THE EFFECT OF NITROGEN DEPRIVATION ON STABLE ISOTOPE 1 5N FRACTIONATION AND PREFERENCE OF AMMONIUM, NITRATE, AND UREA BY MARINE DINOFLAGELLATES AND A RAPHIDOPHYTE by Kugako Sugimoto B. S., University of Hawaii at Manoa, 1994 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Earth and Ocean Sciences) We accept this thesis as conforming to the required standard THTfUNIVERSITY OF BRITISH COLMBIA OCTOBER 1998 © Kugako Sugimoto, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of /WZX O-Kcl Oc^an S^/>C0*L££ and 47 ug-at N L"1 urea were added at 186.5 h. Bars represent + 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 33 Figure 3 N uptake rates of triplicate, batch cultures of Prorocentrum micans grown on NH 4+ , N0 3", and urea and under continuous light plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4+ , N0 3", and urea at t=0 to previously N-starved cultures for 56 h. (b) is an expanded version of the re-addition phase of (a). Cultures were N starved between 130 and 186.5 h and NH 4+ , N0 3 ' , and urea were added at 186.5 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 36 Figure 4 Relative N preference of N H 4+ and N 0 3 ' relative to urea for Prorocentrum micans during the period I before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation 38 Figure 5 Relative N preference of urea relative to N 0 3 ' for Prorocentrum micans during the period I I before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation 38 Figure 6 Growth of Amphidinium carterae grown on NH 4+ , N0 3", and urea under continuous light. Log plot of in vivo fluorescence versus time. Bars represent + 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 39 ix Figure 7 Nitrogen concentrations (urea as p.g-at N L"1) during the growth of Amphidinium carterae grown on NH 4+ , N0 3 ' , and urea under continuous light plotted against a) elapsed time and b) elapsed time since re-addition of NH 4+ , N0 3", and urea at t=0 to previously N-starved cultures for 75 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Cultures were N-starved between 119 and 193.5 h and then 29 uM NH 4 + , 30 uM N0 3 ' , and 30 ug-at N L"1 urea were added at 193.5 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 41 Figure 8 N uptake rates of triplicate, batch cultures of Amphidinium carterae grown on NH 4+ , N0 3 ' , and urea and under continuous light plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4*, NOV, and urea at t=0 to previously N-starved cultures for 75 h. (b) is an expanded version of the re-addition phase of (a). Cultures were N-starved between 119 and 193.5 h and NH 4+ , N0 3", and urea were added at 193.5 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 43 Figure 9 Relative N preference of N H 4+ and urea relative to N0 3 " for Amphidinium carterae during the period I before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation 45 Figure 10 Relative N preference of urea relative to N0 3 " for Amphidinium carterae during the period n before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation 45 Figure 11 Growth of Heterosigma carterae grown on NH 4+ , N0 3", and urea under continuous light. Log plot of in vivo fluorescence versus time. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 47 Figure 12 Nitrogen concentrations (urea as ug-at N L"1) during the growth of Heterosigma carterae grown on NH 4+ , NOV, and urea under continuous light plotted against a) elapsed time and b) elapsed time since re-addition of N H 4 \ N0 3 ' , and urea at t=0 to previously N-starved culture for 0 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). 31 uM N H 4 \ 41 uM N0 3", and 38 ug-at N L'1 were added at 175 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 48 X Figure 13 N uptake rates of triplicate, batch cultures of Heterosigma carterae grown on NH 4+ , N0 3 ' , and urea and under continuous light plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4+ , N0 3", and urea at t=0 to previously N-starved cultures for 0 h. (b) is an expanded version of the re-addition phase of (a). NH 4+ , N0 3", and urea were added at 175 h. Bars represents 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 51 Figure 14 Relative N preference of NH 4* and N 0 3 ' relative to urea for Heterosigma carterae during the period I before (•) and after (•) N starvation. The relative preference for N 0 3 ' is fixed at 1, both before and after N starvation 53 Figure 15 Relative N preference of urea relative to NOV for Heterosigma carterae during the period 11 before and after N starvation. The relative preference for urea is fixed at 1, both before (•) and after (•) N starvation 55 Figure 16 Growth of Heterosigma carterae grown on N 0 3 ' under continuous light. Log plot of in vivo fluorescence versus time. Bars represent + 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 56 Figure 17 N 0 3 ' concentrations during the growth of Heterosigma carterae grown on N0 3 " under continuous light plotted against a) elapsed time and b) elapsed time since re-addition of N0 3 " at t=0. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Cultures were N0 3"-starved between 64 and 116 h and then 88 L I M N0 3 " was added at 116 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 57 Figure 18 N0 3 * uptake rates of triplicate, batch cultures of Heterosigma carterae grown on N0 3 * and under continuous light plotted against a) average time between sampling and b) average time between sampling since re-addition of N0 3 " at t=0 to previously N-starved cultures for 51.5 h. (b) is an expanded version of the re-addition phase of (a). Cultures were N0 3 " starved between 64 and 115 h and then N0 3 " was added at 116 h. Bars represent + 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 59 Figure 19 Growth of Heterosigma carterae grown on N0 3 " and under a 14:10 L:D cycle. Log plot of in vivo fluorescence versus time. Stippled xi areas indicate dark periods. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 60 Figure 20 NOV concentrations during the growth of Heterosigma carterae on NOV and under a 14:10 L D cycle plotted against a) elapsed time and b) elapsed time after the re-addition of 73 uM N 0 3 ' at 215 h to N-starved Heterosigma carterae for 96 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Cultures were N0 3 '-starved from 119 h to 215 h. Stippled areas indicate dark periods. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 62 Figure 21 N0 3 " uptake rates of triplicate, batch cultures of Heterosigma carterae grown on N 0 3 ' and under a 14:10 L D cycle plotted against a) average time between sampling and b) average time between sampling since re-addition of N0 3 " at t=0 to previously N-starved cultures for 96 h. (b) is an expanded version of the re-addition phase of (a). Cultures were N 0 3 ' starved between 119 and 215 h and then N 0 3 ' was added at 215 h. Stippled areas indicate dark periods. Bars represent + 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible 64 Figure 22 Growth of Heterosigma carterae on N H 4+ under continuous light, culture 1. Log plot of in vivo fluorescence versus time 66 Figure 23 N H 4+ concentrations during the growth of Heterosigma carterae grown on N H 4+ under continuous light, culture 1, plotted against a) elapsed time and b) after the re-addition of N H 4+ at t=0 to N-starved cells for 33 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Culture was NH4*-starved between 65 and 98 h and then 77 uM N H 4+ was added at 98 h 67 Figure 24 N H 4+ uptake rates of a batch culture of Heterosigma carterae grown on N H 4+ and under continuous light, culture 1, plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4+ at t=0 to previously