UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Coupled nitrogen and oxygen isotope fractionation of nitrate imparted during its assimilation and dissimilatory… Granger, Julie 2006

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

Item Metadata

Download

Media
831-ubc_2006-199862.pdf [ 11.94MB ]
Metadata
JSON: 831-1.0053159.json
JSON-LD: 831-1.0053159-ld.json
RDF/XML (Pretty): 831-1.0053159-rdf.xml
RDF/JSON: 831-1.0053159-rdf.json
Turtle: 831-1.0053159-turtle.txt
N-Triples: 831-1.0053159-rdf-ntriples.txt
Original Record: 831-1.0053159-source.json
Full Text
831-1.0053159-fulltext.txt
Citation
831-1.0053159.ris

Full Text

C O U P L E D N I T R O G E N A N D O X Y G E N ISOTOPE F R A C T I O N A T I O N OF N I T R A T E I M P A R T E D D U R I N G ITS ASSIMILATION A N D DISSIMILATORY R E D U C T I O N B Y U N I C E L L U L A R P L A N K T O N by JULIE G R A N G E R B . S c , McGi l l University, 1995 M.Sc., McGi l l University, 1998 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S FOR THE D E G R E E OF D O C T O R OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Earth and Ocean Sciences) THE UNIVERSITY OF BRITISH C O L U M B I A A U G U S T 2006 © Julie Granger, 2006 Abstract I report the first measurements of coupled nitrogen (N) and oxygen (O) isotopic variations of nitrate (NO3") during its assimilation and dissimilatory reduction by laboratory cultures of marine and freshwater plankton. I derive the N and O kinetic isotope effects for nitrate assimilation by strains of marine and freshwater phytoplankton, as well as N and O isotope effects for denitrification by marine and freshwater strains of denitrifying bacteria. Large inter- and intra-species variations in the N and O isotope effects were observed among phytoplankton and among denitrifiers. However, the O isotope effect associated with either nitrate consumption or denitrification always co-varied with the N isotope effect, such that the l 8 0 / l 6 0 and 1 5 N / ' 4 N of nitrate changed concomitantly with a consistent ratio of-1:1, regardless of species or culture conditions. A single strain of denitrifiers, Rhodobacter sphaeroides, showed an O-to-N co-variation of 0.6. My results indicate that the dominant driver of the N and O isotope effects is nitrate reductase. The O-to-N ratio of 1 owes to the isotopic signature of the various nitrate reductases involved in nitrate reduction, namely the eukaryotic assimilatory nitrate reductase (eukNR), the prokaryotic assimilatory nitrate reductase (NAS), as well as the prokaryotic respiratory nitrate reductase (NAR). The variability in the observed magnitude of the N and O isotope effects is attributed to changes in the ratio of cellular nitrate uptake and efflux. The unusual O-to-N ratio of 0.6 observed fori?, sphaeroides is imprinted on nitrate by the periplasmic auxiliary nitrate reductase N A P . I also report a novel method to remove nitrite from samples for isotopic analysis of nitrate with the 'denitrifier method,' which measures the isotopic composition of nitrate and nitrite concomitantly. Nitrite is removed with ascorbate while purging with an inert gas. The method is non-toxic (hence compatible with the denitrifier method) and does not alter the concentration or the isotopic composition of nitrate in the samples. The findings reported here provide insight into the physiological mechanisms underlying nitrate isotopic fractionation during its assimilation and dissimilatory reduction. The trends observed in culture underline the biological systematics from which to interpret in situ measurements of coupled N and O isotopes of nitrate. ii Moreover, this work emphasizes the need to investigate the environmental factors that potentially contribute to variations in the magnitude of the N and O isotope effects in situ. iii Table of contents Abstract ii Table of contents iv List of tables vii List of figures viii Acknowledgements x Statement of co-authorship xii C H A P T E R 1: General introduction and thesis overview: Stable isotopes in the study of the marine nitrogen cycle 1 1.1. Preview 2 1.2. The marine nitrogen cycle 2 1.3. N isotopes as tracers of ocean N-processes 5 1.3.1. Isotopic fractionation and the Rayleigh model 5 1.3.2. The physiological mechanism of N isotope fractionation 8 1.3.3. The marine N isotope budget 9 1.3.4. Evaluating nitrate loss from denitrification with the N isotopes of nitrate 12 1.3.5. N isotopes in oligotrophic gyres to determine inputs from N-fixation 14 1.3.6. Inferring water mass transport from N isotopes 16 1.3.7. Sedimentary N isotopes as a paleo-tracer of past nutrient utilization 16 1.3.8. Current limitations of N isotopes as a tracer of ocean N cycling 17 1.4. The 5 1 8 0 of nitrate as a tracer of biological N transformations 19 1.5. Thesis objectives 21 1.6. References 30 C H A P T E R 2: Coupled nitrogen and oxygen isotope fractionation of nitrate during assimilation by cultures of marine phytoplankton 38 2.1. Introduction 39 2.2. Methods 41 2.3. Results 44 i v 2.4. Discussion 46 2.4.1. The mechanism of isotope fractionation during nitrate assimilation 46 2.4.2. The magnitude of the N isotope effect during nitrate assimilation 50 2.4.3. Summary and concluding remarks 52 2.5. References 61 C H A P T E R 3: A method for nitrite removal in nitrate N and O isotope analyses 66 3.1. Introduction 67 3.2. Materials and Procedures 69 3.2.1. Reduction of nitrous acid by ascorbate 69 3.2.2. Methodology 70 3.2.3. Procedure 71 3.3. Assessment and Discussion 74 3.4. Comments and Recommendations 77 3.5. References 85 C H A P T E R 4 : Nitrate N and O isotope fractionation associated with dissimilatory nitrate reduction by denitrifying bacteria 88 4.1. Introduction 89 4.2. Materials and Methods 91 4.3. Results 94 4.3.1 Nitrate N and O isotope fractionation by 'respiratory' denitrifiers 95 4.3.2 Nitrate N and O isotope fractionation during 'auxiliary' denitrification by R. sphaeroides 97 4.4. Discussion 97 4.4.1. Magnitude of the N and O isotope effects by respiratory denitrifiers 99 4.4.2. Nitrate reduction by R. sphaeroides 103 4.4.3. Coupling between nitrogen and oxygen fractionation of nitrate in the ocean.. 104 4.4.4. The freshwater conundrum 105 4.4.5. Isotope fractionation associated with denitrification and nitrate assimilation.. 106 4.4.6. Conclusions 107 v 4.5. References 120 C H A P T E R 5: The fractionation of nitrate N and O isotopes during its assimilation by prokaryotic and eukaryotic phytoplankton 127 5.1. Introduction 128 5.2. Methods 130 5.3. Results and discussion 132 5.3.1. Magnitude of the N and O isotope effects 132 5.3.2. The N : 0 coupling 133 5.3.3. Oceanographic implications 135 5.4. References 142 C H A P T E R 6: Conclusions and Future Outlook 146 6.1. The 1:1 rule 147 6.2. Coupled measurements of nitrate N and O isotopes to infer N biogeochemical cycling 151 6.2.1. Separating N-fixation from denitrification 152 6.2.2. Separating nitrate assimilation from remineralization in the mixed layer 153 6.3. Future research 155 6.4. References 161 Append ix 1: Computa t ion of isotope effects 165 Append ix 2: Physio logical model of nitrate isotope fract ionat ion 170 vi List of tables Table 1.1: Sources and sinks for the global marine nitrogen budget 23 Table 1.2: Compilation of N isotope effects (e) observed in laboratory cultures 24 Table 1.3: Isotope effects (e) from in situ estimates forN cycle processes 25 Table 2.1: Isotopic fractionation of nitrate and particulate N during assimilation 55 Table 2.2: Computed and empirical isotope effects for N - 0 bond of nitrate 56 Table 3.1: N and O isotopic composition of nitrate after nitrite removal 78 Table 3.2: The N and O isotopic composition of IAEA-N3 after nitrite removal 79 Table 3.3: Nitrate 5 I 5 N and 5 l 8 0 values measured for duplicate nitrite removals 80 Table 4.1: List of denitrifying strains used in this study 109 Table 4.2: Nitrate 15e and I 8 E and corresponding 5 I 8 0 : 5 I 5 N relationships 110 Table 4.3: A N C O V A to test for homogeneity among regression slopes 112 Table 4.4: Nitrate N isotope effects by denitrification 113 Table 5.1: Experimental strains and their assimilatory nitrate reductase 138 Table 5.2. Nitrate N and O isotope effects during nitrate assimilation 139 vii L i s t of figures Figure 1.1: The marine N cycle 26 Figure 1.2: The Rayleigh distillation model for N isotope fractionation 27 Figure 1.3: The marine N isotope budget 28 Figure 1.4: Processes that affect N and O atoms in NOV 29 Figure 2.1: NO3" assimilation and isotope fractionation by T. pseudonana 57 Figure 2.2: NO3" 6 1 5 N and 5 1 8 0 during nitrate assimilation 58 Figure 2.3: N and O isotopic composition of internal and external NO3" 59 Figure 2.4: NO3" N and 0 isotope fractionation by a eukaryotic cell 60 Figure 3.1: N O 2 " removal from aqueous samples with ascorbic acid 81 Figure 3.2: Time dependence of NOV removal by ascorbic acid 82 Figure 3.3: NO3" remaining in seawater after NOV removal 83 Figure 3.4: NOV N and O isotope fractionation by P. aureofaciens 84 Figure 4.1: NO3" and N O 2 " concentrations during growth of P. denitrificans 114 Figure 4.2: NO3" 6 1 5 N vs. In/among four strains of denitrifiers 115 Figure 4.3: NO3" S l 8 0 vs. the corresponding 5 1 5 N 116 Figure 4.4: NO3" 6 I 5 N vs. In / in cultures of R. sphaeroides 117 Figure 4.5: NO3" 5 1 8 0 vs. the corresponding 6 I 5 N in cultures R. sphaeroides 118 Figure 4.6: Putative NO3" N and O fractionation mechanisms of denitrifiers 119 Figure 5.1: NO3" 5 I 5 N of vs. In/during assimilation phytoplankton 140 Figure 5.2: NO3" 5 1 8 0 vs. the corresponding 5 1 5 N in phytoplankton cultures 141 Figure 6.1: Station locations from coring cruise O X M Z 0 1 M V 157 viii Figure 6.2: Depth profiles of N 0 3 " 6 ' 8 0, N 0 3 " 5 I 5 N, N * , [NCy], and [0 2] 158 Figure 6.3: N0 3 ~ 5 I 8 0 versus N 0 3 " S I 5 N for all data reported here 159 Figure 6.4: Decoupling of N 0 3 " N and O isotopes in the mixed layer 160 Figure 6.5: A l l A15,18 values plotted on [ N 0 3 ] 161 ix Acknowledgements I have been fortunate to be surrounded by friends, peers and mentors, who have had confidence in my abilities, have appreciated my intellectual input, and who have shown unwavering support throughout my Ph.D. First I thank Phil, who welcomed me in his lab, let me pursue questions tailored to my interests, let me express my intellectual creativity, and let me explore tenuous tangents. Few, I imagine, would entrust such faith in a pupil as to let her explore tangential paths, as I have. Moreover, Phil's help in consolidating my thoughts and my writing has been invaluable during these last stages of my thesis. I also want to express my admiration for Phil, who has boldly directed his research to unfamiliar waters in pursuit of important questions. His wisdom, creativity, and calm have been as source of inspiration to me, as he has achieved a healthy balance between a happy family life and a successful and fruitful career. If I could just get up that early in the morning! Then I must thank Danny, who most definitely owes up to his nickname, as he truly is a "Superstar." I am glad he agreed to my unusual proposal of pursuing a "long-distance" Ph.D. - 'cause who wants to live in Jersey - as it gave me a chance to pursue a fascinating question. Danny possesses an unparalleled wealth of knowledge, uncommon cognitive skills, and streaks of absolute brilliance. And Danny shares his insights as freely and efficiently as I disseminate gossip. I have been privy to moments of tremendous nerdy joy from our interactions. I would have no data were it not for Greg Cane, who performed countless analyses of my samples on the mass spec at Princeton. I am indebted to Greg for taking meticulous care of my samples, and for generating beautiful data for me. Merci. I must also acknowledge previous mentors, Neil Price, Bess Ward, and Francois Morel, whom, in their scientific individuality, each imparted me with invaluable knowledge and confidence to pursue challenging questions. I am grateful to Andy Ridgwell, iconoclast par excellence, and "the undergrads" -Melissa, Amber and Dave - kindred spirits who share my holistic view that good science is a group effort. They revived my intellectual curiosity in all that is interesting, and gave me solace in knowing that I'm not the only nerd around. Finally I want to acknowledge my friends and family, who have shown unwavering support throughout the years: Maite, Phil, Chris, Adrian; Papa, Maman, Eve, Mireille, and now Jasmine - who's barely met her aunt; Grand-Maman, 'cause I'm her favourite; and Handsome Andy, who saw me through the most difficult period of my life. To all I am profoundly grateful, and I feel blessed to have such tremendous friends and kin in my life. Thank you. xi Statement of co-authorship Chapter 1: Was conceived and written by Julie Granger. Chapter 2: A l l of the experiments were executed by Julie Granger. The isotope analyses were performed at Princeton University by Greg Cane, technician in Daniel Sigman's laboratory. Joseph Needoba provided samples that were analyzed at Princeton University by Greg Cane, and that are featured in Figure 2.3. Daniel Sigman and Paul Harrison acted as co-advisors during the project. Chapter 3: The method featured was developed entirely by Julie Granger. Isotope analyses were performed at Princeton University by Greg Cane, technician to Daniel Sigman. Maria Prokopenko tested the method, and her results are featured in Table 3.2. The chapter was written entirely by Julie Granger with editorial input from Daniel Sigman and Philippe Tortell, who acted as co-advisors during the project. Moritz Lehmann provided input in the early stages of the project. Chapter 4: Julie Granger executed all of the experiments featured in this chapter. Isotope analyses were performed at Princeton University by Greg Cane, technician to Daniel Sigman. The chapter was written entirely by Julie Granger, with editorial input from Daniel Sigman and Philippe Tortell. Philippe Tortell and Daniel Sigman acted as co-advisors throughout the project. Chapter 5: Julie Granger executed experiments featured in this chapter. This chapter was written entirely by Julie Granger, with editorial input from Daniel Sigman and Philippe Tortell, who were co-advisors during the project. Chapter 6: This chapter was written entirely by Julie Granger. The data from the eastern North Pacific Margin will be featured in a publication co-authored by Daniel Sigman and Julie Granger. The data from the Equatorial Upwelling wi l l be xii featured in a publication written by Julie Granger and co-authored by Dan Sigman. x i i i Chapter 1 General introduction and thesis overview: Stable isotopes in the study of the marine nitrogen cycle 1 1.1. Preview The nitrogen cycle plays a central role in the biogeochemistry of the oceans. As an essential nutrient, biologically available (or "fixed") N has the potential to limit biological productivity at the surface ocean. Moreover, changes in the N cycle are of growing interest in research at the interface between climate and biogeochemistry as hypotheses have been put forth for indirect influences of the marine N budget on climate. The stable isotopes of various elements have been measured in marine systems in order to obtain integrated measurements of their physical transport, and their chemical and biological transformations. The stable isotopes of nitrogen have been particularly useful in elucidating N processes in the ocean, in part because fixed nitrogen species are chemically dynamic on spatial and temporal scales that elude comprehensive in situ sampling. Moreover, stable isotopes of nitrogen have been useful in paleoceanographic studies to assess nutrient utilization in the paleo-ocean. This introductory chapter consists of a brief overview of the modern N cycle, followed by a comprehensive review of the insights that have been gained from the study of N isotopes in the ocean. I then describe a novel tracer for N cycling, namely the use of coupled measurements of the N and O isotopes of nitrate to trace N biogeochemical transformations. A brief outline of that which we knew about the biological fractionation of O isotopes of nitrate prior to my thesis work is followed by an outline of the research I conducted to investigate systematics that describe the coupled fractionation of the N and O isotopes of nitrate during its biological reduction. 1.2. The marine nitrogen cycle Nitrogen is a major constituent of living mass and thus a chief determinant in metabolism and growth of open ocean algae. The distribution and mean concentration of nitrate in the ocean thus affect the global fertility of the sea and its consequent exchange of gases with the atmosphere. As such, nitrogen has been proposed as a major driver of the atmospheric CO2 changes that characterize glacial/interglacial cycles. Increased nitrate consumption in polar surface waters during the last glacial age is hypothesized to have caused the apparent CO2 decrease (Francois et al. 1997). Enhancement of low-2 latitude productivity due to increased nitrogen fixation at low latitudes also figures as a plausible scenario to explain low CO2 concentrations during the last glaciation (Falkowski 1997). Constraining the pools and fluxes of nitrogen in the modern ocean, as well as understanding the mechanisms that underlie biological nitrogen transformations are thus paramount to expanding current knowledge of ocean biogeochemistry. Ultimately, more intimate knowledge of the ocean's nitrogen cycle may lead to insight into its relation to global climate change. A schematic representation of the oceanic nitrogen cycle is presented in Figure 1.1. Nitrate (NOV), figured at the top of the diagram, is the most oxidized species of nitrogen. Biological reduction of nitrate catalyses the loss of an oxygen atom, resulting in nitrite (NOV)- This transformation is characteristic of two distinct biological reactions termed assimilatory and dissimilatory nitrate reduction. The former refers to the assimilation of nitrate by algae - and heterotrophic bacteria (Allen et al. 2002 and references therein) - for N nutrition: Nitrate is internalized at the cell surface and then reduced intracellularly to ammonia, via nitrite. Ammonia then serves as the primary template for amino acid synthesis. Living mass thus generated at the surface ocean is subject cellular excretion of N-species, to consumption by grazers, or it may senesce as a result of nutrient starvation or viral lysis. These processes engender nitrogen release from grazed and senescent cells, as ammonia (or rather, ammonium, the cationic form at seawater pH) or dissolved organic nitrogen (DON). D O N can further be catabolyzed by bacteria back to ammonium. Ammonium at the surface ocean that originates from excretion, consumption or decomposition of plankton, constitutes a choice source of nitrogen for live phytoplankton. Primary production originating from the utilization of ammonium as an N source is referred to as "regenerated production." "New production," in contrast, is fuelled by nitrate freshly supplied from depth to the surface ocean (Dugdale and Goering 1967). In a hypothetical steady-state system, what enters the euphotic zone (nitrate) must be exported back to depth (organic material), such that new production measurements (e.g., 1 5N-labeled nitrate uptake rates measured for field sample incubations) have been used to estimate of total N export to the deep ocean (Eppley and Peterson 1979). Deeper in the water column, ammonia released during organic matter decomposition encounters a different fate. In the absence of light, nitrifying bacteria, namely ammonia oxidizers and nitrite oxidizers, oxidize ammonia back to nitrate as a 3 means of securing reducing power to synthesize primary sugars from CO2. These organisms do a distinctly thorough job of this, as ammonium (or nitrite) is hardly detectable in deep water. In contrast, low concentrations of ammonium and nitrite accumulate at the top of the nitracline and above in the euphotic zone, where multiple processes may be operating simultaneously. In the mixed layer, the supply of ammonium or nitrite may exceed assimilation or oxidation rates. Phytoplankton cannot keep up with N supply as light becomes progressively limiting with depth, whereas nitrifiers may not be able to use ammonium and nitrite fully because their activity is progressively suppressed with increasing light levels. Nonetheless, significant oxidation rates of ammonium and nitrite are detectable at the nitracline and at shallower depths (Ward et al. 1989). So, in reality, nitrate is not only regenerated from ammonia below the nitracline, but also within the surface mixed layer. This poses a caveat to the "new" vs. "regenerated production" paradigm, which assumes no nitrate regeneration within the mixed layer. Ward et al. (1989) report significant nitrate production within the mixed layer relative to nitrate assimilation in the California current, implying that part of the nitrate assimilated is functionally regenerated instead of new. Furthermore, the new production paradigm assumes consumption of nitrate that is exclusive to photoautotrophs. Mounting evidence reveals that a large fraction of the nitrate may be consumed by heterotrophic bacteria (Allen et al. 2002 and references therein), such that nitrate consumption cannot be equated with carbon fixation. Euphotic zones throughout the oceans represent areas of dynamic N cycling where operative N-processes remain poorly defined. Dissimilatory nitrate reduction, the alternate pathway for biological nitrate reduction, is also termed denitrification. In the absence of oxygen, denitrifying bacteria use nitrate as a final electron acceptor to carry out respiration (Zumft 1997). Nitrite generated from this reaction can further be reduced sequentially to nitric oxide (NO) gas, nitrous oxide (N2O) gas, and finally to dinitrogen (N2) gas (Figure 1.1) - whence each intermediate serves as a terminal electron acceptor, albeit with sequentially increasing redox potentials that provide for moderate to marginal electron gradients within the respiratory chain. The denitrification process is not widespread throughout the ocean, but occurs in localized areas of high surface production and low oxygen source waters. The Arabian Sea, the Eastern Tropical North Pacific, and the Peru Upwelling are known as major areas of active water-column denitrification. Sediments underlying productive coastal 4 areas also host substantial denitrifying activity (Brandes and Devol 2002; Devol 1991; Middelburg et al. 1996; Seitzinger 1988). Denitrification represents the major sink for oceanic fixed nitrogen (Table 1.1). The magnitude of this loss term is of utmost relevance for understanding the modern ocean nitrogen budget. Yet due to the difficulty inherent in measuring and defining the extent of a process that is variable in space and time, the loss of oceanic fixed N incurred from denitrification remains poorly constrained (Brandes and Devol 2002; Codispoti et al. 2001). Denitrification in the ocean is countered by biological N-fixation, which involves the catalytic reduction of dinitrogen gas to ammonia by nitrogen-fixing prokaryotes. Much of the research on N-fixation in the marine environment has focused on the cyanobacterium Trichodesmium. This genus inhabits low nutrient tropical and subtropical seas where it often forms massive near-surface blooms of conspicuous aggregate colonies (Carpenter et al. 1992). Though Trichodesmium likely contributes a significant fraction of total oceanic fixed nitrogen, a number of cyanobacterial groups as well as a-, y-, and fi-proteobacteria are also potentially large perpetrators of oceanic N-fixation (Zehr et al. 2000). Because N-fixation throughout the ocean is spatially heterogenous, temporally stochastic, and thus, undersampled, the generation of accurate estimates for global N-fixation rates has proven even more challenging than for denitrification. Global N-fixation rates have been successively revised upwards as more direct and indirect estimates are generated (reviewed by (Karl et al. 2002). Yet a recent model study by Brandes et al. (2002) suggests that even the latest estimates may grossly underestimate marine nitrogen fixation rates. The budget presented in Table 1.1 clearly illustrates that the sources and sinks of fixed nitrogen to the ocean at present are poorly constrained, to the extent that it is not even clear whether sources and sinks are in relative balance, and if the ocean is progressively losing or gaining fixed nitrogen. 1.3. N isotopes as tracers of ocean N-processes 1.3.1. Isotopic fractionation and The Rayleigh model 5 The study of oceanic nitrogen cycling has been facilitated by the existence of two stable isotopes of nitrogen, namely 1 4 N and 1 5 N. Naturally occurring nitrogen is comprised chiefly of 1 4 N , yet a minute fraction (0.36765 ± 0.00081 %) occurs as l 5 N , which possesses an additional, stable neutron. The isotope generally has little effect on the chemical properties of an element, as these are chiefly determined by electronic configuration. Yet small differences in chemical behaviour of two isotopes of a given element do exist. For a given element in fixed environmental surroundings, two isotopomers of the same element have the same kinetic energy (KE), but different velocities due to their mass difference: An example is water vapour. The lighter molecule has the higher velocity and can more easily escape from the fluid phase. This causes isotopic fractionation, where the vapour phase generated is relatively deplete in the heavier isotope, while the remaining fluid phase is enriched with the heavier isotope. The slight differences in nuclear mass between isotopes also affect the bond energy, in that the bond strength of the heavier isotope is greater. In chemical reactions that involve bond breakage, the energy barrier for the reaction of a molecule bearing a heavier isotope is greater than that for the same molecule bestowed with the lighter isotope. In biological reactions, mass-dependent differences in chemical behaviour often result in isotopic fractionation, wherein molecules harbouring a lighter isotope (say 1 4N) react more quickly than those that have the heavier isotope ( l 5N). Consequently, throughout the course of a biochemical reaction, the substrate being consumed becomes progressively enriched with the heavier isotope, while the resultant product is relatively enriched with the lighter isotope. This process is illustrated in Figure 1.2 for nitrate uptake by a marine diatom in batch culture. On the y-axis, the isotope ratio of 1 5 N to 1 4 N is expressed in 5-notation (in per mil units, %o), as KE = (l/2)mv2 15-KT/14-5 , 5 N ( % 0 ) = [ B ^ - 1 x1000 (1) 6 The standard is atmospheric N 2 , which in this notation has a 8 1 5 N of 0%o. As illustrated in Figure 1.2, the 8 1 5 N of NO3" increases progressively as nitrate is depleted from the culture medium by cellular uptake. The isotope effect e (also called fractionation factor) quantifies the relative magnitude of isotopic enrichment in the reactant pool, e is a function of the ratio of the reaction rates (k14 and k15) of the two isotopes, e =(1-k 1 5 /k 1 4 )x 1000 (2) This value is determined from the integrated expression of the progress of a unidirectional reaction according to the following linear approximation, O , 5N r eactan« = 5 1 5 N i n i t i a l -£ ( ln j 0 (3) The term/designates the fraction of reactant remaining, 8 l 5 Ni n j t j a i is the 8 I 5 N of initial reactant N pool, and e is the kinetic isotope effect of the transformation. The above equation describes the Rayleigh model for isotope fractionation, which applies to reactions occurring in a closed system (Mariotti et al. 1981). In practice, e is the negative slope of the linear relation of 8 1 5 N r e a c t a n t (reactant = nitrate) vs. the natural logarithm of the fraction of reactant remaining (f: nitrate/nitrateinitiai). As shown in Figure 1.2, total cell mass, i.e. the integrated product, also becomes isotopically heavier throughout the reaction, since cells are consuming progressively heavier nitrate throughout the course of the reaction. However, at any given moment, the organic N being generated is always isotopically lighter than the reactant NO3" by a difference of e (Figure 2), such that the instantaneous product is defined as 8' 5Ni n s tant = 8' 5Nreactant " £ (4) It follows that the integral of this expression describes the 8 I 5 N of the integrated product, namely that of total accumulated cell mass (see Mariotti et al. 1981), 8 1 5 N i n t e g r a t e d = 8 1 5 N i n i t i a l + e(lny) xfl{\-J) (5) 7 The Rayleigh model has been invaluable in order to make sense of N-isotopic data for processes occurring both in laboratory cultures, as well as in oceanic situations. An alternative to the Rayleigh model is the steady-state model, in which reactant N is continuously supplied and partially consumed, and residual reactant is exported at a steady-state rate. However, this is beyond the scope of this overview. 1.3.2. The physiological mechanism of N isotope fractionation during nitrate assimilation The mechanisms of isotopic fractionation during nitrate assimilation by marine phytoplankton and during respiratory nitrate reduction by denitrifiers are not well understood. Neither lab nor field measurements of isotope effects have provided robust insight into what controls the magnitude of the isotope effect. Lab estimates vary widely (Table 1.2) and show no discernible pattern with respect to culture conditions, save an increase in £ at low light observed for a single diatom species (Needoba and Harrison 2004; Wada and Hattori 1978). Field estimates of e converge around 5%o for nitrate assimilation (Table 1.3), with notable variation (Karsh et al. 2003). As for denitrification, oceanic estimates range from 20 to 30%o and upwards, with variability that may arise in part from the difficulty in deriving isotope effects in physically dynamic systems (Table 1.2). The isotopic fractionation of N during nitrate assimilation is attributed to the catalysis of nitrate reduction by the enzyme nitrate reductase (NR). This hypothesis stems from a number of factors. Measurement of N-isotopic fractionation for purified spinach nitrate reductase post a relatively high isotope effect of 25%o (Ledgard et al. 1985). Nitrate reduction is believed to be the rate-limiting step in nitrate assimilation (Berges and Harrison 1995), causing accumulation of an intracellular pool of l 5 N -erinched nitrate. Indeed, diatoms have been shown to accumulate an internal pool of nitrate that is significantly more isotopically enriched than external nitrate (Needoba et al. 2004). Because isotopic fractionation occurs internally, manifestation of the isotope effect in extracellular nitrate requires that the organism be subject to significant rates of nitrate efflux. Nitrate efflux was assessed in a cyanobacterium by progressively 8 inhibiting nitrate reductase activity and monitoring of the concomitant increase in the N isotope effect, which resulted from increased efflux of un-reacted internal nitrate (Shearer et al. 1991). This work confirmed NR-based fractionation as the primary driver of the N isotope effect on nitrate during its assimilation. Whether uptake or efflux steps during nitrate assimilation impart any isotopic fractionation on nitrate is unknown. Mariotti et al. (1981) proposed that neither the uptake nor efflux step have an intrinsic isotope effect on nitrate for assimilation by Pearl Millet seedlings. Rather, they concluded that the isotopic fractionation observed during nitrate assimilation is caused solely by NR. Whether fractionation occurs at the cell membrane during nitrate influx or efflux has not been determined for eukaryotic phytoplankton. While the large variation in isotope effects observed both among and within plankton species still remains unexplained, Needoba et al. (2004) showed that it correlates with the 8 1 5 N difference between internal and external nitrate, invoking the ratio of cellular nitrate uptake to efflux as the primary modulator of the magnitude of the isotope effect in eukaryotic phytoplankton. 1.3.3. The marine N isotope budget Fractionation of N isotopes in the ocean reflects the biological processes active in the water column. As such, N-isotopes have been used as a tool to elucidate N-cycling in both the modern and paleo-ocean. The 5 I 5 N of particulate organic nitrogen and of nitrate in the water column show variations in magnitude that reflect the biological transformations effected on ambient N. Similarly, organic-N residue stratified in deep-sea sediment is also telling of past history of N-cycling and organic N sedimentation. Figure 1.3 illustrates the current N stable isotope budget of the modern ocean. In deep water resides the bulk of fixed nitrogen in the form of nitrate. Measurements of deep ocean 5 I 5 N throughout the seas converge on a relatively uniform value of 5%o (Sigman et al. 2000). This value reflects the integrated signal of all localized N-isotopic fractionation caused by major (biological) sources and sinks of fixed N in the ocean. Nitrogen fixation, the dominant input term for oceanic fixed nitrogen (Table 1.1), provides new nitrogen with a 5 1 5N of around -1 to 0%o, as measured in cellular N of N-fixer colonies collected at sea (Carpenter et al. 1997). By comparison, the 5 I 5 N of 9 dissolved N 2 is around 0.6%o relative to atmospheric N 2 . Laboratory cultures of N-fixers (Table 1.2) corroborate the apparent lack of N isotope fractionation associated with N-fixation, where fractionation factors (e) around 0%o have also been measured. Consequently, the 5 1 5N of organic material collected in shallow sediment traps in some oligotrophic tropical gyres is relatively low (around 2%o), as particulate nitrogen sinking out of the surface ocean bears the signature of N-fixation (Karl et al. 2002). The plankton mass that incorporates freshly fixed nitrogen and sinks out of the surface ocean is decomposed and nitrified to nitrate that retains the low 5 I 5 N imparted by N-fixation. Were it not for the large isotope effect associated with denitrification (the dominant sink for fixed nitrogen), the 5 1 5N of bulk nitrate in the deep ocean would remain around 0%o. However localized pockets of denitrification throughout the ocean impart a heavy 8 1 5N signal on resident nitrate. Three regions, the Peru Upwelling, the Eastern Tropical North Pacific, and the Arabian Sea, account for most of global water-column denitrification in the ocean. As nitrate (or nitrite) is used in lieu of oxygen to sustain decomposition of organic material, the remaining nitrate pool becomes highly enriched in l 5 N , in this case reaching upwards of 15%o at the oxygen minimum (compared to 5%o for global deep ocean). Laboratory estimates of isotopic fractionation (e) by denitrifiers tend to be high and variable (Table 1.2). Field values are similarly high, with more recent estimates ranging between 20 and 30%o (Table 1.3). The high degree of N isotope discrimination associated with denitrification is thus reflected in the 8 1 5N of ambient nitrate in denitrifying zones. On the whole, the magnitude and N isotopic signature of denitrification, relative to the magnitude of oceanic N-fixation, amount to a global ocean nitrate 8 I 5 N of 5%o, as measured in deep water nitrate. Neglected in the above simplification is the impact of sedimentary denitrification on the global ocean N-isotope budget. The relative importance of sedimentary denitrification as a sink for fixed N has been progressively revised upwards as our understanding of modern N cycle evolves (Brandes and Devol 2002; Middelburg et al. 1996). Unlike water-column denitrification, that in sediment is believed to impart no isotope effect on nitrate because it is limited by the diffusion of nitrate to the sediment. Al l nitrate supplied to the sediment is denitrified, such that no isotopically-enriched nitrate pool remains (Brandes and Devol 1997). The global 8 1 5N of mean ocean nitrate 10 thus quantifies the net signal of fluxes, pools and respective isotope effects for N-fixation relative to sedimentary and water-column denitrification. The internal cycling of oceanic nitrogen, namely the cycle of nitrate uptake, ammonification, and nitrification, has little effect on 5 1 5 N of mean ocean nitrate. Nitrate supplied from the deep ocean to the surface is completely consumed by resident plankton in most of the global surface ocean. Although nitrate assimilation by phytoplankton is associated with potentially large isotope effects (Table 1.2), complete nitrate consumption pre-empts isotopically-enriched nitrate from remaining at the surface ocean (Altabet and McCarthy 1985). Providing there are no alternate sources of fixed N to the surface (e.g., from N-fixation), organic material produced at the surface from nitrate originating from deep water is imparted with the 6 1 5 N of its source. And in a Sisyphean manner, the organic nitrogen exported back to the deep ocean is remineralized to nitrate that has the 5 1 5 N of deep ocean nitrate (Figures 1.1 & 1.3). Evidence of this process was presented by Altabet (1988), who observed isotopic similarity between the annually integrated sinking flux out of the Sargasso Sea mixed layer and thermocline nitrate from that region. Sedimenting particulate nitrogen collected in sediment traps below the euphotic zone showed a 5 1 5 N identical to that of nitrate in the water underlying the euphotic zone, showing close coupling between nitrate supply from deeper water to the surface and sedimenting plankton mass. Complete nitrate consumption at the surface ocean explains the relative uniformity of deep-water nitrate 5 1 5 N (Sigman et al. 2000). The 5 I 5 N measured for deep ocean nitrate is relatively invariant both within and between deep ocean basins, estimated around 4%o in the North Atlantic to 6%o in the North Pacific (Liu and Kaplan 1989; Liu 1979; Liu et al. 1996; Sigman et al. 2000; Wu et al. 1997). Yet at high latitudes, nitrate consumption by phytoplankton is not complete due to iron limitation of primary production (Martin et al. 1990). Resident phytoplankton consume only a fraction of the nitrate supply, such that a residual pool of 515N-enriched nitrate remains at the surface ocean. Unlike denitrification, fractionation of surface nitrate from assimilation does not effect any change in the global 5 1 5 N budget of the ocean because N-isotopes are merely redistributed in different water masses. Isotopically-light fixed nitrogen is not lost as N2, as is the case for denitrification. An example of differential distribution of the N isotopes 11 of nitrate in separate water masses due to assimilation and remineralization is provided by Sigman et al. (2000). The authors determined the summer nitrate concentration at locations in the surface mixed layer of the Antarctic to be around 25 uM, compared to 37 uM for source nitrate in the underlying water layer, indicating incomplete nitrate consumption by phytoplankton. As expected, the 5 1 5N of surface nitrate in the region was enriched in 1 5 N . In contrast, the 5 I 5 N of nitrate measured in the Upper Circumpolar Deep Water directly below was found to be lower than 8 1 5N of nitrate at more northerly latitudes of the same water mass (where nitrate use is greater). The diminished 5 1 5N of Upper Circumpolar Deep Water thus appeared to reflect remineralization of isotopically light sinking organic N, the result of incomplete nitrate use at the surface. Thus, surface processes resulted in a relatively shallow and localized variation of nitrate 5 1 5N (Sigman et al. 2000). Such a variation may be expected to be effaced later on, during deep winter mixing, and the nitrate 5 I 5 N of Upper Circumpolar Deep Water would then be restored to the value observed for the northerly portion of the water mass. Incomplete nitrate consumption can result in local variations in nitrate 5 I 5 N that do not impact the relative homogeneity of deep ocean nitrate. Globally, however, incomplete nitrate utilization incurs no net loss of fixed nitrogen from the water column and thus no change in whole ocean 5 1 5N. Hence, processes proper to the internal biological cycling of N do not act as determinants of the global oceanic N-isotope budget. 1.3.4. Evaluating nitrate loss from denitrification with the N isotopes of nitrate The first measurements of the N-isotopic fractionation imparted on nitrate by denitrifiers yielded estimates that ranged between 20 and 30%o (Delwiche and Steyn 1970; Wellman et al. 1968). Around the same time, the first measurements of the N-isotopic composition of nitrate in the western Pacific showed that nitrate was isotopically enriched in the heavy isotope of nitrogen relative to atmospheric N 2 (Miyake and Wada 1967). They observed that nitrate 5 1 5N ranged from 5.1 to 7.5%o in a vertical profile, with an apparent maximum at 1000m. The authors hypothesized that the isotopic enrichment of nitrate was likely related to "various physical and biochemical processes which affect the concentration and distribution of fixed nitrogen in the sea; for example, the kinetic isotope fractionations associated with nitrification, denitrification, or nitrogen 1 2 fixation." A few years later, Cline and Kaplan (1975) investigated the isotopic fractionation of nitrogen isotopes associated with denitrification in the oxygen-deficient waters of the Eastern Tropical North Pacific (ETNP). They reasoned that by virtue of the extent of oxygen-deplete waters in the region, the magnitude of coincident denitrification likely contributed to a large fraction of fixed nitrogen loss in the ocean, and that its effect on the isotopic composition of nitrate in the ocean must be significant. Their estimates of the isotope effect associated with denitrification were significantly higher than previous lab estimates, ranging between 30 and 40%o. Contemporaneous studies in the Peru Upwelling, the Cariaco Trench, and the Santa Barbara Basin also derived estimates of the N isotope effect imparted by denitrification that were variable and higher than lab estimates, ranging between 30 and 60%o (Cline 1973; Liu 1979). More recent estimates in the Arabian Sea, in the ETNP, and in the Santa Barbara Basin, however, have yielded values that are on par with laboratory estimates (Table 3.3), between 20 and 30%o (Brandes et al. 1998; Naqvi et al. 1998; Sigman et al. 2003; Voss et al. 2001). The discrepancy lies in the physical models used to derive estimates of the isotope effects. Early estimates were generated from vertical or cross-isopycnal diffusion models, consistent with the then current view of ventilation; Oxygen minimum zones were seen as stagnant layers ventilated by mixing with surface and deep oxygenated waters (Wyrtki 1962). Recent investigations of isotopic fractionation in denitrifying zones derive the isotope effect using isopycnal oceanographic models. Overall, studies of N isotope fractionation in denitrifying zones of the ocean have contributed to the generalization that the isotopic fractionation factor for water-column denitrification is consistent and thus generates a distinct and consistent isotopic signature in the remaining nitrate that is not affected by the rate of denitrification or by hydrography. The invariant isotopic signature lends itself to paleo-reconstruction of the N isotopic sedimentary record to derive the extent of water-column denitrification in past oceans (Section 1.3.5). The isotopic composition of nitrate in denitrifying zones could also serve as a tracer of vertical and horizontal mixing around suboxic regions, and away from suboxic regions (Section 1.4.6). Finally, Brandes et al. (1998) proposed that the constancy of the N-isotopic signal imparted on nitrate by denitrification could provide "an enhanced background" for nitrogen fixation studies, by discerning the difference between the isotopic composition of the nitrate advected to the surface to compare to that 13 of sinking particles in the thermocline, thereby providing estimates of regional N-fixation rates (Section 1.3.5). 