Bottger-Vetter, and H. Achenbach. 1980. Flexirubin-type pigments in Flavobacterium. Arch. Microbiol. 126: 291–293. doi:10.1007/BF00409934 Reinthaler, T., and G. J. Herndl. 2005. Seasonal dynamics of bacterial growth efficiencies in relation to phytoplankton in the southern North Sea. Aquat. Microb. Ecol. 39: 7–16. Reisch, C. R., W. M. Crabb, S. M. Gifford, Q. Teng, M. J. Stoudemayer, M. A. Moran, and W. B. Whitman. 2013. Metabolism of dimethylsulphoniopropionate by Ruegeria pomeroyi DSS-3. Mol. Microbiol. 89: 774–791. doi:10.1111/mmi.12314 Reisch, C. R., M. A. Moran, and W. B. Whitman. 2011. Bacterial catabolism of dimethylsulfoniopropionate (DMSP). Froniters Microbiol. 2: 1–12. doi:10.3389/fmicb.2011.00172 Rensing, C., and G. Grass. 2003. Escherichia coli mechanisms of copper homeostasis in a changing environment. FEMS Microbiol. Rev. 27: 197–213. doi:10.1016/S0168-6445(03)00049-4 Rensing, C., and S. F. McDevitt. 2013. The Copper Metallome in Prokaryotic cells., p. 417–450. In L. 153  Banci [ed.], Metallomics and the cell. Springer Netherlands. Ridge, P. G., Y. Zhang, and V. N. Gladyshev. 2008. Comparative Genomic Analyses of Copper Transporters and Cuproproteomes Reveal Evolutionary Dynamics of Copper Utilization and Its Link to Oxygen. PLoS One e1378: 1–9. doi:10.1371/journal.pone.0001378 Riedel, T., L. Gómez-Consarnau, J. Tomasch, and others. 2013. Genomics and Physiology of a Marine Flavobacterium Encoding a Proteorhodopsin and a Xanthorhodopsin-Like Protein. PLoS One 8: 10–18. doi:10.1371/journal.pone.0057487 Riemann, L., G. F. Steward, and F. Azam. 2000. Dynamics of Bacterial Community Composition and Activity during a Mesocosm Diatom Bloom. Appl. Environmantal Microbiol. 66: 578–587. doi:10.1128/AEM.66.2.578-587.2000.Updated Rivkin, R. B., and L. Legendre. 2001. Biogenic Carbon Cycling in the Upper Ocean : Effects of Microbial Respiration. Science (80-. ). 291: 2398–2400. Rodriguez, I. B., S. Lin, J. Ho, and T. Y. Ho. 2016. Effects of trace metal concentrations on the growth of the coral endosymbiont Symbiodinium kawagutii. Front. Microbiol. 7: 1–10. doi:10.3389/fmicb.2016.00082 Roshan, S., and J. Wu. 2015. The distribution of dissolved copper in the tropical-subtropical north Atlantic across the GEOTRACES GA03 transect. Mar. Chem. 176: 189–198. doi:10.1016/j.marchem.2015.09.006 Rowley, D. A., and B. Halliwell. 1983. Superoxide-dependent and ascorbate-dependent formation of hydroxyl radicals in the presence of copper salts: A physiologically significant reaction? Arch. Biochem. Biophys. 225: 279–284. doi:10.1016/0003-9861(83)90031-0 Ruiz-González, C., T. Lefort, R. Massana, R. Simó, and J. M. Gasol. 2012. Diel changes in bulk and single-cell bacterial heterotrophic activity in winter surface waters of the northwestern Mediterranean Sea. Limnol. Oceanogr. 57: 29–42. doi:10.4319/lo.2012.57.1.0029 Russell, J. B., and G. M. Cook. 1995. Energetics of bacterial growth : balance of anabolic and catabolic reactions . Microbiol. Mol. Biol. Rev. 59: 48–62. Saito, M. A., D. M. Sigman, and F. M. M. Morel. 2003. The bioinorganic chemistry of the ancient ocean: The co-evolution of cyanobacterial metal requirements and biogeochemical cycles at the Archean-Proterozoic boundary? Inorganica Chim. Acta 356: 308–318. doi:10.1016/S0020-1693(03)00442-0 Sander, S. G., and A. Koschinsky. 2011. Metal flux from hydrothermal vents increased by organic complexation. Nat. Geosci. 4: 145–150. doi:10.1038/ngeo1088 Schallenberg, C., A. B. Davidson, K. G. Simpson, L. A. Miller, J. T. Cullen, A. Lisa, and J. T. Cullen. 2015. Iron(II) variability in the northeast subarctic Pacific Ocean. Mar. Chem. 177: 33–44. doi:10.1016/j.marchem.2015.04.004 Scott, J. T., J. B. Cotner, and T. M. LaPara. 2012. Variable stoichiometry and homeostatic regulation of bacterial biomass elemental composition. Front. Microbiol. 3: 1–8. doi:10.3389/fmicb.2012.00042 Selje, N., M. Simon, and T. Brinkhoff. 2004. A newly discovered Roseobacter cluster in temperate and polar oceans. Nature 427: 445–448. doi:10.1038/nature02272 Semeniuk, D., R. Bundy, C. Payne, and K. Barbeau. 2015. Acquisition of organically complexed copper by marine phytoplankton and bacteria in the northeast subarctic Pacific Ocean. Mar. Chem. 173: 222–233. doi:http://dx.doi.org/10.1016/j.marchem.2015.01.005 Semeniuk, D. M. 2014. Copper nutrition and transport mechanisms in plankton communities in the Northeast Pacific Ocean. PhD Thesis, University of British Columbia. Semeniuk, D. M., R. M. Bundy, A. M. Posacka, M. Robert, K. A. Barbeau, and M. T. Maldonado. 2016a. Using 67Cu to Study the Biogeochemical Cycling of Copper in the Northeast Subarctic Pacific Ocean. Front. Mar. Sci. 3: 1–19. doi:10.3389/fmars.2016.00078 Semeniuk, D. M., J. T. Cullen, W. K. Johnson, K. Gagnon, T. J. Ruth, and M. T. Maldonado. 2009. Plankton copper requirements and uptake in the subarctic Northeast Pacific Ocean. Deep Sea Res. Part I Oceanogr. Res. Pap. 56: 1130–1142. doi:10.1016/j.dsr.2009.03.003 154  Semeniuk, D. M., R. L. Taylor, and R. Bundy. 2016b. Iron–copper interactions in iron‐limited phytoplankton in the northeast subarctic Pacific Ocean. Limnol. Oceanogr. 61: 279–297. doi:doi: 10.1002/lno.10210 Sendovski, M., M. Kanteev, V. S. Ben-Yosef, N. Adir, and A. Fishman. 2011. First structures of an active bacterial tyrosinase reveal copper plasticity. J. Mol. Biol. 405: 227–237. doi:10.1016/j.jmb.2010.10.048 Shao, Y., K. H. Wyrwoll, A. Chappell, and others. 2011. Dust cycle: An emerging core theme in Earth system science. Aeolian Res. 2: 181–204. doi:10.1016/j.aeolia.2011.02.001 Shaw, T. J., J. M. Gieskes, and R. A. Jahnke. 1990. Early diagenesis in differing depositional environments: The response of transition metals in pore water. Geochim. Cosmochim. Acta 54: 1233–1246. doi:10.1016/0016-7037(90)90149-F Sherry, N. D., P. W. Boyd, K. Sugimoto, and P. J. Harrison. 1999. Seasonal and spatial patterns of heterotrophic bacterial production , respiration , and biomass in the subarctic NE Pacific. Deep Sea Res. Part II 46: 2557–2578. Shimizu, Y., T. Iwao, I. Yasuda, S. I. Ito, T. Watanabe, K. Uehara, N. Shikama, and T. Nakano. 2004. Formation process of North Pacific intermediate water revealed by profiling floats set to drift on 26.7 ???? isopycnal surface. J. Oceanogr. 60: 453–462. doi:10.1023/B:JOCE.0000038061.55914.eb Sholkovitz, E. R., P. N. Sedwick, and T. M. Church. 2010. On the fractional solubility of copper in marine aerosols: Toxicity of aeolian copper revisited. Geophys. Res. Lett. 37: 2–5. doi:10.1029/2010GL044817 Signori, C. N., F. Thomas, A. Enrich-Prast, R. C. G. Pollery, and S. M. Sievert. 