N-starved culture for 33 h. (b) is an expanded version of the re-addition phase of (a). Culture was NH4*-starved between 65 and 98 h and then NH 4+ was added 98 h 68 Figure 25 Growth of Heterosigma carterae on N H 4+ under continuous light, cultures 2 and 3. Log plot of in vivo fluorescence versus time. i J xii Bars represent ± 1 S.D. (n=1-2). Error bars are smaller than symbols where not visible 70 Figure 26 N H 4+ concentrations during the growth of Heterosigma carterae grown on N H 4+ under continuous light, cultures 2 and 3, plotted against a) elapsed time and b) elapsed time after the re-addition of N H 4+ at t=0, to N-starved cells for 18 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Cultures were NH 4+-starved between 88 and 106 h and then 74 uM NH 4 + was added at 106 h. Bars represent + 1 S.D. (n=1-2). Error bars are smaller than symbols where not visible 71 Figure 27 N H 4+ uptake rates of duplicate, cultures of Heterosigma carterae grown on N H 4+ and under continuous light, culture 2+3, plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4+ at t=0 to previously N-starved for 18 h. (b) is an expanded version of the re-addition phase of (a). Cultures were NH 4+-starved between 88 and 106 h and then NH 4* was added at 106 h. Bars represent + 1 S.D. (n=1-2). Error bars are smaller than symbols where not visible 73 Figure 28 Growth of Heterosigma carterae grown on various N sources and light conditions. Log plot of in vivo fluorescence versus time. ...78 Figure 29 Time series of [NCV] ( • ) , [NH 4+ ] (O) and [urea] ( T ) during the growth of Prorocentrum micans grown on NOV, NH 4 + , and urea under continuous light. Culture was N-starved between 105 and 186.5 h and then 36.6 ug-at N L'1 NH 4 + , 48.7 ug-at N L' 1N0 3", and 49.4 ug-at N L'1 urea were added at 186.5 h. The best fits for the concentrations were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentration, b is a co-efficient number, and c is the time when N reached 0, and are indicated by the solid lines 102 Figure 30 Time series of PN (A) and 5 1 5 N P N ( • ) during the growth of Prorocentrum micans grown on N0 3 ' , NH 4 + , and urea under continuous light. Culture was N-starved between 105 and 186.5 h and then 36.6 ug-at N L"1 N H 4 \ 48.7 ug-at N L"1 N0 3", and 49.4 ug-at N L"1 urea were added at 186.5 h. The 6 1 5 N P N as predicted by the multiple N source uptake model is shown by the solid line. The 3 dotted lines indicate times when [NH 4+], [NOV] and [urea] fits, determined by the dissolved N data to the concentration data in Fig. 29, reached 0, respectively 103 xiij Figure 31 Time series of [N0 3 ' ] ( • ) , [NH 4+ ] (O) and [urea] ( T ) during the growth of Amphidinium carterae grown on NOV, NH 4 + , and urea under continuous light. Culture was N-starved between 119 and 193.5 h and then 29.6 ug-at N L 1 NH 4+ , 34.3 ug-at N L"1 N0 3", and 31.2 ug-at N L"1 urea were added at 193.5 h. The best fits for the concentrations were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentration, b is a co-efficient number, and c is the time when N reached 0, and are indicated by the solid lines 107 Figure 32 Time series of PN (A) and 8 1 5 N P N ( • ) during the growth of Amphidinium carterae grown on N 0 3 \ NH 4 + , and urea under continuous light. Culture was N-starved between 119 and 193.5 h and then 29.6 ug-at N L"1 NH 4+ , 34.3 ug-at N L'1 NOV and 31.2 ug-at N L'1 urea were added at 193.5 h. The 8 1 5 N P N as predicted by the multiple N source uptake model is shown by the solid line. The 3 dotted lines indicate times when [NH 4+ ] , [N0 3"] and [urea] fits, determined by the dissolved N data to the concentration data in Fig. 31, reached 0, respectively 108 Figure 33 Time series of [ N 0 3 ] ( • ) , [NH 4+ ] (O) and [urea] ( T ) during the growth of Heterosigma carterae grown on N0 3", NH 4 + , and urea under continuous light. Culture was added 32.2 ug-at N L"1 NH 4+ , 41.2 u.g-at N L"1 NOV and 37 ug-at N L"1 urea at 175 h. The best fits for the concentrations were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentration, b is a co-efficient number, and c is the time when N reached 0, and are indicated by the solid lines 110 Figure 34 Time series of PN (A) and 8 1 5 N P N ( • ) during the growth of Heterosigma carterae grown on N0 3", NH 4 + , and urea under continuous light. Nitrogen additions at 175 h to the culture were 32.2 ug-at N L"1 NH 4+ , 41.2 ug-at N L'1 N0 3", and 37 ug-at N L'1 urea. The 6 1 5 N P N as predicted by the multiple N source uptake model is shown by the solid line. The 3 dotted lines indicate times when [NH 4+], [N0 3"] and [urea] fits, determined by the dissolved N data to the concentration data in Fig. 33, reached 0, respectively. 111 Figure 35 Time series of [ N 0 3 ] during the growth of Heterosigma carterae grown on N 0 3 ' under continuous light. Culture was N-starved between 64 and 115.5 h and then 85.6 ug-at N L'1 N0 3 " was added at 115.5 h. The best fits for the concentrations were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentration, b is a co-xiv efficient number, and c is the time when N reached 0, and are indicated by the solid lines 112 Figure 36 Time series of PN(A) and 8 1 5 N P N (#) during the growth of Heterosigma carterae grown on N03* under continuous light. Culture was N-starved between 64 and 115.5 h and then 85.6 ug-at N L'1 NOV was added at 115.5 h. The 8 1 5 N P N as predicted by the N source uptake model is shown by the solid line. The dotted line indicates times when [N03*] fits, determined by the dissolved N data to the concentration data in Fig. 35, reached 0 113 Figure 37 Response of the model to changes in a) e(NH/), b) e(N03"), and c) PNj in the N-sufficient condition for Prorocentrum micans grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4 116 Figure 38 Response of the model to changes in e(NH 4 +) in the N re-supply phase for Prorocentrum micans grown on three N sources. The solid line represents the response of the model using the base variables indicated in Table 4 117 Figure 39 Response of the model to changes in a) e(NH 4 +) and b) e(N03') in the N-sufficient condition for Amphidinium carterae grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4 119 Figure 40 Response of the model to changes in s(NH 4 +) in the N-sufficient condition for Amphidinium carterae grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4 120 Figure 41 Response of the model to changes in a) s(NH 4 +) and b) e(N03") in the N re-supply phase for Heterosigma carterae grown on three N sources. The solid lines represent the response of the model using the base variables indicated in Table 4 121 Figure 42 Response of the model to change in e(NH 4 +) in the N re-supply phase for Heterosigma carterae grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4 122 Figure 43 Response of the model to changes in s(N03") in the N-sufficient condition for Heterosigma carterae grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4 124 XV Figure 44 Response of the model to changes in e(N03") in the N re-supply phase for Heterosigma carterae grown on three N sources. The solid lines represent the response of the model using the base variables indicated in Table 4 125 xvi ACKNOWLEDGMENTS I would like to express my sincere gratitude to Dr. Paul J. Harrison for all his help in providing direction, encouragement, helpful advice, laboratory space, and funds for the experiments, as well as in reviewing this thesis. I would also like to thank Dr. Nathalie Waser for her direction on isotope fractionation analysis and helpful comments on this thesis. I would like to also thank Dr. F. J. R. Taylor for giving me a chance to study phytoplankton, providing cozy laboratory space with nice laboratory mates, such as Tonny Wagey and Juan Saldarriaga, a partial funding, and his comments on the thesis. I appreciate Dr. Al Lewis for helping me out from several emergencies and providing attention. Finally, I would like to thank Allen Milligan, Mingxing Guo, and Zhiming Yu for answering each of my questions in the laboratory and Bente Nielsen and Joe Needoba for helping with the isotope analysis. 