1.3.5. N isotopes in oligotrophia gyres to determine inputs from N-fixation In the mid-80's, Altabet and colleagues examined the linkages in the cycle of new nitrogen input into the euphotic zone, its utilization by phytoplankton, an transport to the deep sea, using l 5 N / l 4 N isotopic tracers (Altabet 1988; Altabet 1989; Altabet et al. 1991; Altabet and McCarthy 1985; Deuser 1986; Voss et al. 1996). This exercise in mass balance became particularly relevant at BATS, where a drawdown of dissolved inorganic carbon (DIC) from surface waters during the summer months cannot be accounted for in terms of either the nutrients required to support the biological removal of the DIC, or the export production that should result from the DIC decrease (Michaels et al. 1994). In the absence of significant nitrate input during this period, N-fixation was proposed as a means of fueling phytoplankton growth to remove the DIC. This hypothesis is supported by regional geochemical observations, including the high concentration of nitrate relative to phosphate in the North Atlantic thermocline (Gruber and Sarmiento 1997; Michaels et al. 1996), and the low 5 1 5 N of nitrate in the thermocline (Altabet 1988). Despite these regional observations, local measurements of N isotopes have provided arguments against N-fixation as the missing N source for the BATS summertime DIC drawdown. Altabet (1988) observed that sedimenting particles below the euphotic zone consistently displayed a 5 1 5 N identical to that of source nitrate in the thermocline (~3.5%o). Since newly fixed N has a 6 1 5 N in the range of ~ -1 - 0%o (Carpenter et al. 1997; Hoering and Ford 1960; Minagawa and Wada 1986), Altabet (1988) concluded that there was little room in the isotope budget for N-fixation to be quantitatively important as a source of N for new production at BATS. Interestingly, Montoya et al. (2002) reached opposite conclusions from concordant data for the Sargasso Sea than those of Altabet (1988, 1989). Montoya attributed the low 5 1 5N of suspended particles at the surface, which was reported by both parties to be ~0.2%o, to the input of light nitrogen from N-fixation. Altabet, on the other hand, attributed this low 8 l 5 N signal of suspended particles to the preferential export of heavy nitrogen from zooplankton faecal pellets (Altabet and Small 1990), and to the 14 isotopic signature of regenerated production by small particles at the surface (Altabet and McCarthy 1985). (Montoya et al. 2002) also invoked the putative role of zooplankton grazing in creating sinking faecal material, however concluded this export to bear a relatively isotopically-light 5 1 5N signal, with a modelled 8 , 5 N of ~l%o. No attempts were made by Montoya and colleagues to reconcile their conclusions with the sediment trap data of Altabet (1988, 1989). The apparent lack of significant input of new N to the surface from N-fixation at BATS imposes a quandary on the origin of the isotopically deplete nitrate in the thermocline relative to deep-water nitrate (3.5%o versus 5%o, respectively). Altabet hypothesized that the low- , 5N source to the thermocline at BATS could be the preferential remineralization of low-1 5N particulate N as the sinking flux transits through the thermocline. However, the 8 I 5 N of sedimenting particles at different depths revealed that heavy-15N is preferentially remineralized at shallower depths (Altabet 1991). Dissolved organic nitrogen (DON) was then invoked as the missing pool in the N budget at BATS that could likely harbour the isotopic signal of N-fixation, and could account for the missing nutrient source during the summer DIC drawdown. However, recent measurements of the seasonal DON concentration and of its 5 1 5 N revealed that the concentration of DON is invariant throughout seasons, as is its 8 1 5 N, which has an N-isotopic imprint akin to that of the global ocean nitrate 5 I 5 N, ~4.9%o (Knapp et al. 2005). DON at BATS is thus a seemingly recalcitrant pool that does not reflect any input from N-fixation. Knapp et al. (2005) modelled the N budget at BATS and inferred that N-fixation, at most, may account for 10% of new N input to the mixed layer. They reasoned that the non-Redfield nutrient ratios in the thermocline and the low 5 I 5 N of nitrate are likely advected from elsewhere in the tropical Atlantic, where N-fixation is more prevalent. N-isotopic measurements at station Aloha in the North Pacific yield a much different N budget than that at BATS (Karl et al. 1997). Particles collected seasonally for 7 years in bottom-moored sediment traps had an annual mean 8 1 5 N of 3.10%o, compared to a 5 1 5N of nitrate in the thermocline of 6.5%o. Moreover, the summer time export pulse that coincided with Trichodesmium blooms yielded a 5 1 5N of 1.53%o. Based on their observations, Karl et al. (1997) concluded that N-fixation accounted for up to 50% of new production at Aloha during the observational period. 15 1.3.6. Inferring water mass transport from N isotopes On local scales, N stable isotopic tracers can also provide information on the source water(s) of ambient nitrate. Based on measurements of nitrate 8 I 5 N in the Subantarctic, Sigman et al. (1999) determined that Subantarctic surface water was supplied laterally from Antarctic surface water, and not from the Subantarctic thermocline. Moreover, Sigman et al. (2000) reported a 5 1 5 N enrichment of Upper Circumpolar Deep Water of the Southern ocean relative to underlying water masses, which they attributed to exchange with low-latitude water carrying heavy nitrate from denitrification. The relatively low nitrate 6 1 5N in the Subantarctic thermocline was then attributed to exchange with the thermocline of the oligotrophic gyre, itself affected by mixing with low-nitrate surface water or by the oxidation of newly-fixed N. In a similar example, seawater off southern California was enriched in 6 1 5 N, and isotopic maxima were found to coincide with isopycnal levels of 15N-enriched water of the Eastern Tropical North Pacific (Liu and Kaplan 1989). 1.3.7. Sedimentary N isotopes as apaleo-tracer of past nutrient utilization Of significance in the work of Altabet and others, as well as similar prior observations by Miyake and Wada (1967), is the recurring observation of isotopically light suspended organic nitrogen at the base of the euphotic zone, where nitrate is first discernible. This feature was interpreted as indicative of local nitrate utilization, whence the existence of an inverse trend between particle 5 1 5 N and the relative fraction of nitrate utilized was established. Altabet and Francois (1994) explicitly showed an inverse relationship between measurements of surface nitrate and 5 1 5 N of near-surface particulate nitrogen along a north-to-south transect in the equatorial Pacific. They also showed surface nitrate to be inversely related to core top sediment 5 1 5N. Application of sedimentary 6 1 5N as a paleotracer for surface nitrate utilization and depletion had previously been the basis for interpretation of data from the Southern Ocean (Altabet and 16 Francois 1994; Francois et al. 1992; Francois et al. 1993) and the Mediterranean (Calvert et al. 1992). Altabet and Francois (1994) thus provided important ground-truth for the application of sedimentary 5 1 5 N as a paleoceanographic tool. To date, studies of sedimentary 8 1 5N as a paleo-recorder have shown glacial changes in both water column denitrification (Altabet et al. 1995; Ganeshram et al. 1995), and nutrient utilization in the Southern Ocean (Francois et al. 1997; Sigman et al. 1999b) that could have led to reduced atmospheric pC02. Similar studies conducted in the Equatorial Pacific infer a decrease in relative nitrate utilization in the region during the last glacial maximum (Altabet 2001; Farrell et al. 1995). Major determinants of the ultimate 6 1 5N signature of exported particulate N are the 5 1 5N of the initial nitrate supply (5 , 5Ninj tiai), the isotope effect of nitrate uptake by phytoplankton (£), and the fraction of nitrate consumed (/). A number of studies provide estimates of 8 for nitrate assimilation in different oceanic regions. These values are derived from the Rayleigh model, using measurements of euphotic zone nitrate concentrations, compared to the 5 1 5N of suspended or sedimenting particulate nitrogen, or to the 8 1 5N of nitrate (Table 3.3). Many estimates converge on 5%o although there is significant variablility. Note that the similarity between global 8 estimates (5%o) and deep water nitrate 8 1 5N (also 5%o) is incidental. The isotope effects measured for laboratory cultures of marine phytoplankton vary widely among and within species (Table 3.3). The physiological factors that determine given isotope effects remain undefined, although our understanding of the physiological mechanism underlying isotope fractionation during nitrate assimilation by phytoplankton has increased significantly in recent years (Granger et al. 2004; Needoba and Harrison 2004; Needoba et al. 2004). 1.3.8. Current limitations of N isotopes as a tracer of ocean N processes Insights have been gained from the study of ocean N-isotope tracers, and advances have been particularly significant in expanding knowledge of the paleo-ocean, about which we know practically nothing. It has proven more challenging, however, for N-isotopes to provide novel constraints for the modern ocean. In many cases, N-isotopes have provided qualitative confirmation of operative processes, but have generally not 17 resulted in robust quantitative estimates of these processes. In cases where N-isotopes have provided flux estimates, such as work by Altabet on internal N-cycling, the isotope numbers have not corroborated numbers generated from alternate estimates. Though N isotopes in these studies may point to missing components in understanding localized re-cycling, they do not resolve the discrepancies. Hence a disadvantage of using N-isotopic tracers is that they alone provide no information on rates. Complementary use of chlorofluorocarbons or tritium/helium tracers can provide information on water mass age. However, time-dependent tracers, in turn, are inherently limited due to the difficulty of defining physical end-members. N-isotopic tracers can provide qualitative and some quantitative information on processes occurring on local scales. There have also been attempts to use N isotopes in tandem with fixed-N source and sink flux estimates to constrain the modern N budget (Altabet and Curry 1989; Brandes and Devol 2002; Deutsch et al. 2004; Liu and Kaplan 1988; Wada et al. 1975). Notably, in a recent exercise, Brandes and Devol (2002) constructed an isotopic mass balance of the modern N budget based on current estimates of N fluxes as well as isotopic values for the various pools and processes. They concluded from their model that both N-fixation and sedimentary denitrification were underestimated in current budgets, and that these missing fluxes entailed a downward revision of the residence time of fixed N in the ocean, from 3000 to 2000 years. This important study highlights the urgency of constraining the magnitude of sources and sinks of fixed nitrogen in the ocean, implicating a need to refine the various components of the ocean's N-isotopic budget. Our relatively superficial understanding of the physiological mechanisms underlying N isotope fractionation during biological reactions also stands as a limitation in their usage. Paleo-inferences on past nutrient utilization at high latitude during the last glacial maximum rest in part on the assumption that fractionation by phytoplankton at the surface has been invariant in time. However, light as been identified as an environmental factor likely to impact the expression of the N isotope effect on nitrate (Needoba and Harrison 2004), and this is the only environmental modulators of N isotope effects for assimilation of which we know. A more in-depth understanding of fractionation as it pertains to the nitrogen physiology of phytoplankton is required. Finally, N isotopes are limited in that they do not record N transformations that over-print one another. For instance, N isotopes cannot separate the N isotopic input 18 from newly remineralized N-rich organic matter from that of denitrification occurring in the same body of water. Similarly, N isotopes in nitrate do not distinguish between nitrate freshly mixed from the thermocline, versus nitrate that has been regenerated in the mixed layer, in essence constituting "regenerated nitrate." This particular shortcoming of N isotopes constitutes the original motivation that prompted Sigman and colleagues to develop a method with which they could measure both the N and the O isotopes of nitrate in seawater. Sigman reasoned that because the O isotopes are sensitive to different reactions in the N-cycle, their coupled use with N isotopes could potentially help disentangle signals from co-occurring processes. 1.4. The 6 1 8 0 of nitrate as a tracer of biological N transformations An additional tracer for ocean N-processes has recently been proposed by Sigman and colleagues, which, when used in tandem with N isotopes, promises to provide novel insights into the modern ocean N cycle (Casciotti et al. 2002; Lehmann et al. 2003; Lehmann et al. 2004; Sigman et al. 2001; Sigman et al. 2005). A method was developed in which nitrate in seawater samples is converted to nitrous oxide gas by denitrifying bacteria, allowing for measurements of both the 5 1 5N and the 6 I 8 0 of nitrate (i.e., the 1 8 0/ l 6 0) in this gas analyte at concentrations as low as 0.5 uM nitrate (Casciotti et al. 2002; Sigman et al. 2001). The utility of this additional tracer lies in the fact that measurements of the 6 1 8 0 of nitrate may act to complement the processes in the oceanic N-cycle that are not fully captured by the nitrogen isotopes. A comparison of the plight of the l 5 N isotope throughout the N-cycle contrasted 18 to that of the O isotope clarifies the former statement (Figure 1.4). The internal cycling of nitrogen, namely assimilation and remineralization, cause no change in the isotopic composition of the nitrogen atoms involved. Net changes in whole ocean (i.e., not local) 6 1 5N are driven solely by input and output processes, largely dominated by nitrogen fixation and denitrification (Table 1.1; Figure 1.3). Nitrogen fixation tends to lower the 8 N of fixed nitrogen, while denitrification can cause an enrichment of 5 l 5 N . The 5 1 8 0 of nitrate, while involved with fixed nitrogen, contrast those of its nitrogenous partner (Figure 1.4). Oxygen is only transitory within the nitrogen cycle, and its input and output processes are distinct from those of nitrogen. Nitrate assimilation, organic matter 19 decomposition, and subsequent nitrification do not represent internal cycling processes, but rather comprise the input and output of oxygen to the N-cycle. In other words, oxygen comes aboard via ammonia and nitrite oxidation, and is released back as water through nitrate and nitrite reduction. It follows that the 5 1 8 0 of freshly generated nitrate (from nitrification) does not depend on the origin of the nitrogen being nitrified, be it from newly fixed N, from denitrified N (as nitrite), from sedimenting biomass of phytoplankton growing in nitrate-rich environments, or from sedimenting biomass of plankton that completely assimilate the supply of surface nitrate. Because oxygen is insensitive to these processes, its isotopic signature acts to complement that of 1 5 N , which bears the scars of these previous N transformations. Measurements of deep-water nitrate suggest that the 6 1 8 0 of newly oxidized nitrate is similar to that of seawater (Casciotti et. al. 2002). This may appear surprising in light of the fact that biochemical studies of nitrifiers have established that one of the three oxygens added to ammonia to form nitrate comes from dissolved oxygen, while the remaining two originate from water (Andersson et al. 1982). One would then expect 5 1 8 0 values to partly reflect those of dissolved O2 in the ocean interior, which lie between 23.8%o to 35.5%o relative to air (Bender 1990). However these same biochemical studies also demonstrate a strong nitrifier-catalyzed oxygen exchange between nitrite and water. Similarly, Casciotti et al. (2002) observe nitrite-to-water oxygen isotope exchange in cultures of ammonia oxidizers, where less than one in six oxygen atoms in the nitrite produced comes from oxygen. Oxygen exchange with water is also plausible during nitrite oxidation, further replacing 02-oxygen atoms with H20-oxygen atoms. A depth profile of the 5 1 5 N and 6 1 8 0 of nitrate at station PAPA in the eastern subarctic Pacific, the first such profile for ocean nitrate, shows that like the nitrogen atom in nitrate, the oxygen atoms in nitrate are subject to isotopic fractionation, presumably from plankton assimilation of nitrate at the surface (Casciotti et al. 2002). Of further interest is the apparent similarity of the isotope effect observed for 6 1 5N compared to that for 5 , 8 0 . The fractionation of nitrate 1 5 N of appears similar, if not identical, to that of the 18 nitrate O (Casciotti et al. 2002). The 1:1 coupling for assimilation seemingly extends to the O-to-N fractionation ratios observed in conjunction with water-column denitrification (Sigman et al. 2003). Confirming that these ratios are indeed imparted by nitrate assimilation and denitrification became a central theme of my thesis work. 20 1.5. Thesis objectives Coupled measurements of the N and O isotopes of nitrate offer a novel and exciting avenue of research that promises to provide much insight into nitrogen processes in the ocean. However, the behaviour of nitrate with respect to N and O isotopes must clearly be understood before coupled 6 1 80:5 1 5N variations in oceanic nitrate can be used to disentangle operative N-processes in the water column, both qualitatively and quantitatively. The goal of my doctoral thesis was thus to elucidate the patterns and mechanisms of coupled N and O isotopic discrimination observed during nitrate assimilation by phytoplankton and dissimilatory reduction by denitrifying bacteria. Through this work, I have provided the systematics that will allow for pertinent interpretation of field observations of coupled N and O isotope patterns in nitrate, and consequently lead to important qualitative and quantitative constraints on oceanic re-cycling. Moreover, my observations have led to more insight into the physiological mechanism(s) underlying nitrate isotopic fractionation during its assimilation or denitrification. The work presented here consists of four sections. In the initial section, I present measurements of nitrate N and O isotope fractionation by various strains of eukaryotic marine phytoplankton to discern patterns in the coupled isotope effects (Chapter 2). This work not only establishes that the isotopes of nitrate are fractionated in equivalent and constant proportion during assimilation, but also provides insight into the mechanism underlying N and O isotope fractionation by eukaryotic phytoplankton. In the following section, I outline the development of a method to eliminate nitrite interference in our measurements of nitrate N and O isotopes. The 'denitrifier method' does not distinguish between nitrate and nitrite, such that the isotope ratios measured with this method reflect those of both molecules. Because growth of cultures often results in accumulation of nitrite, and moreover, because samples collected in the field often contain both nitrate and nitrite, I developed a method to eliminate nitrite from aqueous samples which is compatible with the denitrifier method and which does not alter the concentration or isotope ratios of incident nitrate. The fourth chapter of this thesis reports on N and O isotope fractionation by cultures of marine and freshwater denitrifiers. Again, as for eukaryotic assimilators, 21 denitrifiers show equivalent N and O isotope fractionation. The putative fractionation mechanism of denitrifier is then discussed in detail. The final part of my work consists of measurements of N and O isotope fractionation for prokaryotic algae assimilating nitrate. This work verifies whether the N and O coupling observed for eukaryotes follows suite in prokaryotes. Moreover, these results suggest that transport of nitrate across the cell membrane does not impart a significant isotope effect. 22 Table 1.1. Sources and sinks in the global marine nitrogen budget. Process Tg N yr 7T sources Pelagic N 2 fixation 110a - 330 Benthic N 2 fixation 15° River input 25b - 76a Atmospheric deposition 30a Total sources 180 - 451 sinks Water column denitrification 80a Sedimentary denitrification 95a - 280d Sedimentation 25 N 2 0 loss 4 Total sinks 204 - 389 (Gruber and Sarmiento 1997) (Brandes and Devol 2002) (Carpenter 1983) (Middelburg et al. 1996) (Nevison et al. 1995) 23 Table 1.2. Compilation of isotope effects (e) observed in laboratory cultures for N cycle processes. Process Isotope effect (e) Details Nitrification 35 - 38%o Nitrosomonas europaea a'b (NH 4 + -> N0 2") 14%, Nitrosomonas marina b 19°/oo Nitrosomonas C-l 13a b 25%o Nitrosospira tenuis b 32%o Nitrosomonas eutropha b Denitrification 28%o Paracoccus denitrificans0 (N0 3" -> N 2) 20 - 30%o Pseudomonas stutzerid 13 -20%. Pseudomonas denitrificansc Nitrogen fixation 0%o Trichodesmium sp. f (N 2 -> NH 4 + ) 0%o Azotobacter vinlandif N H 4 + assimilation 20%o Skeletonema costratum 8 20%o Mixed cultureh 16% Emiliana huxlei (coastal)' 15 - 19%o Emiliana huxlei (oceanic)1 25%o Chaetoceros debilis 1 NO3" assimilation 5 - 7%o Thalassiosira pseudonana ''J 4°/oo Emiliana huxlei (coastal)' 4 - 20% Emiliana huxlei (oceanic)1'-1 5°/oo Chaetoceros debilis 1 7- 11%0 Skeletonema costratumk 2 - 4 %0 Isochrysis galbana k - 6 - 4 %0 Pavlova luterik 1 - 6 %0 Dunaliella tertiolecta k 1 - 3%o Chroomonas salina k 5 - 17%o Thalassiosira weissflogiiJ'k'1 7- 19%o Phaeodactylum tricornutum m •.._' _—: : : n 10-20%o Thalassiosira oceanica3 "(Mariotti et al. 1981);b(Casciotti et al. 2003); c(Barford et al. 1999); d(Wellman et al. 1968); e(Delwiche and Steyn 1970); f(Carpenter et al. 1997); h(Waser et al. 1999); 8(Pennock et al. 1996); 'Waser et al. (1998); J(Granger and Sigman unpublished); k(Montoya and McCarthy 1995); '(Needoba and Harrison 1994); m(Wadaand Hattori 1978). 24 Table 1.3. Isotope effects (e) from in situ estimates for N cycle processes. Process Isotope effect (e) Details Nitrification 15°/oo Chesapeake Bay3 Denitrification 20 - 30%; - 40%o* Eastern Tropical North Pacific b ' 22 -27%, Arabian Sea c ' f 25%o Gulf of California8'11 25 - 29%o Soil 1 Nitrogen fixation -0.4%, Trichodesmium coloniesJ -0.38%o Western Tropical North Pacifick Ammonia assimilation 6.5 - 8%o Chesapeake Bay a 9.1%, Delaware Estuary1 5 - 20%o Bacterial assemblage m Nitrate assimilation 4 - 10%, Southern Ocean n ' 0 , p 5°/oo Subarctic Pacific r 5°/oo Equatorial Pacific r 5%o Eastern North Pacific h "(Horrigan et al. 1990); "(Voss et al. 2001); c(Brandes et al. 1998);d(Liu and Kaplan 1989);*e(Cline and Kaplan 1975); f(Naqvi et al. 1998); g(Sigman et al. 2003); h(Altabet et al. 1999); ^Mariotti et al. 1981); J(Carpenter et al. 1997); k(Karl et al. 1997);'(Cifuentes et al. 1989); m(Hoch et al. 1996); "(Sigman et al. 1999a); °(Karsh et al. 2003); "(Lourey et al. 2003); "(Wu et al. 1997); r(Altabet 2001);;s(Lourey et al. 2003). 25 N 0 3 " (nitrate) a s s i m i l a t i o n denitrification N 0 2 ' (nitrite) (nitric oxide) N O i i (nitrous oxide) N 2 0 nitrogen fixation nitrification (?) (hydroxylamine) NH2OH N H / a n a b o l i s m / a s s i m i l a t i o n = ± P O N / D O N (ammonium) c a t a b o l i s m Figure 1.1. Schematic diagram of the processes and pools of N fundamental to the cycling of N in the ocean. PON: particulate organic nitrogen. DON: dissolved organic nitrogen. Nitrate is the most oxidized N species, while ammonium and organic nitrogen comprise the most reduced species involved in the cycle. Hydroxylamine is an intermediate species within the ammonia oxidation pathway, however it does not accumulate extracelullarly. The dashed line designates a physiological process that has been observed solely in vitro (Beaumont et al. 2002) and whose oceanographic relevance is uncertain. 26 100 - 1 o n i - - 0 -01 80 ™ 60 m~"' "A 40 1 20 A ••-A-0 0 2 days 20 -- 10 -• 0 -• -10 -20 1/3 TD -O-- [ N C y ] •A- • 6 1 5Nreactant -A- • 5 1 5 Nins tan t - X — 5 1 5 Nin teg ra ted Figure 1.2. Consumption o f nitrate during growth o f the marine diatom Thalassiosira weissflogii in batch culture, and the concomitant increase in the 8 1 5 N of nitrate (the reactant). The estimated e for the integrated reaction is 1 l%o. A lso plotted are the calculated instantaneous 5 1 5 N of growing cells (the instantaneous product) as wel l as the calculated 5 I 5 N of accumulated cells (the integrated product). Measurements o f nitrate and 5 I 5 N of nitrate for the growing culture were measured by Granger and Sigman. 27 Surface ocean Particulate N remineralizatione <5%. Nitrate uptake Deep ocean E = -l°/oo N03-l 5 N = 5%» Atmospheric N 2 o , 5 N = 0%^ Dissolved N 2 5 I 5 N = 0.6%o Continental Inputs 6 I 5 N - 3%c N H / , N 0 3 D O N , PN 6 l 5 N 0 3 - = 10-20%o Water column denitrification e = 25 %o Sedimentary denitrification E = 0%„ Figure 1.3. The marine nitrogen isotope budget and the processes that affect the distribution of nitrogen isotopes in the ocean. Sources comprise N-fixation at the surface-ocean and continental inputs. Sinks of fixed N are water-column and sedimentary denitrification. Internal cycling processes include nitrate assimilation at the surface and remineralization at depth. Ammonia generation, assimilation, and oxidation are omitted for simplicity. The isotope effect associated with nitrification is also not included because this process is believed to go to completion. Question marks indicate uncertainties in the estimates due to large variations in reported measurements, or paucity of data. This figure is a reproduction of a similar diagram presented elsewhere (Sigman and Casciotti 2001). 2 8 N i t r o g e n a t o m s in m a r i n e n i t r a t e : N2 fixation 6 1 5 N = -1%„ nitrate assimilation 1 5 E = 5%o nitrification • 1 5 £ = 17% 1 5 s = 25%o ^- denitrification O x y g e n a t o m s in m a r i n e n i t r a t e : 6 1 8 O = 0%o nitrification 1 8 C _ 15, . nitrate assimilation 1 8 E _ 1So ^ denitrification Figure 1.4. Processes that affect N (left) and O (right) atoms in N0 3 \ Red refers to a source or sink of N or O in N03". Typical N and O isotope effects (15e, l8e) and ratios (5) are shown. 29 1.6. References Allen, A. E., M. H. Howard-Jones, M. G. Booth, M. E. Frischer, P. G. Verity, D. A. Bronk, and M. P. Sanderson. 2002. Importance of heterotrophic bacterial assimilation of ammonium and nitrate in the Barents Sea during summer. J. Marine Syst. 38: 93-108. Altabet, M. A. 1988. Variations in nitrogen isotopic composition between sinking and suspended particles: implications for nitrogen cycling and particle transformation in the open ocean. Deep-Sea Res. Part I 35: 535-554. Altabet, M. A. 1989. A time-series study of the vertical structure of nitrogen and particle dynamics in the Sargasso Sea. Limnol. Oceanogr. 24: 1185-1201. Altabet, M. A. 1991. Nitrogen Isotopic Evidence for the Source and Transformation of Sinking Particles in the Open Ocean. Abstr. Pap. Am. Chem. S. 201: 41-Geoc. Altabet, M. A. 2001. Nitrogen isotopic evidence for micronutrient control of fractional NO3" utilization in the equatorial Pacific. Limnol. Oceanogr. 46: 368-380. Altabet, M. A., and W. B. Curry. 1989. Testing models of past ocean chemistry using foraminiferal 15N/14N. Global Biogeochem. Cy. 3. Altabet, M. A., W. G. Deuser, S. Honjo, and C. Stienen. 1991. Seasonal and depth-related changes in the source of sinking particles in the North Atlantic. Nature 354: 136-139. Altabet, M. A., and R. Francois. 1994. The use of nitrogen isotopic ratio for reconstruction of past changes in surface ocean nutrient utilization, p. 281-306. In R. Zahn, M. Kaminski, L. Labeyrie and T. F. Pederson [eds.], Carbon Cycling in the Glacial Ocean: Constraints on the Ocean's Role in Global Change. NATO ASI series. Springer Verlag. Altabet, M. A., R. Francois, D. W. Murray, and W. L. Prell. 1995. Climate-Related Variations in Denitrification in the Arabian Sea from Sediment N-15/N-14 Ratios. Nature 373: 506-509. Altabet, M. A., and J. J. McCarthy. 1985. Temporal and spatial variation in the natural abundance of l 5 N in PON from a warm-core ring. Deep-Sea Res. 32: 755-772. Altabet, M. A., C. Pilskaln, R. Thunell, C. Pride, D. Sigman, F. Chavez, and R. Francois. 1999. The nitrogen isotope biogeochemistry of sinking particles from the margin of the eastern North Pacific. Deep-Sea Res. 46: 655-679. 30 Altabet, M. A., and L. F. Small. 1990. Nitrogen isotopic ratios in fecal pellets produced by marine zooplankton. Geochim. Cosmochim. Acta 54: 155-163. Andersson, K. K., S. B. Philson, and A. B. Hooper. 1982.180 isotope shift in 1 SN NMR analysis of biological N-oxidations: H 20-N0 2" exchange in the ammonia-oxidizing bacterium Nitrosomonas. Proc. Natl. Acad. Science 79: 5871-5875. Barford, C. C , J. P. Montoya, M. A. Altabet, and R. Mitchell. 1999. Steady-state Nitrogen isotope effects of N2 and N20 production in Paracoccus denitrificans. Appl. Environ. Microbiol. 65: 989-994. Beaumont, H. J. E., N. G. Hommes, L. A. Sayavedra-Soto, D. J. Arp, D. M. Arciero, A. B. Hooper, H. V. Westerhoff, and R. J. M. Van Spanning. 2002. Nitrite reductase of Nitrosomonas europaea is not essential for production of gaseous nitrogen oxides and confers tolerance to nitrite. J. Bact. 184: 2557-+. Bender, M. L. 1990. The d , 80 of dissolved 0 2 in seawater - A unique tracer of circulation and respiration in the deep-sea. J.of Geophys. Res. - Oceans 95: 22243-22252. Berges, J. A., and P. J. Harrison. 1995. Nitrate reductase-activity quantitatively predicts the rate of nitrate incorporation under steady-state light limitation - a revised assay and characterization of the enzyme in 3 species of marine-phytoplankton. Limnol. Oceanogr. 40: 82-93. Brandes, J. A., and A. H. Devol. 1997. Isotopic fractionation of oxygen and nitrogen in coastal marine sediments. Geochim. Cosmochim. Ac. 61: 1793-1801. Brandes, J. A., and A. H. Devol. 2002. A global marine fixed nitrogen isotopic budget: Implications for Holocene nitrogen cycling. Global Biogeochem. Cy. 16: 10.1029/2001GB001856. Brandes, J. A., A. H. Devol, T. Yoshinari, D. A. Jayakumar, and S. W. A. Naqvi. 1998. Isotopic composition of nitrate in the central Arabian Sea and eastern tropical North Pacific: a tracer for mixing and nitrogen cycles. Limnol. Oceanogr. 43: 1680-1689. Calvert, S. E., B. Nielsen, and M. R. Fontugne. 1992. Evidence from nitrogen isotope ratios for enhanced productivity during formation of Eastern Mediterranean sapropels. Nature 359: 223-225. Carpenter, E., H. Harvey, B. Fry, and D. Capone. 1997. Biogeochemical tracers of the marine cyanobacterium Trichodesmium. Deep-Sea Res. Part I 44: 27-38. 31 Carpenter, E. J. 1983. Nitrogen fixation by marine Oscillatoria (Trichodesmium) in the world's oceans., p. 65-103. In E. J. Carpenter and D. G. Capone [eds.], Nitrogen in the Marine Environment. Academic Press. Carpenter, E. J., B. Bergman, R. Dawson, P. J. A. Siddiqui, E. Soderback, and D. G. Capone. 1992. Glutamine-synthetase and nitrogen cycling in colonies of the marine diazotrophic cyanobacteria Trichodesmium Spp. Appl. Environ. Microbiol. 58: 3122-3129. Casciotti, K. L., D. M. Sigman, M. G. Hastings, J. K. Bohlke, and A. Hilkert. 2002. Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater using the denitrifier method. Anal. Chem. 74: 4905-4912. Casciotti, K. L., D. M. Sigman, and B. B. Ward. 2003. Linking diversity and stable isotope fractionation in ammonia-oxidizing bacteria. Geomicrobiol. J. 20. Cifuentes, L. A., M. L. Fogel, J. R. Pennock, and J. H. Sharp. 1989. Biogeochemical Factors That Influence the Stable Nitrogen Isotope Ratio of Dissolved Ammonium in the Delaware Estuary. Geochim. Cosmochim. Ac. 53: 2713-2721. Cline, J. D. 1973. Denitrification and isotopic fractionation in two contrasting marine environments: The eastern tropical North Pacific and the Cariaco Trench, Ph.D. Thesis, Univ. of California, Los Angeles. Univ. of California, Los Angeles. Codispoti, L. A., J. A. Brandes, J. P. Christensen, A. H. Devol, S. W. A. Naqvi, H. W. Paerl, and T. Yoshinari. 2001. The oceanic fixed nitrogen and nitrous oxide budgets: Moving targets as we enter the anthropocene? Sci. Mar. 65: 85-101. Delwiche, C. C , and P. L. Steyn. 1970. Nitrogen isotope fractionation in soils and microbial reactions. Env. Sci. Tech. 4: 929-935. Deuser, W. D. 1986. Seasonal and interannual variations in deep-water particle fluxes in the Sargasso Sea and their relationship to surface hydrography. Deep-Sea Res. 53: 225-246. Deutsch, C , D. M. Sigman, R. C. Thunell, A. N. Meckler, and G. H. Haug. 2004. Isotopic constraints on glacial/interglacial changes in the oceanic nitrogen budget. Global Biogeochem. Cy. 18: GB4012. Devol, A. H. 1991. Direct measurement of nitrogen gas fluxes from Continental Shelf sediments. Nature 349: 319-321. Dugdale, V. A., and J. J. Goering. 1967. Uptake of new and regenerated forms of nitrogen in primary productivity. Limnol. Oceanogr. 12: 196-206. 32 Eppley, R. W., and B. J. Peterson. 1979. Particulate organic matter flux and planktonic new production in the deep ocean. Nature 282: 677-680. Falkowski, P. G. 1997. Evolution of the nitrogen cycle and its influence on the biological sequestration of C 0 2 in the ocean. Nature 387: 272-275. Farrell, J. W., T. F. Pedersen, S. E. Calvert, and B. Nielsen. 1995. Glacial-interglacial changes in nutrient utilization in the equatorial Pacific Ocean. Nature 377: 514-517. Francois, R., M. A. Altabet, and L. H. Burkle. 1992. Glacial to interglacial changes in surface nitrate utilization in the Indian sector of the Southern Ocean as recorded by sediment d 1 5 N. Paleoceanogr. 7: 589-606. Francois, R., M. P. Bacon, M. A. Altabet, and L. D. Labeyrie. 1993. Glacial/intergalcial changes in sediment rain rate in the S.W. Indian sector of Subantarctic waters as recorded by 2 3 0Th, 2 3 1Pa, U, and d l 5N. Paleoceanogr. 8: 611-630. Francois, R. F. and others 1997. Water column stratification in the Southern Ocean contributed to the lowering of glacial atmospheric CO2. Nature 389: 929-935. Ganeshram, R. S., T. F. Pedersen, S. E. Calvert, and J. W. Murray. 1995. Large changes in oceanic nutrient inventories from glacial to interglacial periods. Nature 376: 755-758. Gruber, N., and J. L. Sarmiento. 1997. Global patterns of marine nitrogen fixation and denitrification. Global Biogeochem. Cy. 11: 235-266. Hoch, M. P., R. A. Snyder, L. A. Cifuentes, and R. B. Coffin. 1996. Stable isotope dynamics of nitrogen recycled during interactions among marine bacteria and protists. Mar. Ecol.-Prog. Ser. 132: 229-239. Hoering, T., and H. T. Ford. 1960. The isotope effect in the fixation of nitrogen by Azotobacter. J. Amer. Chem. Soc. 82: 376-378. Horrigan, S. G., J. P. Montoya, J. L. Nevins, and J. J. McCarthy. 1990. Natural isotopic composition of dissolved inorganic nitrogen in the Chesapeake Bay. Est. Coast. Shelf Sci. 30: 393-410. Karl, D., R. Letelier, L. Tupas, J. Dore, J. Christian, and D. Hebel. 1997. The role of nitrogen fixation in biogeochemical cycling in the subtropical North Pacific Ocean. Nature 388: 533-538. Karl, D. and others 2002. Dinitrogen fixation in the world's oceans. Biogeochem. 57/58: 47-98. 33 Karsh, K. L., T. W. Trull, A. J. Lourey, and D. M. Sigman. 2003. Relationship of nitrogen isotope fractionation to phytoplankton size and iron availability during the Southern Ocean Iron RElease Experiment (SOIREE). Limnol. Oceanogr. 48: 1058-1068. Knapp, A. N., D. M. Sigman, and F. Lipschultz. 2005. N isotopic composition of dissolved organic nitrogen and nitrate at the Bermuda Atlantic time-series study site. Global Biogeochem. Cy. 19. Ledgard, S. F., K. C. Woo, and F. J. Bergersen. 1985. Isotopic fractionation during reduction of nitrate to nitrite by extracts of spinach leaves. Aust. J. Plant Physiol. 12: 631-640. Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2003. The effect of benthic nitrogen cycling on the delta N-15 and delta 0-18 of water-column nitrate. Geochim. Cosmochim. Ac. 67: A249-A249. Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2004. Coupling the N-15/N-14 and 0- 18/0-16 of nitrate as a constraint on benthic nitrogen cycling. Mar. Chem. 88: 1- 20. Liu, K.-K., and I. R. Kaplan. 1988. Variation in nitrogen isotope fractionation during denitrification and nitrogen isotope balance in the ocean. EOS 69: 1098. Liu, K.-K., and I. R. Kaplan. 1989. The eastern tropical Pacific as a source of 15N-enriched nitrate in seawater off southern California. Limnol. Oceanogr. 34: 820-830. Liu, K. K. 1979. Geochemistry of inorganic nitrogen compounds in two marine environments; the Santa Barbara Basin and the ocean off Peru. University of California, Los Angeles. Liu, K. K., M. J. Su, C. R. Hsueh, and G. C. Gong. 1996. The nitrogen isotopic composition of nitrate in the Kuroshio Water northeast of Taiwan: evidence for nitrogen fixation as a source of isotopically light nitrate. Mar. Chem. 54: 273-292. Lourey, M. J., T. W. Trull, and D. M. Sigman. 2003. Sensitivity of delta N-15 of nitrate, surface suspended and deep sinking particulate nitrogen to seasonal nitrate depletion in the Southern Ocean. Global Biogeochem. Cy. 17. Mariotti, A., J. C. Germon, P. Hubert, P. Kaiser, R. Letolle, A. Tardieux, and P. Tardieux. 1981. Experimental determination of nitrogen kinetic isotope 34 fractionation: some principles; illustration for the denitrification and nitrification processes. Plant Soil 62: 413-430. Martin, J. FL, S. E. Fitzwater, and R. M. Gordon. 1990. Iron deficiency limits growth in Antarctic waters. Global Biogeochem. Cy. 4: 5-12. Michaels, A. F., N. R. Bates, K. O. Buesseler, C. A. Carlson, and A. H. Knap. 1994. Carbon-Cycle Imbalances in the Sargasso Sea. Nature 372: 537-540. Michaels, A. F. and others 1996. Transformations of nitrogen and phosphorus in the deep North Atlantic Ocean. Biogeochem. 35: 181-226. Middelburg, J. J., K. Soetaert, P. M. J. Herman, and C. H. R. Heip. 1996. Denitrification in marine sediments: A model study. Global Biogeochem. Cy. 10: 661-673. Minagawa, M., and E. Wada. 1986. Nitrogen isotope ratios of red tide organisms in the East China Sea: A characterization of biological nitrogen fixation. Mar. Chem. 19: 245-249. Miyake, Y., and E. Wada. 1967. The abundance of 15N/14N in marine environments. Records of Oceanography Works, Japan 9: 37-53. Montoya, J. P., E. J. Carpenter, and D. G. Capone. 2002. Nitrogen fixation and nitrogen isotope abundances in zooplankton of the oligotrophic North Atlantic. Limnol. Oceanogr. 47: 1617-1628. Montoya, J. P., and J. J. McCarthy. 1995. Isotopic fractionation during nitrate uptake by marine phytoplankton grown in continuous culture. J. Plankton Res. 17: 439-464. Naqvi, S. W. A., T. Yoshinari, J. A. Brandes, A. H. Devol, D. A. Jayakumar, P. V. Narvekar, M. A. Altabet, and L. A. Codispoti. 1998. Nitrogen isotopic studies in the suboxic Arabian Sea. Proc. Indian Acad. Sciences 107(4): 367-378.107: 367-378. Needoba, J. A., and P. J. Harrison. 2004. Influence of low light and a light: Dark cycle on NO3" uptake, intracellular NO3", and nitrogen isotope fractionation by marine phytoplankton. J. Phycol. 40: 505-516. Needoba, J. A., D. M. Sigman, and P. J. Harrison. 2004. The mechanism of isotope fractionation during algal nitrate assimilation as illuminated by the N-15/N-14 of intracellular nitrate. J. Phycol. 40: 517-522. Nevison, C. D., R. F. Weiss, and D. J. Erickson. 1995. Global oceanic emissions of nitrous-oxide. J. Geophys. Res.-Oceans 100: 15809-15820. 35 Pennock, J. R., D. J. Velinsky, J. H. Sharp, J. Ludlam, and M. L. Fogel. 1996. Isotope fractionation of ammonium and nitrate during their uptake by Skeletonema costatum.: Implications for the dl5N dynamics under bloom conditions. Limnol. Oceanogr. 41: 451-459. Seitzinger, S. P. 1988. Denitrification in freshwater and coastal marine ecosystems: Ecological and geochemical significance. Limnol. Oceanogr. 33: 702-724. Shearer, G., J. D. Schneider, and D. H. Kohl. 1991. Separating the efflux and influx components of net nitrate uptake by Synechococcus-R2 under steady-state conditions. J. Gen. Microbiol. 137: 1179-1184. Sigman, D. M., M. A. Altabet, R. Francois, D. C. McCorkle, and G. Fischer. 1999a. The d 1 5N of nitrate in the Southern Ocean: Consumption of nitrate in surface waters. Global Biogeochem. Cy. 13: 1149-1166. Sigman, D. M., M. A. Altabet, R. Francois, D. C. McCorkle, and J.-F. Gaillard. 1999b. The isotopic composition of diatom-bound nitrogen in Southern Ocean sediments. Paleoceanogr. 14: 118-134. Sigman, D. M., M. A. Altabet, D. C. McCorkle, R. Francois, and G. Fischer. 2000. The dl5N of nitrate in the Southern Ocean: Nitrogen cycling and circulation in the ocean interior. J. Geophys. Res. 105: 19,599. Sigman, D. M., and K. L. Casciotti. 2001. Nitrogen isotopes in the ocean, p. 1884-1894. In J. H. Steele, K. K. Turekian and S. A. Thorpe [eds.], Encyclopedia of Ocean Sciences. Academic Press. Sigman, D. M., K. L. Casciotti, M. Andreani, C. Barford, M. Galanter, and J. K. Bohlke. 2001. A bacterial method for the nitrogen isotopic analysis of nitrate in seawater and freshwater. Anal. Chem. 73: 4145-4153. Sigman, D. M., J. Granger, P. J. DiFiore, M. M. Lehmann, R. Ho, G. Cane, and A. Van Geen. 2005. Coupled nitrogen and oxygen isotope measurements of nitrate along the eastern North Pacific margin. Global Biogeochem. Cy. 19: GB4022. Sigman, D. M., R. Robinson, A. N. Knapp, A. Van Geen, D. C. McCorkle, J. A. Brandes, and R. C. Thunell. 