2014. Microbial diversity and community structure across environmental gradients in Bransfield Strait, Western Antarctic Peninsula. Front. Microbiol. 5: 1–12. doi:10.3389/fmicb.2014.00647 Sijerčić, A., and N. M. Price. Hydroxamate siderophore secretion by Pseudoalteromonas haloplanktis during steady-state and transient growth in Fe-limited culture. Sohrin, Y., S. Urushihara, S. Nakatsuka, T. Kono, E. Higo, T. Minami, K. Norisuye, and S. Umetani. 2008. Multielemental Determination of GEOTRACES Key Trace Metals in Seawater by ICPMS after Preconcentration Using an Ethylenediaminetriacetic Acid Chelating Resin. Anal. Chem. 80: 6267–6273. doi:10.1021/ac800500f Solioz, M., H. K. Abicht, M. Mermod, and S. Mancini. 2010. Response of gram-positive bacteria to copper stress. J. Biol. Inorg. Chem. 15: 3–14. doi:10.1007/s00775-009-0588-3 Solioz, M., and J. V Stoyanov. 2003. Copper homeostasis in Enterococcus hirae. FEMS Microbiol. Rev. 27: 183–195. doi:10.1016/S0168-6445(03)00053-6 Sunda, G. W., and S. A. Huntsman. 1995. Regulation of copper concentration in the oceanic nutricline by phytoplankton uptake and regeneration cycles. Limnol. Oceanogr. 40: 132–137. Sunda, W. G. 2012. Feedback Interactions between Trace Metal Nutrients and Phytoplankton in the Ocean. Front. Microbiol. 3: 204. doi:10.3389/fmicb.2012.00204 Sunda, W. G. W., and S. a. S. Huntsman. 1998. Processes regulating cellular metal accumulation and physiological effects: Phytoplankton as model systems. Sci. Total Environ. 219: 165–181. doi:10.1016/S0048-9697(98)00226-5 Tainer, J. A., V. A. Roberts, and E. D. Getzoff. 1992. Protein metal-binding sites. Curr. Opin. Biotechnol. 3: 378–387. doi:10.1016/0958-1669(92)90166-G Takano, S., M. Tanimizu, T. Hirata, and Y. Sohrin. 2014. Isotopic constraints on biogeochemical cycling of copper in the ocean. Nat. Commun. 5: 5663. doi:10.1038/ncomms6663 Takeda, S. 2011. Iron and Phytoplankton Growth in the Subarctic North Pacific. Aqua-BioScience Monogr. 4: 41–93. doi:10.5047/absm.2011.00402.0041 Talley, L. D. 1993. Distribution and Formation of North Pacific Intermediate Water. J. Phys. Oceanogr. 23: 517–537. Tang, D., and M. M. Morel. 2006. Distinguishing between cellular and Fe-oxide-associated trace elements 155  in phytoplankton. Mar. Chem. 98: 18–30. doi:10.1016/j.marchem.2005.06.003 Tanita, I., S. Takeda, M. Sato, and K. Furuya. 2015. Surface and middle layer enrichment of dissolved copper in the western subarctic North Pacific. La mer 53: 1–18. Teeling, H., B. M. Fuchs, D. Becher, and others. 2012. Substrate-controlled succession of marine bacterioplankton populations induced by a phytoplankton bloom. Science (80-. ). 336: 608–611. Teixeira de Mattos, M. J., and O. M. Neijssel. 1997. Bioenergetic consequences of microbial adaptation to low-nutrient environments. J. Biotechnol. 59: 117–26. Tempest, D. W., and O. M. Neijssel. 1992. Physiological and energetic aspects of bacterial metabolite overproduction. FEMS Microbiol. Lett. 100: 169–170. doi:10.1016/0378-1097(92)90205-3 Todd, J. D., M. Kirkwood, S. Newton-Payne, and A. W. B. Johnston. 2012. DddW, a third DMSP lyase in a model Roseobacter marine bacterium, Ruegeria pomeroyi DSS-3. ISME J. 6: 223–226. doi:10.1038/ismej.2011.79 Tortell, P. D., M. T. Maldonado, J. Granger, and N. M. Price. 1999. Marine bacteria and biogeochemical cycling of iron in the oceans. FEMS Microbiol. Ecol. 29: 1–11. doi:10.1111/j.1574-6941.1999.tb00593.x Tortell, P. D., M. T. Maldonado, and N. M. Price. 1996. The role of heterotrophic bacteria in iron-limited ocean ecosystems. Lett. to Nat. 383: 330–331. Tottey, S., D. R. Harvie, and N. J. Robinson. 2005. Understanding how cells allocate metals using metal sensors and metallochaperones. Acc. Chem. Res. 38: 775–783. doi:10.1021/ar0300118 Tovar-Sanchez, A., S. A. Sañudo-Wilhelmy, M. Garcia-Vargas, R. S. Weaver, L. C. Popels, and D. A. Hutchins. 2003. A trace metal clean reagent to remove surface-bound iron from marine phytoplankton. Mar. Chem. 82: 91–99. doi:10.1016/j.marchem.2004.01.002 Twining, B. S., and S. B. Baines. 2013. The trace metal composition of marine phytoplankton. Ann. Rev. Mar. Sci. 5: 191–215. doi:10.1146/annurev-marine-121211-172322 Twining, B. S., S. Rauschenberg, P. L. Morton, and S. Vogt. 2015. Metal contents of phytoplankton and labile particulate material in the North Atlantic Ocean. Prog. Oceanogr. 137: 261–283. doi:10.1016/j.pocean.2015.07.001 Uematsu, M., R. A. Duce, J. M. Prospero, L. Chen, J. T. Merrill, and R. L. McDonald. 1983. Transport of Minearl Aerosol From Asia Over the North Pacific Ocean. J. Geophys. Res. 88: 3343–3332. Ueno, H., and I. Yasuda. 2003. Intermediate water circulation in the North Pacific subarctic and northern subtropical regions. J. Geophys. Res. 109: 2222. doi:10.1029/2003JC002222 Venter, J. C., K. Remington, J. F. Heidelberg, and others. 2004. Environmental Genome Shotgun Sequencing of the Sargasso Sea. Science (80-. ). 304: 66–74. Vogel, C., and N. S. Fisher. 2010. Metal accumulation by heterotrophic marine bacterioplankton. Limnol. Oceanogr. 55: 519–528. doi:10.4319/lo.2009.55.2.0519 Vrede, K., M. Heldal, S. Norland, and G. Bratbak. 2002. Elemental Composition ( C , N , P ) and Cell Volume of Exponentially Growing and Elemental Composition ( C , N , P ) and Cell Volume of Exponentially Growing and Nutrient-Limited Bacterioplankton. Appl. 68: 2965–2971. doi:10.1128/AEM.68.6.2965 Vrede, T., D. R. Dobberfuhl, S. A. L. M. Kooijman, and J. J. Elser. 2004. Fundamental connections among organism C : N : P Stoichiometetry, macromolecular composition, and growth. Ecology 85: 1217–1229. doi:10.1890/02-0249 Vu, H. T. D., and Y. Sohrin. 2013. Diverse stoichiometry of dissolved trace metals in the Indian Ocean. Sci. Rep. 3: 1–5. doi:10.1038/srep01745 Walve, J., J. Gelting, and J. Ingri. 2014. Trace metals and nutrients in Baltic Sea cyanobacteria : Internal and external fractions and potential use in nitrogen fixation. Mar. Chem. 158: 27–38. doi:10.1016/j.marchem.2013.11.002 Weaver, R. S., D. L. Kirchman, and D. A. Hutchins. 2003. Utilization of iron/organic ligand complexes by marine bacterioplankton. Aquat. Microb. Ecol. 31: 227–239. 156  Whitman, W. B., D. C. Coleman, and W. J. Wiebe. 1998. Perspective Prokaryotes : The unseen majority. Proc. Natl. Acad. Sci. 95: 6578–6583. Whitney, F. A., W. R. Crawford, and P. J. Harrison. 