1 INTRODUCTION overview Nitrogen (N) limitation determines the amount of primary production by phytoplankton in most coastal (Ryther & Dunstan 1971) and oceanic waters (Goldman et al. 1979). This has been expressed in terms of Liebig's "law of the minimum". The element in the lowest concentration relative to the other elemental requirements limits the production (Liebig 1840) and N is considered the principle limiting nutrient in the ocean. There are two kinds of primary production which are characterized by the form of N used. One is new production which is supported by allochthonous N sources such as N03" which is mixed into surface waters from the N03"-rich deep ocean, N2-fixation, riverine inflow and precipitation. Allochthonous N is defined as the encounter to the system for the first time. The other form of primary production is regenerated production. NH4 +, urea, and amino acids are regenerated N sources. These N sources are released by death, cell lysis, grazing, viral attack (Suttle et al. 1990), and animal excretion. The concentration of regenerated N in coastal regions fluctuates. The maximum natural concentration is ~5 uM and concentrations < 0.5 uM are frequently found (McCarthy 1980; Sharp 1983; Antia et al. 1991). Among new and regenerated N, N03', NH4* and urea (CO(NH2)2) are the three most abundant forms of N in the ocean that are used by phytoplankton. Marine phytoplankton in the nutrient-depleted layer of the ocean may experience rapid N supply by upwelling of N03" (Walsh et al. 1978) or animal excretion of NH 4 + and possibly urea (Dugdale 1967). Ammonium (NH4+) Atmospheric precipitation of NH 4 + is important for N supply into the ocean (Buijsman etal. 1991). NH4 +, however, can be released by biological activities such as zooplankton excretion, bacterial decomposition, and schooling fish. Regenerated NH 4 + contributed a significant portion of the primary production in the study of the spring bloom in the New York Bight (Conway & Whitledge 1979). In this study NH 4 + utilization as a percentage of NH 4 + plus N03" utilization was 59% for the inshore areas and 70% at the shelf-break. The shelf-break had less N compared to the inshore water and primary production depended more on regenerated N. Nitrate (N03~) NOV is the most abundant form of N in coastal regions (Sharp 1983) and the most important N source in a highly productive ocean (Harrison et al. 1987). The abundance of N0 3 ' shows seasonal trends. Surface N03" concentrations of coastal waters increase during winter (> 20 ug-at N L"1) and gradually decrease due to the utilization by phytoplankton during the spring bloom. Urea (CO(NH2)2) Urea contains two atoms of N and therefore 1 uM urea = 2 ug-at L"1 or 2 umol L"1. The concentration of urea also largely depends on biological processes such as excretion by marine organisms, especially fish. Atmospheric precipitation also supplies urea to the marine system (Timperley et al. 1985). 3 Turley (1986) reported the highest urea concentration (23 uM) in the southern German Bight. Eutrophicated waters periodically exceed 3 uM urea. (Remsen 1971; Remsen et al. 1972: Berman 1974; Steinmann 1976; Satoh & Hanya 1981; Ignatiades 1986). Physiology There are two steps in the utilization of N by phytoplankton. The first step is uptake which is the actual transport of the particular form of N across the cell membrane. The second step is assimilation. In the second process, NOV, N02', and urea are reduced to NH 4 + and are incorporated into amino acids during cellular metabolic processes (Lobban et al. 1985). The ambient N concentration, the relative abundance of different N forms, the physiological status of the phytoplankton, the availability of light, and temperature influence both uptake and assimilation of N by phytoplankton. Uptake NH 4 + can be taken up by facilitated diffusion with the indirect expenditure of cellular metabolic energy. This diffusion needs carriers to bind NH 4 + at the outer membrane and assist the passage across the membrane to the inner surface. Ion channels which are macromolecular pores that traverse the cell membrane, have been invoked to explain the high transport rates that occur down an electrochemical-potential gradient (Hedrich & Schroeder 1989). NH 4 + may also be taken up by active transport processes. 4 NOV is actively taken up and hence this process requires energy. Carriers move N03" across the membrane against an electrochemical potential gradient. External concentrations of NOV are in the micromolar NOV range while intracellular concentrations are in the millimolar range. This large concentration difference between the inside and outside of the cell makes passive diffusion along an electrochemical gradient less important (Reed 1990). Active transport typically is much slower than channel-mediated transport because it is possible to transport only 103- 105 ions per second by active transport. In contrast, channels may allow 106 ions per second (Lobban & Harrison 1994). Urea is an uncharged molecule and therefore it diffuses through the cell membrane. However, phytoplankton have an active transport system and are capable of accumulating urea intracellularly. Light stimulates urea uptake and proton ionophores inhibit the uptake of urea. Thus, urea transport by phytoplankton requires ATP. Urea uptake does not directly require light (Syrett 1981). However, the production of ATP for membrane transport (Falkowski 1975) and the assimilation of urea occur during photosynthesis and this makes it appear that urea uptake depends on light. Light intensity and N uptake show a hyperbolic relationship with N uptake showing saturation at high light intensities (Eppley & Rogers 1970). In the ocean, light availability depends on the season, latitude, day/night cycle, and attenuation with depth. Phytoplankton in the ocean experience various light intensities in a daily periodic fashion. Malone et al. (1975) reported the periodicity of N uptake by various species in culture. The maximum N uptake 5 occurs during the day and the minimum uptake occurs at night (Maclsaac 1978; Fisher et al. 1982; Cochlan et al. 1991). Assimilation NH 4* can be directly assimilated into amino acids, usually as glutamine. Incorporation of N H 4 + into amino acids occurs in the chloroplast. Glutamate dehydrogenase (GDH) which has a low affinity for N H 4 \ is used when the internal pool of NH 4* is high. On the other hand, glutamate synthetase (GS) has a high affinity for N H 4 + and is used when the internal pool has low N H 4 + concentration. Glutamine oxoglutarate aminotransferase (GDF) follows the reaction by GS. The control of N H 4 + assimilation is often assumed to be rate-limiting for N incorporation (Syrett 1981). N 0 3 ' must first be reduced to N 0 2 ' by nitrate reductase using NADH or NAD(P)H as the electron donor. Nitrate reductase has been isolated and purified from many microalgae (Syrett 1981; Tischner etal. 1989; Solomonson & Barber 1990). The electron donor is usually NADH. However, NAD(P)H must be the electron donor in some phytoplankton (Lee 1980; Syrett 1981; Solomon & Barber 1990). Secondly, the product, N02", needs to be reduced to N H 4 + by nitrite reductase with the help of ferredoxin