2003. Distinguishing between water column and sedimentary denitrification in the Santa Barbara Basin using the stable isotopes of nitrate. Geochem. Geophys. Geosys. 4: 1040-1059. Voss, M., M. A. Altabet, and B. Vonbodungen. 1996. delta N-15 in sedimenting particles as indicator of euphotic-zone processes. Deep-Sea Res. Pt I 43: 33-47. 36 Voss, M., J. W. Dippner, and J. P. Montoya. 2001. Nitrogen isotope patterns in the oxygen-deficient waters of the Eastern Tropical North Pacific Ocean. Deep-Sea Res. 4 8 : 1905-1921. Wada, E., and A. Hattori. 1978. Nitrogen isotope effects in the assimilation of inorganic nitrogenous compounds. Geomicrobiol. J. 1: 85-101. Wada, E., T. Kadonaga, and S. Matsuo. 1975. 15N abundance in nitrogen of naturally occuring substances and global asswessment of denitrification from isotopic viewpoint. Geochem. Journal 9: 139-148. Ward, B. B., K. A. Kilpatrick, E. H. Renger, and R. W. Eppley. 1989. Biological nitrogen cycling in the nitracline. Limnol. Oceanogr. 3 4 : 493-513. Waser, N. A., Z. M. Yu, K. D. Yin, B. Nielsen, P. J. Harrison, D. H. Turpin, and S. E. Calvert. 1999. Nitrogen isotopic fractionation during a simulated diatom spring bloom: importance of N-starvation in controlling fractionation. Mar. Ecol.-Prog. Ser. 179: 291-296. Wellman, R. P., F. D. Cook, and H. R. Krouse. 1968. Nitrogen-15: microbiological alteration of abundance. Science 161: 269-270. Wu, J., S. E. Calvert, and C. S. Wong. 1997. Nitrogen isotope variations in the subarctic Pacific northeast Pacific: relationships to nitrate utilization and trophic structure. Deep-Sea Res. Part I 4 4 : 287-314. Wyrtki, K. 1962. The oxygen minima in relation to ocean circulation. Deep-Sea Research 9. Zehr, J. P., E. J. Carpenter, and T. A. Villareal. 2000. New perspectives on nitrogen-fixing microorganisms in tropical and subtropical oceans. Trends Microbiol. 8: 68-73. Zumft, W. G. 1997. Cell biology and molecular basis of denitrification. Microbiol. Mol. Biol. R. 61: 533-616. 37 Chapter 2 Coupled nitrogen and oxygen isotope fractionation of nitrate during assimilation by cultures of marine phytoplankton A version of this chapter has been published. Granger, J., D. M. Sigman, J. A. Needoba, and P. J . Harrison. 2004. Coupled nitrogen and oxygen isotope fractionation of nitrate during assimilation by cultures of marine phytoplankton. Limnol. Oceanogr. 49: 1763-1773. 38 2.1. Introduction Stable isotopes are a common tool in studies of the N cycle, both in the ocean and on land. Variations in the 1 5 N/ 1 4 N ratio of inorganic nitrogen pools can provide an integrative picture of the N cycle, from which physical, chemical and biochemical fluxes can be inferred. Environmental variations in 1 5 N/ 1 4 N typically occur because, in most chemical and biochemical transformations, the rate of reaction of' 5N-bearing substrate is slightly lower than that for the same substrate bearing 1 4 N. The extent to which a biological N transformation fractionates between 1 4 N and 1 5 N is given by the isotope effect, 1 5e. In the case of a unidirectional reaction, 1 5 E is referred to as the kinetic isotope effect and is a function of the ratio of the reaction rates (k) for the molecules containing the two isotopes: 15e(%0) = ( ' V 5 k - 1) x 1000. Due to isotope fractionation during nitrate assimilation, the 1 5 N/ ! 4 N of nitrate is observed to increase as it is consumed in the surface ocean (Sigman et al. 1999b; Sigman et al. 1997; Wu et al. 1997). This isotopic change in nitrate is communicated by assimilation into phytoplankton biomass, resulting in meridional gradients in the 1 5 N/ I 4 N of sinking particulate N and surface sedimentary N in the Equatorial Pacific and in the Southern Ocean that are correlated with nitrate utilization (and thus anti-correlated with surface nitrate concentration; (Altabet and Francois 1994; Farrell et al. 1995; Francois et al. 1992; Francois et al. 1997; Sigman et al. 1999a)). Thus, down-core changes in the l 5 N/ 1 4 N of sediment and sedimentary fractions have been interpreted as reflecting past changes in the degree of nitrate consumption by phytoplankton in the surface ocean (Altabet and Francois 1994; Farrell et al. 1995; Francois et al. 1992; Francis et al. 1997; Sigman et al. 1999a). The isotopic signal of nitrate assimilation can also be used to study aspects of the modern ocean N cycle, for instance, the source of nitrate to surface waters in nutrient-rich regions (Sigman et al. 1999b). The isotope effect is a key parameter linking nitrate assimilation to the l 5 N/ ' 4 N of nitrate and particulate N. The 1 5 N/ 1 4 N of nitrate in the upper ocean typically suggests an isotope effect of 5-10%o for nitrate assimilation, with most estimates closer to 5%o (Altabet 2001; Sigman et al. 1999b; Wada 1980; Wu et al. 1997). In contrast, estimates of 1 5 E derived from laboratory studies of marine phytoplankton show a wide range of variation, from 0%o to 20%o (Montoya and McCarthy 1995a; Needoba 2003; Wada and 39 Hattori 1978; Waser et al. 1998a). The observed variation in 15e among and within cultured phytoplankton species is not understood, and environmental controls on 1 5 E are undefined, largely because the N isotope fractionation mechanism for nitrate assimilation remains uncertain. These unknowns limit the utility of the N isotopes in the modern ocean and in the effort to reconstruct past ocean conditions. Previously, nitrate l 5 N / l 4 N in seawater has been measured by reduction to ammonia, followed by extraction of the ammonia by distillation (Cline and Kaplan 1975) or diffusion (Sigman et al. 1997), reaction to N 2 gas (typically by combustion), and mass spectrometric analysis of the N 2 . Here, we take advantage of a recently developed method for nitrate isotopic analysis that uses denitrifying bacteria to convert nitrate (and nitrite) to nitrous oxide (N 20), followed by isotopic analysis of the N 2 0 (Casciotti et al. 2002; Sigman et al. 2001). The 'denitrifier method' for seawater nitrate isotope analysis has advantages relative to the ammonium-based methods that are critical to culture studies of marine phytoplankton. First, a roughly 100-fold reduction in sample size requirement allows for cultures of small to moderate volume. Second, a lack of cross-contamination by dissolved organic N (DON) or ammonium (NH/) and a reduction in reagent blank are both critical for culture studies and allow for the reproducible isotopic analysis of samples with as little as 0.5 pmol L"1 nitrate. Finally, and most central to this study, the denitrifier method is the first to allow O isotope analysis of nitrate in a saline solution such as seawater (Casciotti et al. 2002). We report here a culture study of N and O isotope fractionation of nitrate during its assimilation by marine phytoplankton. To our knowledge, this study represents the first culture-based effort to characterize the relationship between the N and O isotopes of nitrate during an important N transformation. Our measurements of coupled N and O isotope fractionation provide new insight into the mechanism underlying isotopic fractionation during nitrate assimilation by marine phytoplankton. This has significance for the biological chemistry of nitrate assimilation and for paleoceanographic studies, in which assumptions must be made about the isotope effect of nitrate assimilation in the past. Moreover, it provides a basis for use of the coupled N and O isotopes of nitrate to study global ocean N cycling (Lehmann et al. 2004; Sigman et al. 2003). 40 2.2. Methods Experimental algal strains, Thalassiosira weissflogii (acting Thalassiosira oceanica, Thalassiosira pseudonana (3H), and Emiliana huxleyi, were grown in semi-continuous batch cultures in the artificial seawater medium AQUIL (Price 1988/1989) at 20°C under continuous saturating light at 150 umol quanta m"2 s"1. The artificial seawater mixture was amended with 100 umol L"1 silicate, 10 pmol L"1 phosphate, and between 50 to 150 umol L"1 nitrate. It was then passed through Chelex 100 resin to remove contaminating trace-metals (Price et al. 1988/89). Metal-clean artificial seawater was stored in acid-washed polycarbonate bottles, and sterilized by microwaving (Keller et al. 1988). Filter-sterilized f/2 vitamins were added to sterile media, as well as filter-sterilized AQUIL-trace-metals chelated with 100 umol L"1 ethylenediaminetetraacetic acid (EDTA; Price et al. 1988/89). Cells were acclimated to respective media for two culture transfers (ca. 8 generations) and inoculated in 500 ml polycarbonate media bottles. Our focal interest was in the variations of nitrate N and O isotopic fractionation, both among and within species. Two of the experimental strains were thus subjected to a variety of culture conditions in an attempt to effect changes in the isotope effects manifested by a specific strain. In cultures of T. weissflogii and T. oceanica, iron concentrations were modulated to obtain iron-limited growth rates of the strains. Total Fe concentrations in incremental treatments were 1 umol L"1, 100 nmol L"1, 45 nmol L"1, and 12 nmol L'1, corresponding to approximate free ferric iron concentrations (expressed as pFe = -log(Fe3+)) of 19, 20, 20.5 and 21, respectively, calculated with MINEQL -(Westall 1976). A set of experiments with T. weissflogii at various iron concentrations was also conducted in stirred culture vessels to compare with unstirred cultures. Finally, N and O isotopic fractionation by T. weissflogii was monitored in short-term nitrate uptake experiments. In these experiments, cells pre-conditioned in pFel9 or pFe20 medium were harvested in late exponential growth. Whole cultures (500 ml) were filtered onto acid-washed, 5 um pore-size polycarbonate filters, and resuspended in 500 ml of fresh medium (of the same iron concentration). Subsamples of culture filtrate were then collected at short (ca. bi-hourly) time intervals. Cell cultures were sub-sampled throughout exponential growth. Growth was monitored by cell counts on a Z™ Series Coulter® Counter. For particulate N isotopic measurements, 20 ml of cell culture were gently filtered onto a pre-combusted AE glass-41 fibre filter. Filters were then dried at 60°C, pelleted into tin capsules and sent for N isotopic analysis to Dr. David Harris at the Stable Isotope Facility, University of California, Davis. Isotope ratios of the particulate N on the filters were determined by continuous flow combustion/isotope mass spectrometry using a Europa ANCA elemental analyzer on-line with a Europa Hydra 20/20 mass spectrometer. Intracellular nitrate was collected for isotopic comparison with the medium in four T. weissflogii cultures grown under conditions similar to those described above (described in detail by Needoba and Harrison 2004). Briefly, cells were grown in semi-continuous batch cultures in artificial seawater with 200 umol L"1 nitrate. Cultures were initiated in filter-sterilized medium and incubated at 18°C under saturating or sub-saturating light levels, and some cultures were also grown in iron-deplete medium. Exponentially growing cells (300 - 500 ml) were harvested onto a 47 mm pre-combusted GF/F filter, washed with 3% NaCl to remove remnant extracellular nitrate, and boiled in water to extract intracellular nitrate (Thorensen et al. 1982). Nitrate concentrations were then measured as described below, and the N and O isotopic compositions of internal nitrate were then determined with the denitrifier method (Casciotti et al. 2002; Sigman et al. 2001). The nitrate concentrations and N isotope data alone are reported by Needoba et al. (2004). For N and O isotope analysis of nitrate, filtrate of exponentially growing cultures was collected in acid-washed 30 ml polypropylene bottles and immediately frozen until analysis. Nitrate concentrations of thawed samples were measured by conversion to NO (nitric oxide) followed by chemiluminescence detection (Braman and Hendrix 1989). Nitrite was also measured in this manner, though none was detected in any of the experimental samples (detection limit ~ 0.1 umol L"1 nitrite). In the samples of growth medium and extracted intracellular pools, the l 5 N/ 1 4 N and 1 8 0 / 1 6 0 of nitrate were determined following the denitrifier method (Casciotti et al. 2002; Sigman et al. 2001). Isotope ratios are reported using delta (5) notation in units of per mil (%o): 6'5NSample = ((15N/14N)sample/(15N/14N)reference - 1) X 1000 %o (1) 518Osample = (('W60)samp,e/(180/160)reference - 1) X 1000 %o (2) 42 where the l 5 N/ 1 4 N reference is N 2 in air and the l 8 0 / 1 6 0 reference is Vienna Standard Mean Ocean Water (VSMOW). Referencing to air and VSMOW was through comparison to the international potassium nitrate reference material IAEA-N3, with an assigned 8 1 5N of +4.7%o (Gonfiantini et al. 1995) and reported 8 I 8 0 of+22.7 to +25.6%o (Bohlke et al. 2003; Lehmann et al. 2003; Revesz et al. 1997; Silva et al. 2000). We adopted a 5 1 8 0 of 22.7%o, but without consequence, as only isotope ratio differences are used in this study. The N and O isotopic ratios represent the mean of any replicate measurements; the N and O isotopic ratio measurements of roughly 50% of the samples were duplicated within a day's batch of analyses, and individual samples were analyzed in 1 to 3 of the individual batches of analysis. Reproducibility of the replicates was generally consistent with previously reported analysis standard deviations of 0.2%o for 5' 5N and 0.5%o for 5 , 8 0. N and O isotopic ratios for individual experiments were fitted to the Rayleigh isotope fractionation model to determine the isotope effect (Mariotti et al. 1981). Nitrate N and O isotopic measurements were modelled according to the following Rayleigh linearization: 8' 5N (or 5 1 8 0 Wan. = 8 1 5N (or 5 1 8 0 ) i n i t i a, - e(ln/) (3) where nitrate is the reactant and/is the fraction of the initial nitrate pool that remains. In some experiments, e was also determined from the 1 5 N/ 1 4 N of accumulated particulate N in the culture bottles, using the Rayleigh integrated product equation (Mariotti et al. 1981) 8 1 5 N i n t e g r a ted = o 1 5 N i n i t i a , + e(F) (4) where F = In/ xfl(\-f). We refer to the N and O isotope effects derived from nitrate (and Equation 3) as 15e and , 8e, whereas we refer to the N isotope effect derived from particulate nitrogen (and Equation 4) as l5eparticuiate-Regression analyses of the linear models were computed with the statistical software Systat™ to ascertain the significance of the observed linear trends (p < 0.05) and to obtain an estimate of the error associated with respective slopes (e ± standard 43 error). Significant differences between (nitrate-derived) l 5e and 1 8e, as well as between l 5e and 15eParticuiate within individual experiments were uncovered with a Student's f-test for slopes {p < 0.05). 2.3. Results Derivation of the isotope effect, e, based on the Rayleigh model fit of N and O isotopic measurements is illustrated in Figure 2.1 for T. pseudonana grown in pFe 19 medium. During exponential growth, nitrate was depleted from the medium (Figure 2. la), and the 5 1 5 N of the reactant pool (nitrate) and of the integrated product pool (particulate nitrogen) increased accordingly. Also shown in Figure 2.1a is the concomitant increase in the S 1 8 0 of nitrate. The 5 1 5N and 5 1 8 0 of the reactant pool (i.e., nitrate) are plotted over the natural logarithm of the fraction of nitrate consumed (ln f) in Figure 2.1b. The slopes derived from the resultant linear relationship represent the respective N and O isotope effects, 15e and l 8e (Equation 3); it should be noted that they are not significantly different (Table 2.1). Figure lb also illustrates the derivation of 15e from the slope of the linear relationship between the 5 1 5N of the particulate nitrogen (i.e., the integrated product) and F (Equation 4). In this experiment, the isotope effect computed from the reactant pool ( l 5e = 7.1 ± 0.4 %o) and that from the integrated product (15eParticuiate= 5.8 ± 1.0 %o) differ significantly (Table 2.1), though this conclusion is cautionary as it is based on a relatively small sample size for the particulate measurements (i.e., n = 4; Table 2.1). The kinetic isotope effects computed for all experiments are reported in Table 1. Unless flagged (*), the isotope effects reported in Table 2.1 are the coefficients of statistically significant regression analyses (p < 0.05). Strikingly apparent from the data in Table 1 is the large variation in nitrate-derived 1 5e, both among strains and within strains subjected to different experimental treatments. 15e for T. pseudonana was in the lower portion of the observed range, around 7%o to 8%o for the two experiments conducted with this species. The highest values of 15e were observed for cultures of E. huxleyi, at 20%o in two experiments. I 5e values for T. oceanica ranged between 9%o to 13%o for three experiments; a set of experiments intended to test for the impact of iron on the isotope effect showed no obvious trend. 44 The majority of the experiments reported in Table 2.1 were conducted with T. weissflogii. Nitrate-derived I 5 E ranged from 6%o to 17%o. The lower values are associated with iron-limited cultures, however not all iron-limited T. weissflogii cultures showed reduced 1 5e, such that the impact of low iron conditions on the isotope effect (if any) remains unclear. The highest 1 5 E measured for T. weissflogii (ca. 17%o) correspond to short-term nitrate uptake experiments at pFe20. T. weissflogii in stirred cultures showed no apparent differences in 15e across iron treatments compared to unstirred cultures, in spite of markedly higher growth rates in the stirred cultures at respective iron concentrations (Table 2.1). The kinetic isotope effects derived from measurements of particulate N 8 1 5N corresponded only modestly well to those derived from the 5 1 5 N of nitrate (Table 2.1). Significant differences were observed for the two experiments conducted with T. pseudonana (e.g., as in Figure 2.1), but the sense of the differences were not internally consistent, as the nitrate-derived 15e was slightly higher than 15eparticuiate in one case but lower in the other. A large difference between nitrate-derived 15e and '^ particulate was observed in one experiment with T. oceanica grown in pFe 21 medium, where 15e was 13.4%o while '^ particulate was 20.6%o. However, this discrepancy was not replicated in a similar experiment with T. oceanica (Table 2.1). We would not wish to emphasize the above discrepancies, as estimation of '^ particulate was not a priority of this study, so our sampling was incomplete. Like nitrogen, the oxygen atoms of nitrate underwent isotopic fractionation during nitrate assimilation, with an increase in 5 1 8 0 during nitrate assimilation in all experiments (Figure 2.1a, Figure 2.2). Most striking is the observed coupling between the relative change in nitrate 5 1 5Nand that in 5 1 8 0. As illustrated in Figure 2.2b, the linear regression of5 1 5 N vs. 8 1 8Oforthe sum of the points in all of the experiments conducted indicates that the increase in nitrate 5 1 5 N throughout growth was matched by an equivalent increase in the 5 1 8 0 of nitrate (slope = 1.0 ± 0.01; p < 0.01). 1 5 E and 1 8 E were thus nearly similar regardless of algal species, the magnitude of the isotope effect, or culture conditions (Table 2.1). This 1 5 E : 1 8 E of ~1 is indistinguishable from the 1 5 E : I 8 E inferred from the nitrate 5 l 5Nand 8 1 8 0 increase into the surface layer at station PAPA in the Gulf of Alaska (Casciotti et al. 2002). 45 The 5 1 5 N and 5 1 8 0 were also coupled in the intracellular nitrate pool of T. weissflogii (Figure 2.3). The 5 1 5N of intracellular nitrate was elevated relative to medium nitrate (Figure 2.3a; Needoba et al. 2004). Taking the isotope data for intracellular and extracellular (medium) nitrate together, there is a strong correlation between the 6 1 5N and 5 1 80 of nitrate, with linear regression yielding a slope close to 1 (0.93; Figure 2.3a). The isotopic ratios plotted in Figure 3a are not normalized to the 5 1 5N and 5 1 8 0 of the initial nitrate stock, which can vary slightly among experiments. More importantly, part of the correlation in Figure 2.3a is driven by the result described above that medium nitrate tends to increase in 6 , 5 N and 8 , 8 0 by a ~1:1 ratio as consumption proceeds. To focus on the relationship between internal pool and medium nitrate, the difference between the 5 , 5 N and 5 l gO of internal nitrate versus external (medium) nitrate for each sampling of a culture is plotted in Figure 2.3b, with linear regression again yielding a slope close to unity (0.95 ± 0.6). 2.4. Discussion 2.4.1. The mechanism of isotope fractionation during nitrate assimilation As in higher plants, marine unicellular algae import nitrate with active transporters on the cell surface and reduce it to nitrite via an assimilatory nitrate reductase, an enzyme that resides in the cytoplasm (reviewed by Tischner 2000). The nitrite is subsequently reduced to ammonium by assimilatory nitrite reductase, which is located in the chloroplasts. The ammonium then undergoes incorporation into amino acids (Figure 2.4). Different investigators have proposed different causes for isotope fractionation during nitrate assimilation. Wada and Hattori (1978) first argued that N isotope fractionation by phytoplankton occurs during nitrate reduction, while Montoya and McCarthy (1995) ultimately favoured fractionation associated with nitrate transport into the cell (Figure 2.4). Understanding isotope fractionation during nitrate assimilation requires an understanding of nitrate assimilation itself, or at least aspects of it. To maximize the activity of nitrate reductase, algal strains reportedly accumulate millimolar concentrations of nitrate in the cell (Dortch 1984). Moreover, nitrate reductase 46 activity of light-limited phytoplankton is highly correlated to net nitrate assimilation among different algal groups, which suggests that nitrate reduction is the rate limiting step in nitrate assimilation (Berges and Harrison 1995), as seen in higher plants (Tishner 2000 and references therein). Nitrite, the product of nitrate reduction, generally does not accumulate within cells (Dortch et al. 1984), indicating that the subsequent step of nitrite reduction is unlikely to limit assimilation. Thus, we expect that nitrate reduction represents the last possible step for the origin of isotope fractionation during nitrate assimilation. The focus of Wada and Hattori (1978) on nitrate reduction as the driver of fractionation during assimilation was motivated in part by the expectation that transport cannot impart a significant isotope effect because it does not involve bond breakage. While chemical interactions must occur in the transporter, simple theory would suggest that this type of interaction is generally too weak to be important in stable isotope fractionation (Melander and Saunders 1980). Similarly, the slower diffusion of 1 5 N - or 1 8 0- bearing nitrate is also expected to be trivial (< l%o) because of solvation by and the molecular motion of water (Hammond and Prokopenko, in press; O'leary 1984). This view concurs qualitatively with work done on higher plants (Mariotti et al. 1982), cyanobacteria (Shearer et al. 1991), and diatoms (Wada and Hattori 1975). Moreover, the lack of isotope fractionation during assimilation of nitrite by marine phytoplankton (Waser et al. 1998b) also supports the view that transport is a non-fractionating process. Since nitrite does not accumulate intracellularly, only uptake of nitrite at the cell surface has the potential to fractionate; in the above-cited cases, no isotope effect is observed. The enzyme nitrate reductase appears to impart a sizeable N isotope fractionation, with several studies suggesting an isotope effect of 15-30%o (Ledgard et al. 1985; Schmidt and Medina 1991), making nitrate reductase a reasonable candidate for driving the isotope effect associated with nitrate assimilation. However, this mechanism for fractionation requires a significant nitrate efflux from the cells in order to propagate the isotope effect extracellularly (Figure 2.4). Given that energy must be spent to transport nitrate into the cell and that nitrate limits phytoplankton growth in large regions of the ocean, some investigators have doubted that the cell could be so open as to allow much expression of the nitrate reduction isotope effect. This has provided motivation for continued attention to transport as a possible source of nitrate isotope fractionation 47 (Montoya and McCarthy 1995). However, nitrate efflux is routinely observed in higher plants, where it is believed to be dependent on internal nitrate concentrations (Ter Steege etal. 1999). The isotopic fractionation mechanism for nitrate assimilation has remained equivocal in part because of the difficulty inherent in quantifying the isotopic composition of internal pools. Recently, Needoba et al. (2004) measured the concentration and &1 5N of intracellular nitrate of T. weissflogii grown under various environmental conditions. As reported in previous studies, the authors found intracellular nitrate pools to be highly concentrated relative to the external media. More importantly, the intracellular nitrate was highly enriched in 1 5 N relative to extracellular (medium) nitrate. As discussed by the authors, these results indicate that the isotope fractionation of nitrate reduction inside the cell is greater than that of nitrate uptake into the cell; if the rates of net uptake and reduction are equal (i.e., the internal nitrate pool is at steady state), then this implies that the isotope effect of reduction is greater than that of uptake. The authors thence argued that observed N isotope effects were due in large part to N isotope fractionation imparted by nitrate reductase and were manifested extracellularly via nitrate efflux. Interestingly, intra-species isotope effects were seemingly related to intracellular nitrate concentrations (Needoba et al. 2004), suggesting that, as in higher plants, putative nitrate efflux may be dependent on intracellular nitrate concentrations. Moreover, the N isotope effects measured by Needoba et al. (2004) were also directly proportional to the difference in nitrate 5 I 5 N between internal and external pools, strongly incriminating nitrate efflux as the cause of variation in the isotope effect of nitrate uptake by T. weissflogii. In this conceptual model, the degree to which the isotope effect of nitrate reductase is expressed varies with the rate at which the 15N-enriched internal nitrate is effluxed from the cell relative to the rate at which it is reduced within the cell (Shearer et al. 1991). However, it has not yet been explicitly addressed whether the nitrate efflux from the cell is sufficiently high to explain the assimilation isotope effect as the result of nitrate reductase. Hypothetically, if nitrate efflux did not occur, the isotope effect imparted on nitrate by the reductase would not propagate out of the cell, and all of the intracellular nitrate would eventually be consumed. The 1 5 N enrichment of external nitrate during nitrate assimilation would thence be due solely to uptake at the cell surface. 48 Alternatively, while the internal pool 1 5 N data of Needoba et al. (2004) clearly show that the nitrate reductase isotope effect is greater than that of nitrate uptake into the cell, they do not rule out the possibility of a significant isotope effect for nitrate uptake. The coupled N and O isotope data presented here suggest an isotope effect driven solely by the reductase, and not by uptake at the cell surface. The l 5 N and 1 8 0 enrichment of external (medium) nitrate conform to a ~1:1 trend for a broad range of N isotope effects (Figure 2.2b). To the degree that different processes might impart different N:0 ratios of isotopic fractionation, this suggests that a single mechanism is driving the 1 5 N and 1 8 0 enrichment of external nitrate under all conditions and in all phytoplankton strains. Moreover, the N:0 ratio of the heavy isotope enrichment in the intracellular nitrate relative to the medium is very close to 1:1 (Figure 2.3), the same as observed in the temporal variations of the external nitrate. One could posit that fractionation is occurring at both the uptake and reduction steps, only that the isotope effect of nitrate reduction is greater (Needoba et al. 2004). However, for both the intercellular-extracellular isotope difference and the progressive external nitrate isotope enrichment to occur with an 1 5 N/ I 4 N: 1 8 0 / 1 6 0 ratio of 1:1, both uptake and reduction of nitrate would need to have an 1 5 E : 1 8 E ratio of nearly 1:1. We expect that 15e:18£ differs among various processes and even depends on their details. For instance, if molecular diffusion is inversely proportional to the square root of the molecular mass, then the 1 5 E : 1 8 E for nitrate diffusion would be ~ 0.5. A calculation based on Transition State Theory of 1 5 E : I 8 E for nitrate thermic N-0 bond breakage (without any catalysis) yields 1 5e:1 8e ~ 2 (Table 2.2; Appendix 1). In contrast, and quite remarkably, ab-initio calculations using the NAP dissimilatory nitrate reductase from Desulfovibrio desulfuricans (Dias et al. 1999) as a model of the reaction centre indicate a 15e:18e of ~1 (Zaharahiev, F. E. Pers. comm.). In any case, there is no reason to expect that several components of nitrate assimilation will all have an 1 5 E : 1 8 E of ~1. We do not know of measurements that provide insight on the expected 1 5e:1 8e for binding and release of nitrate in a transporter, but our prediction is that the I 5 E : 1 8 E will be low for this process because a modest intermolecular interaction such as this is more likely to depend on the isotope of the O atoms, which are more readily available to interaction with the transporter. In general, the observation of I 5 E : 1 8 E ~ 1 for a wide range in I 5 E suggests that only one process drives the fractionation for this entire range. This fits well with the 49 model posed above, where the fractionation of nitrate N and O isotopes is driven entirely by nitrate reduction and the amplitude of the assimilation isotope effect is modulated by the degree to which intracellular nitrate is able to efflux from the cell before being reduced. The 1 5e:1 8e of -1 that applies at 15e of 20%o also applies at l 5e of 10%o and apparently even lower, arguing for the same fractionation mechanism across this range, such that transport is not a major contributor to fractionation even when the latter is slight. It is worth noting that nitrate reductase is also found embedded in the plasma membrane of plant and algal cells (Tischner 2000). Plasmalemma nitrate reductase has been hypothesized to participate in nitrate uptake, coupling transport across the membrane with nitrate reduction. In such a case, reduction could drive the isotope fractionation without requiring the efflux of nitrate from the cell. However, the large accumulations of nitrate in the intracellular pool suggest that a sizeable amount of the nitrate reduction is occurring within the cytoplasm, arguing against a dominant role for this external nitrate reductase in the isotope effect of nitrate assimilation. Moreover, the putative role of plasmalemma-bound nitrate reductase in nitrate transport has been challenged (Mora et al. 2002) and remains ambiguous. 2.4.2. The magnitude of the N isotope effect during nitrate assimilation We have argued above that nitrate reductase comprises the chief fractionating step in nitrate assimilation, with its expression occurring by the efflux of intracellular nitrate back into the environment. This suggests that the amplitude and variability of the isotope effect will largely be determined by the rate of this efflux relative to the rate of nitrate reduction (the 'efflux/reduction ratio' to which we refer below). One might infer from the low amplitude of l 5e in the open ocean relative to cultures (and its apparently lower degree of variability) that nitrate efflux tends be constant and minimal in the open ocean. However, we do not understand the controls on the nitrate efflux/reduction ratio and thus cannot predict l 5e, even in our controlled cultures. Apparent from our observations and those of other studies (Montoya and McCarthy 1995b; Needoba 2003; Wada and Hattori 1978; Waser et al. 1998a) is a seeming lack of species or algal group specificity relating to 1 5e. While some 15e values obtained in this study for T. pseudonana 50 and T. weissflogii agree with some previous reports (Montoya and McCarthy 1995b; Waser et al. 1998a), the isotope effect of 20%o observed here foris. huxleyi is excessively high compared to previous estimates of 4%o (Needoba 2003; Waser et al. 1998b), though culture conditions were roughly similar. This study presents the first published values of e for T. oceanica. These fell in the range observed here for T. weissflogii While large intra-specific variations in 15e point strongly to an effect of growth conditions, our study does not provide a clear picture of this effect. Our attempt to define a role for iron-limitation in modulating l 5e yielded no conclusive trend. Though iron-limitation appeared to cause a decrease in 15e in some experiments with T. weissflogii, not all experiments concurred (with T. weissflogii or with T. oceanica). Similarly, Needoba et al. (2004) reported no significant changes in isotope effect of T. weissflogii grown in iron-deplete medium. The lower 15e observed in some of the low-iron experiments may be due to a decrease in nitrate uptake rates and lower intracellular nitrate concentrations due to iron stress, resulting in lower nitrate efflux. Shaking of the cultures caused a marked increase in the growth rates of T. weissflogii compared to analogous treatments in unstirred cultures, but no concomitant changes in isotope effects were observed. Though net nitrate uptake was significantly higher in the stirred cultures, the lack of a clear difference in l 5e implies that the efflux/reduction ratio was similar in shaken and unstirred cultures. We observed a higher l 5e for nitrate assimilation by T. weissflogii cells that were resuspended in fresh medium compared to cell cultures initiated from small inocula. We did not expect this result, reasoning that exponential phase cells would be at a physiological steady-state with respect to nitrate assimilation, such that resuspension into fresh medium would cause no change in the isotope effect. It is possible that the cells were experiencing N-limitation in late exponential growth prior to resuspension, or that filtration and resuspension affected the integrity or permeability of the cell membrane. Either of these factors might have increased the nitrate efflux/reduction ratio: previous N-limitation could result in over-expression of nitrate uptake and high intracellular nitrate concentrations following resuspension in nitrate-bearing medium, and hence a higher efflux/reduction ratio, while filtration could also contribute to higher efflux if membrane permeability were affected. 51 Cell geometry, namely cell surface area to volume ratio, is one of a few major controls on carbon isotopic fractionation during inorganic carbon assimilation (Popp et al. 1998). By analogy, a higher surface area to volume ratio could potentially lead to higher cellular efflux of nitrate for a given rate of nitrate reduction. No systematic relationship between surface area to volume ratio and N isotopic fractionation was apparent in our study, whether among the algal species in our experiments or for a set of experiments run with a single species (data not shown). Our cell surface area and volume estimates were inferred from Coulter counter measurements of cell diameter and with consideration of cell shape in only a rudimentary way, so we cannot rule out that future studies will reveal such a relationship. Generally, though, we expect that the isotope effect of nitrate assimilation is the result of multiple additional factors (e.g., vacuolar space, nitrate transporter density, and nitrate reductase concentration) that are regulated by the cells on the basis of resource availability and overall cellular energetics (Needoba and Harrison 2004). As with the isotope dynamics of carbon assimilation, much work is required to develop a fully mechanistic understanding. 2.4.3. Summary and concluding remarks The culture experiments described in this study indicate that phytoplankton fractionate the N and O isotopes of nitrate with a 1 5e:1 8e ratio of ~1, regardless of the absolute isotope effect. A similar 1:1 increase in 5 1 5N and 8 I 8 0 has been observed in surface ocean nitrate that has undergone nitrate assimilation by resident plankton (Casciotti et al. 2002). Isotopic analyses of intracellular nitrate indicate a ~1:1 N:0 isotope enrichment of this nitrate relative to external nitrate. The internal pool observations suggest that assimilatory nitrate reductase has an intrinsic 1 5e:1 8e of around 1. Moreover, the consistency of the internal pool and medium time-course results suggest that a single fractionating mechanism is at work in all studied phytoplankton strains and under all studied growth conditions, regardless of the amplitude of 15e that is expressed. If there were more than two significant fractionating mechanisms, then both would need to have an 1 5e:1 8e of 1, and theoretical calculations of 1 5e:1 8e for various processes indicate that this is highly unlikely (Table 2.2). Together, these data strongly support the scenario that nitrate reductase is the dominant source of isotope fractionation in nitrate 52 assimilation by phytoplankton, even at low 15e (~6%o). If this is correct, then the range of isotope effects observed for nitrate assimilation by phytoplankton cultures is driven by variations in the degree to which nitrate efflux across the cell boundary enables full expression of the large isotope effect intrinsic to nitrate reductase (~15-30%o; Ledgard et al. 1985, Schmidt and Medina 1991). It is noteworthy that this model is analogous to that of isotope fractionation for photosynthetic carbon dioxide fixation by C3 plants (O'leary 1981). Another process likely to impact the N and O isotopic signatures of nitrate in the ocean is denitrification. Though N and O isotopic coupling has not been investigated for laboratory cultures of denitrifying bacteria, nitrate N and O isotopic measurements in denitrifying zones suggest that the 1 5e:1 8e of denitrification is also indistinguishable from 1 (Sigman et al. 2003). One might infer that, as with nitrate assimilation, the isotope effect observed in external nitrate is due to the isotope effect intrinsic to nitrate reduction, with transport of the 1 5 N- and 180-enriched intracellular nitrate back into the environment. Water column data from various regions yield a remarkably consistent estimate of 25-30%o for the isotope effect for denitrification (Brandes et al. 1998; Cline and Kaplan 1975; Liu and Kaplan 1989; Sigman et al. 2003). This value is close to the maximum value from denitrifier cultures, 25-30%o (e.g., Mariotti et al. 1981). One possible explanation for this similarity is that denitrifiers universally allow rapid efflux of nitrate into the environment, so that the isotope effect intrinsic to nitrate reductase is always fully expressed. Such "open system" behaviour for intracellular nitrate is sensible for marine denitrifiers (as opposed to phytoplankton) in that ambient [NO3'] is typically quite high in zones of water column denitrification, weakening the competitive advantage associated with nitrate storage. Though multiple forms of nitrate reductase catalyze the reduction of nitrate to nitrite, and assimilatory and dissimilatory forms differ, all share a molybdenum reaction centre and likely yield similar transition-state structures of Mo-bound nitrate at the active site. It follows that the N and O isotope effects of all nitrate reductases, whether assimilatory or dissimilatory, seem likely to show similar fractionation characteristics (see Melander and Saunders 1960). In this sense, the concordance of the available data with a 1 5e:1 8e of ~1 for both nitrate assimilation and denitrification is consistent with the view that nitrate reductase is the driver of fractionation in both processes. 53 However, in contrast with oceanic results, measurements of N and O isotopes of nitrate in freshwater systems suggest a l 5e: 1 8e of 1.4 to 2 for denitrification (Lehmann et al. 2003). This discrepancy is remarkable and difficult to explain, given the data in hand. We are presently working to characterize the 15e:18e of denitrification in bacteria isolated from marine and terrestrial environments. The use of coupled N and O isotopic measurements of nitrate represents a potentially powerful tool for studying the oceanic N cycle. Nitrate , 5 N/ 1 4 N shares a limitation with other geochemical tools used to the study the N cycle (e.g., nitrate concentration and nitrate-to-phosphate ratio) in that it records regional imbalances in the counteracting fluxes of the N cycle but does not illuminate the gross rates of these fluxes. Notably, in the case of nitrate assimilation, neither the N isotopes nor the other available tracers are able to separate nitrate assimilation from the organic matter remineralization and nitrification that recycles the assimilated N back into nitrate. Similarly, the N isotope signals of nitrogen fixation and denitrification interfere destructively, removing both signals. Separating the impacts of these processes on the nitrate concentration of seawater, both at the surface and in the ocean interior, would greatly advance our understanding of ocean biogeochemistry. The use of the coupled N ad O isotopes of nitrate allows for discrimination between the processes that otherwise overprint one another (Lehmann et al. 2004; Sigman et al. 2003), thereby providing a new and important tool in study of the N cycle. Toward this end, the data reported here provide a fundamental constraint on the behaviour of N and O isotopes of nitrate. 54 Table 2.1. Isotopic fractionation of nitrate N and O and particulate N during nitrate assimilation by marine phytoplankton grown in batch culture under various conditions. Growth N0 3 NOa 15e ,8e 15e rate initial final nitrate nitrate §n particulate Strain Culture pFe (d-1) (umol L"') (umol L"1) (±SE) (±SE) (±SE) n E. huxleyi unstirred 19 1.1 154 83 19.7 ±0.5 19.7 ±0.5 6 n.d. E. huxleyi unstirred 19 1.2 101 22 20.4 ± 0.6 21.0 ±0.5 5 J21.6 ± 4.5 3 T. oceanica unstirred 19 1.2 57 33 10.5 ±0.1 10.4 ±0.3 5 n.d. T. oceanica unstirred 21 0.9 51 4 8.9 ±0.5 7.8 ±0.3 5 *9.0± 1.2 3 T. oceanica unstirred 21 0.7 156 103 13.4 ± 1.8 12.6 ±0.9 7 ^ ^ ± 3 . 