2005. Physical processes that enhance nutrient transport and primary productivity in the coastal and open ocean of the subarctic NE Pacific. Deep. Res. Part II Top. Stud. Oceanogr. 52: 681–706. doi:10.1016/j.dsr2.2004.12.023 Wickham, H. 2009. ggplot2: elegant graphics for data analysis. Springer New York. Wickham, H. 2016. Dplyr: A Grammar of Data Manipulation, version 0.5.0 (R package). https://github.com/tidyverse/dplyr. Wilhelm, S. W., A. L. King, B. S. Twining, and others. 2013. Elemental quotas and physiology of a southwestern pacific ocean plankton community as a function of iron availability. Aquat. Microb. Ecol. 68: 185–194. doi:10.3354/ame01611 Wilke, C. O., and H. Wickham. 2016. cowplot: Streamlined Plot Theme and Plot Annotations for “ggplot2” (R package). https://cran.r-project.org/web/packages/cowplot/index.html. Williams, R. J. P. 2001. Chemical selection of elements by cells. Coord. Chem. Rev. 216–217: 583–595. doi:10.1016/S0010-8545(00)00398-2 Williams, R. J. P., and J. J. R. Fraústo da Silva. 2003. Evolution was Chemically Constrained. J. Theor. Biol. 220: 323–343. doi:10.1006/jtbi.2003.3152 Williams, T. J., D. Wilkins, E. Long, F. Evans, M. Z. Demaere, M. J. Raftery, and R. Cavicchioli. 2013. The role of planktonic Flavobacteria in processing algal organic matter in coastal East Antarctica revealed using metagenomics and metaproteomics. Environ. Microbiol. 15: 1302–1317. doi:10.1111/1462-2920.12017 Wong, C. S., Z. Yu, W. K. Johnson, R. J. Matear, and F. A. Whitney. 2000. Dissolved Organic Carbon in Sub-Arctic Northeast Pacific Ocean, p. 299–309. In E.T. and T.H. N. Handa [ed.], Dynamics and Characterization of Marine Organic Matter. Springer Netherlands. Wright, J. 2013. Microbial community structure and ecology of marine Group A bacteria in the oxygen minimum zone of the Northeast subarctic Pacific Ocean. PhD Thesis, University of British Columbia,pp. 53-54. Xiu, P., A. P. Palacz, F. Chai, E. G. Roy, and M. L. Wells. 2011. Iron flux induced by Haida eddies in the Gulf of Alaska. Geophys. Res. Lett. 38: 1–5. doi:10.1029/2011GL047946 Yiğiterhan, O., J. W. Murray, and S. Tuğrul. 2011. Trace metal composition of suspended particulate matter in the water column of the Black Sea. Mar. Chem. 126: 207–228. doi:10.1016/j.marchem.2011.05.006 Yilmaz, P., P. Yarza, J. Z. Rapp, and F. O. Gl??ckner. 2016. Expanding the world of marine bacterial and archaeal clades. Front. Microbiol. 6: 1–29. doi:10.3389/fmicb.2015.01524 Yokokawa, T., T. Nagata, M. T. Cottrell, and D. L. Kirchman. 2004. Growth rate of the major phylogenetic bacterial groups in the Delaware estuary. Limnol. Oceanogr. 49: 1620–1629. Zamzow, H., K. H. Coale, K. S. Johnson, and C. M. Sakamoto. 1998. Determination of copper complexation in seawater using flow injection analysis with chemiluminescence detection. Anal. Chim. Acta 377: 133–144. Zeng, J., and S. Spiro. 2013. Finely tuned regulation of the aromatic amine degradation pathway in Escherichia coli. J. Bacteriol. 195: 5141–5150. doi:10.1128/JB.00837-13 Zhang, X. Y., R. Arimoto, and Z. S. An. 1997. Dust emission from Chinese desert sources linked to variations in atmospheric circulation. J. Geophys. Res. 102: 28041–28047. Zhang, Y., and V. N. Gladyshev. 2010. General trends in trace element utilization revealed by comparative genomic analyses of Co, Cu, Mo, Ni, and Se. J. Biol. Chem. 285: 3393–405. doi:10.1074/jbc.M109.071746 Zhang, Y., D. A. Rodionov, M. S. Gelfand, and V. N. Gladyshev. 2009. Comparative genomic analyses of nickel, cobalt and vitamin B12 utilization. BMC Genomics 10: 78. doi:10.1186/1471-2164-10-78   157  Appendices  Appendix A  Supplementary material for Chapter 2  Table A.1: Comparison of samples from station P26 (collected during Aug 2011 cruise) after ~ 3 months of acidic storage (analysis Jan‒Feb 2012) and ~ 4 years of acidic storage (analysis in July 2015) without UV (storage only) and with UV treatment (storage + 2 h of UV). Please note that the samples analyzed in Table A.1 and A.2 are from different bottles, and might not be directly comparable.              Sample Storage only (~ 3 months) dCu (nmol kg-1) ± StDev 2011 analysis Storage only (~ 4 years) dCu (nmol kg-1) ± StDev 2015 analysis Storage (~ 4 years) + UV dCu (nmol kg-1) ± StDev 2015 analysis 25 m 1.4 ± 0.03 1.73 ± 0.01 1.79 ± 0.06 40 m 1.4 ± 0.05 1.93 ± 0.03 2.12 ± 0.07 50 m 1.42 ± 0.09 1.97 ± 0.05 2.31 ± 0.08 75 m 1.47 ± 0.1 2.12 ± 0.02 2.43 ± 0.07 100 m 1.51 ± 0.03 2.19 ± 0.02 2.28 ± 0.15 200 m 2.07 ± 0.05 2.60 ± 0.12 2.96 ± 0.01 400 m 2.43 ± 0.02 2.71 ± 0.03 2.92 ± 0.04 1400 m n/a 3.18 ± 0.06 3.06 ± 0.06 1600 m 2.57 ± 0.05 3.27 ± 0.03 3.19 ± 0.01 158   Table A.2:  Dissolved Cu (nmol kg-1) in samples from OSP at different time points after acidification to pH 2 measured using FIA-CL, with and without UV (2 h). Measurements of dCu at 48 h, 2 weeks and 2 months were performed on samples that were collected during Aug 2011-27 cruise, and kept frozen (‒20°C) until 28th of July 2015. These measurements are compared to samples from the same cruise, that were acidified after the cruise and measured in July 2015 with 2 h of UV and (effective storage time at pH 2 of ~ 4 years). We also performed a longer UV treatment of 2 samples stored for 2 months (4 h, marked with asterisk). This dataset is plotted in Fig. 2.3 of Chapter 2. Please note that the samples analyzed in Table A.1 and A.2 are from different bottles, and might not be directly comparable.     Storage time Sample No UV dCu (nmol kg-1) ± StDev UV dCu (nmol kg-1) ± StDev 48 h 50 m 0.76 ± 0.06 1.21 ± 0.017 Aug 1st 2015 75 m 1.07 ± 0.02 1.41 ± 0.013  200 m 1.19 ± 0.03 1.76 ± 0.055  400 m 1.21 ± 0.01 1.83 ± 0.01  1400 m 1.29 ± 0.005 2.02 ± 0.02  1600 m 1.39 ± 0.02 2.19 ± 0.03 2 weeks 50 m 1.19 ± 0.02 1.43 ± 0.07 Aug 17th 2015 75 m 1.43 ± 0.03 1.69 ± 0.008  200 m 1.69 ± 0.007 2.10 ± 0.06  400 m 1.96 ± 0.09 2.20 ± 0.014  1400 m 1.79 ± 0.02 2.47 ± 0.07  1600 m 2.06 ± 0.03 2.71 ± 0.04 2 months 50 m 1.55 ± 0.01 1.76 ± 0.06 Oct 6th 2015 50 m* 1.51 ± 0.01  1.82 ± 0.02*  75 m 2.03 ± 0.02 2.17 ± 0.01  75 m* 2.03 ± 0.02  2.10 ± 0.14*  400 m 2.58 ± 0.03 2.88 ± 0.06  1400 m 2.38 ± 0.02 2.99 ± 0.02  1600 m 3.18 ± 0.06 3.40 ± 0.20 ~ 4 years 50 m  2.31 ± 0.02 July 24th 2015 75 m  2.43 ± 0.17  200 m  2.96 ± 0.01  400 m  2.92 ± 0.08  1400 m  3.06 ± 0.06 159    Table A.3: Dissolved Cu concentrations at OSP for 2010, 2011 and 2012 cruises. The dCu dataset for 2011-27 (FIA-CL method) and 2012-13 cruise (CSV method) are plotted in Fig. 2.4A of Chapter 2 and the dataset from all three cruises are plotted in Fig. 2.8 (samples below 5‒400 m).   