as the electron donor. These steps follow the equation below: 2e" 6e' N0 3 " > N0 2" > N H 4 + 6 Therefore, growth on N03" requires extra energy. In addition to the high energy cost during the assimilation step, the energy cost for N03" uptake is higher than that for NH 4 + (Falkowski 1975; Turpin & Bruce 1990). Since cells grown on N03" require an extra cost in reduction, the demand is increased for reducing power relative to ATP (Turpin & Bruce 1990). Both a higher photosynthetic quotient (Cleveland et al. 1989), and lower growth rate (Paasche 1971; Ward & Wetzel 1980; Rhee & Lederman 1983) occur. Several studies, however, reported that growth rates on NOV were not significantly different from those on NH 4 + (Conover 1975; Syrett 1981; Thompson etal. 1989). Phytoplankton grown on N0 3" might compensate for their higher reductant requirements by adjustments in their biochemical composition, rather than by reducing their growth rates (Thompson et al. 1989). The reason why uptake of N03" is affected by photoperiod is due to the diel periodicity in activity and synthesis of the nitrate reductase enzyme (Syrett 1981). N03'-sufficient cells showed more pronounced diel periodicity of NOV uptake than N-limited cells. When urea is transported into the cell, it is broken down enzymatically using urease or urea carboxylase and converted to NH 4 + as the final N product (Syrett 1981, 1988). The equation for this reaction is: CO(NH2)2 + H 20 + 2H+ -> C 0 2 + 2NH4 + This assimilation requires energy. In terms of reductant, urea assimilation is similar to NH4 +. 7 Kinetic parameters A hyperbolic function of the concentration of the limiting nutrient (similar to the Michaelis-Menten equation for enzyme kinetics) expresses the rate of steady-state N uptake by phytoplankton (Eppley & Coatsworth 1968). Ks is the half-saturation constant that represents the substrate concentration at which the uptake rate is half its maximum. Spatial and seasonal abundance of a particular species partially depends on the value of these N uptake parameters (Maclsaac & Dugdale 1972). For example, species that possess a lower Ks have an advantage because they take up N faster than other species at low N concentrations. Frequently the uptake rate varies with time in the case of N H 4 + limitation (Probyn & Chapman 1982; Rosenberg era/. 1984; Thomas & Harrison 1987) and NOV limitation (Thomas & Harrison 1987). A short incubation period can solve these problems by estimating the maximal uptake rate (Harrison et al. 1989)'. Starvation and rapid uptake Rapid uptake of NH 4* by N-starved phytoplankton is an example of the effect of cellular physiological state on N uptake (Glibert & Goldman 1981). N H 4 + uptake was facilitated by starvation and showed much more rapid uptake of N H 4 + by N-starved phytoplankton than by the N-replete phytoplankton (Suttle & Harrison 1988). Rapid uptake of N0 3 " , however, did not consistently occur after N starvation (Price & Harrison 1988). Increased urea uptake rates were also not observed after N starvation (Bekheet & Syrett 1979; Price & Harrison 1988). 8 Generally, the effect of N starvation causes phytoplankton to take up N H 4 + at rates in excess of those rates required to meet their ordinary N demands during balanced growth. Ecology Culture and field studies exhibit the uptake interactions between the various forms of N (Antia et al. 1991). The utilization of different forms might affect algal growth, species composition, and species succession (Harrison & Turpin 1982). This is because limiting nutrients have selective roles due to the nutrient uptake kinetics of the different species. For example, species that are capable of rapid uptake of a transient supply of N H 4 + and urea (in a patch) would have an advantage because such cells need only to be exposed to an intermittent pulse of N H 4 + in order to acquire their daily ration of N for growth (McCarthy & Goldman 1979). Interactions 1 )NH 4 + and N0 3 " The interaction of NOV and N H 4 + in uptake has been well studied. Commonly, NH 4 * inhibits N 0 3 ' uptake due to its preferential uptake over N 0 3 ' as a N source. Chlorella vulgaris that was grown with NH 4* and N0 3* ceased N0 3 " assimilation after the addition of NH 4* (Syrett & Morris 1963). These cells started to take up N0 3" immediately after N H 4 + was used up in the medium. Chlamydomonas reinhardi also showed a similar phenomenon (Thacker & Syrett 1972a,b; Florencio & Vega 1983). Not only phytoplankton, but also macroalgae such as Gracilaria foliifera and Neoagardhiella baileyi preferred N H 4 + over N0 3 " as a N source (D'Elia & deBoer 1978). The presence of N0 3 " did not decrease N H 4 + uptake by these macroalgae. The N H 4 + concentration affected the degree of inhibition in NOV uptake in the blue-green alga Anabaena cylindrica (Ohmori et al. 1977). In a lab study with Dunaliella tertiolecta and in a field investigation with Skeletonema costatum, N H 4 + inhibited N0 3" uptake (Conway 1977). This inhibition of N0 3 " uptake usually occurred above a certain range of NH 4* concentration. Oyster-pond microalgae, however, showed a long-term elevated N 0 3 ' uptake even in the presence of high NH 4* concentration (Collos et al. 1989). The maximum N0 3 " uptake by these microalgae occurred when N H 4 + concentrations ranged between 10 and 30 uM and this maximum coincided with the maximum N H 4 + uptake rate or with the start of a decrease in NH 4* uptake rate. N0 3 " uptake, then decreased and reached a secondary maximum rate again at low N H 4 + concentrations. 2) Effect of the starvation on interaction between N H 4 + and NOV N-deficient Skeletonema costatum showed strongly inhibited carrier-mediated N 0 3 ' uptake without affecting diffusion (Serra et al. 1978). The N-deficient batch culture of the same species enhanced N0 3" uptake rate and accumulated a large N0 3" internal pool after the addition of a mixture of N H 4 + and N0 3 " (DeManche & Curl 1979). N0 3" uptake ceased following this initial N0 3 " uptake and N0 3" assimilation was inhibited until the ambient N H 4 + was 10 taken up. At this point, NOV uptake and assimilation commenced. In contrast, in the kelp, Laminaria groenlandica, N0 3* uptake was suppressed for 30 min after N0 3 " and N H 4 + were added back together to the N-starved macroalga (Harrison etal. 1986). 3) Effect of light limitation on the interaction between N H 4 + and N 0 3 ' Syrett and Morris (1963) used two different irradiances in investigating the preference of N0 3 " and NH 4 + . The pattern of the inhibition of NOV uptake by NH 4 + , however, did not show any difference between high and low irradiance. However, the higher irradiance promoted the process because N H 4 + disappeared faster and N0 3" uptake commenced earlier in the time course. 