1 6 T. pseudonana unstirred 19 1.4 146 63 6.7 ± 0.2 5.8 ±0.8 7 f 9.9± 1.8 6 T. pseudonana unstirred 19 1.4 103 25 7.1 ±0.4 7.5 ±0.6 7 f 5.8± 1.0 4 T. weissflogii unstirred 19 1.4 70 44 13.0 ±0.2 14.3 ±0.2 4 n.d. T. weissflogii unstirred 19 1.4 68 39 12.2 ±0.4 10.8 ±0.7 4 n.d. T. weissflogii unstirred 20 0.9 76 56 9.1 ±0.4 10.3 ± 1.5 4 n.d. T. weissflogii unstirred 20 1.0 75 41 10.4 ±0.4 11.1 ±0.5 4 n.d. T. weissflogii unstirred 20.5 0.9 50 13 9.2 ± 0.2 10.0 ±0.4 5 { 10.5±4.3 4 T. weissflogii unstirred 20.5 0.6 114 27 10.5 ±0.2 10.7 ±0.3 5 n.d. T. weissflogii unstirred 20.5 0.9 86 *1 11.6 ±0.1 10.7 ±0.1 5 10.8 ± 1.4 5 T. weissflogii unstirred 20.5 0.9 83 *4 11.4 ±0.4 11.5 ±0.6 4 12.8 ± 1.5 4 T. weissflogii stirred 19 1.7 84 25 12.8 ±0.4 14.2 ±0.4 9 n.d. T. weissflogii stirred 19 1.6 85 17 10.8 ±0.9 12.1 ± 1.4 9 n.d. T. weissflogii stirred 20 1.4 74 31 13.2 ±0.7 13.1 ±0.6 8 n.d. T. weissflogii stirred 20 1.5 74 20 13.5 ±0.3 13.7 ±0.4 8 n.d. T. weissflogii stirred 21 0.7 80 65 9.2 ±0.8 6.1 ± 1.0 9 n.d. T. weissflogii stirred 21 0.6 77 65 5.6 ± 1.5 5.1 ± 1.5 9 n.d. T. weissflogii uptake 19 - 72 15 14.9 ±0.2 14.2 ±0.1 6 n.d. T. weissflogii uptake 20 - 75 14 16.3 ±0.2 17.4 ±0.3 7 n.d. T. weissflogii uptake 20 - 78 16 16.9 ±0.3 18.2 ±0.3 7 n.d. u: growth rate; n.d.: not determined; § n refers to the sample number in the regressions for I 5 E as well as1 8e. lowest [n03"] isotopic values were not considered in the regression analyses; *slope (e) of linear regression is not statistically significant; particulate l 5e is significantly different from nitrate I 5 E . 55 Table 2.2. Computed and empirical vibrational frequencies for the N-O bond of nitrate ( 1 5N-and l80-bearing), and corresponding calculated N and O isotope effects for dissociation of a single O atom from nitrate (hypothetical thermic decomposition). Also listed are measurements of the isotope effects for Fe 2 + reduction of nitrate and reduction by nitrate reductases. e(%«) t e(%o)J E(%O) N0 3 " uKcm"1) Di(cm"') e (%o) computed measured NO3" isotopes computed measured* Fe(II) TJi V i reductase l 4 N 1 6 0 3 " 1362 1376 l 5 N 1 6 0 3 " 1340 1344 , 5e: 57 74 75 15-30 1 4 N , 6 0 2 1 8 0 " 1328 - 18e: 29 ''from Hooke's law, with a vibrational force constant for the N-0 bond of nitrate = 8.2 x IO2 N m"' (Brown and Drury 1967); *by (Begun and Fletcher 1960); Computed as in Brown and Drury (1967) with temperature = 298.15°k;+ Brown and Drury (1967); ¥Ledgard et al. (1985), Schmidtt and Medina (1991). 56 ln([NOj ]/[NOj' initial]) 15 e: e-1.00 "+- T. weissflogii -9- E. huxleyi "9- T. pseudonana -M- T. oceanica I — ' I ' 1 0 ,g 10 20 30 A5 O of N 0 3 (%o vs. starting value) Figure 2.2. Increase in nitrate 6 I 5 N and 5 1 8 0 during nitrate assimilation by three species of diatoms and E. huxleyi grown under various environmental conditions (see Table 1). a) Change in nitrate § ' 5 N relative to initial nitrate 5 I 5 N plotted as a function of ln(f). The slopes of individual experiments define the corresponding N isotope effects on NO3" (1 5e) reported in Table 1. Dashed lines show slopes for 15e of 5 and 25%o. b) The 5 I 5 N vs. 5 I 8 0 change in NO3". The slope of the linear regression and its standard error (1.00 ± 0.01) are inclusive of all the experiments combined. The estimate of the standard error for the slope of the regression describes the error computed from the least-squares regression of sample means and does not take into account the measurement error associated with individual samples. Dashed lines show slopes of 1.1 and 0.9, for comparison. 58 Figure 2.3. Relationship between the N and O isotopic composition of nitrate in intracellular vs. extracellular nitrate pools of T. weissflogii cultures grown under various environmental conditions (see Needoba et al. (in press) for details of experimental treatments), a) Nitrate 6 1 5 N vs. S 1 8 0 for intracellular and extracellular nitrate, b) Differences in nitrate 6 I 5 N and 5 I 8 0 between intracellular and extracellular (medium) nitrate. 59 Figure 2.4. Schematic depiction of nitrate uptake and assimilation by a phytoplankton cell. Hypothetically, isotope fractionation of nitrate could occur a) at the diffusive boundary layer, b) during uptake at the cell surface, c) during potential nitrate efflux from the phytoplankton cell, and/or d) during reduction to nitrite by nitrate reductase. 60 2.5. References Altabet, M. A. 2001. Nitrogen isotopic evidence for micronutrient control of fractional NO3" utilization in the equatorial Pacific. Limnol. Oceanogr. 4 6 : 368-380. Altabet, M. A., and R. Francois. 1994. Sedimentary nitrogen isotopic ratio as a recorder for surface ocean nitrate utilization. Global. Biogeochem. Cy. 8: 103-116. Begun, G. M., and W. H. Fletcher. 1960. Vibrational spectra of isotopic nitric acids. J. Chem. Phys. 3 3 : 1083. Berges, J. A., and P. J. Harrison. 1995. Nitrate reductase-activity quantitatively rredicts the rate of nitrate incorporation under steady-state light limitation - a revised assay and characterization of the enzyme in 3 species of marine-phytoplankton. Limnol. Oceanogr. 4 0 : 82-93. Bohlke, J. K., S. J. Mroczkowski, and T. B. Coplen. 2003. Oxygen isotopes in nitrate: new reference materials for 0-18 : 0-17 : 0-16 measurements and observations on nitrate-water equilibration. Rapid. Commun, Mass. Sp. 1 7 : 1835-1846. Braman, R. S., and S. A. Hendrix. 1989. Nanogram nitrite and nitrate determination in environmental and biological materials by V(III) reduction with chemiluminescence detection. Anal. Chem. 6 1 : 2715-2718. Brandes, J. A., A. H. Devol, T. Yoshinari, D. A. Jayakumar, and S. W. A. Naqvi. 1998. Isotopic composition of nitrate in the central Arabian Sea and eastern tropical North Pacific: a tracer for mixing and nitrogen cycles. Limnol. Oceanogr. 4 3 : 1680-1689. Brown, L. L., and J. S. Drury. 1967. Nitrogen-isotope effects in the reduction of nitrate, nitrite, and hydroxylamine to ammonia. I> In sodium hydroxide solution with Fe(II). J. Chem. Phys. 4 6 : 2833-2837. Casciotti, K. L., D. M. Sigman, M. G. Hastings, J. K. Bohlke, and A. Hilkert. 2002. Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater using the denitrifier method. Anal. Chem. 7 4 : 4905-4912. Cline, J. D., and I. R. Kaplan. 1975. Isotopic fractionation of dissolved nitrate during denitrification in the Eastern Tropical North Pacific Ocean. Mar. Chem. 3 : 271-299. 61 Dias, J. M. and others. 1999. Crystal structure of the first dissimilatory nitrate reductase at 1.9 angstrom solved by MAD methods. Struct. Fold. Des. 7: 65-79. Dortch, Q., J.R. Clayton, S.S. Thorensen, S.I. Ahmed. 1984. Species differences in accumulation of nitrogen pools in phytoplankton. Mar. Biol. 81: 237-250. Farrell, J. W., T. F. Pedersen, S. E. Calvert, and B. Nielsen. 1995. Glacial-Interglacial Changes in Nutrient Utilization in the Equatorial Pacific-Ocean. Nature 377: 514-517. Francois, R., M. A. Altabet, and L. H. Burkle. 1992. Glacial to interglacial changes in surface nitrate utilization in the Indian sector of the Southern Ocean as recorded by sediment d l 5 N. Paleoceanogr. 7: 589-606. Francois, R. F. and others 1997. Water column stratification in the Southern Ocean contributed to the lowering of glacial atmospheric CO2. Nature 389: 929-935. Gonfiantini, R., W. Stichler, and K. Rosanski. 1995. Standards and Intercomparison Materials Distributed by the IAEA for Stable Isotope Measurements. International Atomic Energy Agency. Hammond, D. E., and M. P. G. C. Acta, in press. Differential diffusivity of light stable isotopes in water and seawater: 15N/14N in ammonium. Geochim. Cosmochim. Acta. Keller, M. D., W. K. Bellows, and R. R. L. Guillard. 1988. Microwave treatment for sterilization of phytoplankton culture media. J. Exp. Mar. Biol. Ecol. 117: 279-283. Ledgard, S. F., K. C. Woo, and F. J. Bergersen. 1985. Isotopic fractionation during reduction of nitrate to nitrite by extracts of spinach leaves. Aust. J. Plant Physiol. 12: 631-640. Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2003. The effect of benthic nitrogen cycling on the delta N-15 and delta 0-18 of water-column nitrate. Geochim. Cosmochim. Ac. 67: A249-A249. Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2004. Coupling the N-15/N-14 and 0- 18/0-16 of nitrate as a constraint on benthic nitrogen cycling. Mar. Chem. 88: 1- 20. 62 Liu, K.-K., and I. R. Kaplan. 1989. The eastern tropical Pacific as a source of 15N-enriched nitrate in seawater off southern California. Limnol. Oceanogr. 34: 820-830. Mariotti, A., J. C. Germon, P. Hubert, P. Kaiser, R. Letolle, A. Tardieux, and P. Tardieux. 1981. Experimental determination of nitrogen kinetic isotope fractionation: some principles; illustration for the denitrification and nitrification processes. Plant Soil 62: 413-430. Melander, L., and W. H. J. Saunders. 1980. Reaction Rates of Isotopic Molecules. John Wiley. Montoya, J. P., and J. J. McCarthy. 1995. Isotopic fractionation during nitrate uptake by marine phytoplankton grown in continuous culture. J. Plankton Res. 17: 439-464. Mora, C , F. G. Witt, P. J. Aparicio, and M. A. Quinones. 2002. Independent induction of two blue light-dependent monovalennt anion transport systems in the plasma memebrane of Monoraphidium braunii. J. Exp. Bot. 53: 1909-1918. Needoba, J. A., and P. J. Harrison. 2004. Influence of low light and a light: Dark cycle on N03- uptake, intracellular N03-, and nitrogen isotope fractionation by marine phytoplankton. J. Phycol. 40: 505-516. Needoba, J. A., N. A. Waser, P. J. Harrison, and S. E. Calvert. 2003. Nitrogen isotope fractionation in 12 species of marine phytoplankton during growth on nitrate. Mar. Ecol.-Prog. Ser. 255: 81-91. Needoba, J. A., D. M. Sigman, and P. J. Harrison. 2004. The mechanism of isotope fractionation during algal nitrate assimilation as illuminated by the N-15/N-14 of intracellular nitrate. J. Phycol. 40: 517-522. O'leary, M. H. 1981. Carbon isotope fractionation in plants. Phytochemistry 20: 553-567. O'leary, M. H. 1984. Measurement of the isotope fractionation associated with diffusion of carbon dioxide in aqueous solution. J. Phys. Chem 88: 823-825. Popp, B. N., E. A. Laws, R. R. Bidigare, J. E. Dore, K. L. Hanson, and S. G. Wakeham. 1998. Effect of phytoplankton cell geometry on carbon isotopic fractionation. Geochim. Cosmochim. Acta. 62: 69-77. Price, N. M., G. I. Harrison, J. G. Herring, R. J. Hudson, P. M. V. Nirel, B. Palenik and F. M. M. Morel. 1988/1989. Preparation and chemistry of the artificial algal culture medium Aquil. Biol. Oceanogr. 6: 443-461. 63 Revesz, K., J. K. Bohlke, and T. Yoshinari. 1997. Determination of d l80 and dl5N in nitrate. Anal. Chem. 69: 4375-4380. Schmidt, H. L., and R. Medina. 1991. Possibilities and Scope of the Double Isotope Effect Method in the Elucidation of Mechanisms of Enzyme Catalyzed-Reactions. Isotopenpraxis 27: 1-4. Shearer, G., J. D. Schneider, and D. H. Kohl. 1991. Separating the efflux and influx components of net nitrate uptake by Synechococcus-R2 under steady-state conditions. J. Gen. Microbiol. 137: 1179-1184. Sigman, D. M., M. A. Altabet, R. Francois, D. C. McCorkle, and J.-F. Gaillard. 1999a. The isotopic composition of diatom-bound nitrogen in Southern Ocean sediments. Paleoceanography 14: 118-134. Sigman, D. M., M. A. Altabet, D. C. McCorkle, R. Francois, and G. Fischer. 1999b. The delta N-15 of nitrate in the Southern Ocean: Consumption of nitrate in surface waters. Global. Biogeochem. Cy. 13: 1149-1166. Sigman, D. M., M. A. Altabet, R. H. Michener, D. C. McCorkle, B. Fry, and R. M. Holmes. 1997. Natural abundance-level measurement of the Nitrogen isotopic composition of oceanic nitrate: an adaptation of the ammonia diffusion method. Mar. Chem. 57: 227-242. Sigman, D. M., K. L. Casciotti, M. Andreani, C. Barford, M. Galanter, and J. K. Bohlke. 2001. A bacterial method for the nitrogen isotopic analysis of nitrate in seawater and freshwater. Anal. Chem. 73: 4145-4153. Sigman, D. M., R. Robinson, A. N. Knapp, A. Van Geen, D. C. McCorkle, J. A. Brandes, and R. C. Thunell. 2003. Distinguishing between water column and sedimentary denitrification in the Santa Barbara Basin using the stable isotopes of nitrate. Geochemistry, Geophysics, Geosystems 4: 1040-1059. Silva, S. R., C. Kendall, D. H. Wilkinson, A. C. Ziegler, C. C. Y. Chang, and R. J. Avanzino. 2000. A new method for collection of nitrate from fresh water and the analysis of nitrogen and oxygen isotope ratios. J. Hydrol. 228: 22-36. Ter Steege, M. W., I. Stulen, P. K. Wiersema, F. Posthumus, and W. Vaalburg. 1999. Efficiency of nitrate uptake in spinach: impact of external nitrate concentration and relative growth rate on nitrate influx and efflux. Plant Soil 208: 125-134. 64 Thorensen, S. S., Q. Dortch, and S. I. Ahmed. 1982. Comparison of methods for extracting intracellular pools of inorganic nitrogen from marine phytoplankton. J. Plankton Res. 4: 695-704. Tischner, R. 2000. Nitrate uptake and reduction in higher and lower plants. Plant Cell. Environ. 23: 1005-1024. Wada, E. 1980. Nitrogen isotope fractionation and its significance in biogeochemical processes occurring in marine environments, p. 375-398. In E. Goldberg, Y. Horibe and K. Saruhashi [eds.], Isotope Marine Chemistry. Uchida Rokakuho. Wada, E., and A. Hattori. 1978. Nitrogen isotope effects in the assimilation of inorganic nitrogenous compounds. Geomicrobiol. J. 1: 85-101. Waser, N. A., K. D. Yin, Z. M. Yu, K. Tada, P. J. Harrison, D. H. Turpin, and S. E. Calvert. 1998a. Nitrogen isotope fractionation during nitrate, ammonium and urea uptake by marine diatoms and coccolithophores under various conditions of N availability. Mar. Ecol.-Prog. Ser. 169: 29-41. Waser, N. A. D., D. H. Turpin, P. J. Harrison, B. Nielsen, and S. E. Calvert. 1998b. Nitrogen isotope fractionation during the uptake and assimilation of nitrate, nitrite, and urea by a marine diatom. Limnol. Oceanogr. 43: 215-224. Westall, J. C , J.L. Zachary, and F.M.M. Morel. 1976. MINEQL: a computer program for the calculation of chemical equilibrium composition in aqueous systems. Tech. Note 18. MIT. Wu, J., S. E. Calvert, and C. S. Wong. 1997. Nitrogen isotope variations in the subarctic Pacific northeast Pacific: relationships to nitrate utilization and trophic structure. Deep-Sea Res. Part I 44: 287-314. 65 Chapter 3 A method for nitrite removal in nitrate N and O isotope analyses A version of this chapter has been published. Granger, J., D. M. Sigman, M. G. Prokopenko, M. F. Lehmann, and P. D. Tortell. 2004. A method for nitrite removal in nitrate N and O isotope analyses. Limnol. Oceanogr.: methods 4 : 205-212. 66 3.1. Introduction Current methods to determine nitrate (NO3") nitrogen (N) and oxygen (O) isotope ratios ( 1 5N/ 1 4N and 1 8 0/ 1 6 0) in aqueous samples do not distinguish between the isotopic composition of nitrate vs. that of nitrite in a given sample. In the case of the ammonia distillation (Cline and Kaplan 1975) and ammonia diffusion (Sigman et al. 1997) methods for nitrate N isotope analysis, both nitrate and nitrite are converted to ammonia, then the N isotopic composition of extracted ammonia is measured. Similarly, the denitrifier method for coupled nitrate N and O isotopic analysis (Casciotti et al. 2002; Sigman et al. 2001) does not distinguish between the respective signals imparted by nitrate and nitrite, as denitrifying cells convert both nitrate and nitrite to the nitrous oxide (N2O) gas analyte. Finally, anion exchange and pyrolysis of samples for N and O isotope analysis of nitrate in freshwater samples (Amberger and Schmidt 1987; Revesz et al. 1997; Silva et al. 2000) also includes nitrite in a sample. Separation of nitrate from nitrite is particularly important when measuring the O isotopes with the denitrifier method, as the 5 1 8 0 of the N2O product depends on whether nitrate or nitrite is the original substrate (Casciotti et al. 2002). The use of azide to quantitatively convert nitrite to N 2 0 allows for the analysis of nitrite without interference from nitrate, and this method (when combined with nitrate reduction using spongy cadmium) also allows for isotopic analysis of both nitrate and nitrite (Mcllvin and Altabet 2005). Thus, the nitrate isotopes could be derived by difference using this method, but the error associated with this approach will be great when nitrite occurs at high concentration. Nitrate samples collected in the ocean are generally devoid of nitrite, such that interference from nitrite is of no concern. Notable exceptions include oxygen minimum zones, where nitrite accumulates at mid-depths due to denitrification. Concentrations of nitrite in regions such as the Arabian Sea, the Eastern Tropical North Pacific and the Peru Upwelling can be upwards of 13 pmol L"1, while nitrate concentrations in these waters are in the range of 10 to 30 pmol L"1. Nitrite also typically accumulates at the bottom of the ocean's euphotic zone from oxidation of regenerated ammonia (Ward 1987; Ward et al. 1989). Eutrophic freshwater systems characterized by high biological oxygen demand can also have significant concentrations of coincident nitrate and nitrite, as do sediment 67 pore waters. Other instances where nitrite interferes with measurements of nitrate isotopic composition include cultures of denitrifying bacteria, where nitrite often accumulates in large amounts as nitrate is being consumed. Accurate estimates of N and O isotopic fractionation imparted on nitrate by denitrification therefore require that interference from nitrite be eliminated. A number of published methods allow for the removal of nitrite without affecting nitrate concentrations. Binding of nitrite to sulfamic acid has been used previously to measure N isotope ratios of nitrate (Wu et al. 1997). However, sulfamic acid is a potent antibiotic and thus incompatible with the denitrifier method. Binding of nitrite to iodide (Garside 1982), a method commonly used to detect low concentrations of nitrite, was also deemed problematic to measure nitrate isotopes, in part because high concentrations of iodide are likely to be toxic to bacteria, and also because the method requires very low pH, which can lead to oxygen atom exchange between nitrate and water (Bunton et al. 1952). We initially used hydroxylamine to remove nitrite from our samples, the product being N 2 0 (Bothner-By and Friedman 1952). Though this reagent is non-toxic, any hydroxylamine remaining in solution competed with denitrifiers for the nitrite being produced from the sample nitrate, disturbing the isotopic relationships between a nitrate sample and its product N 2 0 . Ascorbic acid ultimately proved to be effective in the removal of nitrite without interfering with any aspect of the denitrifier method for nitrate isotope analysis. We believe that it will prove useful in many analytical applications requiring the removal of nitrite. We have developed an ascorbate-based method to remove nitrite from both freshwater and seawater samples which imparts no change in the concentration or the N and O isotopic composition of coincident nitrate. The procedure relies on the capacity of ascorbate to reduce nitrite to nitric oxide (NO) gas in mildly acidic solution at room temperature. The method is non-toxic to the bacteria used in the denitrifier method of Sigman et al. (2001) and Casciotti et al. (2002) and is cost-effective. The method presented here allows for precise and accurate quantification of nitrate N and O isotopic ratios in aqueous samples where nitrite concentrations are significant. 68 3.2. Materials and Procedures Reduction of nitrous acid by ascorbate The pKai of ascorbic acid (AscH2) is 4.2, around which pH the concentration of nitrous acid (HNO2) becomes sufficiently high to engender its spontaneous reduction by ascorbate (AscH'), generating NO (nitric oxide) gas (Kanda and Taira 2003). A s c H 2 ~ AscH" + H + (pKa, 4.2) (1) H N 0 2 ~ N 0 2 " + H +(pKa3.3) (2) AscH" + 2HN0 2 -» DHA (dehydroxyascorbic acid)+ 2N0" + 2H 2 0 (3) While ascorbate readily reduces nitrous acid, it does not reduce nitrate. Thus, nitrite can be selectively removed from aqueous samples with an ascorbic acid addition sufficient to bring the solution down to a pH of around 3.5. However, diatomic oxygen can readily react with the NO free radical and oxidize it to nitrogen dioxide (N0 2 ; Equation 4). The latter can then react with the ascorbate anion to form nitrite (Equation 5) or can react with water to form both nitrite and nitrate (Equation 6): 2 NO' + 0 2 2 N0 2 - (4) N0 2 " + AscH" -> N0 2 " + Asc- + H + (5) 2 N0 2 - + H 2 0 -» HNO3 + HN0 2 (6) To avoid these reactions, nitric oxide gas is removed continually throughout the reaction by bubbling the solution with an inert gas, maintaining a low oxygen 69 concentration in the sample and sweeping away product NO before it react with any oxygen that is present. 3.2.2. Methodology Nitrate and nitrite concentrations were measured by conversion to NO followed by chemiluminescence detection (Braman and Hendrix 1989) on an Antek 1750 nitrate/nitrite analyzer. Our limit of detection was < 20 nmol L"1 nitrate or nitrite. Nitrite was also quantified colourimetrically by reaction with Greiss reagents (sulfanilamide and NNED: N-l-naphthyleneethylenediamine) and measurement of absorbance at 543 nm. Prior to use, all glassware and plastic-ware were acid washed (10% HCI) and rinsed with milli-Q water (Millipore), or furnace-combusted at 400°C for 2 hours in the case of glassware. The culture medium samples analyzed in this study consisted of synthetic ocean water (Price et al. 1988/89) made from milli-Q water. This mixture did not contain any detectable nitrate or nitrite. Freshwater samples refer to milli-Q water that was not amended with salts. A culture of the denitrifying bacterium Pseudomonas aureofaciens (ATCC 13985 recently reclassified as a strain of P. chlor or aphis), was grown to test nitrite removal by the present method. The strain was grown in artificial seawater medium with -260 pmol L"1 initial nitrate according to the procedure outlined previously (Granger and Ward 2003). The 1 5 N/ 1 4 N and 1 8 0 / l 6 0 of nitrate and/or nitrite were determined following the denitrifier method (Casciotti et al. 2002; Sigman et al. 2001). Isotope ratios are reported using delta (8) notation in units of per mil (%o): 815NSample = [ ( 1 5 N/ , 4 N ) s a m p ,e / ( 1 5 N/ , 4 N ) r e f e r e nc e - 1] X 1000 (9) 8180samp le = [ ( ' 'O / '^sapnpie / ^O/ 'preference - 1] X 1000 (10) where the 1 5 N/ I 4 N reference is N 2 in air and the 1 8 0 / 1 6 0 reference is Vienna standard mean ocean water (VSMOW). Referencing to air and VSMOW was through comparison 70 to the international potassium nitrate reference material IAEA-N3, with an assigned 5 1 5N of+4.7%o (Gonfiantini et al. 1995) and reported 5 I 8 0 of+22.7 to +25.6%o (Bohlke et al. 1997; Bohlke et al. 2003; Revesz et al. 1997; Silva et al. 2000). We adopted a 6 1 8 0 of 25.6%o (Bohlke et al. 2003), but without consequence, since only isotope ratio differences are used in this study. Unless indicated otherwise, the N and O isotopic ratios represent the mean of any replicate measurements; the N and O isotopic ratio measurements of roughly 10% of the samples were duplicated. N and O isotope ratios of nitrate measured for a growing P. aureofaciens culture were fitted to the Rayleigh isotope fractionation model to determine the isotope effect, e (Mariotti et al. 1981). Nitrate N and O isotopic measurements were modeled according to the following approximate Rayleigh linearization: 5 1 5N (or S 1 8 0 ) r e acant = S , 5 N (or 5 1 8 0) i n i t i a l - e (\nf) (11) where nitrate is the reactant and/is the fraction of the initial nitrate pool that remains. 3.2.3. Procedure The protocol devised here for removal of nitrite from aqueous samples involves the reaction of nitrite (as nitrous acid) with ascorbic acid (as ascorbate) to form NO gas that is continuously removed by bubbling with an inert gas (N 2, Ar, He). The concentration of nitrate in the samples containing nitrite was first measured on the nitrate/nitrite analyzer after trapping nitrite with sulfanilamide and NNED (Greiss reagents). The samples were then transferred into serum vials. Either 10 ml or 20 ml of sample was aliquoted into 20 or 30 ml vials, respectively. The vials were capped with (grey) butyl or silicone septa and secured with an aluminium crimp seal. The weight of each capped sample vial was then recorded in order to account for evaporation of water incurred by bubbling. Sealed serum vials containing samples that originated from dense cultures of denitrifying bacteria were gently heated on a hot plate for 10 minutes to temperatures slightly below 100 °C in order to sterilize the mixture. This step prevented bacteria from 71 reactivating in the samples. This heating step appeared to cause no change in the isotopic composition of nitrate (data not shown). Oxygen was then stripped from the samples by gently bubbling the vials with an inert gas for 2 hours at room temperature. The gas inflow consisted of a 1.5", 24-gauge needle (Beckton-Dickinson) perforating the septum and immersed in the sample, while the outflow was a 0.5", 26-gauge needle emerging from the headspace in the vial. A 1.0 mol L"1 solution of ascorbic acid was prepared daily with milli-Q water in a 10 ml serum vial that was capped and sealed as the samples. Each preparation of ascorbic acid was tested for nitrate or nitrite contamination with the nitrate/nitrite analyzer. Ascorbic acid solutions from different stock salts were consistently found to be devoid of either contaminant. The sealed ascorbic acid solution was then bubbled for 2 hours concomitantly with the samples in order to strip it of dissolved (and headspace) oxygen. After gas purging, an aliquot of the ascorbic acid solution was added to each sample to a final concentration of 10 mmol L"1 (which was found to be optimal for nitrite removal, as described below). The solution was transferred using an acid-rinsed gas-tight syringe, taking care not to contaminate the samples with oxygen. While extracting the ascorbic acid, the inflow of inert gas remained immersed in this solution, while the outflow needle was removed to prevent oxygen entry into the solution as it was syringed from the vial. Purging of each sample was continued while injecting the ascorbic acid. The ascorbic acid solution was injected in the samples' headspace as a spray, to minimize the dissolution of any contaminating oxygen in the samples. Once the ascorbic acid was added, the samples were left to bubble from 3 hours to overnight, depending on the initial nitrite concentration (see below). Continuous purging allowed for generated NO gas to escape from the vials, thus shifting the equilibrium of the reaction to NO production. Moreover, the positive pressure imparted by continuous bubbling of the samples ensured that no oxygen leaked into the samples and that NO was efficiently purged away after being produced (as neither butyl nor silicone stoppers are completely gas-tight), thus preventing formation of nitrite or nitrate from NO. For each run, a single sample was split into duplicates, and one of these was amended with 50 pi per 10 ml of 10% atom IiV80 (Medical Isotopes), yielding a sample 72 with water having a 5 1 8 0 of approximately 300%o. A significant positive deviation in the 6 I 8 0 of nitrate in the enriched sample compared to its un-amended duplicate would indicate the formation of nitrate from nitric oxide during nitrite removal, due to contamination with oxygen during the reaction. Also, the possibility of O exchange in nitrate mediated by the acidic pH's generated by the ascorbic acid would be tested by the addition of l 8 0 -H 2 0 , though this side-reaction is of minor concern as it is extremely slow at a pH of 3.5 (Bunton et al. 1952). To ensure complete nitrite removal in the samples by ascorbate at the end of the bubbling phase, an aliquot of each sample was injected into the nitrate/nitrite analyzer (whose vanadium solution rested in a methanol/ice bath to detect nitrite but not nitrate). Samples in which nitrite was still present were immediately further degassed to prevent contamination with oxygen. Samples devoid of nitrite were then weighed individually to account for volume loss during bubbling. The concentration of nitrate in each sample was measured on the nitrate/nitrite analyzer. Though the concentration of nitrate that was measured was generally accurate once we accounted for any loss of volume from purging, our initial measurements of nitrate while trapping nitrite proved more reliable. Approximately 4 pi of a 10 mol L"' solution of sodium hydroxide (NaOH) was then added to 10 mL of sample, restoring the pH to neutral or above. Though NaOH of the purest available quality was used, prepared stock solutions of NaOH were consistently found to be contaminated with at least ~ 0.1 uM nitrate/nitrite (corresponding to an additon of 0.4 nmol L"1 nitrate/nitrite to the samples). Blank samples (milli-Q water + nitrate no base, milli-Q, nitrate + base) were thus run to account for this, however NaOH addition was not found to alter the isotopic composition of nitrate perceptibly. The neutralization step with NaOH ensures that any nitrite generated by the denitrifiers during isotope analysis will not be converted to NO by ascorbate at lower pH, but rather converted to N 2 0 solely by the bacteria. Trial analyses without the NaOH suggested that this step was not necessary, presumably because the sample additions were too small to affect the pH of the bacterial growth medium. However, we cannot rule out an effect under all conditions. The neutralized samples were then transferred to acid-washed polypropylene bottles and frozen awaiting isotope analysis. 73 3.3. Assessment and Discussion To determine the amount of ascorbic acid necessary to remove nitrite, seawater and freshwater samples amended with 300 pmol L'1 nitrite were titrated with increasing concentrations of ascorbic acid. As illustrated in Figure 3.1a, the amount of nitrite removed by ascorbate in solution after four hours of purging was inversely related to the pH of the solution, such that more nitrite was removed at lower pH. The amount of nitrite removed was greatest around a pH of 3.5; explicably, this lies between the pKa's of nitrous acid and ascorbic acid, wherein nitrous acid and the ascorbate anion coexist in sufficiently large concentrations to react with one another. The decrease in pH of the solutions was itself a direct function of the amount of ascorbate added to the solution (Figure 3.1b). In both freshwater and seawater samples, 10 mmol L"1 ascorbic acid reduced the pH of the solution sufficiently to allow for complete nitrite removal (Figure 3.1c). The time dependence of the reaction was monitored to determine the length of time necessary to degas NO from the samples completely (Figure 3.2). Replicate seawater samples were amended with 1 mmol L"1 nitrite, and sub-sampled at short time intervals following the addition of ascorbic acid at a final concentration of 10 mmol L'1. The sub-samples were extracted with a syringe and transferred immediately to Greiss reagents such that nitrite was measured colourimetrically. In this experiment, detection of nitrite with the Greiss reaction was preferable to detection using the nitrate/nitrite analyzer because the latter would also have detected nitric oxide present in solution. As illustrated in Figure 2, over 90% of the nitrite was lost within 2 hours of ascorbic acid addition. However, only after 5 hours were both nitrite and nitric oxide no longer detectable in the sample solution and sample headspace (< 20 nmol L"1 with the nitrate/nitrite analyzer), as the rate of reaction seemingly decreased with decreasing nitrite (and ascorbate) concentrations. Based on these results, we suggest bubbling the samples for at least 3 hours following ascorbic acid addition for nitrite concentrations < 10 umol L"1, and for 5 hours or more for concentrations exceeding 300 umol L"1 nitrite. 74 In many instances, we allowed our samples to bubble overnight for the sake of convenience. To ascertain that nitrate was not formed during nitrite removal, we carried out the reaction in nitrate-free seawater amended with a range of nitrite concentrations, from 0 to 300 pmol L"1. Figure 3.3 illustrates that a small amount of nitrate was detected in the samples after nitrite was removed, except in the control sample to which no nitrite was added. This could possibly indicate that some nitrate was formed in the reaction. However, the concentration of nitrate remaining in the samples was roughly identical between runs for a given nitrite concentration, and was also linear with respect to initial nitrite. This suggests that the "leftover" nitrate is likely a contaminant of the nitrite stock, amounting to a 0.1% molar ratio in the stock. We posit that detectable formation of nitrate due to oxygen contamination would have yielded less consistent nitrate concentrations than those observed here. As a positive control, ascorbic acid was added to seawater samples containing 100 pmol L"1 nitrite, and these samples were left uncapped on the bench top overnight. Though no nitrite remained in the samples the following day, variable amounts of nitrate had formed, ranging from 6 to 20 pmol L"1 (data not shown). Shown in Table 3.1, an increasing range of nitrite additions was removed from a 10 pmol L"1 nitrate solution to verify that the isotopic composition of nitrate was not altered by nitrite removal. No clear isotopic differences were detectable at nitrite concentrations up to 50 pmol L"1. The isotopic composition of standard additions of IAEA-N3 potassium nitrate to seawater, from which a more narrow range of nitrite concentrations were removed, is summarized in Table 3.2. The isotopic composition of nitrate in the treatment samples showed slightly lower 5 1 5 N and 6 , 8 0 than standard, with negative deviations in the treatments averaging 0.2%o and 0.3%o for 6 1 5N and S 1 8 0, respectively. The magnitude of individual deviations was not proportional to the concentration of nitrite removed from samples. A number of samples were collected from various growing cultures of denitrifying bacteria (Granger et al. 2004a). Individual samples were split into duplicates, from which nitrite was removed and nitrate isotopes were analyzed in separate 75 runs (Table 3.3). Duplicate measurements were, in most cases, similar for both N and O isotopes. The reproducibility of the measurements is quite good considering that both nitrate and nitrite spanned a considerable range of concentrations in these samples, from a few umoles L"1 to upwards of 300 pmol L"1. The mean precision based on 15 duplicate analyses was 0.34%o for 5 1 5 N and 0.37%o for 8 1 80. These estimates include variability in flask preparation, nitrite removal, blank variability and isotope analyses. Also shown in Table 3 are nitrate isotopic measurements of duplicate samples where 180-enriched water was added to one of the duplicates prior to nitrite removal. Again, the 6 1 8 0 measured among duplicates were nearly identical, indicating no measurable nitrate formation in these samples. One uncertainty that remains regarding our protocol and all isotope work on nitrite-bearing samples involves the possibility of nitrite loss and conversion (in particular, to nitrate) during sample storage (Dore et al. 1996). We and others have observed that nitrite is converted to nitrate in acidified samples during storage (M. A. Altabet, pers. comm., 2005), and the same may occur in frozen samples (freezing being the preservation method for all samples analyzed here). It is possible that this underlies some of the cases of poor external replication reported in Table 3.3. Preservation and storage tests on dissolved nitrite are badly needed and are underway in another laboratory (K. L. Casciotti, pers. comm., 2005). Until these questions are resolved, we recommend that the nitrite removal step be performed as soon as possible after sample collection. We tested our method on samples collected from a growing culture of the denitrifier P. aureofaciens. As illustrated in Figure 3.4a, nitrate was respired during growth and nitrite was concomitantly produced in nearly stoichiometric amounts to the nitrate consumed. Nitrite, which reached concentrations upwards of 240 umol L"1, was removed from the samples, and the N and O isotopes of nitrate were then measured. Figure 4a shows that, as nitrate concentration decreased, both the , 5 N/ 1 4 N and 1 8 0 / l 6 0 of the nitrate increased due to isotope fractionation imparted by the uptake and reduction of nitrate. On a Rayleigh plot (Figure 3.4b), the nitrate isotope ratios fall on a straight line against the natural logarithm of the fraction of nitrate consumed, yielding slopes approximating respective isotope effects (e) of 23.5%o for N and 22.0%o for O. The N isotope effect observed here falls in range expected from previous observations (Barford 76 et al. 1999). Oxygen isotope effects on nitrate during denitrification have not been reported previously. However, the similarity in the magnitudes of the N and O isotope effects of P. aureofaciens is consistent with our previously reported 1:1 relationship between nitrate N and O isotope fractionation in nitrate assimilation by marine eukaryotic algae (Granger et al. 2004b). 3.4. Comments and Recommendations Presented here is a simple, non-toxic, and cost-effective method to remove nitrite, allowing for accurate quantification of the nitrate N and O isotopic composition of samples with co-occurring nitrite. One weakness of the method is the risk of exposure to oxygen during nitrite removal, where the NO gas generated from nitrite reduction can be oxidized to nitrate. Such mishaps can be monitored with 180-labeled water addition, or simply by replication of the nitrite removal step for individual samples, allowing for detection of outlying measurements. This method is proving highly useful to examine the isotopic composition of nitrate associated with the growth of denitrifying bacteria in culture (Granger et al. 2004a), in which nitrite often accumulates in large amounts as nitrate is consumed. Moreover this method can be used to remove nitrite from natural samples, allowing for accurate measurements of the N and O isotopic composition of nitrate in oxygen-deficient and eutrophic waters. 77 Table 3.1. N and O isotopic composition of nitrate (or nitrite) after nitrite was removed from seawater with ascorbic acid while bubbling with N 2 gas. Nitrate Nitrite Ascorbic acid d 1 5N ± stdev d , 8 0 ± stdev (umol L"1) (umol L"1) (mmol L"1) n (%o) (%o) 10 - - 2 1.63 ±0.01 24.55 ± 0.49 10 - 10 2 1.77 ±0.20 23.99 ±0.40 10 10 10 2 1.71 ±0.21 23.21 ±0.39 10 50 10 2 1.95 ±0.18 23.52 ±0.28 - 100 - 2 0.94 ±0.20 -14.17 ±2.64 78 Table 3.2. The N and O isotopic composition of international potassium nitrate reference material IAEA-N3 (solutions in de-ionized water) with an assigned 5 1 5 N of 4.7 ± 0.05 %o (Gonfiantini et al., 1995) and a reported 6 1 8 0 of 25.6 ± 0.03 %<, (Bohlke et al. 2003) - after nitrite was removed with ascorbic acid while purging with helium gas. Nitrate Nitrite Ascorbic acid S 1 5 N ± stdev 5 1 8 0 ± stdev (pmol L"1) (pmol L"1) (mmol L"1) n (%o) (%o) 20 0 0 3 4.60 ± 0.05 25.77 ±0.12 20 2.7 10 3 4.39 ±0.04 25.35 ±0.15 20 5.4 10 3 4.52 ±0.16 25.29 ±0.30 20 10.8 10 3 4.51 ±0.19 25.30 ±0.11 79 Table 3.3. Nitrate 5 I 5 N and 5 1 8 0 values measured for duplicate nitrite removals of individual samples of denitrifying bacterial cultures. Underlined values point to replicates amended with 180-enriched water. [nitrate] [nitrite] 51 5 N (%0) Difference 5 I 8 0 (%o) Difference Sample (umol L"') (umol L"1) (1) (2) (%o) (1) (2) (%o) 28.3B 3 283 21.68 20.42 1.26 41.23 41.81 0.57 40.5A 53 270 6.53 6.80 0.28 29.86 30.12 0.26 40.5B 132 118 5.97 5.80 0.17 28.96 28.79 0.17 39.7A 144 146 20.91 21.87 0.97 43.70 44.56 0.86 39.5A 187 63 3.73 3.72 0.01 26.80 26.83 0.02 36.7B 244 246 13.23 13.60 0.37 34.94 35.52 0.58 38.3B 252 38 6.93 6.23 0.70 29.35 29.17 0.18 39.5B 262 28 6.21 5.50 0.71 29.18 30.08 0.90 40.1 A 322 0 1.24 1.29 0.05 24.64 24.64 0.01 37.3B 348 142 9.35 9.50 0.15 32.00 32.41 0.41 37.2B 367 123 8.94 8.95 0.01 32.17 31.60 0.57 36.5B 374 139 6.42 6.51 0.10 28.66 28.42 0.24 3 6.5A 425 65 4.24 4.13 0.11 27.25 26.93 0.32 37. IB 515 3 1.34 1.25 0.09 24.45 24.71 0.26 37.1 A 524 0 1.30 1.21 0.09 24.45 24.25 0.20 mean: 0.34 mean: 0.37 ± stdev: 0.39 ± stdev: 0.26 80 100-o •4—* 80-E 'a 60-;s 40-«4-l _ %o 20-0 -QP • O X [ascorbic acid] (mmol L") 10<M [ascorbic acid] (mmol L"') Figure 3.1. Nitrite removal from freshwater and seawater samples with ascorbic acid. Vials containing 10 mL of water and 300 pmol L"1 nitrite were amended with a range of ascorbic acid additions (0 to 10 mmol L"1) and purged with helium gas for 4 hours, a) % of initial nitrite (300 pmol L"1) remaining as a function of pH, b) change in the pH of seawater and freshwater samples with increasing concentrations of ascorbic acid, and c) % of initial nitrite removed as a function of ascorbic acid concentration. 81 100-'B 8 0 -'S 6 0 -' - ^ 4 0 -o 2 0 -0 -0 • • Replicate 1 • Replicate 2 • I 2 Hours Figure 3.2. Time dependence of nitrite removal by ascorbic acid. A 1 mmol L'1 nitrite solution in replicate 10 mL seawater samples amended with 10 mmol L"1 ascorbic acid and bubbled with N 2 gas. Nitrite concentrations were monitored for 5 hours following the addition of ascorbic acid. 82 Figure 3.3. The concentration of nitrate remaining in seawater after the removal of incremental amounts of nitrite. Ten (10) mL samples containing from 0 to 300 pmol L"1 nitrite were amended with 10 mmol L"1 ascorbic acid and bubbled with N 2 gas overnight in three separate experiments. Symbols represent the separate removal experiments. 83 250 -[•]- N 0 3 " 200 - [0]- N 0 2 " a 3 150 — ....0'" £ 1 0 0 -o i O 5 0 -.....-0-0 -- 6 0 - 5 0 CY, - 4 0 o> - 3 0 o o O ? - 2 0 o - 10 o - 0 r 30.0 30.5 31.0 31.5 32.0 Hours after inoculation •1.2 -0.8 -0.4 ln([N0 3-]/[N0 3- i n i t i a l]) 0.0 Figure 3.4. N and O isotope enrichment of nitrate during the growth of P. aureofaciens; a) changes in the concentration of nitrate and nitrite as a function of time, as well as concomitant increases in the 8 I 5N (closed circles) and 5 I 8 0 of nitrate (open circles); b) the 5'5N and 5' 80 of nitrate plotted against the natural logarithm of the fraction of nitrate consumed. The slopes of the fitted regressions represent the respective isotope effects, e, for N and O imparted by the dissimilatory reduction of nitrate by P. aureofaciens. 84 3.7. References Amberger, A., and H.-L. Schmidt. 1987. Natiirliche isotopengehalte von nitrat als Indikatoren fur dessen herkunft. Geochim. Cosmochim. Ac. 51 : 2699-2705. Barford, C. C , J. P. Montoya, M. A. Altabet, and R. Mitchell. 1999. Steady-state Nitrogen isotope effects of N 2 and N 2 0 production in Paracoccus denitrificans. Appl. Environ. Microbiol. 65: 989-994. Bohlke, J. K., G. E. Ericksen, and K. Revesz. 1997. Stable isotope evidence for an atmospheric origin of desert nitrate deposits in northern Chile and southern California. Chem. Geol. 136: 135-152. Bohlke, J. K., S. J. Mroczkowski, and T. B. Coplen. 