Cruise pH Method UV Date of analysis/ storage Depth (m) dCu (nmol kg-1) ± StDev 2010-14 2 FIA-CL 2 h 25/07/2015 5 2.38 ± 0.03     ~ 5 years acidic 10 2.32 ± 0.04     storage 25 1.72 ± 0.03      40 1.88 ± 0.05      75 1.85 ± 0.03      100 2.55 ± 0.03      200 2.04 ± 0.01      300 2.84 ± 0.13      400 3.06 ± 0.02      600 2.95 ± 0.02      800 2.97 ± 0.02 2011-27 2 FIA-CL 2 h 24/07/2015 10 1.83 ± 0.05     ~ 4 years acidic 25 1.79 ± 0.06     storage 40 2.12 ± 0.07      50 2.31 ± 0.02      75 2.43 ± 0.17      100 2.28 ± 0.15      130 2.29 ± 0.04      150 2.43 ± 0.07      185 2.78 ± 0.10      200   2.96 ± 0.01  160   Cruise pH Method UV Date of analysis/ storage Depth (m) dCu (nmol kg-1) ± StDev 2011-27 2 FIA-CL 2 h 24/07/2015 300 2.74 ± 0.08     ~ 4 years acidic 400 2.92 ± 0.08     storage 600 2.90 ± 0.19      800 2.83 ± 0.02      1100 2.94 ± 0.09      1200 2.95 ± 0.04      1400 3.06 ± 0.06      1600 3.19 ± 0.01      1800 3.22 ± 0.10      2000 3.30 ± 0.06 2012-13 natural CSV 45 min No acidic storage 10 2.32 ± 0.1      25 2.25 ± 0.1      50 2.02 ± 0.1      75 2.21 ± 0.1      100 2.35 ± 0.1      200 2.95 ± 0.1      400 2.93 ± 0.1      600 2.93 ± 0.3      800 2.83 ± 0.2      1000 3.02 ± 0.1      1200 3.12 ± 0.4      1400 3.21 ± 0.2      161  Table A.4: Station locations and methodologies dCu studies in the North Pacific plotted in Fig. 2.5 of Chapter 2 (Storage period was included where reported) Study Station Latitude Longitude Date sampled Cruise Method  Tanita et al., 2015 5 47.00°N 160.07°E Aug-Sep, 2008 KH-08-2 CSVa, UV for some samples  22 11.50°N 155.00°E Aug-Sep, 2008  16 months of acidic sample storage Fujishima et al., 2001 17 44.98°N 140.01°W Aug, 1997 KH-97-2 Solid-phase extraction  19 47.02°N 160.00°W Aug, 1997  MAF-8HQb /ICP-MS analysis, at least 3 months at low pH Ezoe et al., 2004 BO01 39.98°N 160.00°E June, 2000 KH-00-3  Solid-phase extraction   BO02 39.98°N 170.98°W June, 2000  MAF-8HQb /ICP-MS analysis  BO03 30.00°N 170.98°W July, 2000  1-3 years of sample storage  BO04 16.98°N 160.00°W July, 2000    BO05 19.98°N 175.00°W July, 2000    BO06 20.00°N 168.98°E July, 2000    BO07 21.98°N 150.98°E July, 2000   Moffett and Dupont, 2007 4 47.02°N 170.50°W July, 2003  Isotope dilution with Mg(OH)2  6 47.00°N 170.33°E July, 2003  co-precipitation, ICP-MS analysis       (after Wu & Boyle (1998)       and further modified by Saito & Schneider (2006) Martin et al., 1989 T5 39.60°N 140.77°W July, 1987 VERTEX  APDC/DDDCc- chloroform extraction  T6 45.00°N 142.87°W Aug, 1987 VII (after Bruland et al 1979)  T7  50.00°N 145.00°W Aug, 1987  GFAASd analysis  T8 55.50°N 147.50°W Aug, 1987    T9 58.68°N 147.95°W Aug, 1987   Coale & Bruland, 1990 T4 33.00°N 139.00°W July, 1987 VERTEX APDC/DDDCc- chloroform extraction  T5 39.36°N 140.77°W July, 1987 VII (after Bruland et al 1979, 1985)  T6 45.00°N 142.87°W Aug, 1987    T7  50.00°N 145.00°W Aug, 1987    T8 55.50°N 147.50°W Aug, 1987   Takano et al., 2014 TR7 29.99°N 165.01°E June,2011 KH-11-7 Solid-phase extraction    TR-15 51.00°N 165.01°E June,2011 KH-11-7 NOBIAS-PA1 & AG MP1/MC-ICP-MS 162  Study Station Latitude Longitude Date sampled Cruise Method Takano et al. ,2014 CR27 46.97°N 159.99°E June, 2010 KH-10-2   BD 17 43.00°N 132.40°W Sept, 2012 KH-12-4   BD 11 48.45°N 128.43°W Sept/Oct 2012 KH-12-4  Boyle et al., 1977 219 53.10°N 177.28°W Oct, 1973 GEOSECS Pre-concentration with cobalt   226 30.57°N 170.60°E Nov,1973  Pyrrolidine, dithiocarbamate/ FAAe analysis  340 10.47°N 123.63°W June, 1974  > 1 year of acidified sample storage  345 22.52°N 122.20°W June, 1974    202 34.10°N 139.57°W Aug, 1973          Bruland, 1980 H-17 32.68°N 144.98°W Sept, 1977 H-77  APDC/DDDCcchloroform extraction       FAAe analysis Biller and Bruland, 2012 SAFe 30.00°N 140.00°W Summer 2009 U.S. GEOTRACES Solid-phase-extraction Nobias PA1 resin/MC-ICP-MS analysis                a Cathodic Stripping Voltammetry (CSV) with Salicylaldoxime (SA) method of Campos & van den Berg, 1994 b column extraction with 8-hydroxyquinoline immobilized on fluorinated metal alkoxide glass (MAF-8HQ) c ammonium 1-pyrrolidinedithiocarbamate (APDC) and diethylammonium diethyldithiocarbamate (DDDC) d Graphite Furnace Atomic Absorption Spectrophotometry e Flameless Atomic Absorption  References to Table A.4  Biller, D. V., and K. W. Bruland. 2012. Analysis of Mn, Fe, Co, Ni, Cu, Zn, Cd, and Pb in seawater using the Nobias-chelate PA1 resin and magnetic sector inductively coupled plasma mass spectrometry (ICP-MS). Mar. Chem. 130–131: 12–20.  Boyle, E.., F. R. Sclater, and J. M. Edmond. 1977. The distribution of dissolved copper in the Pacific. Earth Planet. Sci. Lett. 37: 38–54.  Bruland, K. K. W. 1980. Oceanographic distributions of cadmium, zinc, nickel, and copper in the North Pacific. Earth Planet. Sci. Lett. 47: 176–198.  Coale, K. H., and K. W. Bruland. 1990. Spatial and temporal variability in copper complexation in the North Pacific. Deep Sea Res. Part A. Oceanogr. Res. Pap. 37: 317–336. N  163  Ezoe, M., T. Ishita, M. Kinugasa, X. Lai, K. Norisuye, and Y. Sohrin. 2004. Distributions of dissolved and acid-dissolvable bioactive trace metals in the North Pacific. Geochem. J. 38: 535–550.  Fujishima, Y., K. Ueda, M. Mauro, E. Nakayama, C. Tokutome, H. Hasegawa, M. Matsui, and Y. Sohrin. 2001. Distribution of Trace Bioelements in the Subarctic North Pacific Ocean and the Bering Sea (the R/V Hakuho Maru Cruise KH-97-2). J. Oceanogr. 57: 261–273.  Martin, J. H., R. M. Gordon, S. Fitzwater, and W. William. 1989. VERTEX: phytoplankton / iron studies in the Gulf of Alaska. Deep Sea Res. 36.  Moffett, J. W., and C. Dupont. 2007. Cu complexation by organic ligands in the sub-arctic NW Pacific and Bering Sea. Deep Sea Res. Part I Oceanogr. Res. Pap. 54: 586–595.  Takano, S., M. Tanimizu, T. Hirata, and Y. Sohrin. 2014. Isotopic constraints on biogeochemical cycling of copper in the ocean. Nat. Commun. 5: 5663.  Tanita, I., S. Takeda, M. Sato, and K. Furuya. 2015. Surface and middle layer enrichment of dissolved copper in the western subarctic North Pacific. La mer 53: 1–18.                        