4) Interactions between N H 4 + and urea The first study of the interaction between inorganic N (NH 4 + and N 0 3 ) and organic N (urea) was done by Lui and Roels (1970). N H 4 + repressed the rate of urea uptake by the chrysmonad Ochromonas malhamensis in their study. Later, Horrigan and McCarthy (1982) showed that N-replete Thalassiosira pseudonana did not decrease urea uptake during the interaction with 10 uM NH 4* for the first minute. In addition, urea uptake increased 5 min after these cells were exposed to NH 4 + . Skeletonema costatum, however, showed urea uptake inhibition by N H 4 + (Horrigan & MaCarthy 1982). This inhibition required 30 min exposure to NH 4 + . According to another report by Lund (1987), inhibition of urea uptake occurred immediately after exposure to NH 4 + . Both Thalassiosira pseudonana 11 and Skeletonema costatum showed increased urea uptake inhibit ion by N H 4+ under N-deplete condit ions. The extent of inhibition of urea uptake may be a funct ion of N H 4 * concentrat ion, the N status of the cells and the species (Molly & Syrett 1988a; 1988b). N H 4+ at 350 uM strongly inhibited urea uptake by Chlorella emersonii, but not by Phaeodactylum thcornutum. Urea uptake at N H 4+ > 500 a M was strongly inhibited, particularly for Phaeodactylum thcornutum that was N-starved for 48 h. In contrast, 4 mM N H 4+ was sufficient to inhibit urea uptake if the cells were N-replete. The addit ion of urea also depressed N H 4+ uptake by Pheodactylum thcornutum (Molloy & Syrett 1988a). The uptake rate of u rea-grown cells decreased to near zero when such cells were cultured and resuspended in 5 mM N H 4+ for 24 h (Rees & Syrett 1979). Probably, N H 4+ suppressed the development of the urea uptake mechanism in Phaeodactylum thcornutum. 5) Interactions between N03" and urea Urea inhibited NOV uptake in most experiments. In particular, such inhibit ion clearly occurred under dark and N-starved condit ions (Mol loy & Syrett 1988b). Nitrate reductase activity of the diatom, Cyclotella cryptica decreased due to the presence of urea (Liu & Hellebust 1974). 1 mM urea inhibited N 0 3 " uptake in the prasinophyte Platymonas (Ricketts 1988). However, 1 m M of NOV totally inhibited urea uptake (Ricketts 1988). These observat ions agree wi th the 12 general idea that urea suppresses N 0 3 ' uptake, but at a lower level than N H 4+ (Grand etal. 1967; McCarthy & Eppley 1972). Interestingly, NCV-sufficient cells showed enhanced uptake rates of both urea and N H 4+ (Horrigan & McCarthy 1982). Later, however, Price and Harrison (1988) reported that N0 3 '-replete Thalassiosira pseudonana did not exhibit an enhanced rate of urea uptake. Lund (1987), however, reported urea did not suppress NOV uptake by Skeletonema costatum. On the other hand, N0 3 " suppressed urea uptake by the same species. In general, urea uptake is a constitutive property of phytoplankton growing on N 0 3 ' as the sole N source. In the presence of NH 4+ , N0 3", and urea, NH 4* are preferentially taken up, followed by N0 3 "and urea (Dugdale & Goering 1967; Paasche & Kristiansen 1982; Levasseur et al. 1990). Phytoplankton prefer N H 4+ (more reduced form) over NOV and urea (more oxidized forms) because N H 4+ does not require a supply of electrons for reduction in the assimilation process which means less energy is required (Dugdale & Goering 1967; Eppley etal. 1973; McCarthy etal. 1977). 6) Field study In a field study of the interactions among NH 4+ , N0 3", and urea, all three N sources were taken up simultaneously in a natural spring phytoplankton assemblage from oyster ponds (Robert & Maestrini 1986). The urea uptake rate in this study, however, increased almost 10-fold with depletion of both N H 4+ and NOV Interestingly, algal species composition changed within this assemblage. 13 Inorganic N (NH 4+ and NOV) promoted diatoms. On the other hand, urea supported the growth of dinoflagellates. Simultaneous uptake of urea and other N sources by phytoplankton communities were reported in other field studies (McCarthy & Eppley 1972; Price etal. 1985). Isotopic fractionation N uptake and assimilation by phytoplankton is associated with N isotopic fractionation. The two naturally occurring stable isotopes of N are qualitatively similar in chemical behavior. Their reaction, however, occurs at slightly different rates and generally, 1 4 N reacts more quickly than 1 5 N due to lighter atomic mass. The isotopic composition of the product of a reaction could be affected by that of the substrate even though the difference in reaction rate is small. The product has the same isotopic composition as the original substrate if the reaction is complete. On the other hand, if the reaction is not complete, the resulting product pool is depleted in 1 5 N relative to the original substrate. This is one of the important fractionation processes in the biogeochemical cycle of N in the ocean. Isotope expression The relative abundance of the stable isotopes of N is expressed as a ratio. For example, 1 5 N/ 1 4 N = 0.00677. The isotope ratio of a sample is measured relative to a standard. The del value expresses this relative difference and is defined as 14 8 in % o - ( R sample " R standard )/(R standard) X 1000 where R is the isotope ratio with the most abundant isotope in the denominator. Isotope fractionations Variations in relative isotope abundance are derived from the preferential reaction of one of the isotopes. Fractionation occurs during physicochemical processes. The fractionation factor (a) expresses the degree of segregation of one isotope as a result of a particular process and a is defined as: CL = R products reactants This equation can be expressed as: (a - 1 ) 1000 « 8Products - 8reactants and a is dependent on the environmental conditions. The isotopic composition of the product varies as a result of the extent of reaction. This process is referred to as Rayleigh distillation. The effect of Rayleigh distillation on the isotopic composition of the remaining reactant is expressed by Rf/R0 = F( a 1 ) where f is the fraction of remaining reactant. R f is the del value of the reactant at a particular f. Ro refers to the del value when f = 1, which indicates any reactant that is not removed. e (i.e. the per mil enrichment factor of the substrate relative to the product) is typically > 0 and expressed as: e = (a - 1 ) x 1 0 0 0 15 Rayleigh equation (Mariotti etal. 1981) provides the magnitude of fractionation for a single-step, unidirectional reaction: 51 5Nproduct = 515NSubstrate, t=0 " E*- f/(1 - f)*ln f A positive e indicates that the light isotope accumulates more quickly than the heavy one. Montoya and McCarthy (1995) reported interesting results in which four flagellates, such as Isochrysis galbana, Pavlova lutheri, Dunaliella tertiolecta, and Chroomonas salina showed a lower fractionation factor (1 - 3%o) for NOV uptake and diatoms such as Skeletonema costatum and Thalassiosira weissflogii had higher fractionation factors (9 - 12%o). These authors suggested that the mobility of flagellates might result in a smaller diffusion zone around the cell, which would reduce the potential of a fractionating step during the uptake for N0 3 " across the outer cell membrane. Nitrogen isotope fractionation by phytoplankton Recently the use of natural stable isotopes of N has provided a new way to examine the relationships between N sources and primary production (Francois etal. 1992; Calvert etal. 1992; Farrell etal. 1995). The isotope fractionation in various oceanic regimes can differ in terms of N availability, N source and species composition. The sedimentary 1 5 N/ 1 4 N ratio underlying nutrient-rich oceanic regimes can be used as a tracer of past relative N0 3 " utilization and under certain assumptions, of past productivity (Calvert et al. 1992; Farrell et al. 1995). Low particulate 1 5 N/ 1 4 N ratios of bulk sedimentary 16 organic matter and N0 3 '-rich waters and vice versa showed correspondence (Francois etal. 1992; Altabet & Francois 1994; Farrell etal. 1995). This correspondence results from the isotopic fractionation during the incorporation of NOV by phytoplankton. However, the factors which control fractionation by phytoplankton require further investigation (Goericke et al. 1994). The isotopic fractionation associated with the uptake of N0 3", N02~, and N H 4+ has been studied in the laboratory (Wada & Hattori 1978; Wada 1980; Montoya & McCarthy 1995; Pennock et al. 1996). The degree of isotopic fractionation during N uptake by the diatom Phaeodactylum