2003. Oxygen isotopes in nitrate: new reference materials for 0-18 : 0-17 : 0-16 measurements and observations on nitrate-water equilibration. Rapid Comm. Mass Sp. 17: 1835-1846. Bothner-by, A., and L. Friedman. 1952. The reaction of nitrous acid with hydroxylamine. J. of Chem. Phys. 20: 459-462. Braman, R. S., and S. A. Hendrix. 1989. Nanogram nitrite and nitrate determination in environmental and biological materials by V(III) reduction with chemiluminescence detection. Anal. Chem. 61 : 2715-2718. Bunton, C. A., E. A. Halevi, and D. R. Llewellyn. 1952. Oxygen exchange between nitric acid and water. Part I. J. Chem. Soc: 4913-4917. Casciotti, K. L., D. M. Sigman, M. G. Hastings, J. K. Bohlke, and A. Hilkert. 2002. Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater using the denitrifier method. Anal. Chem. 74: 4905-4912. Cline, J. D., and I. R. Kaplan. 1975. Isotopic fractionation of dissolved nitrate during denitrification in the Eastern Tropical North Pacific Ocean. Mar. Chem. 3: 271-299. 85 Dore, J. E., T. Houlihan, D. V. Hebel, G. Tien, L. Tupas, and D. M. Karl. 1996. Freezing as a method of sample preservation for the analysis of dissolved inorganic nutrients in seawater. Mar. Chem. 53: 173-185. Garside, C. 1982. A chemiluminescent technique for the determination of nanomolar concentrations of nitrate, nitrite, or nitrite alone in seawater. Mar. Chem. 11: 159-167. Gonfiantini, R., W. Stichler, and K. Rosanski. 1995. Standards and Intercomparison Materials Distributed by the IAEA for Stable Isotope Measurements. International Atomic Energy Agency. Granger, J., D. M. Sigman, M. F. Lehmann, and P. D. Tortell. 2004a. Nitrogen and oxygen isotope effects associated with nitrate assimilation and denitrification by laboratory cultures of marine plankton. Eos Trans. AGU 85: Abstract H51E-052. Granger, J., D. M. Sigman, J. A. Needoba, and P. J. Harrison. 2004b. Coupled nitrogen and oxygen isotope fractionation of nitrate during assimilation by cultures of marine phytoplankton. Limnol. Oceanogr. 49: 1763-1773. Granger, J., and B. B. Ward. 2003. Accumulation of nitrogen oxides in copper-limited cultures of denitrifying bacteria. Limnol. and Oceanogr. 48: 313-318. Kanda, Y., and M. Taira. 2003. Flow-injection analysis method for the determination of nitrite and nitrate in natural water samples using a chemiluminescence NOx monitor. Anal. Sciences 19: 695-699. Mariotti, A., J. C. Germon, P. Hubert, P. Kaiser, R. Letolle, A. Tardieux, and P. Tardieux. 1981. Experimental determination of nitrogen kinetic isotope fractionation: some principles; illustration for the denitrification and nitrification processes. Plant Soil 62: 413-430. Mcllvin, M. R., and M. A. Altabet. 2005. Chemical conversion of nitrate and nitrite to nitrous oxide for nitrogen and oxygen isotopic analysis in freshwater and seawater. Anal. Chem. 77: 5589-5595. 86 Price, N. M., G. I. Harrison, J. G. Herring, R. J. Hudson, P. M. V. Nirel, B. Palenik, and F. M. M. Morel. 1988/89. Preparation and chemistry of the artificial algal culture medium Aquil. Biol. Oceanogr. 6: 443-461. Revesz, K., J. K. Bohlke, and Y. Yoshinari. 1997. Determination of dl80 and dl5N in nitrate. Anal. Chem. 69: 4375-4380. Sigman, D. M., M. A. Altabet, R. H. Michener, D. C. McCorkle, B. Fry, and R. M. Holmes. 1997. Natural abundance-level measurement of the Nitrogen isotopic composition of oceanic nitrate: an adaptation of the ammonia diffusion method. Mar. Chem. 57: 227-242. Sigman, D. M., K. L. Casciotti, M. Andreani, C. Barford, M. Galanter, and J. K. Bohlke. 2001. A bacterial method for the nitrogen isotopic analysis of nitrate in seawater and freshwater. Anal. Chem. 73: 4145-4153. Silva, S. R., C. Kendall, D. H. Wilkinson, A. C. Ziegler, C. C. Y. Chang, and R. J. Avanzino. 2000. A new method for collection of nitrate from fresh water and the analysis of nitrogen and oxygen isotope ratios. J. Hydrol. 228: 22-36. Ward, B. B. 1987. Nitrogen transformations in the Southern California Bight. Deep-Sea Res. 34: 785-805. Ward, B. B., K. A. Kilpatrick, E. H. Renger, and R. W. Eppley. 1989. Biological nitrogen cycling in the nitracline. Limnol. Oceanogr. 34: 493-513. Wu, J., S. E. Calvert, and C. S. Wong. 1997. Nitrogen isotope variations in the subarctic Pacific northeast Pacific: relationships to nitrate utilization and trophic structure. Deep-Sea Res. Part I 44: 287-314. 87 Chapter 4 Nitrate N and O isotope fractionation associated with dissimilatory nitrate reduction by denitrifying bacteria A version of this chapter will be submitted for publication. Granger, J., D. M. Sigman, J., P. D. Tortell. In prep. Nitrate N and O isotope fractionation associated with dissimilatory nitrate reduction by denitrifying bacteria. Applied and Environmental Microbiology. 88 4.1. Introduction Denitrification is the principal mechanism on land and in the sea by which fixed nitrogen is lost to the atmosphere. It involves the bacterially mediated reduction of nitrate (NOV) and nitrite (NO2") to nitrous oxide gas (N2O) then dinitrogen (N2) gas. Denitrification exhibits significant biological isotope fractionation, where 1 4 N - and 1 60-bearing nitrate is consumed more rapidly than that bearing l 5 N and 1 8 0, leaving the remaining substrate pool with elevated ratios of 1 5 N/ 1 4 N and 1 8 0/ 1 6 0. The extent against which heavy isotopes are discriminated in a unidirectional biological reaction such as denitrification can be quantified in terms of the kinetic isotope effect, e. This effect is a measure of the ratio of the reaction rate coefficients for the light versus heavy isotope: e = (knght/kheavy) - 1 x 1000, when expressed in per mil (%o). Culture studies of denitrifying bacterial isolates have yielded a broad range of nitrate N isotope effects (15e) for various denitrifiers, from 2%o to 30%o (summarized in Table 4.4). While field data from freshwater and marine systems reveal that both the N and O isotopes of nitrate are strongly fractionated during denitrification (Lehmann et al. 2003 and references therein; Sigman et al. 2005), coupled nitrate N and O isotope fractionation associated with denitrification has not been verified in culture, and the kinetic O isotope effect of denitrification (18e) is poorly known. Furthermore, the physiological mechanism of nitrate isotopic fractionation by denitrifiers has not been studied, and few environmental factors that may contribute to variations in the magnitude of observed N (and O) isotope effects in culture or in the environment have been considered. Until recently, no measurements of nitrate O isotopes in seawater were available because the methodology used to extract nitrate from freshwater is not applicable to seawater. In these methods, nitrate is extracted from freshwater with a non-selective cation exchange resin that is not effective in saline water (e.g., Silva et al. 2000). Recently, Sigman et al. (2001) and Casciotti et al. (2002) devised a method to measure N and O isotopes of nitrate in both freshwater and seawater, wherein denitrifiers convert nitrate in a sample to N2O gas, allowing for mass spectrometric determination of the N and O isotope ratios of this gas analyte. In addition, a chemical method has been developed for conversion of nitrate to N2O (Mcllvin and Altabet 2005). With these new 89 methods, investigation of the coupled fractionation of the N and O isotopes of nitrate in the ocean are beginning (Lehmann et al. 2004; Lehmann et al. 2005; Sigman et al. 2005). The first published ocean depth profile of nitrate N and O isotope ratios from the Subarctic Pacific revealed 1:1 co-variation of the N and O isotope ratios associated with nitrate assimilation in the euphotic zone (Casciotti et al. 2002). Subsequent lab studies then showed that phytoplankton cultures fractionate the N and O isotopes of nitrate to the same extent during nitrate assimilation (Granger et al. 2004), such that nitrate in the mixed layer waters likely reflects the 1 8 0 / l 6 0 to 1 5 N/ 1 4 N ratio of 1:1 imparted during assimilation by indigenous plankton communities. Oceanic data in low O2 waters, where nitrate is consumed in the subsurface by denitrifiers, show increases in nitrate 1 80/ 1 60-to-1 5 N/ 1 4 N associated with denitrification that are in a ratio of 1:1 or greater, with deviations above unity attributed to additional N-transforming processes (Sigman et al. 2005; Sigman et al. 2003). In contrast, observations in oceanic environments appear inconsistent with the relative nitrate N and O isotope fractionation observed for denitrification in terrestrial systems, where denitrification in groundwater and in lakes is associated with a l 8 0 / 1 6 0 to l 5 N/ 1 4 N fractionation ratio o f - 0.55 (Lehmann et al. 2003 and references therein). This striking difference between freshwater and marine systems is unexplained. As of yet, there are no laboratory studies that investigate the coupled nitrate N and O isotope effects imparted by freshwater or marine strains of denitrifiers to address the origin of the nitrate isotopic signatures in these environments. In this study, we investigate the fractionation of N and O isotopes imparted on nitrate during its dissimilatory reduction in cultures of both freshwater and marine denitrifying bacteria. This work aims to constrain the relationship between N and O isotope effects associated with the fractionation process in freshwater and seawater conditions. We interpret our observations in the context of a putative physiological mechanism for nitrate isotopic fractionation during denitrification and consider the implications for coupled measurements of nitrate N and O isotope ratios as a tracer for biogeochemical N cycling. 90 4.2. Materials and Methods The experimental strains chosen for this study are summarized in Table 4.1. Four strains of facultative aerobic chemo-heterotrophic denitrifying bacteria were examined, Pseudomonas stutzeri, Ochrobacter sp., Paracoccus denitrifican, Pseudomonas aureofaciens, as well as an anoxygenic photo-heterotrophic bacterial strain, Rhodobacter sphaeroides. The latter is distinguished by possessing only an auxiliary periplasmic dissimilatory nitrate reductase, NAP (Ferguson et al. 1987), while the remaining strains presumably possess both the respiratory trans-membrane nitrate reductase, NAR, and the periplasmic NAP (this is not confirmed for Ochrobacter sp. because denitrification by this strain has not been investigated previously). Of the experimental strains, P. denitrijicans, P. aureofaciens, and R. sphaeroides are freshwater/soil isolates, while P. stutzeri and Ochrobacter sp. are seawater isolates. Seawater medium consisted of the artificial seawater medium AQUIL (Price 1988/1989) amended with 10 pmol L"1 phosphate, between 100 pmol L"1 and 2 mmol L'1 nitrate, 0.2 g L"1 casein hydrolysate, and 0.2 g L"1 bacteptone. The medium was sterilized by microwaving (Keller et al. 1988), then supplemented with AQUIL trace metals and f/2 vitamins. RCV medium was used as the freshwater medium (Weaver et al. 1975). RCV salts were supplemented with 0.2 g L"1 casein hydrolysate and 0.2 g L"1 bactopeptone. The autoclaved medium was then supplemented with AQUIL trace metals and f/2 vitamins. Cells were grown in batch culture at room temperature. Cultures were initiated from frozen stocks and acclimated to experimental medium for 10 generations. Overnight cultures were then inoculated in 250 ml opaque, tri-laminate, polyethylene-lined, TEDLAR™ gas-tight bags (Granger and Ward 2003). These had been previously rinsed with 10% HCI and milli-Q water. The bags may not have been sterile, so cultures were initiated immediately following introduction of the medium into the bags. Freshwater and seawater isolates were grown in their respective media. To investigate whether strains fractionate nitrate isotopes differently in freshwater than in seawater, the freshwater/soil isolates P. denitrificans, P. aureofaciens andi?. sphaeroides were also cultured in seawater medium, as they proved to be halo-tolerant. Growth of all 91 cultures was monitored by measuring the accumulation of nitrite in the culture bags, or by measuring the disappearance of nitrate. Nitrite was quantified colourimetrically by reaction with Greiss reagents (Parsons et al. 1984). Nitrate was measured by conversion to NO followed by chemiluminescence detection (Braman and Hendrix 1989) on an Antek 1750 nitrate/nitrite analyzer. Chemiluminescence detection simultaneously measures both nitrate and nitrite in a sample; to measure nitrate without the interference of nitrite, we trapped nitrite with sulfanilamide prior to sample injection in the Antek 1750. Subsamples (20 ml) of growing culture were sequentially extracted from the bags for isotope analysis of the nitrate as it was progressively depleted from cultures. The culture sub-samples were stored in acid-washed polypropylene bottles and frozen immediately. Prior to nitrate N and O isotope analysis, nitrite in the culture samples was removed with ascorbate because nitrite interferes with nitrate N and O isotope analysis (Granger et al. in press). The 'ascorbate method' is effective in removing all detectable nitrite without altering the isotopic composition of incident nitrate in treated samples (Granger et al. 2006). Moreover, it is non-toxic, and thus compatible with the denitrifier method for isotope analysis. Samples were divided into duplicates, and nitrite was removed from these in separate batches, to eliminate potential systematic error associated with individual batch processing. We noticed that during prolonged sample storage, nitrite concentrations decreased in some samples. This resulted in N and O isotopic shifts at lower nitrate concentrations that were likely due to the spontaneous decomposition of nitrite to NO and re-oxidation of NO to nitrate; we observed this isotopic phenomenon previously when developing the ascorbate method. We thus carefully screened samples and omitted cultures that showed significant and haphazard non-linearity in their N and O isotope ratios towards lower nitrate concentrations. In most discarded samples and cultures, nitrite loss during storage was verified. In ongoing work, we now remove nitrite immediately after sample collection. 92 The 1 5 N/ 1 4 N and , 8 0/ 1 6 0 of nitrate were determined following the denitrifier method (Casciotti et al. 2002; Sigman et al. 2001). Isotope ratios are reported using the delta (6) notation in units of per mil (%o): o'Xample = [(15N/14N)samp,e/(15N/,4N)reference - 1] X 1000 (1) 61 8O s a m p le = [(,80/'60)sarnp,e/(,80/'60)reference " 1] X 1000 (2) The 1 5 N/ I 4 N reference is N 2 in air, and the 1 8 0/ 1 6 0 reference is Vienna standard mean ocean water (VSMOW). Individual analyses were referenced to injections of N 2 0 from a pure gas cylinder and then standardized through comparison to the international potassium nitrate reference materials IAEA-N3 with an assigned 6 1 5N of +4.7 versus atmospheric N 2 (Gonfiantini et al. 1995) and most recently reported 5 , 8 0 of 25.6%o versus VSMOW (Bohlke et al. 2003). The size of the culture blank was determined by running a prepared vial to which no sample was added. Potential exchange of the oxygen atoms with water during reduction of nitrate to N 2 0 was calculated for each batch of measurements from the slope of the regression between 8 , 8 0 of IAEA-N3 in normal versus 180-enriched water (S 1 80H20 = 3 00%o). The latter half of our sample measurements of the 5'80 of nitrate were corrected for oxygen exchange against the international nitrate isotopic standard USGS-34 (6I80 = -27.9%o; Bohlke et al. 2003). Use of either correction scheme did not change the reported values appreciably. The N and O isotopic ratios represent the mean of any replicate measurements. The N and O isotopic measurements of roughly 10% of the samples were duplicated within a day's batch of analyses, and replicate isotopic measurements of individual samples were generally consistent with previously reported analysis standard deviations of 0.2%o for l 5 N and 0.5%o for 1 80. As mentioned above, the majority of the samples were also processed in duplicates for nitrite removal. Reproducibility among duplicate nitrite removals was around 0.4% for , 5 N and 0.8%o for 1 80. To derive estimates of the N and O isotope effects imparted on nitrate (15e and I 8 E ) during growth, the nitrate 6 1 5N and 8 I 80 measurements were fit to the following linear equations (Mariotti et al. 1981): 93 ln (5 1 5N + 1) = ln (5 1 5 N i n i t i a | + 1) + {(,5e/1000)ln(/)} (3) ln (8 I 8 0 + 1) = ln (6 1 80i ni t i a, + 1) + {(18e /1000)ln(/)} (4) where/= [N03"]/[N03"]initiai. Regression of ln (6 I 5N +1) or ln (5 1 80 +1) on ln(/) yields respective slopes of 15E/1000 or l8e/1000. The simplified linearization of the Rayleigh model was used to plot the isotope ratio measurements. Namely, the 6 1 5N of nitrate was plotted on ln([NC>3"]) and ln(/) according to the more intuitive (but less accurate) linear derivation (Mariotti et al. 1981), where the slope of the line approximates 1 5 E : Given the relatively large isotopic ranges encountered in our culture experiments, this simplified, approximate form of the Rayleigh model is not used for derivation of the isotope effects, as errors could have been significant (Mariotti et al. 1981). 4.3. Results Cultures initiated in gas-tight bags from small inocula fuelled their initial growth using the residual O2 in the bags for respiration. As cell densities increased, the O2 tension presumably decreased, and cells switched to nitrate respiration. Formost of the cultures grown in this study, nitrite accumulated concomitantly as nitrate was depleted from the cultures, after which nitrite was depleted from the culture medium (Figure 4.1, Table 4.2). The R. sphaeroides cultures are an exception in several regards. First, they rarely approached complete nitrate consumption, which is probably related to the fact that this strain's periplasmic NAP does not fuel respiration. Second, none of the nitrite produced was consumed thereafter, because the strain of R. sphaeroides used in this 6 1 5N =5 l 5N i n i t i a,+ 15£{ln([N03-])} (5a) 6 1 5 N =6 1 5N i n i t i a,+ ,5e{ln(/)} (5b) 94 study does not have a nitrite reductase enzyme (Gavira et al. 2002). Nitrite concentrations were easily measurable with the Greiss reaction and thus served as a proxy to estimate the extent of nitrate consumption in the cultures. However, Ochrobacter sp., as well as P. denitrificans grown in freshwater medium, showed little to no nitrite accumulation during nitrate consumption (except culture Ochrobacter sp. 33A, see Table 4.2), implying that any nitrite generated was respired concomitantly with nitrate. For these cultures, nitrate was measured in order to monitor growth. The nitrate N and O isotope data generally conformed to a linear relationship with the natural logarithm of the fraction of nitrate remaining, as predicted by the Rayleigh model (equations 3 and 4; r2 values consistently greater than 0.97), with the regression coefficients yielding estimates of the N and O isotope effects for the dissimilatory reduction of nitrate (Table 4.2). However, as described below, a characteristic deviation from linearity developed at lower concentrations. With regard to nitrate N and O isotope coupling, our results indicate a distinction between the 'true' respiratory denitrifiers that use nitrate as a terminal electron acceptor during respiration with NAR (P. stutzeri, Ochrobacter sp., P. aureofaciens, and P. denitrificans) and R. sphaeroides, which reduces nitrate during aerobic growth with the auxiliary nitrate reductase enzyme, NAP. The experimental results are described below according to these two apparent categories. 4.3.1. Nitrate N and O isotope fractionation by 'respiratory' denitrifiers Nitrate N isotope effects (15e) observed among four of the five experimental strains, the 'true' denitrifiers, spanned a broad range, between 5%o and 25%o, though the majority of these estimates fell above 15%o (Table 4.2). Similarly, corresponding 18e estimates ranged between 5%o and 23%o, and most of these estimates exceeded 15%o. An analysis of covariance (ANCOVA) was performed for each of the respiratory strains to determine whether the isotope effects determined for replicate cultures of a given strain differed significantly from one another (Table 4.3). The covariance analyses confirmed that isotope effects differed significantly between experiments with the same strain, reflecting genuine variations in the isotope effects expressed by a given strain. Growth of 95 the respiratory denitrifiers P. denitrificans and P. aureofaciens in seawater versus freshwater media had no discernible effect on 15e or 1 8e, such that differences in isotope effects were not readily attributable to medium type. However, we only present data for a single culture of P. aureofaciens in freshwater medium, such that differences in isotope effects due to medium salinity could become apparent given more replicates of this strain (Table 4.2). In Figure 4.2, all of the 5 1 5N values for the afore-mentioned experiments are plotted against the natural logarithm of nitrate concentration (Figure 4.2a) and that of fractional nitrate consumption (Figure 4.2b). The respective slopes of the resulting lines approximate the 15e of the experiments (Table 4.2). A conspicuous feature in some of the experiments is the apparent asymptotic behaviour of the Rayleigh plot at lower nitrate concentrations (Figure 4.2a). For example, P. stutzeri (28B) shows no further increase in nitrate 8 1 5N at concentrations lower than 3 pmol L"1. On the same figure, a similar deviation of I 5e from linearity is discernable for P. aureofaciens (19B) at 6 pmol L"1 nitrate. Hence the 15e associated with the denitrifying cultures decreased or disappeared at lower nitrate concentrations. This trend is not apparent for all cultures that were sampled at lower nitrate concentrations, though this likely results from the lack of sampling intervals at intermediate concentrations for these cultures. While 1 8 E and l 5e differed significantly between experiments for a given strain (Table 4.3), all estimates of , 8 s were similar to their corresponding 15e within experiments. Regression of 8 , 8 0 on the 6 1 5N of corresponding samples for individual experiments yielded slopes that were consistently close to 1, although most commonly slightly below it (Table 4.2). The lowest slopes were observed among the four P. sturzeri cultures (-0.88), while Ochrobacter sp. consistently showed high 5 , 8 0:5 I 5 N ratios, averaging 0.98 for 6 cultures. Analyses of covariance (ANCOVA) indicate that the differences in the 5 1 8 0:5 1 5 N relationships within replicate cultures of the same strain are not statistically significant (Table 4.3). However, ANCOVA among all slopes computed for the four strains combined reveals that these differ significantly, with P. stutzeri comprising the outlying group (F0.05(i),22,73 = 129, P < 0.0001). Nonetheless, among the four strains, 8 1 8 0:5 1 5 N covariance ratios range from 0.88 to 0.98, with a pooled slope of 0.95. 96 4.3.2. Nitrate N and O isotope fractionation during 'auxiliary' denitrification by R. sphaeroides The photo-heterotroph R. sphaeroides was distinct with respect to nitrate isotope fractionation in that 15e and 1 8 E covered a narrower range of magnitudes. Observed 15e values were between 13%o and 20%o (Table 4.2) and differed significantly from each other (Table 4.3), showing a modal value of ~ 15%o (Figure 4.4). Similarly, 18e estimates were significantly different from each other (Table 4.3), ranging between 8%o and 13%o (Table 4.2), with a modal value of ~ 9%o. Finally, growth of R. sphaeroides in seawater (for 2 cultures only) showed no consistent difference compared to freshwater cultures. As described above for two other strains, nitrate isotope fractionation by R. sphaeroides decreased at lower nitrate concentrations in one culture (Exp. 27A; Figure 4.4a). In this case, however, inflection of the Rayleigh slope was apparent between 30 and 12 pmol L"1 (Figure 4.4a). R. sphaeroides is also distinct from the other experimental strains in its pooled 6 1 8 0:5 1 5 N covariance ratio of 0.62, as opposed to 0.95 (Figure 4.5). This ratio was relatively invariant, showing no statistically significant differences among cultures (Table 4.3). 4.4. Discussion The observations of our study confirm that respiratory denitrification imparts nearly identical isotope effects on the N and O isotopes of nitrate, consistent with the interpretation of ocean field data (Sigman et al. 2005; Sigman et al. 2003). Although the N and O isotope effects were robustly coupled, their individual values varied, highlighting the plasticity of the isotope effect magnitude. The constancy of the N-to-0 isotope coupling implicates a single enzyme, the respiratory NAR, as the dominant driver of coupled N and O isotope effect imprinted on nitrate, with variations in the magnitude of the fractionation reflecting the degree to which the intrinsic isotope effect of NAR is propagated to the extracellular medium. We thus examine the physiology underlying respiratory denitrification, and delineate a putative fractionation mechanism to uncover 97 environmental factors that may contribute to the variable magnitude of the isotope effects. The difference in the l 8e: 1 5e ratio that separates R. sphaeroides from the other experimental strains also stands as a striking result in this study. This observation likely reflects the different nitrate reductases involved in bulk nitrate dissimilation by the two groups: While nitrate reduction by the respiratory denitrifiers is largely undertaken by the membrane-bound respiratory reductase NAR, R. sphaeroides denitrifies solely via its auxiliary periplasmic reductase, NAP (Table 4.1). Thus we proceed to interpret our experimental observations in the context of the activity of NAR versus NAP in denitrifiers. Putative fractionation mechanism associated with NAR Respiratory denitrifiers possess up to three kinds of nitrate reductases: a membrane bound NAR, a periplasmic NAP, and a cytosolic NAS. The latter is an assimilatory enzyme that was likely not active in our cultures because our growth media were replete with organic N substrates, discouraging the costly consumption of nitrate for N nutrition. Both NAR and NAP are denitrifying enzymes; however, NAP does not generate a significant proton motive force across the cytoplasmic membrane, and thus does not qualify as a respiratory enzyme per-se. Its physiological role, though not entirely clear, has been attributed to maintenance of electrochemical balance in growing cells (Berks et al. 1995). Thus, bulk nitrate reduction by respiratory denitrifiers is likely incurred by the activity of NAR. The putative physiological mechanism of nitrate N and O isotope fractionation during denitrification is illustrated in Figure 6. Nitrate is actively transported into the cell from the periplasm via a nitrate-specific transporter that works against extant membrane potential (Alefounder and Ferguson 1980). Nitrate is thus concentrated inside the cell relative to external concentrations (Kucera et al. 1996), such that nitrate reductase (NAR) activity is presumably saturated. Nitrate diffusion through the pores of the bacterial peptidoglycan wall into the periplasm and subsequent uptake by a transporter likely do not impart significant isotope fractionation on nitrate because neither process involves 98 substantial disruption of the covalent bonds of the nitrate molecule (Melander and Saunders 1980). Isotopic fractionation of nitrate likely occurs during bond breakage at the catalytic site of NAR, an integral membrane protein whose catalytic site is oriented toward the cytoplasm. The intrinsic N and O isotope effects imparted on nitrate by NAR have not been determined specifically, but intrinsic N isotope effects measured for eukaryotic assimilatory nitrate reductases (eukNR) range between 15%o for spinach (Ledgard et al. 1985) to 30%o for Chlorella and maize (Schmidt and Medina 1991). The upper magnitudes observed among eukaryotic nitrate reductases are likely equivalent to those expected for NAR, since ~ 30%o corresponds to the highest N isotope effects observed for denitrifiers (Table 4.4). Propagation of the isotope effect imparted on nitrate by NAR into the external nitrate pool requires significant nitrate efflux from the cells. Nitrate efflux has been observed in higher plants and its putative role relates to the maintenance of cellular electrochemical balance or simply to cell leakiness (Glass 2003). Cellular nitrate efflux has not been specifically documented for denitrifiers, though its occurrence is mandatory to explain observations of nitrate isotope effects imparted by NAR (Mariotti et al. 1982; Shearer et al. 1991). Nitrate efflux by denitrifiers may serve to maintain electrochemical balance, may result from leakiness, or may be inadvertently transported from the cells along with nitrite. Nitrite, which is produced upon nitrate reduction by NAR, is highly toxic, and is thus assiduously effluxed from cells through specific transporters (Rowe et al. 1994). 4.4.1. Magnitude of the N and O isotope effects by respiratory denitrifiers We can interpret the observed isotope effects in our cultures on the basis of the fractionation mechanisms outlined above. In terms of the intrinsic fractionation imparted by NAR, this enzymatic isotope effect may differ among strains due to potential structural differences among enzymes, although the upper limit of 15e around 25%o - 30%o observed in culture and in the environment suggests that NAR imparts similar intrinsic isotope effects in spite of potential structural differences (Table 4.4). 99 Yet, the enzymatic expression of the isotope effect intrinsic to NAR may be variable within a strain depending on physiological state. For instance, Bryan et al. (1983) showed that the N isotope effect intrinsic to nitrite reductase (NiR) in vitro is sensitive to both the concentration of (internal) nitrite and to concentration of reductant provided to the enzyme. The N isotope effect decreased with reduced nitrite supply to the enzyme, while the N isotope effect increased when additional reductant was provided. The mechanistic explanation of Bryan et al. (1983) for this effect is that the expression of the enzyme isotope effect decreases with a decrease in the fraction of nitrate bound into the enzyme that is released back into solution, as opposed to undergoing reduction. In direct analogy, the isotope effect intrinsic to nitrate reductase could be sensitive to the internal nitrate concentration and to the availability of reductant to the enzyme. While the magnitude of the isotope effect is potentially sensitive to NAR and to the physiological determinants of NAR activity, the process that likely imparts the largest variations in the isotope effect is the ratio of cellular nitrate uptake to efflux, as it determines the extent to which internal nitrate is propagated to the external pool. Given this understanding, the asymptotic behaviour of the nitrate 5 1 5 N and 5 1 8 0 at lower nitrate concentrations (Figure 4.2a) is best explained as a reduction of cellular nitrate N and O isotope effects due to a decrease in nitrate efflux as a fraction of the gross nitrate influx. The point of inflection, which is apparent in the range of 3 to 6 pmol L"1 nitrate, is consistent with the half-saturation constant for nitrate uptake of ~5umol L"1, as determined for P. denitrificans (Parsonage et al. 1985). At these lower external nitrate concentrations, the supply of nitrate to the cell membrane no longer saturates transport, such that delivery of nitrate to NAR is reduced. As internal nitrate becomes depleted, nitrate efflux decreases, and the isotope effect no longer propagates to external nitrate. The inflection in the Rayleigh plots seen here for denitrifiers was not observed for cultures of marine eukaryotic phytoplankton (Granger et al. 2004). This likely reflects the 10-fold difference in the half-saturation constant for nitrate uptake of marine eukaryotic algae, which lies around ~ 0.4 umol L"1 (Berges and Harrison 1995). Given the exponential growth of the algal cultures, nitrate concentrations in this range were not captured by the samples collected. 100 The magnitude of the isotope effect is ostensibly sensitive to environmental conditions that could modulate the balance between cellular nitrate uptake and efflux. We tested whether salinity of the growth media modulated the nitrate N and O isotope effects, as the imposition of a different osmotic balance could potentially have modified relative rates of nitrate uptake and efflux. However the data indicate that changes in medium salinity did not result in consistent differences in nitrate N and O isotope effects. Aside from differences in media salinity, culture conditions were intended to be similar among experiments. Yet the observed variability in the N and O isotope effects of the respiratory denitrifiers suggests that nitrate uptake and efflux by the cells were sensitive to unforeseen differences in experimental conditions. For example, replicate cultures of P. stutzeri (Exp. 28) showed relatively low N isotope effects of around 5%o and 9%o at one time, then later fractionated at ~18%o among another set of replicates (Exp. 36; Table 4.2). Some of this variability could be linked to a potential sensitivity of the uptake-to-efflux ratio to growth rate, which may vary as a function of temperature. Temperature was not formally controlled in our experiments because cultures were grown on the bench top at ambient room temperature, which may have varied by ± 5°C between experiments. Mariotti et al. (1981) observed a 4%o decrease in the nitrate N isotope effects associated with denitrification during soil incubations at 30 °C compared to 20°C incubations. This could result from an increase in nitrate demand by NAR that exceeds the concomitant increase in nitrate uptake when temperature is increased, which would act to decrease the isotope effect imparted on internal nitrate by NAR, as well as decrease the relative efflux of nitrate relative to its uptake. Hence, temperature differences could have given rise to some of the observed variations in isotope effects between cultures of given strains grown at different times. For one culture of Ochrobacter sp. (Exp. 33 A) grown at relatively high initial nitrate (2.8 mmol L"1), nitrite accumulated in proportion to nitrate consumption in the culture (Table 1), such that nitrite toxicity may have changed the relative ratio of uptake to efflux. Nitrate uptake by denitrifiers is inhibited at high intracellular nitrite (Rowe et al. 1994), and nitrous acid can passively diffuse through the lipid bilayer, resulting in intracellular accumulation of nitrite. Thus, the relatively low 15e of 6%o in the identified 101 Ochrobacter sp. culture could be the result of inhibition of nitrate uptake due to the high nitrite. A final possibility for the modulation isotope effects by efflux and uptake could be oxygen contamination. Though the cells were cultured in gas-tight bags, repeated use of these may have introduced leaks of oxygen into some cultures. The expression of both the nitrate transporter and NAR is regulated by ambient oxygen. However, only the activity of the transporter is reversibly inhibited by oxygen, while that of NAR is insensitive to the presence of oxygen (Alefounder and Ferguson 1980; reviewed by Berks et al. 1995). Hence, oxygen contamination in our cultures could plausibly have lead to a lowering of the isotope effects due to diminished nitrate transport into the cells relative to NAR activity. Putative evidence of oxygen control on N isotope effects was observed in carbon-limited chemostats of P. denitrificans grown at different oxygen tension (Barford et al. 1999). At the highest experimental oxygen concentration of 1 pmol L"1, the N isotope effect imparted on nitrate reportedly dropped to 12%o at 1 pmol L"1 O2, from 28%o at O2 < 0.3 pmol L"1. Although the authors attributed this decrease to atmospheric N 2 contamination of the N2 gas product, computation of the N isotope effect from the isotopic composition of the initial and final nitrate 5 1 5N (instead of derivation from the isotopic composition of the N2 gas product) yields an isotope effect of ~7%o (calculated as outlined by Sigman and Casciotti 2001). Though we are unable to assess whether oxygen contamination accounts for any variation in our observed isotope effects, the putative role of oxygen in modulating expression of the isotope effect is of particular interest. There are only a few estimates of the isotope effect imparted on nitrate during denitrification in oxygen minimum zones of the ocean (Table 4.4). At the heart of the oxygen minimum, estimates of the N isotope effect imparted on nitrate by water-column denitrification range between 20-30%o. Given this isotope effect for water column denitrification, the observation of minimal expressed fractionation during sedimentary denitrification (Brandes and Devol 1997), and a modern mean ocean nitrate 5 1 5N of 5%o, the N isotope budget of the ocean requires that sedimentary denitrification dominates N loss from the ocean (Brandes and Devol 2002; Deutsch et al. 2004). This is problematic, as it suggests a rate of N loss from the ocean that is greater than current estimates of the input, which is dominated by N 2 fixation (Gruber and Sarmiento 1997). An 102 environmental control acting to reduce the isotope effect associated with nitrate loss from denitrification in the water column would lead to a more equal share of N loss by water column and sedimentary denitrification. This would lower the estimate of total N loss, bringing it closer to the N input estimates, thus helping to reconcile current discrepancies in the ocean N budget. A potential source of variation for nitrate isotope effects in our cultures that is not mediated by nitrate uptake and efflux is the differential expression of NAP activity among cultures. If the isotope effects measured here for R. sphaeroides are representative of nitrate fractionation by NAP in general (namely, 15e ~ 15%o and 18e ~ 9%o), then increased NAP activity would act to reduce net nitrate N and O isotope effects for nitrate consumption, as well as lowering the O-to-N isotope effect ratio. This possibility is discussed further in a later section. 4.4.2. Nitrate reduction by R. sphaeroides Nitrate reduction by R. sphaeroides in our cultures was ostensibly restricted to NAP activity. As mentioned earlier, the putative role of the soluble periplasmic NAP nitrate reductase is ascribed to the maintenance of cellular redox balance during growth (reviewed by Berks et al. 1995). NAP activity is sometimes associated with some proton motive force, yet it is insufficient to sustain growth. The extent to which NAP is expressed, or active, during photo-heterotrophic growth by R. sphaeroides and during chemo-heterotrophic growth by denitrifiers appears to depend on several environmental variables, including luminosity (in the case of R. sphaeroides only), nitrate concentration, oxygen tension, and the type(s) of reduced substrate(s) fuelling growth (Gavira et al. 2002). Hence, we are uncertain whether culture conditions in this study favoured the expression or the activation of NAP during aerobic or anaerobic growth of the respiratory denitrifiers. R. sphaeroides, however, seemingly respired available oxygen for growth, all the while shuttling excess electrons to NAP. Indeed, nitrate was rarely drawn to zero by R. sphaeroides cultures, unless air was deliberately added to the culture bags (data not shown). This evokes the association of NAP with so-called 'aerobic' denitrification (Robertson and Kuenen 1990). 103 Unlike NAR, NAP is located in the periplasm of the cells, such that N and O isotope effects are presumably «oraffected by the relative rates of cellular nitrate uptake and efflux. The supply of nitrate to the enzyme is determined by external nitrate concentrations, the rate of nitrate reduction in the periplasm, and the nitrate concentration gradient that thus develops, driving the diffusion of nitrate to the periplasm. Compared to some of the other experimental strains, both nitrate N and O isotope effects were less variable, roughly ~15%o and ~9%o, respectively, although they were not invariant. The observed variations in isotope effects cannot be ascribed to cellular uptake or efflux, and were thus dependent on the availability of substrate or reductant to the enzyme. The sensitivity of fractionation to substrate concentration was apparent between 12 and 30 umol L'1 nitrate, when 15e (and 18e) seemed to asymptote as nitrate concentrations decreased. This range in concentration is consistent with the half saturation constant of NAP estimated at 32 umol L"1 (Bursakov et al. 1997). The dampening of the N and O isotope effects then likely reflects substrate-limitation of enzymatic activity in the periplasm (Bryan et al. 1983). At saturating substrate concentrations, the observed variation in the magnitude of the N and O isotope effect among cultures might be due to variation in the supply of reductant to the enzyme. This could alter expression of the enzyme isotope effect by changing the fraction of enzyme-bound nitrate that is released back into solution without undergoing reaction (Bryan et al. 1983). 4.4.3. Coupling between nitrogen and oxygen fractionation of nitrate in the ocean The I 8£:1 5e ratio of ~ 0.95 observed here for the respiratory denitrifiers likely reflects that intrinsic to NAR, while the value of ~ 0.62 observed for R. sphaeroides corresponds to the ratio imprinted on nitrate by NAP. Interestingly, the N and O isotope coupling of respiratory denitrifiers is similar to that observed for nitrate assimilation by eukaryotic phytoplankton, which show a consistent l 8e: l 5£ of ~ 1.00 (Granger et al. 2004). The constancy of this coupling provides an important constraint that facilitates interpretation of coupled measurements of nitrate N and O isotopes in the water column. Such coupled measurements can serve as an oceanographic tracer of N-transformations, allowing identification of biological N-transformations that overprint one another in the 104 water column (Sigman etal. 2005). Arguably, the 1 8 E : 1 5 E observed here is not ~1 but rather < 1, as it ranged between 0.88 and 1.02. This inter-culture variation in 1 8 E : ' 5 E may stem from simultaneous nitrate reduction by both NAR and NAP. If we assume the 1 8 E : 1 5 E of NAR to be 1, then nitrate reduction by NAP could lower the ratio towards its putative intrinsic 1 8 E : 1 5 E of ~ 0.6. Differential expression of NAR vs. NAP by denitrifiers needs to be examined more closely to determine whether NAP modulates nitrate N and O isotope effects expressed by respiratory denitrifiers. 