164  Table A.5: Comparison of protocols used in seawater dCu determination. GEOTRACES Inter-calibration standards and certified reference materials were included where reported. The consensus values for the GEOTRACES standards are shown in Supplementary Table A.6. Study Study Region/Cruise Sample storage  Filter Analytical method Sample pH UV conditions Reference material or sample nmol kg-1, nmol L-1 Sohrin et al., 2008 North Pacific KH-99-3 KH-04-3 Not reported 0.2 µm Nucleopore  NOBIAS PA1 resin Dynamic Reaction Cell (DRC)- ICP-MS pH 2.0 no UV SAFe S = 0.404 ± 0.01 nmol kg-1 SAFe D2 =1.83 ± 0.06 nmol kg-1  Vu and Sohrin, 2013  Bay of Bengal, Arabian Sea, Indian Ocean GEOTRACES KH-09-5   Not reported  0.2 µm Nucleopore   NOBIAS PA1 resin High Resolution (HR)-ICP-MS (method of Sohrin et al. 2008)  pH 2.2  no UV  SAFe S = 0.56 ± 0.01 nmol kg-1 SAFe D1 = 2.37 ± 0.07 nmol kg-1 Boye et al., 2012 Southeastern Atlantic Southern Ocean MD166 BONUS- Good Hope cruise GEOTRACES, GIPY4 ~ 22 months 0.22 µm Sartobran 300 Toyopearl-AF-650M resin/ Isotope Dilution/ICP-MS (method of Milne et al. 2010)  pH 1.9 no UV SAFe S = 0.56 ± 0.03 nmol L-1 SAFe D2 = 1.39 ± 0.13 nmol L-1 GS = 0.72 ± 0.07 nmol L-1 GD = 1.12 ± 0.04 nmol L-1  Butler et al., 2013 Australian sector of the Southern Ocean, SAZ-Sense voyage GEOTRACES GIPY6/AU806 > 3 months 0.2 µm Pall Supor membrane, Acropack 200  Toyopearl-AF-650M resin/ ICP-MS pH 1.9 no UV SAFe S = 0.56 ± 0.07 nmol L-1 SAFe D2 = 2.26 ± 0.11 nmol L-1 NASS-5 = 4.82 ± 0.25 nmol L-1  Jacquot et al., 2013 Eastern South Pacific AT-15-61 at least 1 month 0.2 µm Pall  Acropack  NTA resin/ isotope dilution/ICPM-MS (method of Lee et al. 2011)  < pH 2.0 no UV SAFe S = 0.54 ± 0.05 nmol L-1 SAFe D1 = 2.35 ± 0.19 nmol L-1  Jacquot and Moffett, 2015 Tropical-subtropical Atlantic Ocean GEOTRACES GA03 KN199-4 KN204-1 2 months 0.2 µm Pall Acropack, Supor capsule filter NTA resin/isotope dilution/ ICP-MS (method of Lee et al. 2011)  pH 1.7 no UV GS = 0.872 ± 0.014 nmol L-1 GD = 1.72 ± 0.029 nmol L-1  Roshan and Wu, 2015 Tropical-subtropical Atlantic Ocean GEOTRACES GA03 KN199-4 KN204-1  Not reported 0.2 µm Pall Acropack, Supor capsule Isotope Dilution with Mg(OH)2 co-precipitation/ MC-HR-ICPMS pH 1.7‒2.0 no UV SAFe S = 0.51 ± 0.04 nmol L-1 SAFe D2 = 2.35 ± 0.09 nmol L-1  Lee et al., 2011  BATS  2008 GEOTRACES IC1 inter-calibration  Not reported 0.4 µm Nucleopore NTA-type Superflow resin/ isotope dilution/ Quadrupole ICP-MS  SAFe MIT-pH 2.0 Station samples pH 2 no UV SAFe MIT “A”= 0.58 ± 0.05 nmol L-1 SAFe D2 = 2.49 ± 0.1 nmol L-1 165  Study Study Region/Cruise Sample storage  Filter Analytical method Sample pH UV conditions Reference material or sample nmol kg-1, nmol L-1 Takano et al., 2014 Eastern & Western Pacific, Indian Ocean KH-09-4 KH-10-2 KH-11-7  Not reported 0.2 µm Acropack (Pall) NOBIAS PA1 chelate followed by AG-MP1/ICP-MS (full method in Takano et al, 2013) pH 1.7‒2.2 no UV NASS-6 = 0.224 ± 0.004 μg L−1 CASS-5 = 0.366 ± 0.004 μg L−1 (in Takano et al., 2013)  Lagerström et al., 2013  Atlantic (BATS) and Pacific (SAFe) GEOTRACES stations Profile samples collected in 2008 (BATS) and 2009 (SAFe)  0.2 μm  capsule filters NOBIAS PA1 resin,  Isotope Dilution/ and comparison with external calibration/ On-line FIA ICP-MS  pH 2.0 No UV SAFe S = 0.48 ± 0.03 nmol kg-1 SAFe D2 = 2.33 ± 0.15 nmol kg-1 GS = 0.86 ± 0.05 nmol kg-1 GS = 1.59 ± 0.08 nmol kg-1 O’Sullivan et al., 2013 SAFe samples  Not reported  In-line FIA using Toyopearl- AF-650M resin, ICP-MS  pH 1.7 no UV SAFe S = 0.51 ± 0.07 nmol L-1 SAFe D2 = 1.99 ± 0.13 nmol L-1 NASS-5 = 4.91 ± 0.25 nmol L-1  This study   Subarctic Northeast Pacific, Line P  ~ 3 months  ~ 4 years  0.22 µm AcroPak FIA-CL a (Zamzow et al., 1998)  pH 2.0  2 h 125 W High-pressure Hg lamp  SAFe D1= 2.24 ± 0.19 nmol kg-1* SAFe D1= 2.31 ± 0.11 nmol kg-1 SAFe D2 = 2.33 ± 0.12 nmol kg-1* SAFe D2 = 2.27 ± 0.15 nmol kg-1 NASS 6 = 3.88 ± 0.15 nmol kg-1 *analyzed with UV Biller and Bruland, 2012 U.S. GEOTRACES Atlantic (near Bermuda Atlantic Time series [BATS]) and Pacific (near SAFe) Baseline stations, offshore California Current Seawater (CCS) for UV experiments Atlantic samples collected in 2008 Pacific samples collected in 2009 CCS-May 2010  0.45 µm Osmotics (Atlantic samples) 0.2 µm Acropack (Pacific samples) NOBIAS PA1 chelate/ Magnetic sector ICP-MS pH 1.8 90 min 18 mW cm-2          SAFe S = 0.48 ± 0.02 nmol kg-1 SAFe D2 = 2.21 ± 0.02 nmol kg-1 GS= 0.87 ± 0.013 nmol kg-1 GD= 1.55 ± 0.04 nmol kg-1  In CCS ~ 30% increase in labile Cu when sample was UV oxidized prior to analysis  Milne et al., 2010   BATS, GEOTRACES inter-calibration  June 2008 AcroPak 200 capsule filters with 0.8µm/0.2µm Supor membranes (Pall)  Toypearl AF- 650M resin/ Isotope Dilution/ HR-ICP-MS pH 1.7 1 h 119 mW cm-2 SAFe S = 0.6 ± 0.04 nmol L-1* SAFe D2 = 2.35 ± 0.02 nmol L-1* SAFe D2 = 2.12 & 2.09 (NO UV) *analyzed with UV 166  Study Study Region/Cruise Sample storage  Filter Analytical method Sample pH UV conditions Reference material or sample nmol kg-1, nmol L-1 Tanita et al., 2015      Western Pacific Ocean 16 months 0.22 µm Millipack 100 (Millipore)    CSVb  (Campos & van den Berg, 1994)  pH 1.7  4 h 150 W lamp-for some samples, not reference materials  NASS-5 = 5.16 ± 0.19 nmol L-1 SAFe S = 0.5 ± 0.03 nmol L-1 SAFe D2 = 2.35 ± 0.15 nmol L-1     Middag et al., 2015 Bermuda Atlantic Time series (BATS) 2 years ‒ Netherlands GEOTRACES samples  1 year ‒  U.S GEOTRACES samples  0.2 µm Sartobran ‒ Netherlands   0.2 µm  Acropak ‒ U.S. NOBIAS PA1 chelate/ ICP-MS (method of Biller and Bruland, 2012) pH ~ 1.7 (US) pH ~ 1.85 (Netherlands) Yes, but conditions not reported  SAFe S = 0.53 ± 0.05 nmol kg-1 SAFe D2 = 2.2 ± 0.1 nmol kg-1 GS = 0.8 ± 0.1 nmol kg-1 GD = 1.5 ± 0.1 nmol kg-1  DCu profile at BATS: US team: dCu increased on average by 13% after UV in samples above 1000 m and 12 % below 1000 m. 7 out of 29 samples that were UV oxidized were < 5% different  Netherlands team: dCu increased on average 22 % after UV in samples above 1000 m and 9 % below 1000 m. 7 out of 24 samples were < 5% different.  