thcornutum clearly depended on the culture conditions (Wada & Hattori 1978). The isotope fractionation factor for N0 3 " uptake by light-limited Phaeodacytylum thcornutum was inversely related to the specific growth rate of the cells (Wada & Hattori 1978). The same results were obtained by Wada (1980) in the study of fractionation factors for NH 4+ uptake by the diatom Chaetoceros sp. Field studies on fractionation factors for the uptake of NH 4+ and N 0 3 ' by phytoplankton agreed with those from culture studies (Altabet & McCarthy 1985; Cifuentes et al. 1989; Montoya etal. 1991). The isotopic fractionation factor shows species differences. The diatom, Thalassiosira weissflogii exhibited a decline in the magnitude of the isotopic fraction factor for N 0 3 ' uptake with growth rate (Montoya & McCarthy 1995). Other tested species did not vary consistently with growth rate. These results suggest that the net isotopic fractionation during growth on N0 3 " is insensitive to growth rate. These authors concluded that the net isotopic fractionation may vary with photon flux density, which is supported 17 by the earlier studies by Wada (1980). Recently, Pennock et al. (1996) reported that fractionation was independent of the NOV concentration. The NH 4+-grown diatoms, however, showed a dependency on the NH 4+ concentration and fractionation increased as the NH 4 * concentration increased. N u t i l i z a t i on by d i no f l age l l a t e s a n d f l age l l a t es Like diatoms, dinoflagellates can use NH 4+ , N0 3", and urea. As in other phytoplankton, NOV and urea must be converted to N H 4+ before incorporation into amino acids (Parsons & Harrison 1983). Dinoflagellates can store N 0 3 ' internally. Internal concentrations of inorganic N were 1.8 mM N 0 3 ' and 8.2 mM N H 4+ in N 0 3 ' and NH 4+-grown Amphidinium carterae (Dortch et al. 1984). When the same species was grown on NH 4+ , the internal pool was 92 mM. The activity of nitrate reductase depends on the presence of N 0 3 ' (Dortch & Maske 1982). Hersey and Swift (1976) reported the highest activity of nitrate reductase in Amphidinium carterae was twice as high at mid-day as at night. Some dinoflagellates did not exhibit this difference in laboratory and field studies (Packard & Blasco 1974; Eppley & Harrison 1975; Harrison 1976). Some dinoflagellates and flagellates migrate vertically downward to take up N0 3 " at night below the nutricline and upward during the day for photosynthesis. These species possess high dark respiration activity that may account for this uptake activity. In addition, these cells used NADH (predominantly respiratory) rather than NADPH (chiefly from photosynthesis) for reduction of N0 3 * (Eppley et al. 1969a). Nitrate reductase activity was less 18 under low light condition than under high light condition (Maclsaac 1978). The same authors reported little or no rhythmicity of nitrate reductase activity. However, migrating species such as Gymnodinium sanguineum (Dortch & Maske 1982) and Gonyaulax polyedra (Eppley et al. 1969b) did take up and reduce NOV in low light or darkness. Dinoflagellates possess both enzyme pathways, GDH and GS/GOGAT, for N H 4 + incorporation. Amphidinium carterae, however, exhibited no GDH (Turpin & Harrison 1978). Such species might have an advantage in an N H 4 + -limited environment because GS has a higher affinity for N H 4 + than GDH (Falkowski & Rivkin 1976). The species studied 1) Amphidinium carterae Amphidinium carterae Hulburt is a temperate naked marine dinoflagellate which occurs in shallow waters (Hulburt 1957). The average cell volume of Amphidinium carterae is 152 um 3 (Needoba 1995). A growth rate of 0.76 d' 1 was reported for this species grown under a 14:10 L:D cycle at 20 °C when 50 uM N H 4 + was given to the culture at the beginning of the light period (Wheeler et al. 1983.) Some species of the genus Amphidinium have been associated with red tides and high concentrations have been reported to discolor sands in subtidal areas (Martin 1929; Herdman 1924a,b). Halstead (1965) described this species as a toxic dinoflagellate. Cells of Amphidinium carterae are toxic to fish and to mice (Thurberg & Sasner 1973). Because this toxin inhibited heart activities of 19 several marine animals, an acetylcholine analog was considered to be present in this organisms (Wangersky & Guillard 1960). Ikawa and Sasner (1975) identified toxins such as acrylycholine and choline o-sulfate in Amphidinium carterae. Seasonal fish kills by Amphidinium carterae blooms in the Sado estuary in Portugal was suspected (Sampayo 1985). Cyst formation has been reported and cycts remain healthy under a wide range of conditions (Sampayo 1985). Encysted stages survive conditions that would destroy the corresponding motile stage, enabling this species to repeatedly occur every year. 2) Heterosigma carterae The raphidophyte Heterosigma carterae is a marine naked unicellular biflagellate phototrophic flagellate (Yamochi 1983; Yokote et al. 1985). The classification of this organism was reviewed by Taylor (1992) and renamed from Heterosigma akashiwo. The cells were 8 - 25 um in length and 6 -15 pirn in width (Hara 1990). The growth rate of this species was < 0.8 d'1 under a 14:10 L D photoperiod of 160 umol photon m"2 s"1 at 20°C with 100 uM of each N source (Hoe Chang & Page 1995). A growth rate of 0.9 d'1 was reported when the culture was grown at 18°C on N H / of 75% E S A W ; N H 4 + and N0 3 " quotas grown on the same condition ranged from 46 to 47 and from 43.3 to 53 pg cell' 1 respectively during the exponential phase (Wood & Flynn 1995). Heterosigma carterae has been identified as a major causative organism of noxious brown tide blooms in western Canada (Taylor & Haigh 1993), temperate and subtropical embayments in Japan (Yamochi 1989), Korea (Park 1989), 20 Singapore (Chang 1993), New Zealand (Larsen & Moestrup1992), England (Larsen & Moestrup 1989), western areas of North America (Tomas 1982) and Bermuda (Tomas 1978). Heterosigma carterae possesses toxins that causes massive fish kills such as salmon, yellowtails, sea bass and other fish in cages (Larsen & Moestrup 1989; Hallegraeff 1991; Honjo 1992). Four neurotoxic components of Heterosigma carterae were identified (Khan et al. 1997). This species is known to exhibit diel vertical migration (Nagasaki et al. 1996). There are several characteristics in the termination process of Heterosigma carterae brown tides. The population ceases diel vertical migration in situ and the brown tide disappears suddenly (Nagasaki etal. 1996). There is a size transition such that smaller-sized cells dominate toward the end of a simulated brown tide in outdoor tanks (Honjo & Tabata 1985). Lytic activities of the viruses suggest that they can be involved in regulating the bloom dynamics. Virus-like particles (VPLs) from infected Heterosigma carterae were isolated and showed strain-specific infection to Heterosigma carterae (Nagasaki & Yamaguchi 1997). Heterosigma carterae is known to possess benthic stages, including cysts, in the life cycle (Tomas 1978). This stage plays an important role as a seed population for brown tides. Cycts excyst between 10 and 15°C (French et al. 1995). 