1 8 E : I 5 E ratios of less than 1 have thus far not been identified in marine denitrifying systems (Sigman et al. 2003, Sigman et al. 2005), suggesting that auxiliary denitrification by NAP is not a significant sink of nitrate in marine suboxic zones. This is consistent with the observation that, in oxygen minimum zones of the ocean, growth of denitrifiers is limited by the supply of organic substrates (Naqvi et al. 1982), such that reductant is presumably not in "excess," and electrons need not be shuttled to NAP. 4.4.4. The freshwater conundrum The apparent difference in 1 8 e : 1 5 E documented in marine versus freshwater systems, with ~1 in the former and ~ 0.6 in the latter, remains a puzzle. In our culture experiments, differences in medium salinity (and concomitantly pH) did not seem to change the observed coupling between N and O isotopes in our cultures of respiratory denitrifiers, and thus do not serve as an explanation for the difference. The 1 8 E : 1 5 E of ~ 0.6 reported for freshwater systems has thus far been attributed to isotopic fractionation imparted by respiratory denitrification. This hypothesis is reportedly supported by in vitro measurements of nitrate N and O isotope fractionation by a dissimilatory nitrate reductase showing coupled fractionation ( 1 8 E : I 5 E ) of 0.6 (Olleros (1983) as cited by Amberger and Schmidt 1987). However, the afore-mentioned work does not distinguish whether the measured nitrate N and O isotope effects were those intrinsic to NAP or NAR. Based on our findings, the 1 8 E : 1 5 E ratio observed in freshwaters is conspicuously similar to that observed here for the photo-heterotroph R. sphaeroides, and hence close to the 1 8 E : ' 5 E ratio of NAP. But NAP is not a respiratory enzyme per se, and hence not likely to be of great significance with respect to nitrate loss in terrestrial systems at low 105 oxygen tension. Thus, the discrepancy between the ratio observed here for denitrifiers possessing NAR (close to 1.0) to the 0.6 observed in freshwater is not reconcilable at this point. This conundrum presents an exciting avenue of research, as it may portend of unidentified yet fundamental differences between the terrestrial and marine N cycles. 4.4.5. Difference in isotope fractionation associated with denitrification and nitrate assimilation The fractionation mechanism proposed here for respiratory denitrifiers is akin to that for nitrate isotope fractionation during its assimilation by eukaryotic and prokaryotic algae (Granger et al. 2004; Needoba et al. 2004; Shearer et al. 1991; Wada and Hattori 1978). However, the isotope effects associated by denitrification versus assimilation tend to differ, in culture and in the environment. In culture, phytoplankton fractionate between 2 and 10%o (Montoya and McCarthy 1995; Needoba et al. 2003; Pennock et al. 1996; Waser et al. 1998; Waser et al. 1997), though some estimates range up to 15 and 20%o (Granger et al. 2004; Needoba and Harrison 2004; Wada and Hattori 1978). Field estimates similarly range between 4 and 10%o (Altabet 2001; Karsh et al. 2003; Sigman et at. 1999; Wu et al. 1997). In contrast, denitrification in culture is generally associated with isotope effects ranging between 20 and 30%o (this work; Table 4.4), albeit with significant variability towards lower values, while water column denitrification is associated with high isotope effects, between 20 and 30%o (Table 4.4). Assuming that the isotope effects intrinsic to the assimilatory and respiratory nitrate reductases are similar, denitrifiers then express the enzymatic isotope effect to a much greater extent. Based on our mechanistic model (Figure 4.6), this implies that denitrifiers are subject to a greater ratio of cellular nitrate efflux to uptake than assimilators. Although there is no definite explanation for this putative difference, the data in hand argue that it relates to the difference in the physiological function of nitrate reduction by denitrifiers versus that by assimilators. The supply of nitrate to phytoplankton at the surface ocean is often limiting, whereas at depths where water column denitrification occurs, nitrate is ample while organic substrates are limiting. A clear adaptation to this difference in nitrate supply is 106 reflected in the 10-fold lower half saturation constant for nitrate uptake in assimilators than denitrifiers. This portends of a trade-off between the affinity of the respective transporters for nitrate, versus the maximum achievable rates of nitrate uptake by the respective transporters (K vs. R selection). While assimilators may have evolved to acquire nitrate at low concentrations, denitrifiers are likely specialized in rapid nitrate acquisition for respiration while nitrate is ample. It follows that because nitrate can be limiting at the surface ocean, assimilators likely minimize the propensity of nitrate to leak out of the cell, ostensibly by sequestering it in vacuolar space (in the case of eukaryotes), and possibly by maintaining a lower cytosolic nitrate concentration than denitrifiers. Denitrifiers, in contrast, may harbour high internal nitrate concentrations to satisfy internal demand, and thereby become subject to leakage - although measurements of internal nitrate in P. denitrificans suggest that this is not the case, as intracellular concentrations have been shown to be relatively low (Kucera et al. 1996). To mitigate nitrate loss in the presence of a steep concentration gradient in nitrate between the cells interior and exterior, as well as a substantial electrochemical gradient that favours leakage of anions (Gradmann and Boyd 1995), cells that assimilate nitrate likely expend substantial energy to prevent its loss. Because nitrate is ample in environments where denitrifiers grow, they may simply forego the energetic expense of minimizing nitrate leakage as it would not confer a competitive advantage. While this analysis is arguably highly speculative, it remains that denitrifiers efflux nitrate to a much greater extent than assimilators, and that this differences is certainly related to the different functional use of nitrate by two groups. 4.4.6. Conclusions Respiratory denitrification fractionates the N and 0 atoms of nitrate with a systematic ratio of ~1. The consistency of this coupling is convenient because it establishes a robust constraint from which to interpret coupled measurements of nitrate N and O isotopes in the environment. While our data reveals some variability, with O-to-N fractionation ratios that fell slightly below 1, the ratio did not exceed 1. As such, positive deviations from a 1 8e:1 5e of 1 in the sub-oxic water column may be telling of N-107 transformations that overlap one another. Sigman et al. (2005) observed such positive deviations in the thermocline of the eastern North Pacific margin, which were interpreted as reflecting the isotopic signal of new nitrate from N-fixation in co-occurrence with the isotopic imprint of denitrification on nitrate. The peculiar coupling of nitrate N and O isotope fractionation during denitrification is identical to that observed for nitrate assimilation by eukaryotic phytoplankton in culture (Granger et al. 2004), as well as that seemingly associated with nitrate assimilation at the surface ocean (Casciotti et al. 2002). The constant coupling reflects the imprint of the nitrate reductase that is driving the isotope effects, namely, the imprint of the assimilatory nitrate reductase of phytoplankton and that of the dissimilatory NAR of denitrifiers. However, this similarity is somewhat unexpected, since the two enzymes are sufficiently different from each other to warrant classification in separate groups of molybdo-enzymes, namely the sulfite-oxidase family for the eukaryotic assimilatory enzyme, and the DMSO (dimethyl sulfoxide)-reductase family for NAR (Campbell 1999; Moreno-Vivian et al. 1999). The distinctive behaviour of the auxiliary NAP nitrate reductase thus stands in sharp contrast to that associated with other nitrate reductases, as it does not conform to the 1 8e: l 5e ratios of ~1 for both respiratory denitrification and nitrate assimilation by eukaryotic phytoplankton (Granger et al. 2004). This difference in the coupling of the N and O isotopes likely reflects a difference in the activation energy required for bond breakage at the catalytic sites of the respective enzymes, suggesting a fundamental difference between NAR and eukNR versus NAP in their respective mechanisms of nitrate reduction. This distinctive signature of NAP may prove useful to elucidate the putative role of aerobic denitrification in aquatic systems. Although N and O isotopes fractionated equally during denitrification, the magnitude of the coupled isotope effects is clearly mutable. The environmental dynamics that potentially underlie such variations remain undefined. Although we cannot incriminate specific factors to explain the variations observed in our cultures, our analysis provides hypotheses amenable to more rigorous investigation of putative controls on the magnitude of the N and O isotope effects imprinted on nitrate by denitrification in the environment. 108 Strain Type Habitat N A R N A P N A S Pseudomonas stutzeri (ATCC 14405) heterotrophic facultative anaerobe marine + +? + Ochrobacter sp.' heterotrophic facultative anaerobe marine +? ? ? Paracoccus denitrificans ( A T C C 19367) heterotrophic facultative anaerobe soil + + + Pseudomonas aureofaciens ( A T C C 13985)2 heterotrophic facultative anaerobe soil + + 9 Rhodobacter sphaeroides ( D S M 158) anoxigenic photo-heterotroph freshwater - + + Table 4.1. List of denitrifying strains used in this study: habitat of origin and types of nitrate reductase enzymes possessed by individual strains. NAR is a trans-membrane respiratory nitrate reductase, NAP an auxiliary periplasmic dissimilatory nitrate reductase, and NAS a cytosolic assimilatory nitrate reductase. 'Ochrobacter sp. was isolated from the Chesapeake estuary by B. Song. 2This strain of P. aureofaciens was recently re-classified as P. chlororaphis. 109 NO/ ,„„ N0 2 Strain Exp Medium (jimol L"') 1 5 E (%o) 18e (%o) 5,80:815N Ochrobacter sp. 26B SW 101 - 22.9 ± 1.7 21.1 ± 1.6 0.94 ± 0.02 33A SW 2779 + 6.7 ±0.7 6.8 ± 0.4 1.02 ±0.06 37A SW 510 - 21.8 ±2.9 20.2 ± 3.3 0.96 ± 0.03 37B SW 500 - 22.8 ±0.8 22.8 ± 1.9 1.02 ±0.03 40A SW 315 - 15.0 ±0.3 14.4 ± 0.6 0.98 ±0.01 40B SW 315 - 17.6 ±0.1 17.2 ± 0.2 1.00 ±0.01 P. aureofaciens 19A SW 208 + 20.5 ± 1.5 19.7 ± 0.4 0.94 ± 0.02 19B SW 166 + 23.0 ± 1.2 21.1 ± 1.1 0.97 ±0.01 21A SW 256 + 20.9 ± 0.9 18.9 ±0 .1 0.91 ± 0.03 21B SW 261 + 22.5 ± 1.7 20.8 ± 0.9 0.94 ± 0.04 48A FW 310 + 16.9 ±0.5 16.4 ±0.4 0.99 ±0.01 P. denitrificans 7A SW 202 + 17.6 ± 1.5 16.5 ± 1.4 0.95 ± 0.04 7B SW 180 + 20.0 ±2 .1 17.7 ± 2.9 0.92 ± 0.02 39A SW 315 + 24.8 ± 0.4 22.6 ±0 .4 0.92 ±0.01 39B SW 310 + 26.6 ± 0.5 22.5 ± 0.9 0.91 ± 0.05 30A FW 95 - 23.5 ± 0.6 20.7 ±0.3 0.90 ± 0.02 38B F W 314 - 18.8 ±0.3 17.9 ± 0.3 0.97 ±0.01 P. stutzeri 28A SW 89 + 9.7 ± 0.2 8.1 ±0.2 0.86 ± 0.04 28B SW 85 + 5.4 ± 0.3 4.8 ±0.1 0.92 ± 0.02 36A SW 500 + 19.7 ± 1.3 17-7 ±1.1 0.88 ± 0.02 36B SW 513 + 17.7 ± 1.8 15.6 ± 1.7 0.91 ± 0.02 R. sphaeroides 27A F W 100 + 12.6 ±0.2 7.9 ± 0.4 0.60 ± 0.02 27B F W 109 + 13.7 ±0.2 8.6 ±0.1 0.63 ±0.01 41A F W 303 + 14.7 ±0.3 8.7 ± 1.0 0.60 ± 0.02 41B F W 301 + 14.5 ±0.1 8.6 ± 0.3 0.61 ± 0.02 49A F W 325 + 18.1 ±0.2 10.3 ± 0.2 0.59 ± 0.03 49B F W 325 + 16.0 ± 0.2 8.91 ±0.1 0.57 ± 0.00 49C F W 290 + 15.7 ± 0.4 9.5 ±0 .1 0.61 ± 0.00 49D F W 328 + 14.8 ± 2.7 9.7 ±2.3 0.68 ± 0.04 50A SW 295 + 15.9 ±0.2 8.9 ± 0.2 0.58 ± 0.03 50B SW 285 + 19.9 ± 0.2 13.1 ±0.2 0.66 ± 0.06 Table 4.2. The caption is on following page. 110 Table 4.2. Nitrate N and O isotope effects (15e and 1 8e, respectively) computed for individual experiments with 5 denitrifying bacterial strains, and corresponding 5 1 8 0:5 I 5 N relationship. 15e and 1 8 E were derived from the slopes of individual regression analyses (Equations 3 and 4), and are reported ± the standard deviation associated with respective slopes. Similarly, 5 I 8 0:5 1 5 N refers the slope of linear regression o fS 1 8 0 versus the corresponding 5 , 5 N and the standard deviation associated with the slope. Note that some estimates of 15e and l s O were no longer linear at relatively low nitrate concentrations (Figures 4.2 and 4.4), such that asymptotic points were omitted from respective regression analyses. I l l strain , 5 E Fo.05(l), k-l,N-k P ,8e Fo.05(l), k-l,N-k P v 8 , s O : 8 , 5 N Pooled Fo.os(i), k-i,N-k P slope Ochromonas sp. P. denitrificans P. aureofaciens P. stutzeri R. sphaeroides^ 48.07 ,5 ,9) < 0 .01* 18.27 (5,20) < 0 . 0 1 * 6.47 (4, , 6 ) 0 .04* 68.66 (3,13, < 0 .01* 6 . 1 0 ( 8 , 2 8 ) < 0 . 0 1 * 17.97(5,,o) < 0 .01* 10.60(5,20) < 0 .01* 9.42 (4, , 6 ) 0 .02* 57.19 (3, , 3 ) < 0 . 0 1 * 1-86 (8,29, 0.10 0.20 ( 5 , 1 0 ) n.s. 0.98 3.26 (5,23) n.s. 0.96 1.65(4,17) n.s. 0.97 2.23 (j, l 8 ) n.s. 0.88 0.90(8,31) n.s. 0.62 Table 4.3. Analyses of covariance (ANCOVA) to test for homogeneity among regression slopes. The critical F value was computed for all the regression functions among individual strains to derive the probability, P, that the slopes are equal. We tested for strain-specific differences in the values derived for e, e, and 5 0:5 N, respectively. Strains for which significant differences were detected among slopes are identified with a star (*), indicating that the probability of all the slopes in the group being equal is less than 5%. 'Experiment 49D was removed from the statistical analyses because the large error associated with l 5e and l 8e obscured potentially significant differences among the remaining experiments. Not significantly different = n.s. 112 E x p e r i m e n t a l s y s t e m C o n d i t i o n s l 5 £ ( % o ) R e f e r e n c e Pseudomonas stutzeri B a t c h c u l t u r e 2 0 - 3 0 ( W e l l m a n e t a l . 1 9 6 8 ) Paracoccus denitrificans B a t c h c u l t u r e 1 3 - 2 0 ( D e l w i c h e a n d S t e y n 1 9 7 0 ) Paracoccus denitrificans C h e m o s t a t 0 2 < 0 .3 u M 2 8 . 6 ± 1.9 ( B a r f o r d e t a l . 1 9 9 9 ) Paracoccus denitrificans C h e m o s t a t 02 ~ 1.2 u M 12 ( B a r f o r d et a l . 1 9 9 9 ) M a r i n e d e n i t r i f y i n g b a c t e r i a 1 4 - 2 1 ( M i y a k e a n d W a d a 1 9 7 1 ) F r e s h w a t e r d e n i t r i f i e r 2 - 1 2 ( W a d a e t a l . 1 9 7 5 ) S o i l G l u c o s e a d d e d 1 4 - 2 3 ( B l a c k m e r a n d B r e m n e r 1 9 7 7 ) S o i l 2 0 ° C 2 9 . 4 ± 2 . 4 ( M a r i o t t i e t a l . 1 9 8 1 ) S o i l 3 0 ° C 2 4 . 6 ± 0 .9 ( M a r i o t t i e t a l . 1 9 8 1 ) G r o u n d w a t e r K a l a h a r i 3 0 ± 6 ( V o g e l e t a l . 1 9 8 1 ) G r o u n d w a t e r C h a l k a q u i f e r 4 - 8 ( M a r i o t t i e t a l . 1 9 8 8 ) G r o u n d w a t e r G r a v e l l y s a n d s 15 .9 ( B o t t c h e r et a l . 1 9 9 0 ) G r o u n d w a t e r S e p t i c s a n d s 2 2 . 9 ( A r a v e n a a n d R o b e r t s o n 1 9 9 8 ) E a s t e r n T r o p i c a l N o r t h P a c i f i c 2 0 - 3 0 ( B r a n d e s e t a l . 1 9 9 8 ) E a s t e r n T r o p i c a l N o r t h P a c i f i c 3 0 ( V o s s et a l . 2 0 0 1 ) C e n t r a l A r a b i a n Sea 2 2 - 2 5 ( B r a n d e s e t a l . 1 9 9 8 ) M a r i n e s e d i m e n t 0 - 3 ( B r a n d e s a n d D e v o l 1 9 9 7 ) Table 4.4. Summary of existing values measured for N isotope effects imparted on nitrate by denitrification in laboratory cultures and in environmental samples. 113 2 0 0 ^ 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Days Figure 4.1. Change in the nitrate and nitrite concentrations in a growing culture of P. denitrificans, from the time of inoculation. 114 1000 Nitrate (umol L' ) 100 10 J IIII 11 i i i 1 • Ochrobacter sp. (n = 6) A P. denitrificans sw (n = 4) A P. denitrificans fw (n = 2) * P. denitrificans sw (n = 4) * P. aureofaciens fw (n = 1)/ * P. stutzeri (n = 4) " i — i — i — r 5 4 3 2 In ([NO,"]) 80 -r 60-^ 40-20-04 i r -1 -2 -3 -4 -5 ln([N03]/[N03- i n i U a ]]) Figure 4.2. Rayleigh plots of the change in the N isotopic composition of nitrate (615N) as a function of the natural logarithm of nitrate use for four strains of respiratory denitrifiers grown in freshwater (fw) and seawater (sw) media, a) 5 I 5N is plotted over the In of nitrate concentrations (reverse scale) to highlight the asymptotic behaviour of nitrate N isotopic fractionation at lower nitrate concentrations, b) 61 5N is plotted over the In of fractional nitrate use (reverse scale) to show the range in N isotope effects (slopes) observed for the experimental strains. 115 0 20 40 60 A5 1 5 N of NO," (%o vs. starting value) Figure 4.3. The 5 1 8 0 of nitrate plotted against the corresponding 5 1 5 N of nitrate for cultures of four strains of respiratory denitrifiers grown in freshwater and seawater media. The pooled slope of all measurements yields a value of 0.95. 116 25 •a 2<H 15 ^ d" to-z o z p o sphaeroides ••-<>••• freshwater seawater 25 -r 20 H \5-\ 1<M 5 4 3 ln ([N03D P ' ,0 " 3e= 15 %c b n 1 1 1 0-0 -0.5 -1.0 -1.5 -2.0 -2 5 ln([N03-]/[N03-milial]) Figure 4.4. Rayleigh plots of the change in the N isotopic composition of nitrate (815N) as a function of the natural logarithm of nitrate use for freshwater and seawater cultures of the photo-heterotroph R. sphaeroides. a) 6 I 5 N is plotted over the ln of nitrate concentrations to highlight the asymptotic behaviour of nitrate N isotopic fractionation at lower nitrate concentrations, b) 5 1 5N is plotted over the ln of fractional nitrate use to show the cluster in N isotope effects (slopes) observed for R. sphaeroides. 117 25 H •3 2 0 ^ 15 -A KM 5-A 0-1 o freshwater • seawater 1:1 / .Q,0 ' / .••''..•il' 5 1 8 0 :6 , 5 N = 0.6 R. sphaeroides 1 1 0 5 I I I I 10 15 20 25 A6 1 5 N of NO," (%c vs. starting value) Figure 4.5. The 5 I 8 0 of nitrate plotted against the corresponding 6 1 5 N of nitrate for cultures R. sphaeroides grown in freshwater and seawater media. The pooled slope of all measurements yields a value of 0.62. 118 Figure 4.6. The putative mechanisms of N (and O) isotope fractionation of nitrate during denitrification mediated by either NAR or NAP. Nitrate (a) diffuses through the peptidoglycan wall into the periplasm, (b) Nitrate is actively transported into the cell where it is concentrated relative to external nitrate, (c) Internal nitrate can be reduced to nitrite by the membrane-bound NAR. NAR fractionates the N and O atoms of nitrate, such that intracellular nitrate becomes isotopically enriched, (d) Isotopically enriched nitrate can efflux from the cell back into the periplasm, and diffuse out of the periplasm into the external medium, (e) NAP is located in the periplasm and it can fractionate the N and O isotopes of periplasmic nitrate. 119 4.5. References Alefounder, P. R., and S. J. Ferguson. 1980. The location of dissimilatory nitrite reductase and the control of dissimilatory nitrate reductase by oxygen in Paracoccus-denitrificans. Biochem. J. 192: 231-240. Altabet, M. A. 2001. Nitrogen isotopic evidence for micronutrient control of fractional NOV utilization in the equatorial Pacific. Limnol. Oceanogr. 46: 368-380. Amberger, A., and H.-L. Schmidt. 1987. Naturliche isotopengehalte von nitrat als Indikatoren fur dessen herkunft. Geochim. Cosmochim. Ac. 51: 2699-2705. Aravena, R., and W. D. Robertson. 1998. Use of multiple isotope tracers to evaluate denitrification in ground water: Study of nitrate from a large-flux septic system plume. Ground Water 36: 975-982. Barford, C. C , J. P. Montoya, M. A. Altabet, and R. Mitchell. 1999. Steady-state Nitrogen isotope effects of N2 and N2O production in Paracoccus denitrificans. Appl. Environ. Microbiol. 65: 989-994. Berges, J. A., and P. J. Harrison. 1995. Nitrate reductase-activity quantitatively rredicts the rate of nitrate incorporation under steady-state light limitation - a revised assay and characterization of the enzyme in 3 species of marine-phytoplankton. Limnol. Oceanogr. 40: 82-93. Berks, B. C , S. J. Ferguson, J. W. B. Moir, and D. J. Richardson. 1995. Enzymes and associated electron transport systems that catalyse the respiratory reduction of nitrogen oxides and oxyanions. Bba-Bioenergetics 1232: 97-173. Blackmer, A. M., and J. M. Bremner. 1977. Nitrogen Isotope Discrimination in Denitrification of Nitrate in Soils. Soil Biol. Biochem. 9: 73-77. Bohlke, J. K., S. J. Mroczkowski, and T. B. Coplen. 2003. Oxygen isotopes in nitrate: new reference materials for 0-18 : 0-17 : 0-16 measurements and observations on nitrate-water equilibration. Rapid. Commun. Mass. Sp. 17: 1835-1846. Bottcher, J., O. Strebel, S. Voerkelius, and H. L. Schmidt. 1990. Using isotope fractionation of nitrate nitrogen and nitrate oxygen for evaluation of microbial denitrification in a sandy aquifer. J. of Hydrol. 114: 413-424. 120 Braman, R. S., and S. A. Hendrix. 1989. Nanogram nitrite and nitrate determination in environmental and biological materials by V(III) reduction with chemiluminescence detection. Anal. Chem. 61: 2715-2718. Brandes, J. A., and A. H. Devol. 1997. Isotopic fractionation of oxygen and nitrogen in coastal marine sediments. Geochim. Cosmochim. Ac. 61: 1793-1801. Brandes, J. A., and A. H. Devol. 2002. A global marine fixed nitrogen isotopic budget: Implications for Holocene nitrogen cycling. Global Biogeochem. Cy. 16: 10.1029/2001GB001856. Brandes, J. A., A. H. Devol, T. Yoshinari, D. A. Jayakumar, and S. W. A. Naqvi. 1998. Isotopic composition of nitrate in the central Arabian Sea and eastern tropical North Pacific: a tracer for mixing and nitrogen cycles. Limnol. Oceanogr. 43: 1680-1689. Bryan, B. A., G. Shearer, J. L. Skeeters, and D. H. Kohl. 1983. Variable expression of the nitrogen isotope effect associated with denitrification of nitrite. J. Biol. Chem. 258: 8613-8617. Bursakov, S. A., C. Carneiro, M. J. Almendra, R. O. Duarte, J. Caldeira, I. Moura, and J. J. Moura. 1997. Enzymatic properties and effect of ionic strength on periplasmic nitrate reductase (NAP) from Desulfovibrio desulfuricans ATCC 2111 A. Biochem. Biophys. Res. Commun. 239: 816-822. Campbell, W. H. 1999. Nitrate reductase structure, function and regulation: bridging the gap between biochemistry and physiology. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50: 277-303. Casciotti, K. L , D. M. Sigman, M. G. Hastings, J. K. Bohlke, and A. Hilkert. 2002. Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater using the denitrifier method. Anal. Chem. 74: 4905-4912. Delwiche, C. C , and P. L. Steyn. 1970. Nitrogen isotope fractionation in soils and microbial reactions. Env. Sci. Tech. 4: 929-935. Deutsch, C , D. M. Sigman, R. C. Thunell, A. N. Meckler, and G. H. Haug. 2004. Isotopic constraints on glacial/interglacial changes in the oceanic nitrogen budget. Global Biogeochem. Cy. 18: GB4012. 121 Ferguson, S. J., J. B. Jackson, and A. G. Mcewan. 1987. Anaerobic respiration in the Rhodospirillacea: characterization of pathways and evaluation of roles in redox balancing during photosynthesis. FEMS Microbiol. Rev.: 117-143. Gavira, M., M. D. Roldan, F. Castillo, and C. Moreno-Vivian. 2002. Regulation of nap gene expression and periplasmic nitrate reductase activity in the phototrophic bacterium Rhodobacter sphaeroides DSM158. J. Bact. 184: 1693-1702. Glass, A. M. 2003. Nitrogen use efficiency of crop plants: physiological constraints upon nitrogen absorption. Crit. Rev. Plant Sci. 22: 453-470. Gonfiantini, R., W. Stichler, and K. Rosanski. 1995. Standards and Intercomparison Materials Distributed by the IAEA for Stable Isotope Measurements. International Atomic Energy Agency. Gradmann, D., and C. M. Boyd. 1995. Membrane voltage of marine-phytoplankton, measured in the diatom Coscinodiscus radiatus. Mar. Biol. 123: 645-650. Granger, J., D. M. Sigman, J. A. Needoba, and P. J. Harrison. 2004. Coupled nitrogen and oxygen isotope fractionation of nitrate during assimilation by cultures of marine phytoplankton. Limnol. Oceanogr. 49: 1763-1773. Granger, J., D. M. Sigman, M. D. Prokopenko, M. F. Lehmann, and P. D. Tortell. in press. A method for nitrite removal in nitrate N and O isotope analyses. Limnol. Oceanogr.: Methods 4: 202-212. Granger, J., and B. B. Ward. 2003. Accumulation of nitrogen oxides in copper-limited cultures of denitrifying bacteria. Limnol. Oceanogr. 48: 313-318. Gruber, N., and J. L. Sarmiento. 1997. Global patterns of marine nitrogen fixation and denitrification. Global Biogeochem. Cy. 11: 235-266. Karsh, K. L., T. W. Trull, A. J. Lourey, and D. M. Sigman. 2003. Relationship of nitrogen isotope fractionation to phytoplankton size and iron availability during the Southern Ocean Iron RElease Experiment (SOIREE). Limnol. Oceanogr. 48: 1058-1068. Keller, M. D., W. K. Bellows, and R. R. L. Guillard. 1988. Microwave treatment for sterilization of phytoplankton culture media. J. Exp. Mar. Biol. Ecol. 117: 279-283. 122 Kucera, I., P. Kaplan, and A. Zeman. 1996. Oxygen increases the steady-state level of nitrate in denitifying cells of Paracoccus denitrificans. FEMS Microbiol. Letters: 163-166. Ledgard, S. F., K. C. Woo, and F. J. Bergersen. 1985. Isotopic fractionation during reduction of nitrate to nitrite by extracts of spinach leaves. Aust. J. Plant Physiol. 12: 631-640. Lehmann, M. F., P. Reichert, S. M. Bernasconi, A. Barbieri, and J. A. Mckenzie. 2003. Modelling nitrogen and oxygen isotope fractionation during nitrate reduction in a hypolimnetic redox transition zone. Geochim. Cosmochim. Ac. 67: 2529-2542. Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2004. Coupling the N-15/N-14 and 0- 18/0-16 of nitrate as a constraint on benthic nitrogen cycling. Mar. Chem. 8 8 : 1- 20. Lehmann, M. F., D. M. Sigman, D. C. Mccorkle, B. G. Brunelle, S. Hoffmann, M. Kienast, G. Cane, and J. Clement. 2005. Origin of the deep Bering Sea nitrate deficit: Constraints from the nitrogen and oxygen isotopic composition of water column nitrate and benthic nitrate fluxes. Global Biogeochem. Cycles 19: GB4005. Mariotti, A., J. C. Germon, P. Hubert, P. Kaiser, R. Letolle, A. Tardieux, and P. Tardieux. 1981. Experimental determination of nitrogen kinetic isotope fractionation: some principles; illustration for the denitrification and nitrification processes. Plant Soil 62: 413-430. Mariotti, A., A. Landreau, and B. Simon. 1988. 1 5 N isotope biogeochemistry and natural denitrification process in groundwater: Application to the chalk aquifer of northern France. Geochim. Cosmochim. Ac. 52: 1869-1878. Mariotti, A., F. Mariotti, M. L. Champigny, N. Amarger, and A. Moyse. 1982. Nitrogen isotope fractionation associated with nitrate reductase-activity and uptake of NO3" by Pearl-Millet. Plant Physiol. 69: 880-884. Mcllvin, M. R., and M. A. Altabet. 2005. Chemical conversion of nitrate and nitrite to nitrous oxide for nitrogen and oxygen isotopic analysis in freshwater and seawater. Anal. Chem. 77: 5589-5595. 123 Melander, L., and W. H. J. Saunders. 1980. Reaction Rates of Isotopic Molecules. John Wiley. Miyake, Y., and E. Wada. 1971. The isotope effect on the nitrogen in biochemical oxidation-reduction reactions. Rec. Oceanogr. Works Japan 11: 1-6. Montoya, J. P., and J. J. McCarthy. 1995. Isotopic fractionation during nitrate uptake by phytoplankton grown in continuous-culture. J. Plankton Res. 17: 439-464. Moreno-Vivian, C , P. Cabello, M. Martinez-Luque, R. Blasco, and F. Castillo. 1999. Prokaryotic nitrate reduction: Molecular properties and functional distinction among bacterial nitrate reductases. J. of Bact. 181: 6573-6584. Naqvi, S. W. A., R. J. Noronha, and C. V. G. Reddy. 1982. Denitrification in the Arabian Sea. Deep-Sea Res. 29: 459-469. Needoba, J. A., and P. J. Harrison. 2004. Influence of low light and a light: Dark cycle on N03- uptake, intracellular N03-, and nitrogen isotope fractionation by marine phytoplankton. J. Phycol. 40: 505-516. Needoba, J. A., N. A. Waser, P. J. Harrison, and S. E. Calvert. 2003. Nitrogen isotope fractionation in 12 species of marine phytoplankton during growth on nitrate. Mar. Ecol.-Prog. Ser. 255: 81-91. Parsonage, D., A. J. Greenfield, and S. J. Ferguson. 1985. The high-affinity of Paracoccus-Denitrificans cells for nitrate as an electron-acceptor - analysis of possible mechanisms of nitrate and nitrite movement across the plasma-membrane and the basis for inhibition by added nitrite of oxidase activity in permeabilized cells. Biochim. Biophys. Acta 807: 81-95. Parsons, T. R., Y. Maita, and L. C. M. 1984. A manual of biological and chemical methods for seawater analysis. Publ. Pergamon Press. Pennock, J. R., D. J. Velinsky, J. H. Sharp, J. Ludlam, and M. L. Fogel. 1996. Isotope fractionation of ammonium and nitrate during their uptake by Skeletonema costatum:. Implications for the dl5N dynamics under bloom conditions. Limnol. Oceanogr. 41: 451-459. Price, N. M., G. I. Harrison, J. G. Herring, R. J. Hudson, P. M. V. Nirel, B. Palenik and F. M. M. Morel. 1988/1989. Preparation and chemistry of the artificial algal culture medium Aquil. Biol. Oceanogr. 6: 443-461. 124 Robertson, L. A., and J. G. Kuenen. 1990. Combined heterotrophic nitrification and aerobic denitrification in Thiosphaera-Pantotropha and other bacteria. Antonie Van Leeuwenhoek Int. J. Gen. Mol. Microbiol. 57: 139-152. Rowe, J. J., T. Ubbinkkok, D. Molenaar, W. N. Konings, and A. J. M. Driessen. 1994. Nark is a nitrite-extrusion system involved in anaerobic nitrate respiration by Escherichia-Coli. Mol. Microbiol. 12: 579-586. Schmidt, H. L., and R. Medina. 1991. Possibilities and scope of the double isotope effect method in the elucidation of mechanisms of enzyme catalyzed-reactions. Isotopenpraxis 27: 1-4. Shearer, G., J. D. Schneider, and D. H. Kohl. 1991. Separating the efflux and influx components of net nitrate uptake by Synechococcus-R2 under steady-state conditions. J. Gen. Microbiol. 137: 1179-1184. Sigman, D. M., M. A. Altabet, R. Francois, D. C. McCorkle, and G. Fischer. 1999. The 6 1 5N of nitrate in the Southern Ocean: Consumption of nitrate in surface waters. Global Biogeochem. Cy. 13: 1149-1166. Sigman, D. M., and K. L. Casciotti. 2001. Nitrogen isotopes in the ocean, p. 1884-1894. In J. H. Steele, K. K. Turekian and S. A. Thorpe [eds.], Encyclopedia of Ocean Sciences. Academic Press. Sigman, D. M., K. L. Casciotti, M. Andreani, C. Barford, M. Galanter, and J. K. Bohlke. 2001. A bacterial method for the nitrogen isotopic analysis of nitrate in seawater and freshwater. Anal. Chem. 73: 4145-4153. Sigman, D. M., J. Granger, P. J. DiFiore, M. F. Lehmann, R. Ho, G. Cane, and A. Van Geen. 2005. Coupled nitrogen and oxygen isotope measurements of nitrate along the eastern North Pacific margin. Global Biogeochem. Cy. 19: GB4022. Sigman, D. M., R. Robinson, A. N. Knapp, A. Van Geen, D. C. McCorkle, J. A. Brandes, and R. C. Thunell. 2003. Distinguishing between water column and sedimentary denitrification in the Santa Barbara Basin using the stable isotopes of nitrate. Geochemistry, Geophysics, Geosystems 4: 1040-1059. Silva, S. R, C. Kendall, D. H. Wilkinson, A. C. Ziegler, C. C. Y. Chang, and R. J. Avanzino. 2000. A new method for collection of nitrate from fresh water and the analysis of nitrogen and oxygen isotope ratios. J. Hydrol. 228: 22-36. 125 Vogel, J. C , A. S. Talma, and T. H. E. Heaton. 1981. Gaseous nitrogen as evidence for denitrification in groundwater. J. Hydrol. 50: 191-200. Voss, M., J. W. Dippner, and J. P. Montoya. 2001. Nitrogen isotope patterns in the oxygen-deficient waters of the Eastern Tropical North Pacific Ocean. Deep-Sea Res. 48: 1905-1921. Wada, E., and A. Hattori. 1978. Nitrogen isotope effects in the assimilation of inorganic nitrogenous compounds. Geomicrobiol. J. 1: 85-101. Wada, E., T. Kadonaga, and S. Matsuo. 1975. 1 5 N abundance in nitrogen of naturally occuring substances and global asswessment of denitrification from isotopic viewpoint. Geochem. Jo. 9: 139-148. Waser, N. A., K. D. Yin, Z. M. Yu, K. Tada, P. J. Harrison, D. H. Turpin, and S. E. Calvert. 1998. Nitrogen isotope fractionation during nitrate, ammonium and urea uptake by marine diatoms and coccolithophores under various conditions of N availability. Mar. Ecol.-Prog. Ser. 169: 29-41. Waser, N. A. D., J. Needoba, S. E. Calvert, and P. J. Harrison. 1997. Nitrogen isotope fractionation by 4 groups of marine microalgae during batch culture growth on nitrate, p. 337, ASLO Aquatic Sciences Meeting. Weaver, P. F., J. D. Wall, and H. Gest. 1975. Characterisation of Rhodopseudomonas capsulata. Arch. Microbiol. 105. Wellman, R. P., F. D. Cook, and H. R. Krouse. 1968. Nitrogen-15: microbiological alteration of abundance. Science 161: 269-270. Wu, J., S. E. Calvert, and C. S. Wong. 1997. Nitrogen isotope variations in the subarctic Pacific northeast Pacific: relationships to nitrate utilization and trophic structure. Deep-Sea Res. Part I 44: 287-314. 126 Chapter 5 The fractionation of nitrate N and O isotopes during its assimilation by prokaryotic and eukaryotic phytoplankton A version of this chapter wil l be submitted for publication. Granger, J., D. M . Sigman, and P. D. Tortell. In prep. The fractionation of nitrate N and O isotopes during its assimilation by prokaryotic and eukaryotic phytoplankton. J. Phycol. 1 2 7 5.1. Introduction Natural abundance stable isotopes provide a powerful tracer to elucidate N biogeochemical cycling on land and in the sea. Nitrate stands among the more abundant species of bioavailable nitrogen, and it is well-suited for investigation with stable isotopes because nitrate molecules have two elements that can potentially be fractionated, namely nitrogen and oxygen. The N isotopes of nitrate have served as a geochemical tracer to determine biological sources and sinks of nitrate, from N-fixation, nitrate assimilation, and water-column or sedimentary denitrification (reviewed by Sigman and Casciotti 2001). Each of these processes leaves a distinct isotopic imprint on nitrate, such that the biological process implicated in the addition or removal of nitrate can be identified. Until recently, only N isotopes of nitrate were measured in marine systems due to limitations in methodology (e.g., Silva et al. 2000). However, novel methods now enable accurate quantification of both the N and O isotope ratios of nitrate in relatively small freshwater or saline samples (Casciotti et al. 2002; Mcllvin and Altabet 2005; Sigman et al. 2001). The emergence of these methods has yielded coupled measurements of nitrate N and O isotope ratios ( 1 5N/ 1 4N and 1 8 0 / ' 6 0 , respectively) in various ocean systems (Casciotti et al. 2002; Lehmann et al. 2004; Lehmann et al. 2005; Sigman et al. 2005). The utility of coupled measurements stems from the distinct biogeochemical origin of the N and O atoms of nitrate within the N cycle; the N atom originates from N-fixation, while the O atoms enter the N-cycle via nitrification (Casciotti et al. 2002). Coupled measurements of nitrate N and O isotopes thus allow for separation of biological N transformations whose N isotopic signals over-print one another in a body of water. For instance, coupled measurements of nitrate N and O isotopes helped identify relative input of new nitrate to the thermocline from N-fixation versus in situ loss from denitrification at the eastern North Pacific margin (Sigman et al. 2005). Coupled measurements have also served to separate nitrification from sedimentary denitrification (Lehmann et al. 2004; Lehmann et al. 2005). Efforts to interpret field measurements of nitrate N and O isotope ratios have led to studies of nitrate N and O isotope fractionation in mono-cultures during nitrate 128 assimilation by marine unicellular eukaryotic algae (Granger et al. 2004), as well as during denitrification by marine and freshwater denitrifiers (Chapter 4). From these studies emerged the unexpected finding that nitrate N and O isotopes are fractionated to the same extent, such that nitrate N and O isotope ratios co-vary linearly with a ratio of 1 during either assimilation or respiratory denitrification. Studies to date implicate bond breakage by nitrate reductase as the dominant fractionating step during assimilation or denitrification (Granger et al. 2004; Needoba et al. 2004; Shearer et al. 1991; Wada and Hattori 1978). Nitrate reductase is thus the major driver of N and O isotope effects imparted on nitrate, while the extent to which the N and O isotope effects are propagated extracellularly is determined by ratio of cellular nitrate uptake and nitrate efflux from the cell (Needoba et al. 2004; Shearer et al. 1991). Hence, the coupling between N and O isotope effects during assimilation and denitrification reflects the isotopic signature of the respective nitrate reductases involved. The eukaryotic assimilatory nitrate reductase (eukNR) and the prokaryotic respiratory nitrate reductase (NAR) thus impart nitrate N and O isotope effects that co-vary with a ratio of 1 (i.e, 18e=15e), whereas the auxiliary prokaryotic nitrate reductase (NAP) has been shown to impart an O-to-N ratio of 0.6 (Granger et al. 2004; Chapter 4). Nitrate N and O isotope fractionation during assimilation by prokaryotes, which is mediated by the cytosolic nitrate reductase NAS, has not been investigated. While all nitrate reductases are molybdo-enzymes, they share little sequence similarity among clades and functional enzyme types, and also differ with respect to the coordination sphere at the Mo active site (Table 5.1). The most closely related nitrate reductases are the periplasmic NAP and the cytosolic NAS of prokaryotes that share quasi-identical coordination spheres around the Mo atom at the catalytic site (Table 5.1). The structural similarity between NAS and NAP could translate to a similar O-to-N isotope coupling of 0.6 for NAS. Nitrate consumption in the ocean mixed layer is not restricted to eukaryotic phytoplankton but is also effectuated by prokaryotes (Allen et al. 2002 and references therein). Hence, it is important to verify N and O isotopic coupling associated with nitrate assimilation by prokaryotes in order to interpret measurements of nitrate N and O isotope distributions at the surface ocean, where nitrate is actively consumed by both eukaryotic and prokaryotic plankton. 129 In this study, we present measurements of nitrate N and O isotopic fractionation during nitrate assimilation by autotrophic prokaryotic plankton. We also measure nitrate N and O isotope ratios in growing cultures of eukaryotes, namely two species of marine diatoms, and a marine and a freshwater chlorophyte, in order to verify that the 1:1 coupling postulated for the eukaryotic nitrate reductase extends to other species and clades of unicellular eukaryotes. 5.2. Methods We conducted experiments with three stains of marine cyanobacteria, a freshwater chlorophyte, a marine chlorophyte, and two marine diatom species (Table 1). Marine isolates were grown in AQUIL seawater medium (Price 1988/1989). AQUIL salts were supplemented with 300 umol L"1 nitrate, 10 umol L"' phosphate, and 100 umol L"1 silicate. Media were prepared in acid-washed polycarbonate bottles and sterilized by microwaving (Keller et al. 1988). These were then supplemented with filter-sterilized AQUIL EDTA- (ethylenediaminetetraacetic acid) trace metals and f/2 vitamins. Copper was omitted from cyanobacterial media as even small addition proved toxic and stunted growth. The diatom Pseudonitzchia hemii, a recent field isolate (Table 1), proved to be fastidious as it only grew in filtered seawater (collected at Station PAPA in the subArctic Pacific), which we supplemented with the above additions of nutrients, EDTA trace metals and vitamins. The freshwater chlorophyte, Chlorella pyrenoidosa, was cultured in medium consisting of de-ionized water amended with 5 mmol L~' KH2PO4, 5 mmol L"1 K 2 HP0 4 , 10 mmol L"1 Mg 2 S0 4 -7H 2 0, 0.5 mmol L"' CaCl2«2H20, 11 umol L'1 H 2 B0 3 , and supplemented with 300 umol L"' nitrate, AQUIL EDTA-trace metals and f/2 vitamins. The diatoms and chlorophytes were grown under continuous light at 20°C. The cyanobacteria cultures failed to grow under continuous light and were thus grown on the bench-top, illuminated by a single neon bulb, and subject to daily changes in light intensity and temperature in the lab. Throughout growth of the cultures, 20 mL samples were collected for nitrate N and O isotope analysis. These were filtered through a combusted AE glass fiber filter to 130 remove the cells, and filtrates were stored in acid-washed polypropylene bottles that were frozen until nitrate analysis and subsequent nitrate N and O isotope analyses. Nitrate was measured by NO (nitric oxide) chemiluminescence detection on an Antek 1750 nitrate/nitrite analyzer (Braman and Hendrix 1989). We also checked for the presence of nitrite in the samples, which was measured colourimetrically (Parsons et al. 1984). Prior to nitrate N and O isotope analysis, nitrite was removed from the samples with ascorbate (Granger et al. 2006) because it interferes with the nitrate isotopic measurements. However, samples in which nitrite was detected after prolonged storage in the freezer were discarded, because nitrite decomposition and the consequent formation of nitrate compromised the isotopic measurements. In later samples, nitrite was measured upon sample collection, and it was removed immediately with ascorbate in order to avoid this problem. The 1 5 N/ 1 4 N and 1 8 0 / l 6 0 of nitrate were determined following the denitrifier method (Casciotti et al. 2002; Sigman et al. 2001). Isotope ratios are reported using the delta (6) notation in units of per mil (%o): 6 l 5 N s a m p i c = [( 1 5N/ 1 4N) s a m p l c / ( l 5N/ 1 4N) r c f c r c n c e - 1] x 1000 (1) 5 ' 8 O s a m p l e = [ ( 1 8 0/ 1 6 0) s a m p , c / ( ' 8 0/ 1 6 0) r e ferencc - 1] X 1000 (2) The 1 5 N/ ' 4 N reference is N 2 in air, and the l 8 0 / l 6 0 is Vienna standard mean ocean water (VSMOW). Individual analyses were referenced to injections of N 2 0 from a pure gas cylinder and then standardized through comparison to the international potassium nitrate reference materials IAEA-N3 with an assigned 5 I 5 N of +4.7%o versus atmospheric N 2 (Gonfiantini et al. 1995) and most recently reported S 1 8 0 of 25.6%o versus SMOW (Bohlke et al. 2003). The size of the culture blank was determined by running a prepared vial to which no sample was added. The 5 I 8 0 measurements were corrected for oxygen exchange against the international nitrate isotopic standard USGS-34 (S , 8 0 = -27.9%0; Bohlke et al. 2003). The N and O isotopic measurements of roughly 10% of the samples were duplicated within a day's batch of analyses, and replicate isotopic measurements of 131 individual samples were generally consistent with previously reported analysis standard deviations of 0.2%o for 5 1 5N and 0.5%o for S l 8 0 . To derive estimates of the N and O isotope effects imparted on nitrate ( 1 5E and 1 8e, respectively) during growth, the nitrate 5 I 5 N and 6 I 8 0 measurements were fit to the following Rayleigh linearization (Mariotti et al. 1981): 6' 5N = 5 l 5 N i n i t i a l + l5e{ln ([N03-]/[N03-]i„i l i ai)} (3) 5 I 8 0 = 6 1 8 O i n i t i a l + 18e{ln ([N03-]/[N03"]initiai)} (4) 5.3. Results and Discussion 5.3.1. Magnitude of the N and O isotope effects The N and O isotope effects measured for the organisms in this study ranged between 3%o and 8%o, with the exception of replicate cultures of Synechococcus sp. SNC1, which expressed isotope effects of ~ 17%o (Table 5.2, Figure 5.1). Values observed in the literature for N isotope fractionation by unicellular algae tend towards the lower values observed here, though some are as high as 15 to 20%o (Granger et al. 2004; Montoya and McCarthy 1995; Needoba and Harrison 2004; Needoba et al. 2003; Wada and Hattori 1978). Among the higher values for the N isotope effect in the literature, some have been shown to be sensitive to changes in culture conditions. Needoba and Harrison (2004) and Wada and Hattori (1978) demonstrated that the N isotope effect in some diatoms and coccolithophores increases in response to light limitation. The very high values for the N isotope effect observed here may be specific to this particular strain of cyanobacteria (SNC1), or may otherwise have resulted from lower light intensity during growth of these replicate cultures. The cyanobacteria cultures were grown on the benchtop such that light was not formally controlled, so that higher isotope effects for SNC1 may have been induced by light limitation. We observed no size dependence of the isotope effects, as the relatively small cyanobacteria and the large pennate diatom Ps. hemii (100 um in length) fractionated at ~ 132 5%o. Similarly, literature values for the N isotope effect do not correlate with cell size. While the physiological role of cellular efflux is not defined, a likely explanation is that cells leak nitrate through unidentified pores embedded in the membrane because of the high internal concentrations they accumulate (Dortch et al. 1984), and also due to the strong electrochemical gradient against which nitrate is concentrated inside the cell (Gradmann and Boyd 1995). If so, smaller cells may be expected to fractionate more by function of their higher surface area to volume ratio (Raven 1980), assuming that the putative pores through which nitrate can escape have a similar surface density on the membrane. However, the constrained isotope effect among cell sizes suggests that smaller cells may mitigate leakiness, perhaps by maintaining lower internal nitrate concentrations. Eukaryotic cells can accumulate very high internal nitrate in vacuoles, and thereby presumably maintain relatively low cytoplasmic concentrations that only suffice to saturate nitrate reductase activity. In support of the contention that cellular efflux is merely nitrate leakage, Needoba and Harrison (2004) observed that the N isotope effect in eukaryotes increases only in response to environmental changes that work to increase internal nitrate concentrations. Cultures growth at lower temperature or subjected to iron limitation did not result in higher internal nitrate concentration, and consequently no significant changes in the N isotope effect of nitrate assimilation were observed. However, light limitation resulted in a considerable increase in intracellular nitrate, and in a concomitant increase in the in N isotope effect, suggesting that the cells leaked more nitrate because cytoplasmic concentrations were presumably higher. Alternatively, nitrate efflux may be a tightly regulated cellular process that works to maintain electrochemical balance of the cell. In this case, the isotope effect may not be directly sensitive to changes in cellular allometry. Until a clear understanding of the nitrate efflux mechanism emerges, physiological controls on the magnitude of isotopic discrimination associated with nitrate assimilation will remain uncertain. 