Achterberg et al., 2001 Tamar Estuary & coastal Celtic Sea Tamar sample processed immediately upon sampling  Celtic Sea samples stored at -20°C and acidified prior to analysis 0.45 µm cellulose nitrate, (Whatman) – Tamar samples  0.4 μm Nucleopore ‒ Celtic Sea samples 8-hydroxyquinoline with FIA-CL,  CSV pH 2.0 0‒12 h of UV was tested 15 mM H2O2 Online & batch UV methods  400 W lamp   NASS-4  SLEW-2  (both within 5% of certified values) After online UV oxidation (with 10 Mn H2O2) the concentrations of dCu were ~ 5‒29% higher than in non-UV oxidized, acidified samples (acidified prior to analysis)  For batch UV oxidation, estuarine samples required at least 4 h of treatment in the presence of 15 mM H2O2 167  Study Study Region/Cruise Sample storage  Filter Analytical method Sample pH UV conditions Reference material or sample nmol kg-1, nmol L-1 Ndung’u et al, 2003       San Francisco Bay Estuary ≥ 1 month for PDC/DDCc   7- 9 years for Solid-phase extraction method 0.22 μm PDC/DDCc with GFAASd   Solid-phase extraction Toyopearl AF chelate 650M resin/ ICP-MS pH 1.7‒1.8 15 min‒4h 9.2 mWcm-2 1200 W  Medium pressure Hg lamp  SLEW-3 (value not reported)  Samples analyzed with solid phase extraction – ICPMS were 10-20%| lower than PDC/DDC with GFAAS if not UV oxidized. After UV oxidation of samples prior to resin extraction-ICPMS both methods agreed a Flow Injection Analysis with chemiluminescence detection (FIA-CL) b Cathodic Stripping Voltammetry (CSV) with Salicylaldoxime (SA) method of Campos & van den Berg, 1994 c 1-pyrrolidinedithiocarbamate (PDC)/ diethyldithiocarbamate (DDC) liquid-liquid extraction d Graphite Furnace Atomic Adsorption Spectroscopy  References to information presented in Table A.5 Achterberg, E. P., C. B. Braungardt, R. C. Sandford, and P. J. Worsfold. 2001. UV digestion of seawater samples prior to the determination of copper using flow injection with chemiluminescence detection. Anal. Chim. Acta 440: 27–36n.  Biller, D. V., and K. W. Bruland. 2012. Analysis of Mn, Fe, Co, Ni, Cu, Zn, Cd, and Pb in seawater using the Nobias-chelate PA1 resin and magnetic sector inductively coupled plasma mass spectrometry (ICP-MS). Mar. Chem. 130–131: 12–20.  Boye, B. D. Wake, P. Lopez Garcia, and others. 2012. Distributions of dissolved trace metals (Cd, Cu, Mn, Pb, Ag) in the southeastern Atlantic and the Southern Ocean. Biogeosciences 9: 3231–3246.  Butler, E. C. V, J. E. O’Sullivan, R. J. Watson, A. R. Bowie, T. A. Remenyi, and D. Lannuzel. 2013. Trace metals Cd, Co, Cu, Ni, and Zn in waters of the subantarctic and Polar Frontal Zones south of Tasmania during the “SAZ-Sense” project. Mar. Chem. 148: 63–76.  Jacquot, J. E., Y. Kondo, A. N. Knapp, and J. W. Moffett. 2013. The speciation of copper across active gradients in nitrogen-cycle processes in the eastern tropical South Pacific. Limnol. Oceanogr. 58: 1387–1394.  Jacquot, J. E., and J. W. Moffett. 2015. Copper distribution and speciation across the International GEOTRACES Section GA03. Deep. Res. Part II Top. Stud. Oceanogr. 116: 187–207.  Lagerström, M. E., M. P. Field, M. Séguret, L. Fischer, S. Hann, and R. M. Sherrell. 2013. Automated on-line flow-injection ICP-MS determination of trace metals (Mn, Fe, Co, Ni, Cu and Zn) in open ocean seawater: Application to the GEOTRACES program. Mar. Chem. 155: 71–80.  Lee, J.-M., E. a Boyle, Y. Echegoyen-Sanz, J. N. Fitzsimmons, R. Zhang, and R. a Kayser. 2011. Analysis of trace metals (Cu, Cd, Pb, and Fe) in seawater using single batch nitrilotriacetate resin extraction and isotope dilution inductively coupled plasma mass spectrometry. Anal. Chim. Acta 686: 93–101.  Milne, A., W. Landing, M. Bizimis, and P. Morton. 2010. Determination of Mn, Fe, Co, Ni, Cu, Zn, Cd and Pb in seawater using high resolution magnetic sector inductively coupled mass spectrometry (HR-ICP-MS). Anal. Chim. Acta 665: 200–207.  Ndung’u, K., R.P. Franks, K.W. Bruland, and A.R. Flegal. 2003. Organic complexation and total dissolved trace metal analysis in estuarine waters: Comparison of solvent-extraction graphite furnace atomic absorption spectrometric and chelating resin flow injection inductively coupled plasma-mass spectrometric analysis. Anal. Chim. Acta 481: 127–138. O’Sullivan, J. E., R. J. Watson, and E. C. V Butler. 2013. An ICP-MS procedure to determine Cd, Co, Cu, Ni, Pb and Zn in oceanic waters using in-line flow-injection with solid-phase extraction for preconcentration. Talanta 115: 999–1010.  168  Roshan, S., and J. Wu. 2015. Water mass mixing: The dominant control on the zinc distribution in the North Atlantic Ocean. Global Biogeochem. Cycles 29:1060–1074. Sohrin, Y., S. Urushihara, S. Nakatsuka, T. Kono, E. Higo, T. Minami, K. Norisuye, and S. Umetani. 2008. Multielemental Determination of GEOTRACES Key Trace Metals in Seawater by ICPMS after Preconcentration Using an Ethylenediaminetriacetic Acid Chelating Resin. Anal. Chem. 80: 6267–6273.  Takano, S., M. Tanimizu, T. Hirata, and Y. Sohrin. 2014. Isotopic constraints on biogeochemical cycling of copper in the ocean. Nat. Commun. 5: 5663. Tanita, I., S. Takeda, M. Sato, and K. Furuya. 2015. Surface and middle layer enrichment of dissolved copper in the western subarctic North Pacific. La mer 53: 1–18.  Vu, H. T. D., and Y. Sohrin. 2013. Diverse stoichiometry of dissolved trace metals in the Indian Ocean. Sci. Rep. 3: 1745.     Table A.6: GEOTRACES reference material consensus values as of May 2013. North Pacific samples - SAFe were collected in 2004 (S-surface, D1, and D2‒1000 m), and North Atlantic (BATS) samples - GS (surface) and GD (2000 m) were collected in 2008 and acidified to pH 1.7. (http://es.ucsc.edu/~kbruland/GeotracesSaFe/kwbGeotracesSaFe.html)   Reference material Consensus as of May 2013 (nmol kg-1) SAFe S 0.52 ± 0.05 nmol kg-1 SAFe D1 2.28 ± 0.15 nmol kg-1 SAFe D2 2.27 ± 0.11 nmol kg-1 GS 0.84 ± 0.06 nmol kg-1 GD 1.62 ± 0.07 nmol kg-1    169  Table A.7: Calculations of vertical copper fluxes at Ocean Station Papa (OSP).     We used one dimensional flux equation and same parameters that have been applied for determining fluxes of Fe at OSP (Johnson et al. 2005; Xiu et al. 2011):  Ftotal = wR + Kz (dR/dz), where w is the vertical velocity (0.05 m d-1 for OSP, Xiu et al., 2011), R is the Cu concentration in deep water (at 200 m), Kz is the vertical diffusion coefficient (0.6 m2 d-1, Johnson et al., 2005; de Baar et al., (1995)), and dR/dz is the vertical Cu concentration gradient across the pycnocline)   References  de Baar, H. J. W., J. T. M. de Jong, D. C. E. Bakker, B. M. Löscher, C. Veth, U. V. Bathmann, and V. Smetacek. 1995. Importance of iron for plankton blooms and carbon dioxide drawdown in the Southern Ocean. Nature 373: 412–415.   Johnson, W. K., L. A. Miller, N. E. Sutherland, and C. S. Wong. 2005. Iron transport by mesoscale Haida eddies in the Gulf of Alaska. Deep. Res. Part II Top. Stud. Oceanogr. 52: 933–953.   Xiu, P., A. P. Palacz, F. Chai, E. G. Roy, and M. L. Wells. 