3) Prorocentrum micans Prorocentrum micans is a common constituent of coastal phytoplankton (Reid et al. 1985). Non-toxic blooms of Prorocentrum micans have been 21 reported in several coastal waters (Sweeney 1975; Cassie 1981; Avaria 1982; Hata et al. 1982; Pybus 1984). The cells were measured 35 - 70 um in length and 20 - 50 um in width (Horiguchi 1990). Growth rates of 0.59 - 0.99 d' 1 were reported for this species grown in f/2 medium under continuous light and 19°C (Costas 1990). The minimum cell quota for N was estimated to be 0.74*10' 1 2 mol cell"1 (Wang et al. 1996). A distinct phototactic response is characteristic to these species. The accumulation of Prorocentrum micans occurs in surface waters (Harvey 1966) and in the upper layer of the water column (Hattori et al. 1983). Therefore, they are exposed to a wide range in the intensity and quality of solar radiation. UV-absorbing compounds and yellow xanthophyll, and diadinoxanthin might protect cells by screening them from harmful radiation (Vernet et al. 1989). However, these UV absorbing compounds did not prevent photoinhibition and a decrease in chlorophyll per cell and a decrease in activities of Rubisco (Lesser 1996). Diel migration is another important characteristic of this species. Eppley and Harrison (1975) reported the ability to migrate through the thermocline on a diel basis. Prorocentrum micans migrated 6.3 m at a speed of 0.85 m h"1 (Edler & Olsson 1985). However, at this speed, Prorocentrum micans could not cross the halocline to reach NOV-rich water in Laholm Bay in Sweden. 22 Objectives The main objectives of this thesis were as follows: 1) Determine the impact of N deprivation on preference of three forms of N (NH 4+ , N0 3 " and urea) by an ecologically important raphidophyte (Heterosigma carterae) and two dinoflagellates, Amphidinium carterae and Prorocentrum micans, isolated from British Columbia coastal waters. 2) Investigate responses by N-replete and N-deplete marine phytoplankton on N uptake. 3) Determine interactions between N0 3 " uptake and light by Heterosigma carterae. 4) Predict 5 1 5 N of particulate N using the multiple N source uptake model by assuming the sum of the Rayleigh contributions associated with the incorporation of each N source. 23 MATERIALS AND METHODS Culture methods - Amphidinium carterae Hulbert (NPCC 629), Heterosigma carterae Hulbert (NPCC 522R), Prorocentrum micans Ehrenberg (NPCC33) were obtained from the Northeast Pacific Culture Collection (NPCC), Department of Earth and Ocean Sciences, University of British Columbia. These cultures were grown on artificial seawater (ESAW) following the recipe of Harrison et al. (1980) as modified by Price et al. (1987). The medium contained a combination of three N sources, NO3", NH 4 + , and urea, which varied in concentration depending on the experiment (Table 1). The initial N 0 3 ' and N H 4 + concentrations in the medium ranged from 64 to 76 uM for Experiment 1 (Exp. 1), 2 (Exp. 2), and 3 (Exp. 3). Urea concentration ranged 14 to 22 uM since a molecule of CO(NH2)2 contains 2 atoms of N. The initial concentration of N for Experiment 4 (Exp. 4) and 5 (Exp. 5) ranged from 78 to 79 and from 77 to 83 uM respectively. The initial N H 4 + concentration of Experiment 6 (Exp. 6) ranged from 70 to 78 uM. The N:P ratio in the medium was 4:1 to ensure that N was limiting biomass during stationary phase. Bicarbonate (NaHC0 3) was initially 2 mM and was added occasionally to the cultures to prevent C limitation and keep the concentration near 2 mM as in ESAW. The pH of the medium initially ranged from 8.2 to 8.4 and increased to 8.8 to 8.9 during the stationary phase for Exp. 1 and Exp. 4. HCI (6N) was occasionally added to prevent an increase in pH due to carbon uptake during growth for all experiments except Exp. 2. Culture medium was filter sterilized (0.22 um Millipore) and transferred into 1-liter 24 Table 1 Summary of experimental conditions. Nitrogen sources, species name, and average N starvation periods of triplicate batch cultures listed. Experiment Species N Source Light Condition Starvation Period (h) Added Back N Source 1 Prorocentrum micans NH4* continuous light 56 NH4* N0 3 N0 3" urea urea 2 Amphidinium carterae NH 4* continuous light 75 NH4* N03" N0 3" urea urea 3 Heterosigma carterae NH4* continuous light 0 NH4* N0 3 ' N0 3-urea urea 4 Heterosigma carterae N0 3- continuous light 51.5 N0 3" 5 Heterosigma carterae N03" light:dark cycle light 96 N0 3-6 Heterosigma carterae NH4* continuous light 33 NH4* (culture 1) 18 NH4* (culture 2+3) 2 5 sterilized flasks, inoculated with the stock culture and placed in a cold room at 17.0±0.5°C. The cultures were gently stirred by hand twice a day. Autoclaving the flasks and inoculating the culture under the flow hood minimized bacterial contamination. The cultures were continuously illuminated with 1 7 0 - 1 8 5 umol photons m" 2s' 1 for all experiments except Exp. 5 which was grown on a 14L:10D cycle. The cells were acclimated to the N substrate in 1-L flasks for at least 6 days and when samples were first collected, the cells had been growing exponentially for a minimum of about three generations. After inoculation, the cultures were grown first in N-sufficient conditions. Then, they experienced N starvation for 0 - 1 0 1 h, and all three N sources were added back into the medium for Exp. 1, 2, and 3 (Table 1). Only N0 3 " was added back for Exp. 4 and 5. Only N H 4+ was added back for Exp. 6. Trace metals, vitamins, and macronutrients (Si, P, Fe, Bo, Se, and C) were added to the culture after N starvation so that their concentrations were the same as in ESAW. The experiment was designed to contrast a normal exponential growth (i.e. phase 1) and growth following the addition of N after a N starvation period, (i.e. phase 2). Growth conditions were designed to correspond to eutrophic and oligotrophic and/or temporarily N-depleted surface oceans. Therefore, dinoflagellates and the raphidophyte were grown under both-N-sufficient and N-depleted conditions. The N-sufficient phase simulated bloom conditions and coastal environments where the concentration of dissolved N is large relative to its biological uptake. On the other hand, the N re-supply phase was designed to 26 simulate ol igotrophic oceans where new and regenerated N is suppl ied to the N-depleted surface ocean during episodic events. Biomass, particulate N and nutrient analysis - Samples for nutr ients and part iculate matter were mostly col lected at relatively high cell densit ies at t ime intervals ranging up to 268.4 h. 8 1 5 N analysis required a minimum of 1.5 umol N. T h e growth rate, u, was calculated from the fol lowing relationship: u = In (F2/F1) / (t2-t2), where F2 and F1 are the in vivo f luorescence values at t ime 2 (t2) and t ime 1 (t1) during log phase. Fluorescence was measured on Turner Designs model 10 f luorometer. Particulate nitrogen (PN) samples were col lected by vacuum filtration at 0.5 atm onto pre-combusted (450 °C) glass-f iber f i l ters (GF/F) measured with a F isons™ automated CHN analyzer (model: NA 1500) on-l ine with the mass spectrometer which was used for 1 5 N isotope analysis. The precision of each PN analysis was 1 - 2%. Al iquots of phytoplankton culture (30 to 480 ml) were withdrawn from the f lasks at des ignated t ime intervals. The filtrate was used for NOV, N H / and urea analyses. Manual N 0 3 " analyses were made using a spongy cadmium method (Jones 1984). N H 4+ was analyzed according to Slawyk & Maclsaac (1972) and urea was measured manual ly according to the diacetyl monoxime method (Price & Harr ison 1987). Rate measurements - Previously acid-washed polypropylene bott les were used to collect the samples and were rinsed once with the sample. The rate of 27 decrease of dissolved N concentration was calculated from the slope of external N concentration against time. This N disappearance rate is expressed as p.g-at urea N L*1 h'1 or uM h"1. To indicate the change in the preference for the three N sources, comparison of the phases of the uptake rate was calculated by using the ratio of each disappearance rate. Uptake rate - Uptake