5.3.2. The N:0 coupling Evidence to date implicates nitrate N-0 bond breakage at the active site of nitrate reductase as the fractionating step during nitrate assimilation. This was concluded in 133 early studies based on the principle that significant bond disruption is required to generate a substantial isotope effect, such that transport of nitrate into the cell is unlikely to represent a significant contribution to observed isotope effect (Delwiche and Steyn 1970; Mariotti et al. 1982; Shearer et al. 1991; Wada and Hattori 1978). Expression of the N isotope effect in extracellular nitrate was thus ascribed to cellular nitrate efflux. Shearer et al. (1991) later demonstrated that the N isotope effect observed during nitrate assimilation by a cyanobacterium could be progressively increased by selectively inhibiting nitrate reductase activity with tungstate, which interferes with the molybdenum catalytic site of the enzyme. As gross nitrate uptake remained constant over the range of inhibition of nitrate reductase activity, amplification of the N isotope effect was attributed to progressive augmentation of cellular nitrate efflux with increasing internal nitrate. Needoba et al. (2004) further observed that the intracellular nitrate pool of the diatoms T. weissflogii was more N-isotopically enriched than extracellular nitrate, confirming that isotope discrimination occurs inside the cell, evidently at the site of nitrate reductase. Granger et al. (2004) also found that the nitrate N and O isotopic enrichment for T, weisflogii was coupled in both internal and external pools of nitrate, showing a uniform l 8e:'5£ of 1, thus corroborating that fractionation originates predominantly from a single process, namely, nitrate reduction by nitrate reductase. Hence, nitrate N and O isotope effects measured in external nitrate reflect the balance between gross nitrate uptake and cellular nitrate efflux (Needoba et al. 2004; Shearer et al. 1991), while coupling between the N and O isotopes of nitrate is the fractionating signature of the enzyme (Granger et al. 2004). Our observations show that nitrate assimilation among the strains studied here is consistent with , 8e: 1 5e of 1.03 ± 0.01 (Table 5.2, Figure 5.2). Hence, the prokaryotic assimilatory nitrate reductase, NAS, seemingly imparts equivalent isotopic fractionation to the N and O atoms of nitrate. Moreover, the 1:1 coupling observed previously among diatoms and a coccolithophore (Granger et al. 2004) extends to the additional diatom species studied here, as well as to an additional clade, namely Chlorophycea, strengthening the evidence that the eukaryotic nitrate reductase fractionates uniformly among phylogenetic groups. These findings, along with those in previous studies, establish that the assimilatory nitrate reductases of prokaryotes and eukaryotes (NAS and 134 eukNR, respectively), as well as the respiratory NAR of denitrifiers (Chapter 4), propagate equivalent 18e and 15e during nitrate reduction. Only the auxiliary nitrate reductase NAP of denitrifiers takes exception to the 1:1 rule, displaying l 8e: 1 5e of 0.6 (Chapter 4). The 1:1 coupling inferred here for NAS is surprising in light of the fact that NAS and NAP are close relatives that possess nearly identical Mo active sites (Moreno-Vivian et al. 1999). Although the respective enzyme mechanism(s) of nitrate reductases are not yet elucidated, the difference in N and O isotope coupling between NAS and NAP implies that, in spite of the relatedness and structural similarities, the two enzymes have dissimilar catalytic mechanisms that translate to dissimilar activation energies required for bond breakage of the various nitrate isotopomers. Conversely, the fact that the 1:1 coupling is common to NAS, NAR and eukNR implies that, in spite of their structural differences, these enzymes have similar reaction mechanisms that translate to similar activation energies required for bond breakage of the various nitrate isotopomers. While bulk nitrate isotopic fractionation is undeniably undertaken by nitrate reductase, a question still remains as to whether uptake of nitrate at the cell surface imparts a small isotope effect. If transport indeed caused isotopic fractionation of nitrate, one might expect this to be restricted to the oxygen atoms on the molecule, as only they are likely to be subject to hydrogen bonding in the transporter. One would then observe a progressively greater decoupling of N and 0 isotope effects as their respective magnitudes decreased, and as the proportional influence of transport to total l 8e increased (Appendix 2). However, among the relatively low isotope effects measured here, there appears to be no evidence of decoupling towards lower magnitudes of e. Hence, the transport isotope effect, if significant, would also need to have a 18e:'5e close to 1. Since this seems highly unlikely to us, we favour the interpretation that the transport occurs without any significant fractionation, at least for the organisms studied here. 5.3.3. Oceanographic implications A major limitation on nitrate §'5N as a tool for modern upper ocean studies as well as for paleo-reconstruction of surface N budgets from sedimentary N isotopes is the 135 uncertainty in the amplitude of N isotope discrimination associated with nitrate assimilation. Current estimates of , 5e from the ocean range between 4 and 10%o (Karsh et al. 2003; Lourey et al. 2003; Pennock et al. 1996; Sigman et al. 1999; Wu et al. 1997), while the range observed in lab cultures is greater, from 0 to 20 %o (Granger et al. 2004; Montoya and McCarthy 1995; Needoba and Harrison 2004; Needoba 2003; Waser et al. 1998). Perhaps fortuitously, the most recent field isolate cultured in this study, Ps. hemii isolated from the subarctic Pacific in August 2003, adheres to a canonical N isotope effect of 5%o. While most lab and field estimates tend around 5%o, the variability towards higher values remains largely unexplained. Yet, the observations in this study collectively present an analogous conumdrum, where most of the isotope effects were constrained around 5%o while one strain showed very high isotopic discrimination. Isotope effects are not necessarily species-specific, as the same species often yields very different isotope effects when grown in different labs under similar culture conditions (Granger et al. 2004; Montoya and McCarthy 1995; Needoba 2003). As observed previously by Needoba and Harrison (2004), light limitation can be invoked here, albeit tenuously, to explain the high isotope effects for SNC1. The modulation of the N isotope effect by light in lab cultures is of direct relevance to interpretation of field observations of nitrate S I 5 N, as isotope effects are indeed seemingly higher in regions subject to light limitation, such as the Subantarctic zone of the Southern Ocean (Karsh et al. 2003). The N isotope effects compiled here add to the existing list of species examined thus far, however, our observations arguably offer little insight into the control of the magnitude of the isotope effect. Clearly, elucidation of the controls on the magnitude of N isotope discrimination during assimilation requires a better understanding of environmental and physiological controls on nitrate assimilation, particularly as these pertain to the concomitant response of cellular nitrate efflux. A very clear result of this study is that nitrate N and O isotope fractionation during nitrate assimilation by prokaryotes is conveniently coupled with a ratio of ~1. Hence, based on this and previous work (Granger et al. 2004), we feel confident in asserting that assimilation at the surface ocean leaves an N and O isotopic imprint on nitrate that co-varies in a proportion of 1:1. Given the invariance of this relationship, any de-coupling of the nitrate N and O isotopic signatures in the ocean surface layer must 136 then be attributed to co-occurring N-transformations. While few measurements of the coupled isotopes of nitrate at the surface ocean exist, preliminary data suggest that nitrate 5 I 8 0 may deviate positively from the expected 1:1 relationship with 8 I 5 N when nitrate tends to lower concentrations in the surface mixed layer (Rafter et al. personal communication; Conclusions). In theory, the oxidation of regenerated ammonium to nitrate by nitrifiers in the mixed layer could act to increase the 8 1 8 0 versus 5 , 5 N, such that coupled measurements of the N and O isotopes may potentially provide a tracer for nitrification occurring in the surface ocean's mixed layer. Although this work is in its infancy, this novel integrative tracer promises to track the extent to which nitrate is regenerated in the mixed layer, in turn providing a clearer understanding of new and regenerated production at the surface ocean. Of further interest is the fact that the N to 0 coupling among nitrate isotopes is invariant in freshwater compared to seawater strains, as attested by the two strains of chlorophytes. Although there were no a priori reasons to expect differences in the N and O isotopic signatures of nitrate reductase between a freshwater and a seawater isolate of the same clade, one could have postulated that the coupling for the freshwater strain would be akin to that observed for nitrate N and O isotopes in freshwater systems, namely 6 1 8 0:5 1 5 N of 0.6 (Lehmann et al. 2003). Based on our observations, assimilation by freshwater algae cannot be invoked as the source of the N and O isotopic signal of 0.6 common to freshwater systems. 137 Table 5.1. Description of the experimental strains and their associated type of assimilatory nitrate reductase. Strain Type E n v i r o n m e n t NR Mo-enzyme familyt Synechococcus sp. DC2 prokaryote marine N A S D M S O reductase Synechococcus sp. SNC1 prokaryote marine N A S D M S O reductase Synechococcus sp. WH8102 prokaryote marine N A S D M S O reductase Pseudonitzchia hemii eukaryote subarctic Pacific eukNR Sulfite oxidase Phaeodactylum tricornutum eukaryote marine eukNR Sulfite oxidase Chlorella pyrenoidosa eukaryote freshwater eukNR Sulfite oxidase Chlorella sp. eukaryote marine eukNR Sulfite oxidase •fThe prokaryotic nitrate reductases all belong to the D M S O reductase family (NAS, N A R and N A P ) , which is characterized by 4 sulfur groups coordinating the Mo atom from 2 M G D (molybdo-guanine dinucleotide) functional groups, and one sulfur from cysteine residue (Moreno-Vivian et al. 1999). The sulfite oxidase family harbours the eukaryotic assimilatory nitrate reductases, which possess only one M P T (molybdopterin) group that provide two coordinating sulfurs to the Mo atom (Campbell 1985). 138 Table 5.2. Nitrate N and O isotope effects during nitrate assimilation by the experimental strains, as well as the regression coefficient of the relative O vs. N isotope fractionation of nitrate for individual cultures (± standard deviation). Strain n 15e (%o) l se (%») 5 , ! iO:5 , 5 N Synechococcus sp. DC2 8 4.6 ±0.3 5.4 ±0.3 1.15 ±0.07 10 5.0 ±0.2 5.9 ±0.2 1.16 ± 0.05 Synechococcus sp. SNC1 11 17.3 ±0.3 16.8 ±0.3 0.97 ±0.01 11 16.3 ±0.3 16.8 ±0.5 1.03 ±0.02 Synechococcus sp. WH8102 7 3.8 ±0.3 3.9 ±0.3 1.01 ±0.02 4 3.3 ±0.5 3.3 ±0.1 0.96 ±0.14 Pseudonitzchia hemii 6 6.0 ±0.3 6.8 ±0.4 1.13 ±0.07 6 4.7 ±0.3 5.6 ±0.3 1.19 ±0.04 6 5.3 ±0.1 5.9 ±0.6 1.08 ± 0.13 Phaeodactylum tricornutum 3 7.8 ± 1.3 8.0 ± 1.0 1.02 ±0.04 4 5.3 ±0.6 4.7 ±1.1 0.90 ± 0.11 5 4.5 ±0.2 5.2 ± 1.1 1.16 ± 0.19 Chlorella pyrenoidosa 6 4.9 ±0.5 4.7 ±0.6 0.96 ± 0.04 Chlorella sp. 4 5.1 ±0.2 4.9 ±0.3 0.95 ±0.04 4 5.7 ±1.1 5.9 ±1.1 1.02 ±0.09 Pooled value = 1.03 ±0.01 139 •4 I < u '3 V -O z < o-L 16 14 • 12-10-8 • 6 • 4 • 2-0 -Ps. hemii • P. tricornutum Chlorella sp. A C. pyrenoidosa -] 1 1 1 r -1.0 -0.8 -0.6 -0.4 -0.2 0.0 In ([NO/MHO/],,^,) + V 15 e = 5 %c \ \ l 5 e = 20% o \ \ + SNCl O DC2 • WH8102 1 \ n i i r --2.0 -1.5 -1.0 -0.5 ln([N03-]/[N03-] initia,) 0.0 Figure 5.1. Rayleigh plots of the change in the 5 I 5 N of nitrate versus the ln of fractional nitrate consumption during its assimilation by monocultures of a) eukaryotic and b) prokaryotic unicellular phytoplankton. Note that panel b is truncated (SNCl expands beyond the axes) to facilitate visualization of the points at lower 5 I 5 N values. 140 Figure 5.2. The relative change in the 5 I S 0 of nitrate plotted over the corresponding change in 5 1 5 N for a) eukaryotic and b) prokaryotic phytoplankton cultures, as well as for c) all experimental cultures combined. The value posted for the slope of the line in panel c was obtained from the least-squares regression among all the points. 141 5.4. References Allen, A. E., M. H. Howard-Jones, M. G. Booth, M. E. Frischer, P. G. Verity, D. A. Bronk, and M. P. Sanderson. 2002. Importance of heterotrophic bacterial assimilation of ammonium and nitrate in the Barents Sea during summer. J. Marine Syst. 3 8 : 93-108. Bohlke, J. K., S. J. Mroczkowski, and T. B. Coplen. 2003. Oxygen isotopes in nitrate: new reference materials for 0-18 : 0-17 : 0-16 measurements and observations on nitrate-water equilibration. Rapid Commun. Mass Sp. 17: 1835-1846. Braman, R. S., and S. A. Hendrix. 1989. Nanogram nitrite and nitrate determination in environmental and biological materials by V(III) reduction with chemiluminescence detection. Anal. Chem. 6 1 : 2715-2718. Campbell, W. H. 1985. The biochemistry of higher plant nitrate reductases, p. 143-151. In P. W. L. a. J. E. Burris [ed.], Nitrogen Fixation and C 0 2 Metabolism. Elsevier. Casciotti, K. L., D. M. Sigman, M. G. Hastings, J. K. Bohlke, and A. Hilkert. 2002. Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater using the denitrifier method. Anal. Chem. 74: 4905-4912. Delwiche, C. C , and P. L. Steyn. 1970. Nitrogen isotope fractionation in soils and microbial reactions. Env. Science Tech. 4: 929-935. Dortch, Q., J. R. Clayton, S. S. Thorensen, and S. I. Ahmed. 1984. Species differences in accumulation of nitrogen pools in phytoplankton. Mar. Biol. 81 : 237-250. Gonfiantini, R., W. Stichler, and K. Rosanski. 1995. Standards and Intercomparison Materials Distributed by the IAEA for Stable Isotope Measurements. International Atomic Energy Agency. Gradmann, D., and C. M. Boyd. 1995. Membrane voltage of marine-phytoplankton, measured in the diatom Coscinodiscus radiatus. Mar. Biol. 123: 645-650. Granger, J., D. M. Sigman, J. A. Needoba, and P. J. Harrison. 2004. Coupled nitrogen and oxygen isotope fractionation of nitrate during assimilation by cultures of marine phytoplankton. Limnol. Oceanogr. 4 9 : 1763-1773. 142 Granger, J., D. M. Sigman, M. D. Prokopenko, M. F. Lehmann, and P. D. Tortell. in press. A method for nitrite removal in nitrate N and O isotope analyses. Limnol. Oceanogr:. Methods 4: 202-212. Karsh, K. L., T. W. Trull, A. J. Lourey, and D. M. Sigman. 2003. Relationship of nitrogen isotope fractionation to phytoplankton size and iron availability during the Southern Ocean Iron RElease Experiment (SOIREE). Limnol. Oceanogr. 48: 1058-1068. Keller, M. D., W. K. Bellows, and R. R. L. Guillard. 1988. Microwave treatment for sterilization of phytoplankton culture media. J. Exp. Mar. Biol. Ecol. 117: 279-283. Lehmann, M. F., P. Reichert, S. M. Bernasconi, A. Barbieri, and J. A. Mckenzie. 2003. Modelling nitrogen and oxygen isotope fractionation during nitrate reduction in a hypolimnetic redox transition zone. Geochim. Cosmochim. Ac. 67: 2529-2542. Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2004. Coupling the N-15/N-14 and 0- 18/0-16 of nitrate as a constraint on benthic nitrogen cycling. Mar. Chem. 88: 1- 20. Lehmann, M. F., D. M. Sigman, D. C. McCorkle, B. G. Brunelle, S. Hoffmann, M. Kienast, G. Cane, and J. Clement. 2005. Origin of the deep Bering Sea nitrate deficit: Constraints from the nitrogen and oxygen isotopic composition of water column nitrate and benthic nitrate fluxes. Global Biogeochem. Cy. 19: GB4005. Lourey, M. J., T. W. Trull, and D. M. Sigman. 2003. Sensitivity of delta N-15 of nitrate, surface suspended and deep sinking particulate nitrogen to seasonal nitrate depletion in the Southern Ocean. Global Biogeochem. Cy. 17. Mariotti, A., J. C. Germon, P. Hubert, P. Kaiser, R. Letolle, A. Tardieux, and P. Tardieux. 1981. Experimental determination of nitrogen kinetic isotope fractionation: some principles; illustration for the denitrification and nitrification processes. Plant Soil 62: 413-430. Mariotti, A., F. Mariotti, M. L. Champigny, N. Amarger, and A. Moyse. 1982. Nitrogen isotope fractionation associated with nitrate reductase-activity and uptake of NO3" by Pearl-Millet. Plant. Physiol. 69: 880-884. 143 Mcllvin, M. R., and M. A. Altabet. 2005. Chemical conversion of nitrate and nitrite to nitrous oxide for nitrogen and oxygen isotopic analysis in freshwater and seawater. Anal. Chem. 77: 5589-5595. Montoya, J. P., and J. J. McCarthy. 1995. Isotopic fractionation during nitrate uptake by marine phytoplankton grown in continuous culture. J. Plankton. Res. 17: 439-464. Moreno-Vivian, C , P. Cabello, M. Martinez-Luque, R. Blasco, and F. Castillo. 1999. Prokaryotic nitrate reduction: Molecular properties and functional distinction among bacterial nitrate reductases. J. Bact. 181: 6573-6584. Needoba, J. A., and P. J. Harrison. 2004. Influence of low light and a light: Dark cycle on NOV uptake, intracellular NO3", and nitrogen isotope fractionation by marine phytoplankton. J. Phycol. 40: 505-516. Needoba, J. A., N. A. Waser, P. J. Harrison, and S. E. Calvert. 2003. Nitrogen isotope fractionation in 12 species of marine phytoplankton during growth on nitrate. Mar. Ecol.-Prog. Ser. 255: 81-91. Needoba, J. A., D. M. Sigman, and P. J. Harrison. 2004. The mechanism of isotope fractionation during algal nitrate assimilation as illuminated by the N-15/N-14 of intracellular nitrate. J. Phycol. 40: 517-522. Needoba, J. A., N. A. Waser, P. J. Harrison, and S. E. Calvert. 2003. Nitrogen isotope fractionation in 12 species of marine phytoplankton during growth on nitrate. Mar. Ecol.-Prog. Ser. 255: 81-91. Parsons, T. R., Y. Maita, and L. C. M. 1984. A manual of biological and chemical methods for seawater analysis. Publ. Pergamon Press. Pennock, J. R., D. J. Velinsky, J. H. Sharp, J. Ludlam, and M. L. Fogel. 1996. Isotope fractionation of ammonium and nitrate during their uptake by Skeletonema costatum.: Implications for the 515N dynamics under bloom conditions. Limnol. Oceanogr. 41: 451-459. Price, N. M., G. I. Harrison, J. G. Herring, R. J. Hudson, P. M. V. Nirel, B. Palenik and F. M. M. Morel. 1988/1989. Preparation and chemistry of the artificial algal culture medium Aquil. Biol. Oceanogr. 6: 443-461. Raven, J. A. 1980. Nutrient transport in microalgae. Adv. Microb. Physiol. 21: 47-226. 144 Shearer, G., J. D. Schneider, and D. H. Kohl. 1991. Separating the efflux and influx components of net nitrate uptake by Synechococcus-R2 under steady-state conditions. J Gen. Microbiol. 137: 1179-1184. Sigman, D. M., M. A. Altabet, R. Francois, D. C. McCorkle, and G. Fischer. 1999. The 5 I 5 N of nitrate in the Southern Ocean: Consumption of nitrate in surface waters. Global Biogeochem. Cy. 13: 1149-1166. Sigman, D. M., and K. L. Casciotti. 2001. Nitrogen isotopes in the ocean, p. 1884-1894. In J. H. Steele, K. K. Turekian and S. A. Thorpe [eds.], Encyclopedia of Ocean Sciences. Academic Press. Sigman, D. M., K. L. Casciotti, M. Andreani, C. Barford, M. Galanter, and J. K. Bohlke. 2001. A bacterial method for the nitrogen isotopic analysis of nitrate in seawater and freshwater. Anal. Chem. 73: 4145-4153. Sigman, D. M., J. Granger, P. J. DiFiore, M. M. Lehmann, R. Ho, G. Cane, and A. Van Geen. 2005. Coupled nitrogen and oxygen isotope measurements of nitrate along the eastern North Pacific margin. Global Biogeochem. Cy. 19: GB4022. Silva, S. R., C. Kendall, D. H. Wilkinson, A. C. Ziegler, C. C. Y. Chang, and R. J. Avanzino. 2000. A new method for collection of nitrate from fresh water and the analysis of nitrogen and oxygen isotope ratios. J. Hydrol. 228: 22-36. Wada, E., and A. Hattori. 1978. Nitrogen isotope effects in the assimilation of inorganic nitrogenous compounds. Geomicrobiol. J. 1: 85-101. Waser, N. A., K. D. Yin, Z. M. Yu, K. Tada, P. J. Harrison, D. H. Turpin, and S. E. Calvert. 1998. Nitrogen isotope fractionation during nitrate, ammonium and urea uptake by marine diatoms and coccolithophores under various conditions of N availability. Mar. Ecol.-Prog. Ser. 169: 29-41. Wu, J., S. E. Calvert, and C. S. Wong. 1997. Nitrogen isotope variations in the subarctic Pacific northeast Pacific: relationships to nitrate utilization and trophic structure. Deep-Sea Res. Part I 44: 287-314. 145 Chapter 6 Conclusions and Future Outlook 146 6.1. The 1:1 rule The foremost conclusion of my thesis work is that nitrate reductase is the driver of the N and O isotope effects imparted on nitrate during its biological reduction. The N and O isotopes of nitrate are fractionated in equivalent proportion during reduction by various types of nitrate reductases. The 1:1 trend in the O-to-N fractionation ratio is consistent among eukaryotic microalgae that assimilate nitrate with the eukaryotic assimilatory nitrate reductase, among prokaryotic algae that have the assimilatory NAS, and denitrifiers who reduce nitrate with the respiratory NAR. The constancy of the coupling among enzyme types is remarkable, not only by function of being the distinctive ratio of 1, but also because these enzymes have little sequence homology among each other, and even have different coordination spheres at the molybdenum active site. Hence the 1:1 trend can be considered ubiquitous for all biological nitrate reductions, except those catalyzed by NAP. In spite of having a coordination sphere nearly identical to that of NAS, NAP fractionates the N and O isotopes of nitrate with an O-to-N fractionation ratio of 0.6. This could be considered as a thorn in the side of a global 1:1 rule for biological nitrate reduction, however, NAP serves as an auxiliary reductase that is privy to electrons only when these are in excess during growth. Though we cannot discount NAP as an active player in some ecosystems, its activity, if present, appears insignificant in oxygen minimum zones of the ocean water column where the O-to-N fractionation ratio has, thus far, not been observed to stray below 1 (Lehmann et al. 2004; Lehmann et al. 2005; Sigman et al. 2005). However, NAP could conceivably be active in organic-rich sediment, in eutrophic lakes where photo-heterotrophs show high activity, or perhaps in the surface-ocean's mixed layer. When Daniel Sigman and Karen Casciotti measured their first profile of nitrate N and O isotopes from water collected at station PAPA in the Subarctic Pacific (Casciotti et al. 2002), they were surprised to find that the N and O isotopes co-varied in equal proportion. They carefully reviewed their novel method for measuring nitrate isotopes in seawater (Casciotti et al. 2002; Sigman et al. 2001), and were then compelled to conclude that this 1:1 trend was not a methodological artifact, but that the isotopes indeed co-varied concomitantly, in equal proportion. Their original surprise at observing a 1:1 ratio 147 was not necessarily due the peculiarity of this co-variation being equal to unity, but rather that the expected relative fractionation of O-to-N isotopes ought to have reflected the relative difference in the reduced masses of the two kinds of isotopomers. Namely, the bond strength of isotopomers is roughly proportional to the square root of the ratio of the reduced masses of the molecules that get cleaved in the reduction reaction (Appendix 1). And bond strength, in turn, is roughly proportional to the isotope effect associated with a given chemical reaction. Thus, simple computation of the reduced masses of the nitrate isotopomers predicted that the nitrate isotopes would be fractionated in O-to-N ratio of ~ 0.6: u = m|in2/(mi + m2) 'V= H-9365079 1 8u = 12.9375 l 8e/ 1 5e ~ ((12.9375 + 3)/l 1.9365079)"2 -0.6 The symbols u refer to the reduced masses of l5N-nitrate and 180-nitrate, respectively. To compute the ratio of the reduced masses, the reduced mass of the l80-nitrate must be divided by 3, because there is only 1 in 3 chances that the breaking bond in the nitrate molecule will be that of the l 8 0 atom. Around the same time, while working as a technician in another lab at Princeton, I approached then post-doctoral fellow Daniel Sigman about culturing phytoplankton to look at the potential change in the N-isotope effect on nitrate when cultures were grown under iron-limitation. I then had only a faint understanding of stable isotope fractionation, but somehow, after seeing a talk by Mark Altabet, I had pieced together that knowing whether phytoplankton fractionated the N isotopes of nitrate differently when iron-limited could help in paleo-reconstruction from the sedimentary record to assess whether the Southern Ocean was relived of iron-limited during the last glacial maximum. And the novel method that Karen Casciotti and Daniel Sigman had just developed seemed like a good opportunity to test a simple hypothesis. Unbeknownst to me, Joe Needoba at UBC had enterprised similar work for his doctoral work. Daniel 148 Sigman promptly accepted my offer and mentioned that growth experiments with phytoplankton would also be important to assess fractionation of the oxygen isotopes of nitrate. Although I remember his mentioning the oxygen isotopes, I understood too little about stable isotopic tracers at the time to appreciate this implication. A couple of years later I had begun my doctorate at UBC, and had decided to "finish" the work I had started with Daniel Sigman. From the few cultures I had grown at Princeton, it had become clear that nitrate assimilation by eukaryotic phytoplankton, or rather, by diatoms, imparted equivalent isotope effects to the N and the O isotopes of nitrate. Sigman, while confident with the numbers that were being generated, kept reiterating that it made no sense, and that the isotopes should fractionate in a ratio of 0.6. In the meantime, however, Joe Needoba was finishing his thesis work at UBC, and his observations indicated that the major fractionating step could only be attributed to bond breakage by nitrate reductase, such that the isotopic enrichment imprinted on nitrate internally was ostensibly propagated in the external pool by cellular efflux (Needoba and Harrison 2004; Needoba et al. 2004). He observed that the highly concentrated internal nitrate pool of diatoms was substantially more enriched in the heavy isotope of N than external nitrate. Moreover, he determined that the difference in the 8 , 5 N between internal and external nitrate pools was a direct predictor of the observed N isotope effect: When the difference between external and internal nitrate was small, a large isotope effect ensued, and vice versa. Hence, the isotope effect is a measure of the ratio of cellular nitrate uptake to efflux, reflecting the extent to which the isotope effect intrinsic to nitrate reductase is propagated in the external nitrate pool. Although this idea was not new (Wada and Hattori 1978), and although nitrate efflux had previously been shown to propagate the enzymatic isotope effect to the extracellular nitrate pool in cyanobacteria (Shearer et al. 1991), the oceanography literature seemed ambivalent on the physiological fractionating step during nitrate assimilation; N isotopic fractionation of nitrate was sometimes attributed to the nitrate uptake step (Montoya and McCarthy 1995). The reluctance to invoke nitrate reductase as the fractionating step stemmed from the counter-intuitive concept of cellular nitrate efflux occurring in nitrogen-limited ecosystems. However, in light of Needoba's results, and considering concurrent evidence reported previously in the literature (Shearer et al. 1991; Wada and Hattori 1978), the 1:1 149 trend observed for O-to-N fractionation ratio during nitrate assimilation then likely reflected the isotopic signature of nitrate reductase. Additional supporting evidence was provided from measurements of the coupled N and O isotope ratios of nitrate in samples of internal nitrate from diatoms that had been collected previously by Needoba (Granger et al. 2004). Not only was the internal pool enriched in both N and O isotopes relative to external nitrate, but the O-to-N isotopic enrichment also showed a ratio of 1:1. This implied that the fractionating process for the internal and external nitrate pools was the same (i.e., showed the same O-to-N isotopic imprint), discounting any significant role of uptake in imparting an isotopic imprint on nitrate, thus singling out nitrate reductase as the dominant fractionating step in the assimilation mechanism. The explanation underlying the apparent quandary between the observed 1:1 trend versus the expected trend of 0.6 turned out to be simple. Namely, the isotope effect associated with bond breakage in a unidirectional chemical reaction is proportional to reduced masses of the isotopomers involved, because reduced mass is proportional to bond strength. Hence the square root of the ratios in reduced mass stands as a good predictor for the isotope effect imparted on the substrate of an uncatalyzed unidirectional reaction. However, enzymes reduce the activation energy required to catalyze a given reaction by de-stabilizing the bonds of the substrate molecule within the coordination sphere of the reactive site, hence decreasing bond strengths. The decrease in bond strength of the isotopomers at the enzymatic site is not predictable from respective reduced masses because it results from complex re-distribution of bonding energies within the transition state structure of nitrate bound in the coordination sphere of the catalytic site. While this apparent "mass-independence" of enzymatic bond breakage would come as no surprise to an enzymologist, it did to some geochemists, and still does when I give presentations on the topic. In light of the direct link between the isotope effect and enzyme mechanism, the similarity in O-to-N isotope fractionation of nitrate by different functional and structural nitrate reductases is fascinating. Even more so is the difference in the coupling of NAP and NAS, which have quasi-identical coordination spheres. While I did not determine the isotope effect intrinsic to the various nitrate reductases in vitro, the trends observed in 150 this work may serve as constraints to elucidate enzyme mechanism from computational models of the respective catalytic sites. 6.2. Coupled measurements of nitrate N and O isotopes to infer N biogeochemical cycling The motivation for my Ph.D. work was to provide the systematics from which to interpret field measurements of coupled nitrate N and O isotopes. The 1:1 rule thus provides a serendipitous marker of biological nitrate reduction. Deviations from this trend may then be interpreted as N transformations co-occurring in a given body of water. In a general sense, the complement between N and O isotope ratios involves the processes that are not captured by the O isotopes. For the N atom in nitrate, nitrate assimilation and nitrification are part of an internal cycle within the ocean that should not cause a net change in the mean 6 1 5N of ocean nitrate over time. N-fixation and denitrification comprise the input/output budget of fixed N and control the mean 5 1 5N of ocean nitrate. In contrast, for the O atoms in nitrate, nitrification is an absolute input, while both nitrate assimilation and denitrification are absolute sinks. The §'sO of newly produced nitrate does not depend on the origin of the ammonia being nitrified, be it from newly fixed N, from the biomass of phytoplankton growing in a nitrate-rich environment, or from biomass of phytoplankton that assimilate the entire supply of upwelled nitrate. To date, Sigman and colleagues have made measurements of coupled nitrate N and O isotopes in various ocean ecosystems, and from these have separated the isotopic imprints of co-occurring processes, such as (1) nitrification from denitrification at the water-sediment interface (Lehmann et al. 2003; Lehmann et al. 2004; Lehmann et al. 2005), (2) N-fixation and denitrification in the water column of the eastern North Pacific margin (Sigman et al. 2005), and (3) the regeneration of ammonia to nitrate within the mixed layer in north-south transects at the Equatorial Upwelling (Granger et al. work in progress). The last two of these applications are discussed in detail below. 151 6.2.1. Separating N-fixation from denitrification The deviation of nitrate concentrations from expectations based on phosphate concentrations and the global Redfield N:P trend, N*, has proven useful in quantifying both the processes of N-fixation and denitrification. The N* of the ocean interior is increased by nitrification of newly fixed N, decreased by denitrification, and homogenized by ocean circulation. However, N* alone cannot be used to disentangle the impacts of N-fixation and denitrification in regions where their effects on these tracers overlap or are mixed together. N* is equally sensitive to the input of newly fixed N and denitrification; if N addition and loss by these processes are equal, N* does not change. Nitrate 5 I 5 N has somewhat different sensitivities to the two processes. In most of the ocean, denitrification increases the 5 I 5 N of nitrate more than an equivalent input of newly fixed N will decrease it. Nitrate 5 1 80 has still a different ratio of sensitivity to N-fixation and denitrification; denitrification causes an increase by the same degree as for nitrate 5 I 5 N (the 1:1 rule), whereas addition of newly fixed N to the nitrate pool has a very minor effect on its S 1 8 0. Nitrification of newly fixed N produces nitrate with a 5 I 8 0 that is not different from that when ammonia of another origin is nitrified. A set of water column profiles from the eastern North Pacific margin (Figure 6.1), off the coast of Baja California, illustrates the approach. The strongly negative N* reflects the loss of nitrate to denitrification, suggesting that -25% of the nitrate has been lost to denitrification (Figure 6.2). The N* minimum at 200-400 m is accompanied by maxima in the 5 1 5N and 6 1 8 0 of nitrate, which is expected given the fractionation associated with denitrification. However the nitrate 5 1 5N increase to its maximum is smaller than the 8 1 8 0 increase, and the 5 1 5N maximum is also slightly deeper. The derived parameter 5 , 5 N - (5 I 80 + 5.5%o) - abbreviated as A15,18 - compares nitrate 5 I 5 N and 5 I 8 0 after normalizing for the 5 I 5 N vs. 5 1 80 difference of the "background" nitrate and the culture-derived 1 8e:'5e of 1 for denitrification. It indicates that nitrate 5 1 5N is as much as 2.5%o lower than one would predict from 5 1 80. The minimum in Al5,18 is shallower than the isotope maxima and the N* minimum so that it is not well-explained by a uniform deviation in the l 8e: l 5e from the 1:1 (Figure 6.3). 152 In regions without nearby denitrification, such as the north Atlantic, there is substantial 6 , 5 N depletion in the shallow thermocline, presumably due to the oxidation of newly fixed N (Altabet 1988; Knapp et al. 2005). Thus the upward change in 5 I 8 0 and 5 1 5N relationship in the North Pacific may be due to oxidation of low 6 1 5 N, newly fixed N to nitrate. That is, the shallower 8 1 80 maximum suggests that the nitrification of newly fixed N is "eroding" the top of the nitrate o l 5 N maximum, consistent with a previous inference based on nitrate 8 I 5 N alone (Brandes et al. 1998). Comparison of profiles at stations further north along the coast, as well as from the central and western North Pacific gyre, suggest that the N-fixation signal is generated locally (Sigman et al. 2005). The full implications of these findings are addressed in Sigman et al. (2005). However, the tentative conclusion that N-fixation and denitrification occur in the same region is controversial, because N-fixation is not believed to occur in regions with high surface nitrate, such as the upwelling at the North Pacific margin. Yet, if the coupled N and O isotopes indeed reflect the isotopic imprint of the co-occuring signals of N-fixation and denitrification, this implies that denitrification and N-fixation in the Pacific Ocean are more closely coupled, both spatially and temporally, than was originally believed. 6.2.2. Separating nitrate assimilation from remineralization in the mixed layer Nitrate in the euphotic zone has been considered as a "new" source of N to the surface ocean, being supplied to the surface ocean by physical processes rather than by regeneration in the surface ocean (Dugdale and Goering 1967). This view has been justified on the basis of the inhibition of nitrification by light, which was believed to restrict nitrifiers to the ocean interior. If the origin of all nitrate in the euphotic zone is indeed mixing and upwelling from the deep ocean, then the rate of nitrate assimilation in a steady-state system is equivalent to "export production," namely the export of organic matter from the euphotic zone (Eppley and Peterson 1979). However, the greatest nitrification rates are actually measured near the bottom of the euphotic zone, where light intensity is around 5 to 10% of surface light intensity. This implies that nitrate in the euphotic zone can be functionally "regenerated," and is not necessarily "new." The supply of nitrate from nitrification in the euphotic zone can 153 reportedly exceed phytoplankton demand (Bianchi et al. 1997; Dore and Karl 1996; Ward et al. 1989), thus decoupling gross nitrate assimilation from export production and from net nitrate consumption (Ward et al. 1989). A potential application for coupled measurement of nitrate 8 1 5N and 5 1 80 is to determine the relative magnitude of nitrate assimilation and nitrification within the euphotic zone. This is illustrated schematically in Figure 6.4. In this hypothetical example, the initial nitrate supplied to the euphotic zone has a nominal 8'5N of 5%o (vs. air) and a 6'80 of 0%o (vs. SMOW), values that are roughly typical of deep water (Casciotti et al. 2002). Based on the ' 1:1 rule,' phytoplankton assimilation of nitrate imparts equal isotope effects on the N and O atoms of nitrate, designated with a canonical l 5e (and l 8e) of 5%o. If only half of the initial nitrate is consumed (f = 0.5), the remaining nitrate pool becomes isotopically enriched, posting a 5 I 5N of 8.5%o and a 5 I 80 of 3.5%o (an additional 3.5%o for both isotope ratios). The resulting particulate nitrogen has a 6 1 5N of 1.5%o. This particulate nitrogen decomposes and is then completely nitrified to nitrate, which produces "regenerated" nitrate with a 5 I 5N of 1.5%o and 5 1 80 of 0%o (identical to seawater). Ammonification and nitrification do not impart any fractionation to this reoxidized nitrate because the reactant pools (the organic nitrogen and the ammonia) are completely reacted (f= 1). Physical mixing of the unconsumed and the regenerated nitrate pools results in nitrate that has a 8 1 5N of 5%o and a 5 1 80 of 1.75%o. Hence, while the N isotopes are only sensitive to nitrate assimilation, the O isotopes carry the imprint of both assimilation and regeneration. Significant positive departure of the O-to-N fractionation ratio from 1:1 signals regeneration of nitrate in the mixed layer. Coupled measurements of nitrate N and O isotopes along two north-south transects across the Equatorial Upwelling provide preliminary evidence that Al5,18 is sensitive to nitrification in the mixed layer. These measurements show positive departures of the O isotope ratios from the expected 1:1 in the surface mixed layer, potentially indicating regeneration of nitrate in situ (Figure 6.5). The A15,18 becomes increasingly negative as nitrate tends to lower concentrations due to nitrate assimilation at the surface. Although this work is in its infancy, this novel integrative tracer promises to track the extent to which nitrate is regenerated in the mixed layer, in turn providing a clearer understanding of new and regenerated production at surface ocean. 