2011. Iron flux induced by Haida eddies in the Gulf of Alaska. Geophys. Res. Lett. 38: 1–5.Parameter Units dCu   Mean dCu deep water concentration (200-2000 m)   µmol m-3   3.0 Vertical gradient of dCu (25-200 m)  µmol m-4  0.018 Advective flux (w = 0.05 m d-1)  µmol m-2 d-1  0.150 Diffusive flux (Kz= 0.6 m2 d-1)  µmol m-2 d-1  0.011 Total vertical flux  µmol m-2 d-1 0.161 Cu supplied to the euphotic zone    Winter mixed layer (~100m)  nmol L-1 y-1  0.58 Summer mixed layer (~ 30 m) nmol L-1 y-1 1.95 170     Figure A.1: Comparison of dCu datasets obtained with FIA-CL between Jan-Feb 2011 (after ~ 3 months of storage, no UV) and July-August 2015 (after ~ 4 years of storage, with 2 h of UV) at along the Line P transect.   171                            Figure A.2: Hovmoller-longitude plot of area averaged Aerosol Optical Thickness (AOD, dark target at 0.55 microns for both Ocean and Land, 1ᵒ spatial resolution monthly means) (a).  Map of the Gulf of Alaska showing the boxed area that was used to perform the AOD time series (b). Hovmoller-longitude plots were produced using analysis and visualization of the Giovanni online data system, developed and maintained by the NASA GES DISC (http://giovanni.gsfc.nasa.gov/giovanni/)       172                                               Figure A.3: Comparison of dissolved Cu profiles in the offshore waters in the central Atlantic (Stn 14, 50°E, 27.58° N, Jacquot and Moffett, 2015), and Northeast Pacific Ocean (Stn P26, 140.60° W, 39.60°N, Martin et al. 1989).      173     Figure A.4: Dissolved Cu [dCu (nmol kg-1)] and silicic acid [Si(OH)4 (μmol kg-1)] profiles at different stations along the Line P transect. The dCu dataset is for Aug 2011 cruise (values represent those measured in UV oxidized samples after ~ 4 years of acidic storage).  174   Appendix B   Supplementary material for Chapter 3 Table B.1: Mean (± SE) macronutrient quotas (C, N, P, S, fmol cell-1) and elemental molar stoichiometry (C:N, S:P) of marine heterotrophic bacteria as a function of Cu (nmol L-1) in the culture media.  Bacterial strain  Cu  n  Carbon  Nitrogen  C:N  n  Phosphorous  Sulfur  S:P  (nmol L-1)  (fmol cell-1) (fmol cell-  1) (mol:mol)     (fmol cell-1) (fmol cell-1) (mol:mol)           Dokdonia sp, strain Dokd-P16 0.6 6 9.3 ± 0.5 1.9 ± 0.2 4.8 ± 0.2 4 0.20 ± 0.03 0.16 ± 0.03 0.78 ± 0.1  2 10     11 ± 1.2 2.5 ± 0.3 4.4 ±0.2 6 0.20 ± 0.09 0.16 ± 0.07 1.09 ± 0.2  10 15 7.5 ± 0.3 1.8 ± 0.1 4.3 ± 0.1 7 0.12 ± 0.01 0.10 ± 0.03 0.89 ± 0.2  25 9 8.3 ± 0.3 1.8 ± 0.2 4.5 ± 0.1 6 0.15 ± 0.01 0.12 ± 0.02 0.83 ± 0.2  50 9 8.0 ± 0.5 1.8 ± 0.1 4.5 ± 0.1 3 0.11 ± 0.01 0.07 ± 0.01 0.61 ± 0.03 Pseudoalteromonas sp  0.6 6 26.3 ± 2.0 6.7 ± 0.2 3.9 ± 0.2 4 0.56 ± 0.06 0.29 ± 0.05 0.76 ± 0.2 strain PAlt-P26 (oceanic) 2 3 28.5 ± 2.1 6.5 ± 1.0 4.6 ± 0.5 4 0.57 ± 0.08 0.29 ± 0.05 0.51 ± 0.1  10 6 30.2 ± 6.0 7.1 ± 1.3 4.2 ± 0.1 4 0.37 ± 0.09 0.20 ± 0.02 0.60 ± 0.1  25 3 22.1 ± 0.6 5.3 ± 0.1 4.1 ± 0.03 nd nd nd nd  50 6 26.5 ± 3.3 6.5 ± 0.8 4.0 ± 0.05 2 0.51 ±0.02 0.20 ± 0.02 0.40 ± 0.01 Pseudoalteromonas sp  0.6 3 18.8 ± 5.9 3.9 ± 1.3 5.0 ± 1.0 4 0.53 ± 0.03 0.22 ± 0.01 0.42 ± 0.03 strain PAtl-P2 (coastal) 10 3 24.7 ± 2.6 6.7 ± 0.6 3.7 ± 0.1 3 0.48 ± 0.02 0.19 ± 0.02 0.39 ± 0.04  25 3 21.6 ± 0.7 6.2 ± 0.5 3.5 ± 0.2 4 0.58 ± 0.07 0.26 ± 0.03 0.45 ± 0.05 Ruegeria pomeroyi 0.1 7 18.5 ± 1.6 4.7 ± 0.4 4.0 ± 0.1 3 0.21 ± 0.02 0.18 ± 0.03 0.81 ± 0.04  2 4 15.3 ± 0.7 3.7 ± 0.2 4.2 ± 0.1 4 0.14 ± 0.03 0.13 ± 0.01 1.08 ± 0.2  10 7 17.8 ± 1.4 4.3 ± 0.3 4.1 ± 0.02 3 0.21 ± 0.12 0.12 ± 0.15 0.58 ± 0.05  25 3 14.1 ± 0.3 3.4 ± 0.1 4.2 ± 0.2 nd nd nd nd  50 6 15.1 ± 0.2 3.6 ± 0.2 4.2 ± 0.2 3 0.24 ± 0.02 0.17 ± 0.025 0.71 ± 0.07 175  Table B.2: One-way ANOVA values for the effect of Cu concentration in media on major nutrients quota: C, N, P, and S (fmol cell-1) and the stoichiometric ratios of C:N and S:P (mol:mol) in 4 bacterial strains.                                                                                ANOVA results  Dokd-P16 PAlt-P2 PAlt-P26 R. pomeroyi Variable F statistic p-value F statistic p-value F statistic p-value F statistic p-value C quota  2.88 0.097 0.17 0.688 0.18 0.688 3.39 0.078 N quota 3.07 0.085 2.06 0.194 0.15 0.706 4.00 0.056 P quota 0.85 0.498 1.00 0.418 1.23 0.366 3.74 0.054 S quota 0.93 0.459 1.66 0.279 1.16 0.390 2.20 0.165 C:N  0.004 0.949 2.74 0.141 0.12 0.734 0.98 0.332 S:P 1.29 0.324 0.41 0.683 0.76 0.541 1.75 0.234 176  Table B.3: Metabolic rates of marine heterotrophic bacteria at different levels of Cu (nmol L-1) in growth media. Cell normalized bacterial respiration (BRcell, fmol O2 cell-1), carbon normalized bacterial respiration (BRcarb, fmol C cell-1), bacterial productivity (BP, fmol C cell-1 d-1), bacterial carbon demand (BCD, fmol C cell-1 d-1) and bacterial growth efficiency (BGE, unitless). Calculations of BP, BCD, and BGE were performed as described in the methods section. All values are mean (±SE) of biological triplicates.   Strain Cu BRcell BRcarb BP BCD BGE  (nmol L-1) (fmol O2 cell-1) (fmol C cell-1) (fmol C cell-1 d-1) (fmol C cell-1 d-1) unitless Dokdonia sp, strain Dokd-P16 0.6 44.6 ± 2.1 5.0± 0.2 19.1 ± 2.8 63.7 ± 6.3 0.30 ± 0.3  2 43.7 ± 1.8 3.7 ± 0.2 31.4 ±3.9 75.2 ± 10 0.41 ± 0.06  10 30.6 ± 1.2 4.8 ± 0.6 44.4 ± 9.0 75.0 ± 3.2 0.58 ± 0.04  25 69.1 ± 7.5 8.7 ± 2.3 69.5 ± 9.4 138.7 ± 0.8 0.50 ± 0.05  50 57.6 ± 1.8 8.1 ± 0.1 89.8 ± 4.9 144.4 ± 4.5 0.60 ± 0.04 Pseudoalteromonas sp  0.6 434.3± 11.9 17.9 ± 0.6 344.5 ± 13 778.8 ± 26 0.44 ± 0.03 strain PAlt-P26 (oceanic) 10 298.4 ± 4.0 14.1 ± 0.5 389.2 ± 36 687.6 ± 36 0.56 ± 0.04  25 388.7 ± 110.7 17.5 ± 4.8 377.8 ± 2.7 766.4 ± 107 0.51 ± 0.06  50 358.7 ± 30.6 13.6 ± 2.0 449.3 ± 147 807.9 ± 50 0.53 ± 0.04 R. pomeroyi 0.6 144.0 ± 16.5 8.7 ± 0.4 88.2 ± 4.9 232.2 ± 8.5 0.38 ± 0.02  10 99.2 ± 68.3 6.2 ± 1.0 90.5 ± 10.9 189.6 ± 17 0.47 ± 0.05  50 86.8 ± 3.6 5.8 ± 0.2 82.0 ± 7.5 168.8 ± 8.8 0.48 ± 0.04 177                                  Figure B.1: Growth rates (d-1) of Pseudoalteromonas sp strains isolated from different stations along line P (station labels on x axis). Bars represent mean growth rates (± SE) of 6 or more biological replicates. Cu deplete and replete treatments contained 0.6 nmol L-1 and 10 nmol L-1 Cu, respectively.                       Figure B.2: Regression of phosphorus values (ppb) determined in same digest samples with ICP-Q-MS and ICP-OES.   y = 1.0568x - 45.493R² = 0.9725050010001500200025000 500 1000 1500 2000 2500Phosphorous (ppb)ICP-Q-MSPhosphrous (ppb) - ICP-OES178   Figure B.3: Trends in phosphorus (mmol Cu: mol P, top panel), carbon (μmol Cu: mol C, middle panel) and cell number normalized Cu quotas (zmol Cu cell-1, bottom panel) of 4 bacterial strains: Dokdonia sp, strain Dokd-P16, coastal Pseudoalteromonas sp, strain PAlt-P2, oceanic Pseudoalteromonas sp, strain PAlt-P26 and Ruegeria pomeroyi. Open circles are means (± SE) and the gray circles are the data points.   179    Figure B.4: Effects of oxalate and DTPA washes on the 14C accumulation of Pseudoalteromonas sp., strain PAlt-P26. Each pair represents identical bacterial culture treated with (A) oxalate and SOW or (B) DTPA and SOW in parallel. Bars represent mean (± SD) 14C activity from duplicate measurements from a single culture.                                 Figure B.5: Relationship between growth rate (d-1) and respiration rate (BRcell [fmol O2 cell-1 day-1]) of Dokdonia sp strain Dokd-P16 measured at different levels of Cu (nmol L-1) represented by different symbols. Data were fitted with least-square linear model and includes all values. The shaded area represents 95% confidence intervals180   Appendix C  Supplementary material for Chapter 4 Table C.1: Cellular trace metal quotas (10-20 mol cell-1) of marine heterotrophic bacteria. Values represent mean metal quotas (± SE) at different levels of Cu (nmol L-1) in culture media.  Strain Cu (nmol L-1) n Fe quota Zn quota Mn quota Cu quota Co quota Dokdonia sp, strain Dokd-16 0.6 4 106 ± 36 24.5 ± 9 6.6 ± 1.0 1.2 ± 0.2 0.4 ± 0.1  2 6 74.9 ± 39 23.5 ± 15 8.7 ± 3.7 1.6 ± 0.7 0.4 ± 0.2  10 7 33.4 ± 3.5 6.40 ± 2 4.6 ± 0.2 0.9 ± 0.1 0.2 ± 0.01  25 6 43.1 ± 12 34.5 ± 14 5.9 ± 0.8 2.1 ± 0.3 0.3 ± 0.03  50 3 24.2 ± 0.3 10.0 ± 5 3.3 ± 0.2 1.3 ± 0.1 0.2 ± 0.001 Pseudoalteromonas sp,  0.6 4 90.1 ± 2.2 72.8 ± 28 4.2 ± 2.0 1.8 ± 0.6 0.6 ± 0.06 strain PAlt-P2 (coastal) 10 3 94.8 ± 6.7 54.6 ± 6 0.9 ± 0.1 1.8 ± 0.04 0.6 ± 0.07  25 4 95.9 ± 12 75.7 ± 10 5.0 ± 1.7 1.9 ± 0.1 0.7 ± 0.08 Pseudoalteromonas sp, 0.1 4 107 ± 16 73.7 ± 16 0.3 ± 0.1 3.6 ± 0.9 1.0 ± 0.2 strain PAlt-P26 (oceanic) 2 4 121 ± 13 102 ± 11 bd 2.9 ± 0.8 0.4 ± 0.04  10 3 88.6 ± 24 38.7 ± 13 bd 1.7 ± 0.3 0.3 ± 0.07  50 3 110 ± 11 25.3 ± 1 bd 1.8 ± 0.3 0.3 ± 0.03 Ruegeria pomeroyi 0.1 3 47.7 ± 2.5 33.5 ± 24 21 ± 1.2 1.1 ± 0.1 0.3 ± 0.04  2 4 75.0 ± 18 18.9 ± 7 15 ± 3.1 1.6 ± 0.3 0.3 ± 0.05  10 3 46.2 ± 0.7 9.7 ± 2 19 ± 0.3 3.1 ± 0.1 0.8 ± 0.02  50 3 55.9 ± 3.4 12.3 ± 3 20 ± 1.6 3.7 ± 0.3 0.7 ± 0.02 181  Table C.2: Carbon normalized trace metal quotas (μmol Me:mol C) of marine heterotrophic bacteria. Values represent mean quotas (± SE) at different levels of Cu (nmol L-1) in culture media. Metal concentration per C were estimated using cell normalized metals (Appendix C, Table C1) and cellular carbon measured during each experiment.    Strain     Cu (nmol L-1) n Fe:C Zn:C Mn:C Cu:C Co:C Dokdonia sp,  0.1 4 109 ± 29 26 ± 9 6.9 ± 1 1.3 ± 0.3 0.4 ± 0.1 strain Dokd-P16 2 6 80 ± 43 26 ± 16 9.8 ± 4 1.4 ± 1 0.4 ± 0.2  10 7 47 ± 7 7 ± 3 6.3 ± 0.4 1.2 ± 0.2 0.3 ± 0.01  25 6 55 ± 15 47 ± 19 7.6 ± 1 2.7 ± 0.4 0.3 ± 0.04  50 3 32 ± 1 13 ± 6 4.3 ± 0.4 1.6 ± 0.1 0.2 ± 0.01 Pseudoalteromonas sp, 0.1 4 41 ± 10 29 ± 10 0.1 ± 0.1 1.4 ± 0.4 0.4 ± 0.1 strain PAlt-P26 (oceanic) 2 4 42 ± 2 37 ± 6 bd 1.0 ± 0.2 0.2 ± 0.02  10 3 26 ± 9.4 11 ± 5 bd 0.5 ± 0.1 0.1 ± 0.03  50 3 42 ± 0.5 10 ± 1 bd 0.7 ± 0.1 0.1 ± 0.001 Pseudoalteromonas sp,  0.1 4 42 ± 1 34 ± 13 1.9 ± 1 0.8 ± 0.3 0.3 ± 0.03 strain PAlt-P2 (coastal) 10 3 44 ± 3 25 ± 3 0.4 ± 0.1 0.9 ± 0.02 0.3 ± 0.03  25 4 44 ± 6 35 ± 5 2.3 ± 1 0.9 ± 0.1 0.3 ± 0.04 Ruegeria pomeroyi 0.1 3 24 ± 4.4 15 ± 10 10 ± 2 0.6 ± 0.1 0.2 ± 0.04  2 4 49 ± 12 12 ± 5 10 ± 2 1.0 ± 0.2 0.2 ± 0.03  10 3 28 ± 4 6 ± 2 11 ± 1.4 1.9 ± 0.3 0.5 ± 0.1  50 3 40 ± 3 8 ± 2 14 ± 1.3 2.5 ± 0.3 0.5 ± 0.02 182    Table C.3: Comparison of trace metal concentrations of Aquil used in marine phytoplankton studies (Ho et al, 2003, Quigg et al, 2011) and this study. The inorganic metal concentrations [Meʹ] in our study was estimated without taking the organic substrate into account. Concentrations are in nmol L-1.               Ho et al., 2003 & Quigg et al., 2011 This study Metal Total  (nmol L-1) [Meʹ]  (mol L-1) Total  (nmol L-1) [Meʹ]  (mol L-1) Mn 120 10 x 10-9 110 7.6 x 10-9 Fe 82  1.4 x 10-10 1370 2.8 x 10-10 Zn 80 2.0 x 10-11 70 2.7 x 10-11 Co 50 2.0 x 10-11 44.2 0.83 x 10-11 Cu 20 2.0 x 10-13 0.6‒50 4.6 x 10-15 – 3.9 x 10-13 183  Table C.4: Medians and ranges of metal contents (μmol Me:mol C) of different microorganisms. This data is plotted in Figure 4.5.   nd (not determined) 1 Data from Cameron et al. 2012 (whole cell digests). Metal contents were estimated assuming 50% C by dry mass. 2 Data for heterotrophic bacteria: mean metal quotas of bacteria in our study (all Cu treatments), the literature values for E. coli (Fe, Zn, Cu and Co, Cameron et al. 2011 & Fe, Zn,    Cu, Mn, Co Outten & O’Halloran 2001) and marine heterotrophic bacteria (Fe quotas determined in Fe-replete cultures by Granger & Price, 1999, Fourquez et al. 2014a). E. coli     data from Outten & O’Halloran, 2001 was estimated assuming 50% C by dry mass. 3 Data for photosynthetic bacteria includes diazotrophic and non-diazotrophic species from Nuester et al. 2012, Quigg et al., 2011, Guo et al.,2012 4 Copper data from Sunda & Huntsman, 1995 (range of Cuʹ in media: 18-707 fmol L-1), Annett et al. 2010 and Guo et al. 2012 (Fe-replete cultures growing at Cuʹ of 23.9 and 239     fmol L-1). Data for Fe, Zn, Mn, Cu and Co from Ho et al., (2003), and Quigg et al., (2011). 5 Field data for natural phytoplankton assemblages estimated by Ho et al., 2003 (using data of Martin & Knauer, 1973, Martin et al. 1976, Coliers & Edmonton, 1984, Kuss &       Kremling, 1999   6 Field data for single cell phytoplankton from North Atlantic (diatoms, autotrophic picoplankton, and flagellates) measured by synchrotron X-ray fluorescence (SXRF) reported by Twining et al. 2015 Group (μmol Me: mol C) Fe:C Zn:C Mn:C Cu:C Co:C Archaea1 Median  352 148 nd 2.25 0.75  Range 177-528 20-276 nd 0.2-4.3 0.3-1.2  n 2 2 nd 2 2 Heterotrophic bacteria2 Median  37 15 6.3 1.3 0.28  Range 1.5-154 6-47 0.4-14 0.5-4.7 0.01- 0.48  n 29 18 15 20 18 Photosynthetic bacteria3 Median  40 21 8.7 5 0.29  Range 21-677 6-38 5-46 0.1-13 0.2-1.7  n 9 9 9 13 4 Eukaryotic phytoplankton4 Median 78 6.4 29 1.7 1.2  Range 6.7-786 0.87-116 7-110 0.04-30 0.05-6  n 25 25 25 67 25 Field5 Median 43 18 3.5 3.5 1.7  Range 34-70 8.1-29 3-15 1.7-4.9   n 4 4 4 4 1 Field_SXRF6 Diatoms 57 ± 9 9 ± 2 2.3 ± 0.9 4.2 ± 2.3 0.42 ± 0.23 (means ± 1 SE) Autotrophic flagellate  35 ± 2 13 ± 1 3.2 ± 0.4 25 ± 3 0.92 ± 0.11  Autotrophic picoplankton  33 ± 6 10 ± 3 2.4 ± 0.7 33 ± 15 1.5 ± 0.5 

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.24.1-0355246/manifest

Comment

Related Items