154 6.3. Future research Numerous directions can be taken to develop the coupled N and O isotopes of nitrate as a tool to study the ocean N cycle. However, a robust understanding the physiological controls on the isotope effects imparted by biological N transformations is vital to the interpretation of nitrate isotope distributions in the environment. Research to further investigate the physiological bases of N and O isotope fractionation of nitrate would constitute a natural extension of the work that was presented here. As seen from the variability of the isotope effects during assimilation by eukaryotic algae (Chapter 2), elucidating the physiological controls on the magnitude of isotope effect during nitrate assimilation is foremost, as interpretation of N isotopes in the paleo-record rest on understanding the role of the environment in modulating the N isotope effect. Needoba has provided important insight into putative controls, such as the role of light in modifying observed N isotope fractionation (Needoba et al. 2004). His work also offers a wealth of data on internal nitrate concentrations under different growth conditions, and the concomitant N isotopic composition of both internal and external nitrate. In order to better understand the controls on cellular nitrate efflux, as well as the putative importance of vacuoles in nitrate assimilation, I have started to construct a physiological model of nitrate N (and O) isotope fractionation constrained by the observations of Needoba (Appendix 2). This work is in its preliminary stage, yet it may yield novel insights into cellular N physiology. The observations presented in Chapter 4 of this thesis highlight the mutability of the N and O isotope effects on nitrate during denitrification. Because we have little insight into environmental variables that can modulate the isotope effect associated with denitrification, future work should aim to identify such variables. Of particular interest is putative role of oxygen as modulator of isotope effect in lab cultures, and in the environment, for reasons outlined in Chapter 4. Another avenue of research involves testing the validity of the 1:1 rule in field incubations, to verify that nitrate assimilation and denitrification do indeed impart identical isotope effects on nitrate. Although I attempted to cover a broad range of 155 organisms that have different nitrate reductases, the conformity of the 1:1 trend could be widespread among microbes, however not universal. Hence, confirmation of the O-to-N ratio during assimilation and denitrification ultimately needs to be obtained in situ. Other questions of interest that stem from the work presented could involve investigation of the putative importance of NAP in imparting N and O isotope effects on nitrate in various environments, or so-called "aerobic denitrification." This could also link to the role of photo-heterotrophs in various aquatic systems, as the latter seemed to have increased propensity to reduce nitrate via NAP than did the denitrifiers (in culture). Although photo-heterotrophs are widely distributed in various environments, their contribution to bulk nitrate reduction is undetermined. The difference in the O-to-N fractionation ratio between marine and freshwater environments is utterly puzzling. While the distribution of nitrate N and O isotopes in the ocean seems to be reflected by the trends observed in culture, nitrate in freshwater ecosystems has an O-to-N ratio that is not directly explained by lab observations. This apparent discrepancy may stem from a fundamental difference in terrestrial and marine N cycles that is undetermined. Finally, further work should focus on examination of the intrinsic isotope effects of various nitrate reductases in vitro. Enzymatic isotope effects can yield important insights into catalytic mechanisms, and this is especially interesting given the striking difference in the N and O isotope effects imparted on nitrate by NAP versus all the other nitrate reductases investigated in this work. 156 Figure 6.1. Station locations from coring cruise OXMZ01MV aboard the RV Melville in November of 1999. Stations 8 to 17 have mid-depth [02] minima reaching below 3 JAM, while station 7 reaches a minimum [02] of ~5 pM; the empirical upper limit [02] for active water column denitrification is ~4-5 fiM (Cline and Richards 1972). The colour coding is used in subsequent Figures 6.2 and 6.3. Station 17 samples waters in the Soledad Basin, which has a sill depth of ~300 m. Station 9 samples waters of an unnamed basin with a sill depth of 460 m. 157 F i g u r e 6.2. For all stations collected during OXMZ01MV, depth profiles of nitrate (NCy) 5 I 8 0 (a), nitrate 8 I 5N (b), N* (c), [NO,] (d), and [02] (e). Colours follow Figure 6.1. Stations 3 (Santa Barbara Basin, red), 17 (Soledad Basin, purple), and 9 (unnamed basin, deep blue) are apparent from their low N* and high 5 I 5 N and 5' 80 at their bottoms N* in Santa Barbara Basin extends beyond the scale used here (Sigman et al. 2003). 158 Figure 6.3. N0 3 " 5 1 8 0 versus N0 3 " 5 1 5 N for all data reported here for OXMZ01MV stations 7-16 (grey circles) and a depth-binned average profile (bold black circles), and the station 7-16 average (bold black circles) and the individual OXMZ01MV stations (2-8, 17-20) further North of the tip of Baja along the California margin. The symbol size is proportional to water depth, scaled to a maximum water depth of 1450 m. 159 f = 0.5 ; £ = 5%o N C V N03" 5%o 0%o £ = 5 % o \ 8.5%o 3.5%o H 2 0 • N 0 3 " f 5%o 1 .75% 0 \ 0 % o N H 4 + — • N 0 3 " 1.5%o 1.5%o 0°/oo -Nitea-Gline-Figure 6.4. Decoupling of nitrate N and O isotopes under the cycle of nitrate assimilation and remineralization. (a) Nitrate assimilation followed by export of organic N out of the surface mixed layer. In this example, half of the initial nitrate is consumed (f = 0.5) and e for both N and O fractionation of nitrate is assigned a value of 5%o. 6 I 5N and 8 I 8 0 of unconsumed nitrate are equally enriched by an increment of 3.5%o, following a 1:1 trend, (b) Nitrate assimilation followed by remineralization in situ. Half of the nitrate is consumed, and e = 5%o for both N and O isotopes of nitrate. Regeneration of organic N and subsequent nitrification to nitrate are complete. If all regenerated nitrate is mixed with the remaining "new" nitrate, mass balance then requires that the final 5 , 5N of nitrate will be the same as 5'5N (5%o). However the 8 I 8 0 of the final nitrate is intermediate between that of the unused nitrate, which has been enriched in 8 I 8 0 by nitrate assimilation, and that of the new nitrate, which has a 5' 80 like that of ambient water (0%0; Casciotti et al. 2002). 160 -10 0 5 10 15 3 25 Z 30 35 40 45 50 -8 A15,18 (5180-2.9) -4 -2 * • Figure 6.5. All A15.18 values measured in depth profiles in August 2003 at stations in the Equatorial Pacific along two north-to-south transects (8°N to 1°S) at 95°W and 110°W, respectively. Lower nitrate concentrations correspond to measurements in the surface mixed layer where nitrate is consumed by phytoplankton. The negative incursion of A15,18 in and around 30 uM nitrate is along an isopycnal contiguous with the thermocline of the Eastern Tropical North Pacific. 161 References Altabet, M. A. 1988. Variations in nitrogen isotopic composition between sinking and suspended particles: implications for nitrogen cycling and particle transformation in the open ocean. Deep-Sea Res. Part I 35: 535-554. Bianchi, M , F. Feliatra, P. Treguer, M.-A. Vincendeau, and J. Morvan. 1997. Nitrification rates, ammonium and nitrate distribution in upper layers of the water column and in sediments of the Indian sector of the Southern Ocean. Deep-Sea Res. Part II 44: 1017-1032. Brandes, J. A., A. H. Devol, T. Yoshinari, D. A. Jayakumar, and S. W. A. Naqvi. 1998. Isotopic composition of nitrate in the central Arabian Sea and eastern tropical North Pacific: a tracer for mixing and nitrogen cycles. Limnol. Oceanogr. 43: 1680-1689. Casciotti, K. L., D. M. Sigman, M. G. Hastings, J. K. Bohlke, and A. Hilkert. 2002. Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater using the denitrifier method. Anal. Chem. 74: 4905-4912. Cline, J. D., and F. A. Richards. 1972. Oxygen deficient conditions and nitrate reduction in the eastern tropical North Pacific Ocean. Limnol. Oceanogr. 17: 885-900. Dore, J. E., and D. M. Karl. 1996. Nitrification in the euphotic zone as a source for nitrite, nitrate, and nitrous oxide at station ALOHA. Limnol. Oceanogr. 41: 1619-1628. Dugdale, V. A., and J. J. Goering. 1967. Uptake of new and regenerated forms of nitrogen in primary productivity. Limnol. Oceanogr. 12: 196-206. Eppley, R. W., and B. J. Peterson. 1979. Particulate organic matter flux and planktonic new production in the deep ocean. Nature 282: 677-680. Granger, J., D. M. Sigman, J. A. Needoba, and P. J. Harrison. 2004. Coupled nitrogen and oxygen isotope fractionation of nitrate during assimilation by cultures of marine phytoplankton. Limnol. Oceanogr. 49: 1763-1773. Knapp, A. N., D. M. Sigman, and F. Lipschultz. 2005. N isotopic composition of dissolved organic nitrogen and nitrate at the Bermuda Atlantic time-series study site. Global Biogeochem. Cy. 19. 162 Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2003. The effect of benthic nitrogen cycling on the delta N-l 5 and delta 0-18 of water-column nitrate. Geochim. Cosmochim. Ac. 67: A249-A249. Lehmann, M. F., D. M. Sigman, and W. M. Berelson. 2004. Coupling the N-15/N-14 and 0- 18/0-16 of nitrate as a constraint on benthic nitrogen cycling. Mar. Chem. 88: 1- 20. Lehmann, M. F., D. M. Sigman, D. C. Mccorkle, B. G. Brunelle, S. Hoffmann, M. Kienast, G. Cane, and J. Clement. 2005. Origin of the deep Bering Sea nitrate deficit: Constraints from the nitrogen and oxygen isotopic composition of water column nitrate and benthic nitrate fluxes. Global Biogeochem. Cy. 19: GB4005. Montoya, J. P., and J. J. McCarthy. 1995. Isotopic fractionation during nitrate uptake by marine phytoplankton grown in continuous culture. J. Plankton Res. 17: 439-464. Needoba, J. A., and P. J. Harrison. 2004. Influence of low light and a light: Dark cycle on N03- uptake, intracellular N03-, and nitrogen isotope fractionation by marine phytoplankton. J. Phycol. 40: 505-516. Needoba, J. A., D. M. Sigman, and P. J. Harrison. 2004. The mechanism of isotope fractionation during algal nitrate assimilation as illuminated by the N-15/N-14 of intracellular nitrate. J. Phycol. 40: 517-522. Shearer, G., J. D. Schneider, and D. H. Kohl. 1991. Separating the efflux and influx components of net nitrate uptake by Synechococcus-R2 under steady-state conditions. J. Gen. Microbiol. 137: 1179-1184. Sigman, D. M., K. L. Casciotti, M. Andreani, C. Barford, M. Galanter, and J. K. Bohlke. 2001. A bacterial method for the nitrogen isotopic analysis of nitrate in seawater and freshwater. Anal. Chem. 73: 4145-4153. Sigman, D. M., J. Granger, P. J. DiFiore, M. M. Lehmann, R. Ho, G. Cane, and A. Van Geen. 2005. Coupled nitrogen and oxygen isotope measurements of nitrate along the eastern North Pacific margin. Global Biogeochem. Cy. 19: GB4022. Sigman, D. M., R. Robinson, A. N. Knapp, A. Van Geen, D. C. McCorkle, J. A. Brandes, and R. C. Thunell. 2003. Distinguishing between water column and sedimentary denitrification in the Santa Barbara Basin using the stable isotopes of nitrate. Geochemistry, Geophysics, Geosystems 4: 1040-1059. 163 Wada, E., and A. Hattori. 1978. Nitrogen isotope effects in the assimilation of inorganic nitrogenous compounds. Geomicrobiol. J. 1: 85-101. Ward, B. B., K. A. Kilpatrick, E. H. Renger, and R. W. Eppley. 1989. Biological nitrogen Cycling in the nitracline. Limnol. Oceanogr. 34: 493-513. 164 Appendix 1: Computation of isotope effects The kinetic isotope effect is a direct function of the rates of reaction (kj) of the isotopically light versus the heavier substrate, where e (%o) = (k,/k2 - l )x 1000 (1) The subscripts 1 and 2 refer to the light and heavy isotopomers, respectively. The tendency for isotopically heavier molecules to display slower reaction rates results because heavier atoms for stronger bonds, and stronger bonds require more energy to disrupt and break. The (activation) energy (E) required to break a bond can be expressed as E = (l/2)hu (2) where h is Planck's constant and u is the vibrational frequency of the molecule along the decomposition coordinate. Molecules exhibit a number of different vibrational modes (Melander and Saunders 1980), and in simplified models (of Transition State Theory), a single vibrational mode along the decomposition coordinate is assumed to be dominant in bond rupture. The movement along the decomposition coordinate can be compared to a kind of vibrational movement that tears the molecule apart. Isotopically heavier molecules exhibit lower vibrational frequencies, such that they have lower potential energy (Equation 3). Thus, for heavy molecules, more energy is required to overcome the activation energy needed to instigate bond breakage than for light molecules; more energy means slower reaction rates. The theoretical isotope effect for a given chemical reaction can be approximated from first principles following algorithms that define Transition State Theory (TST). 165 With a highly simplified model of TST, ki/k2 can be calculated (Bigeleisen and Wolfsberg 1958): k,/k2 = iA,/iA2 [ 1 + G ( U , ) ( A U , . 2 ) ] (3) Ui = hc(-Ui)/kBT G ( U , ) = 1/2 - 1/Ui + l/e u ' " ' where c is the speed of light, kg is Boltzman's constant, T is the absolute temperature in degrees Kelvin, and v\ the vibrational frequency along the decomposition coordinate of the molecule at absolute zero (at the "zero point energy" level). The term v^u^L2 is the ratio of the vibrational frequencies of the respective transition state structures of the isotopomers. Because the transition state is a concept and not necessarily a tangible molecular structure (Thorton and Thorton 1980), it is not possible to measure directly. Instead, it can be approximated from the masses of the isotopic molecules, U t L | / u i L 2 = ( U 2 / U | ) ' / 2 where ui and u 2 are the reduced masses of the light and heavy molecule, respectively. Another unknown in Equation 4 is the vibrational frequency of respective molecules at ground state (v,). This can be measured spectroscopically, or approximated from Hooke's law, v,= l/(2rcc)x(kF/u) , / 2 kp is the vibrational force constant of the bond, in this case the N-0 bond of nitrate. The value of kp is the same for isotopes of a given molecule, and this quantity cannot be approximated from first principles, thus must be measured. 166 The vibrational force constant for the N-0 bond of nitrate is known (Mcgraw et al. 1965), such that we can derive the theoretical N and O isotope effects for nitrate reduction from Equation 3. Moreover, the vibrational frequencies for 1 4 N- and l 5 N -nitrate (u^ and v\s) are also known, which allows us to compute the theoretical N isotope effect for nitrate reduction not only from approximated frequencies, but also from empirical vibrational frequencies. This exercise was executed previously by Brown and Drury (1967). Table 2.2 summarizes the theoretical N and O isotope effects obtained from solving Equation 3 for dissociation of a single oxygen from nitrate. Our computed vibrational frequencies are presented in Table 2.2, along with empirical frequencies. Though the computed values appear close to empirical measurements, the small differences between corresponding frequencies yield isotope effects that are largely different, in biological terms ( l5e of 57%o compared 74%o from computed versus empirical frequencies, respectively). Clearly, isotope effects derived from computed frequencies must be interpreted with caution as they bear large uncertainty. Nevertheless, the theoretical N and O isotope effects calculated with computed frequencies indicate that the isotopes fractionate with a ratio of 1:2 (18e:'5e), and not 1:1, as prevails in this study. Also striking is that both of the theoretical N isotope effects presented in Table 2.2 are much higher than those that have been measured for nitrate reduction by nitrate reductase (at 15 to 30%o; Table 2.2). 15e calculated from empirical frequencies (74%o) rather resembles that measured for nitrate reduction catalyzed by Fe 2 + (75%o; Table 2.2). This similarity could be fortuitous, owing to the uncertainty associated with derivation of theoretical isotope effects from an arguably over-simplified model of TST. Still, 74%o 167 may be close to the "intrinsic" isotope effect, that which would be observed for thermic (i.e., uncatalyzed) decompositon of nitrate. The intrinsic isotope effect is the highest possible isotope effect for a given reaction, because the (hypothetical) transition state structure during thermic decomposition corresponds to the most unstable of the transition states, that with the highest potential energy. Often, as is the case for nitrate, the transition state of a reaction involves charge separation, which is energetically unfavourable. A participating catalyst, such as nitrate reductase, spreads the charge more widely, thus stabilizing the transition state and lowering the energy of activation. In doing so, the transition state is structurally closer to reactants or products, such that vibrational motions of the atoms in the transition-state molecule are similar to vibrations at ground state. The catalyzed isotope effect of a chemical reaction is then generally lower than its corresponding intrinsic isotope effect, because differences in vibrational frequencies at ground state versus transition state are smaller for more stable transition state structures. In other words, an enzyme lowers the isotope effect of a chemical reaction by stabilizing the transition state structure. That the isotope effect here is seemingly not sensitive to Fe 2 +, the reducing agent which forms a bond with the soon-to-be-extracted oxygen atom in the transition state complex, indicates that Fe2+-bonding likely does not act to stabilize the transition state structure. The ratio of l 8e:' 5e is 1:2 in Table 2.2 is hence somewhat meaningless, not only because our simplified model of TST is inadequate to compute the true ratio of reaction rates, but also because it is intended to model the thermal decomposition of nitrate, and not nitrate at the active site of nitrate reductase. The intimate role of nitrate reductase in effecting a decrease >in the N isotope effect compared to the intrinsic isotope effect, and 168 presumably a change in the O isotope effect as well, is too complex to predict from simplified models of TST. References Bigeleisen, J., and M. Wolfsberg. 1958. Theoretical and experimental aspects of isotope effects in chemical kinetics, p. 15-76. In I. Prigogine [ed.], Advances in chemistry and physics. Interscience. Brown, L. L., and J. S. Drury. 1967. Nitrogen-isotope effects in the reduction of nitrate, nitrite, and hydroxylamine to ammonia. I> In sodium hydroxide solution with Fe(II). J. Chem. Phys. 46: 2833-2837. Mcgraw, G. E., D. L. Benitt, and I. C. Hisatsune. 1965. J. Chem. Phys. 42: 237. Melander, L., and W. H. J. Saunders. 1980. Reaction Rates of Isotopic Molecules. John Wiley. Thorton, E. K., and E. R. Thorton. 1980. Scope and limitations of the concept of the transition state. Methods Enzymol. 64: 3-76. 169 Appendix 2 Physiological model of nitrate N (and O) isotope fractionation In order to conceptualize the role of nitrate uptake and efflux in modulating the expression of the N and O isotope effects imparted on nitrate by NR, I constructed a physiological model that reproduces the growth of a cell culture that is assimilating nitrate for growth. The model culture involves cells of the size and the time-dependent nitrogen requirements of T. weissflogii. The sensitivity of the model output is tested against ascribed changes in the rate of cellular nitrate uptake, the rate of nitrate reduction by nitrate reductase (NR), and the rate of cellular efflux: [NOs'lext - ku - [N0 3-] l n, — k j v j R * [PN] ( 1 ) « - k E -The variables [NCy] e x t, [N03~]mt, and [PN] refer to the molar concentrations of external nitrate, internal (intracellular) nitrate, and particulate nitrogen in the culture, respectively. The values for ku, k N R , and k g define the nitrate uptake rate, the rate of nitrate assimilation mediated by NR, and the rate of cellular nitrate efflux, respectively. Nitrate reduction by NR is the rate-limiting step during growth of eukaryotic phytoplankton (Berges and Harrison 1995) and thus constrains the growth rate of the cells. The assimilation rate, k j s j R , is defined according to the Michaelis-Menten model with an ascribed half saturation constant, KM(NR) of 4 5 pmol L"1, typical of diatom NR (Berges and Harrison 1995) . The maximum ascribed catalytic rate for NR, V M A X ( N R ) , ranges between 0.4 to 2 d"' between model runs. Equation 2 describes the assimilation rate, k^R, in terms of the above parameters and in terms of its sensitivity to the internal nitrate concentration, [NCy];™: k N R = V M A X ( N R ) X [NOyjim / ( K M ( N R ) + [NCyjim) (2) 1 7 0 The rate of assimilation for l4N03~, l 4 k N R , is equivalent to kwR defined in Equation 2. In order to impart N (or O) isotope fractionation at this step, NR is assigned an intrinsic N (or O) isotope effect of 25%o, in the range measured for eukaryotic NR (Ledgard et al. 1985). The rate of assimilation for , 5 N 0 3 " , 1 5k N R , is thus l 4kN R/1.0025. The rate of nitrate uptake into the cell is also specified according to the Michaelis-Menten model: ku = V m a x ( U ) X [N0 3 " ]ext / ( K M ( U ) + [N03"]ext) (4) The half saturation constant for nitrate uptake by eukaryotic phytoplankton is -0.4 umol L"', and 4 umol L"1 for denitrifiers (Berges and Harrison 1995; Parsonage and Ferguson 1983). The model's sensitivity to changes in KM(u> was tested for KMOJ) from 0.1 umol L" 1 to 4 umol L"'. Because gross nitrate uptake rates for unicellular plankton are not known, the maximum nitrate uptake rate, Vmax(u), is expressed as a multiplier of the maximum rate of nitrate assimilation, such that: VmaX(u) = M x V m a X ( N R ) (5) The value of the multiplier M in model runs ranges from 1.1 to 5. Uptake exceeds V max(NR), which defines conditions where assimilation does not exceed uptake, such that isotopic fractionation is always expressed in external nitrate. It follows that the rate of efflux from the cell cannot exceed the difference in the rates of uptake and assimilation of nitrate. Thus we defined the maximum rate of efflux, V , n a x ( E ) , as the difference between the effective rates of nitrate uptake and assimilation: V m a x ( E ) = ku - k N R (6) Cellular nitrate efflux in plants has been shown to increase with intracellular nitrate concentrations. The physiological mechanism(s) underlying cellular nitrate efflux is not known, so we defined the rate of efflux as follows: 171 k E = V m a X (E) x [N03"]i„, / (KM ( E)+ [N0 3"]i„t) (7) Efflux is thus sensitive to internal nitrate in a Michaelis-Menten fashion, though V M A X {E) is variable throughout growth. A range of KM<E) values is tested between model runs, from 0.1 mmol L"1 to 10 mmol L"1. This range spans the measured nitrate concentration at which NR enzyme activity is saturated, namely between 0.5 mmol L"1 to 2 mmol L~' for respective phytoplankton species (Berges and Harrison 1995). As will be shown in the following section, the concentration of nitrate surrounding NR remains saturating during exponential (steady-state) growth within the selected range of K M ( E ) -Model results Figure 1 depicts growth by a culture as defined in our model. Exponential growth of the cells (particulate nitrogen, PN) is synchronous with depletion of external nitrate. In this example, the isotopic enrichment of the nitrate pool expresses an isotope effect of ~6%o that is accompanied by an inversely equivalent isotope effect for the particulate pool. Figure 2 shows two mechanisms by which isotope effects can be modified in growing cultures. In Figure 2a, the nitrate N isotope effects increase with uptake rates. A specified maximum uptake rate (VmaX(u)) of only 1.1 times greater than the maximum rate of assimilation (Vm a x(NR)) yields a relatively small isotope effect of 1.8%o, while at 5-fold difference in VmaX(u) compared to V m a x (NR) results in an isotope effect close to 17%o. While only changes in uptake are specified to change the resulting isotope effect, efflux rates adjust according to Equation 7, which results in increased efflux-to-uptake as uptake rates increase. Changes in growth rate can also modify the isotope effect (Figure 2b), in the context of a constant Vmax(u). The maximum rate of assimilation, V m a x (NR), which is directly proportional to growth rate, is decreased while a constant rate of uptake is maintained. The observed isotope effect then increases with lower growth. This scenario may be analogous to light-limited growth of some phytoplankton (Needoba et al. 2004), 172 where decreased growth rates at low light were accompanied with an increase in the N isotope effect, from 6%o at high light to 15%o for T. weissflogii. Figure 3 illustrates the change in internal nitrate associated with the variations ascribed in Figure 2. We defined an initial internal nitrate concentration of 20 mmol L " 1 for the cells, which equilibrates within ~ 5 to 30 minutes to a steady-state concentration defined by the relative fluxes o f nitrate to and from the internal nitrate pool. Increasing or decreasing the size o f the initial internal nitrate pool in separate model runs changes the time required for cells to equilibrate slightly, following which the internal nitrate concentration remains invariant during exponential growth until exhaustion of the external nitrate pool. Figure 3a shows that the increase in nitrate uptake is accompanied by increase in intracellular nitrate, from less than 20 mmol L" ' when maximum uptake exceeds assimilation by a factor o f 1.1, to 110 mmol L"1 for a 5-fold increase in maximum uptake relative to maximum assimilation. Similarly, decreased growth and its concomitant increase in isotope effects is characterized by an increase in internal nitrate, from around 20 mmol L " 1 at V m a X ( N R ) O f 1 d"', to above 50 mmol L " ' at 0.4 d"1 (Figure 3b). Note that while a 5-fold increase in uptake and a decrease in assimilation to 0.4 d" 1 yield similar isotope effects (16%o), the treatments differ in their respective internal nitrate concentration (110 mmol L " 1 vs. 50 mmol L " 1 ) . This apparent discrepancy originates from a difference in the half saturation constant for efflux ( K M ( E J ) that was selected for the two sets of experiment (namely Figures 2a and 3a vs. Figures 2b and 3b). The higher K M ( E ) of 2 mmol L " 1 in 2a and 3a yields a lower relative rate of cellular efflux (Equation 7) that results in higher intracellular nitrate. Alternatively, the lower intracellular nitrate in 2b and 3b owes to the lower selected K M ( E ) of 0.5 mmol L " 1 . The effect of relative efflux on internal nitrate for a range of isotope effects is illustrated in Figure 4. For a given observed isotope effect, variations in K M ( E J result in a broad range in internal nitrate concentration, with progressively higher K M ( E ) values resulting in higher internal nitrate concentrations. Thus, similar isotope effects can result from different internal nitrate. To determine the K M ( E ) that yields internal nitrate concentrations consistent with empirical observations, N isotope effects and their corresponding internal nitrate, as measured for three strains of diatoms cultured in high vs. low light conditions, are also plotted in Figure 4 (Needoba et al. 2004). Seemingly, 173 no single KM(E) value replicates all of Needoba et al.'s observations: while the highest isotope effects observed for T. weisflogii at low light correspond to the highest KM(E) of 10 mmol L " 1 , and hence to low relative efflux compared to uptake, the observations for the other two cultures do not adhere to specific a KM(E). Though efflux is seemingly sensitive to internal nitrate levels, no robust relationship systematically explains empirical observations. Cellular nitrate efflux is thus arguably more complex than the equation to which it is ascribed in the model. This may be explained by storage of nitrate in vacuoles, which would effectively lower the internal nitrate concentration to which efflux is sensitive. In spite of our inability to constrain the regulation of internal nitrate, all model runs indicate that the isotope effect imparted on nitrate is ultimately determined by the relative rates of cellular nitrate efflux compared to uptake, regardless of internal nitrate. In Figure 5a, isotope effects obtained for all of our model runs correlate linearly with the ratio of cellular nitrate efflux to uptake, displaying a slope of 21.5, close to intrinsic N isotope effect of 25%o ascribed to NR. The points plotted here were derived for a range of values of V M A X ( U ) , KM(U ) , V M A X ( N R ) , and K M ( E ) - Hence a low relative rate of nitrate efflux compared to uptake yields a lower observed isotope effect. And conversely, higher relative efflux results in higher isotope effects. The canonical isotope effect of 5%o then corresponds to a ratio of nitrate efflux to uptake of 0.25. Similarly, Shearer et al. (1991) experimentally derived ratio of 0.2 for N isotope effect of 3.7%o for cyanobacteria, which falls squarely on the linear equation produced by our model. Another predictor of the observed isotope effect is also the difference in the 5 I 5 N (or 5 I 80) between internal and external nitrate. This value corresponds to the observed e with a slope of nearly -1. The value of the intercept, 21.6, is close to the intrinsic isotope effect of NR. In the Rayleigh model, the difference between the instantaneous 8 I 5 N of PN vs. that of external nitrate corresponds to the isotope effect. Similarly, the difference between the instantaneous 5 1 5N of internal vs. external nitrate in our model predicts the observed N isotope effect minus a term that is likely set by the intrinsic isotope effect of NR. Interestingly, the slope and intercept of our linear model reach -1 and 25%o, respectively, if we remove the model runs for which K M ( E ) equals or exceeds 2 mmol L " 1 . In order to reconcile the very high internal nitrate concentrations measured in some algae 174 with a consistent efflux mechanism (namely, one with a fixed K M ( E ) ) , we need to invoke the role of vacuoles in maintaining relatively lower cytoplasmic nitrate concentrations, which favours lower values for K M ( E ) . Nonetheless, the relationship derived from our model is comfortingly close to that obtained from the empirical observations of Needoba et al. (2004) for T. weissflogii. The latter slope is also close to - 1 , and the intercept seemingly close to that of NR (20.3). Note that the ascribed value of 25%o for NR of T. weissflogii has not been measured per se. The difference between the model intercept and that from Needoba's suggests that the intrinsic isotope effect for T. weissflogii NR may be somewhat lower than 25%o. The relationships depicted in Figure 5 are still only slightly sensitive to internal nitrate set by the value of K M ( E ) . Figure 6 shows how the internal nitrate concentration and the residence time of nitrate in the cell co-vary with decreasing efflux (higher K M ( E ) ) , such that for a given isotope effect (~6%o for the points plotted in Figure 6), the dilution rate of intracellular nitrate is constant across the range concentrations of internal nitrate. The isotope effect measured in external nitrate is then a direct function of the dilution rate of the internal nitrate pool. The half saturation constant for uptake, K M ( U ) , does not modulate expression of the isotope effect; rather, its magnitude determines the point of inflexion of Rayleigh model as substrate concentrations near full consumption (Figure 7). When the concentration of external nitrate approaches the value of KM(U)> nitrate 8 I 5 N is no longer linear with respect to the In (NOV), and the relationship asymptotes. At this point, the rate of nitrate transport into the cell is lower than nitrate demand by NR, and the rate of nitrate assimilation by NR becomes equivalent to the rate of supply by uptake. Cellular efflux is concomitantly reduced to zero. The K M ( U ) of marine phytoplankton is around 0.4 pmol L"1, such that inflexion in the Rayleigh plot would only be significant at relatively low concentrations of nitrate (~ 1 pmol L"1). In contrast, the K M ( U ) for denitrifiers is higher (4 pmol L"1), which is consistent with the asymptotic behaviour of N and O isotope effects observed by Granger et al. (in prep) for cultures of denitrifiers; inflexion of the Rayleigh plot hovered around external nitrate concentrations of 5 pmol I/1. 175 Figure 7 also illustrates the effect of growth rate on the point of inflexion of the Rayleigh plot. Higher growth rates result in higher fractionation of external nitrate owing to the corresponding change in V m a X ( u ) (Equation 5). In terms of Michaelis-Menten kinetics, increasing V m a X ( u ) while maintaining K M ( U ) results in a steeper initial slope, which means that cells can acquire nitrate faster at lower substrate concentrations. Though the Michaelis-Menten curve that describes nitrate demand by NR is also steeper at higher growth rates ( V M A X ( N R ) ) , the relative change in nitrate demand is smaller for NR than for uptake owing to the higher half saturation constant for NR ( K M ( N R ) = 45 umol L " 1 compared to 0.1 umol L " 1 for uptake in Figure 7). Hence nitrate supply by uptake exceeds demand for NR down to lower external nitrate concentrations, resulting in more extensive fractionation of external nitrate. Such patterns have been noted for denitrifiers (Chapter 4) where the difference in the asymptotic S I 5 N value in the Rayleigh plot for two cultures of a single strain could only be explained by differences in growth rate. The measurements presented earlier for nitrate N and O isotope fractionation of various plankton cultures hint at plausible O isotope fractionation associated with nitrate transport, though not all the data were internally consistent. Nevertheless, Figure 8 illustrates the impact of an isotope effect of 2%o imparted on the oxygen atoms of nitrate during transport of nitrate into or out of the cell (uptake or efflux). As hypothesized earlier, the impact of this isotope effect is more marked at lower overall isotope effects due to lower efflux, and becomes less significant as isotope effects increase. Hence N and O isotope fractionation is seemingly de-coupled at lower isotope effects, while 1 8e: l 5e approaches 1:1 as the respective magnitude of the N and O isotope effects increases. Thus l 8e: l 5e exceeding 1:1 could plausibly indicate O isotope fractionation associated with transport at the cell membrane, and this putative transport O isotope effect may be on the order of 2%o or less. Conclusions The results derived from our physiological model clearly demonstrate that the isotope effect imparted on nitrate N and O atoms by NR is propagated in external nitrate by cellular nitrate efflux. The isotope effects computed in various iterations of the model 176 are a direct function of to the ratio of cellular nitrate efflux to uptake, which ultimately determines the dilution rate of the internal nitrate pool. As such, higher dilution of the internal nitrate pool (larger relative efflux compared to uptake) results in a higher isotope effect propagated in external nitrate. We propose that the canonical isotope effect of 5%o observed for nitrate assimilation at the surface ocean and in plankton cultures reflects a stringency in the physiological mechanism underlying nitrate assimilation, wherein cells actively maintain a steady-state concentration of cytoplasmic (not vacuolar) nitrate that is subject to regulated cellular efflux. However, to back this claim, a more robust understanding of the regulation of nitrate storage in vacuoles is required. Moreover, our results highlight the potential for O isotope fractionation of nitrate imparted by cellular transport. If transport decouples nitrate N and O isotope fractionation from a 1:1 relationship at lower isotope effects, this needs to be considered when interpreting N and O isotope profiles in the mixed layer, where positive deviations in l 8e: l 5e may also reflect the degree to which nitrate has been regenerated in the mixed layer (Granger et al. 2004). 177 I I I I 0 2 4 6 8 Days 3 0 -2 0 ->? 1 0 -co 0 -0 2 4 6 8 Days Figure 1. Example of the patterns in time-dependent a) growth (PN- particulate nitrate) and nitrate depletion, as well as b) N isotopic fractionation of nitrate, derived from the theoretical model construct of a cell culture. In this example, the V m a x of NR is set at 1 d" ', V m a x for nitrate uptake is 1.4 times the V m a x of NR, K M for uptake is 0 .4 pmol L , and the K M for efflux is 0 .5 mmol L ' . Derivation of the N isotope effect from a Rayleigh plot yields an N isotope effect of 6.8%o for both nitrate depletion and particulate nitrogen (PN) accumulation. 178 Figure 2. Modulations in the isotope effect derived from the physiological model effectuated by a) varying the maximum rate of cellular nitrate uptake (Vm a x(u)) or by b) decreasing the maximum catalytic rate of NR (VM A X (NR)) while uptake rates are unchanged. Conditions in a) are set with V M A X (NR) at 1 d"1, K M ( U ) at 0.4 umol L " , and KM(E) at 2 mmol L " 1 . Conditions in b) involve VmaX(u) remaining at 1.4 times Id" 1, KM<u) at 0.4 umol L ' , and K M ( E ) of 0.5 mmol L " . 179 120 — 100 — 'h-l "o 80-o 60-p 40-J—1 <D •j—» G 20-0 6 0 -5 0 -"o g 4 0 -m o 3 0 -z 2 0 -1 0 -0 -l . l x 1.4x ' 1.5x ' 2.5x ' 5x v m a x u P t a k e (multiplier of V m a x NR) 0.75 Vmax NR (d"1) 0.40 Figure 3 . Differences in the internal nitrate concentration resulting from the modulations in Figure 3. (a) Internal nitrate corresponding to selected values for the V m a x of uptake, and (b) internal nitrate corresponding to decreased rates of nitrate assimilation by NR while uptake remains constant. The permil values are the isotope effects on external nitrate imparted by the respective treatments. Note that the difference in internal nitrate between plots (a) and (b) for isotope effects of the same magnitude reflect the difference in the KM(E) values used to compute the isotope effects in (a) vs. (b). 180 K M efflux o - 0.1 mmol L 1 • O - 0 . 5 mmol L 1 -o- 1 mmol L ' 2 mmol L 1 -o- 1 0 mmol L 1 -0 - T. r. HL • LL - C h T. p. HL • LL —A— T. w. HL A LL o 1 0 e (%0) 1 5 2 0 Figure 4 . Internal concentrations of nitrate observed for computed isotope effects as modulated by the K M for efflux. Also plotted are internal concentrations of nitrate vs. corresponding N isotope effects, as measured by Needoba et al. ( 2 0 0 4 ) for three diatoms species grown in light-replete and light-limited conditions. 1 8 1 2 0 -1 5 -£ 1 0 -5 -0.0 0.2 0.4 0.6 0.8 1.0 Ratio of efflux to uptake 20 15 £ 1 0 -i i - 1 — ~ i 1 r 0 5 10 15 20 25 N O ; 5 l 5 N i m - b]\M(%o) Figure 5. The isotope effects computed for all model runs plotted against a) their corresponding ratio of cellular nitrate efflux to uptake, and b) the difference in the S I 5 N of the corresponding internal and external nitrate. Also plotted is the regression line obtained by Needoba et al. (2004) for the diatom T. weissflogii grown under different light regimes. 182 0 2 4 6 8 10 K M efflux (mmol L"1) Figure 6. The residence time and the concentration of internal nitrate for cells fractionating at ~6%o as a function of KM<E), and the corresponding dilution rates of internal nitrate with increasing K M ( E ) -183 - 4 - 2 0 2 I n [N03"] F i g u r e 7. The Rayleigh plot of 6 I 5 N of nitrate vs. the In of nitrate for isotope effects of 6.7%o and 16%o (obtained from V m a x ( U ) of 1.4 and 5 times the V M A X ( N R ) , respectively), showing the effect of changes in the value of the KM(U> The closed symbols represent cells growing at a faster rate (VM A X(NR) = at 2 d"1), compared to open symbols for cells where V M A X ( N R ) = 1 d"1. The KM(E> for these model runs was arbitrarily set at 0.5 mmol L' . 184 transport = 2 % c ' transport = 0 % c transport = 2 % 0 ^transport = 0 % c 20 h- 15 e = 1.8 % 0 i i 1 r 2.0 2.5 3.0 3.5 4.0 4.5 l n [N03"] 10 r-5 5.0 Z o M O 00 o O •*) z O Figure 8. Impact of a 2%o isotope effect imposed on the oxygen atoms of nitrate from uptake and efflux on the l 8e:' 5e. a) Rayleigh plot of the 5 I 5 N and 6 1 8 0 associated with a V m ax(U) of 1.1 x V m a x ( N R ) (blue lines) and V m a X ( U ) of 1.4 x V m a x (NR) (red lines), respectively. K M ( E ) was set at 0.5 mmol L"\ b) Nitrate 5 1 8 0 vs. the corresponding 6 I 5 N. 185 References Berges, J. A., and P. J. Harrison. 1995. Nitrate reductase-activity quantitatively predicts the rate of nitrate incorporation under steady-state light limitation - a revised assay and characterization of the enzyme in 3 species of marine-phytoplankton. Limnol. Oceanogr. 40: 82-93. Ledgard, S. F., K. C. Woo, and F. J. Bergersen. 1985. Isotopic fractionation during reduction of nitrate to nitrite by extracts of spinach leaves. Aust. J. Plant Physiol. 12: 631-640. Needoba, J. A., D. M. Sigman, and P. J. Harrison. 2004. The mechanism of isotope fractionation during algal nitrate assimilation as illuminated by the N-15/N-14 of intracellular nitrate. J. Phycol. 40: 517-522. Parsonage, D., and S. J. Ferguson. 1983. Reassessment of pathways of electron flow to nitrate reductase that are coupled to energy-conservation in Paracoccus-denitrificans. FEBS Letters 153: 108-112. Shearer, G., J. D. Schneider, and D. H. Kohl. 1991. Separating the efflux and influx components of net nitrate uptake by Synechococcus-Kl under steady-state conditions. J. Gen. Microbiol. 137: 1179-1184. 186 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0053159/manifest

Comment

Related Items