rate was calculated using the following equation: V=(C 0-C 1)/N*t when Co and Ci are the concentrations (ug-at N L*1) of N at the beginning and end of the time period. N is the cell density (cells L'1), t is the time period of uptake (h) and V is the uptake rate (ug-at 10"8 cells'1 h"1). Cell density was calculated from the fluorescence value which corresponded with the cell numbers at a particular time. Fluorescence and cell numbers were assumed to change in a linear relationship. Nitrogen isotope analysis - N isotope abundance was determined with a VG PRISM™ mass spectrometer. Particulate samples were prepared by rolling a GF/F filter in tin foil and compressing the foil into small pellets. Then, they were combusted in a stream of oxygen at an oven temperature of 1020° C. Results are expressed in the delta notation: 8 1 5 N = (R s a m p l e/Rstd-1)x1000 -(1) where R is the 1 5 N / 1 4 N ratio and the standard (std) is N 2 gas (NBS-14). The 8 1 5N of N1 and N2 standard, i.e. (NhUfeSO* was 1.34 and 20.85%o relative to NBS-14 28 respectively. It was 0.54 and 20.05%o relative to air. Internationally air is accepted as the standard. However, in the present studies, 8 1 5N relative to N6S-14 is used as a standard because absolute values are not critical for this study. The difference between the 8 1 5N of PN and the source is important other than the absolute value of either of them as explained below. The precision of 8 1 5N of the N source was 0.17%o (±SD of replicate pairs, n = 44 pairs). The 8 1 5N of the three N sources was: 3.82±0.18%o (mean ± 1SE, n = 12) for NaN0 3 , -0.34±0.20%o (mean ± 1SE, n = 2) for NH4CI, and 0.06±0.12%o (mean ± 1SE, n = 3) for CO(NH2)2- Correction for carry-over of 15N-enriched N0 3 " from the inoculum (0.5 mM NOV) to the culture medium was estimated to be small (i.e., < 0.2%o) due to the small volume of inoculum which was used. A small 1 5 N enrichment might be introduced during culture dilution. Calculations of the fractionation factor - The isotope fractionation factor a was calculated using the accumulated product equation (Mariotti et al. 1981). Note that contrary to the definition adopted by Mariotti et al. (1981), a is typically > 1 and s (i.e. the per mil enrichment factor of the substrate relative to the product) is typically > 0; with e = (a -1) x 1000. In a closed system, as in a batch culture, the 8 1 5N of the first accumulated product is given by: 815NpN0 = 8 1 5NDNO - E - (2) 29 where PNo and DNo are the initial particulate N and dissolved N, respectively. Provided that s was constant during the consumption of the substrate, e was then derived from the accumulated product equation: 8 1 5 Np N = 8 1 5N ( D N)o - s • - f*ln f/(1 -f) - (3) where f is the fraction of unreacted substrate at any time during exponential growth. A curve of 8 1 5 N P N versus F (F is defined as [-f/[1-f)] In f) was fitted with a linear function. In order to be applicable for equation (2), only data in log phase were analyzed for the calculation of s. The reason for this is it was only during this phase that N incorporation could be approximated by a one step unidirectional reaction. It means conversion of DN to PN . Once DN is exhausted, cell lysis will start and release various forms of N. Multiple N source uptake model - A model simulates the 8 1 5 N P N which results from batch culture growth on multiple N sources. The model was designed to test if 815NPN can be described as the sum of the Rayleigh contributions associated with the incorporation of each N source. The dependent expressions for concentration of N H / , NOV, and urea are determined by the fits of the dissolved N data. The 8 1 5 N P N t is described as follows: 6 1 5N P N , = (8 1 5N P N*ZPNX + 8 1 5 N P N i * P N i ) / ( I P N x + PN,) 30 Par t i RESULTS During batch culture experiments, N was used up and then cells were N-starved for up to 96 h before they were re-supplied with N. The first period when NH 4 + , NOV, and urea were all in the medium together is termed before I (bef. i ) and the period after N re-supply is termed after I (aft. I ) . When only N0 3 " and urea were present in the medium (i.e. N H 4 + was already taken up), this period is termed before I I (bef. n ) and after N re-supply, after n (aft. I I ) . When only urea was left in the medium, this period is termed before i n (bef. i n ) and after i n (aft. i n ) . The order of N uptake preference was not changed in any experiments by N starvation (Table 2). However, the degree of preference of each N source changed. This will be illustrated in the following sections. Prorocentrum micans grown on and re-supplied with NH 4 + , N 0 3 + , and urea (Experiment 1) The growth curve of Prorocentrum micans is shown in Fig. 1. The growth rate, u calculated from 33 to 149 h was 0.30 d'1. This growth rate was the lowest among the three species used in the present study. The cell number was 940,000 cells L"1 at the fluorescence value of 9.4, which was used for the calculation of uptake rates (i.e. 1 fluorescence value corresponds to 100,000 cells L' 1). The N concentration in the medium during the entire incubation period and after the re-addition of N sources is shown in Figs. 2a and 2b respectively. 31 Table 2 Comparison of the order of N preference before and after re-addition of NH 4 + , NOV, and urea into triplicate batch cultures of P. micans, A. carterae, and H. carterae. Period I indicates the period when NH 4 + , NOV, and urea were all present in the medium. Period n indicates the period when NOV and urea were present in the medium. Period i n indicates the period when only urea was left in the medium. Period I Period n Period i n N H 4 + N sources in medium N0 3" Urea N0 3" Urea Urea Order Order 1st 2nd 3rd 1st 2nd Prorocentrum micans before N starvation NH, t + N0 3 - Urea N0 3" Urea Urea after N starvation N H , + NO3- Urea N 0 3 ' Urea Urea Order Order 1st 2nd 3rd 1st 2nd Amphidinium carterae before N starvation N H : NO3- Urea N0 3" Urea Urea after N starvation N H : NO3- Urea N0 3" Urea Urea Order Order 1st 2nd 3rd 1st 2nd Heterosigma carterae before N starvation N H 4 + N 0 3 ' Urea N0 3* Urea Urea after N starvation N H 4 + N 0 3 ' Urea N 0 3 ' Urea Urea 32 0 -I 1 r 0 50 100 150 200 250 Time (hours) Figure 1 Growth of Prorocentrum micans grown on NH 4+ , N0 3", and urea under continuous light. Log plot of in vivo fluorescence versus time. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 33 b 60 n Time (hours) Figure 2 Nitrogen concentrations (urea as ug-at N L"1) during the growth of Prorocentrum micans grown on NH 4 + , NOV, and urea under continuous light plotted against a) elapsed time and b) elapsed time since re-addition of NH 4 + , N03", and urea at t=0 to previously N-starved cultures for 56 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Cultures were N-starved between 130 and 186.5 h and then 42 uM NH 4 + , 50 uM N03", and 47 ug-at N L"1 urea were added at 186.5 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 34 N H 4 + disappeared from the medium within 49 h at a rate of 0.97 umol L"1 h"1. N H 4 + showed a slow disappearance rate of 0.29 umol L"1 h"1 between 0 and 9 h, followed by a faster rate at 2.5 umol L"1 h'1 (9 -17 h) and 0.64 umol L"1 h"1 (17 -33 h). During that period (bef. I ) , urea was also taken up at a rate of 0.34 ug-at N L"1 h"1. Between 0 and 17 h, there was a slow uptake of 0.18 ug-at N L"1 h'1 followed by a much faster rate of 0.77 ug-at N L' 1 h"1. However, NOV showed little uptake during this period. The disappearance rate of NOV during this period ( I ) was 0.21 umol L"1 h'1. In particular, N0 3" was taken up at a very slow rate of 0.07 umol L' 1 h"1 for the first 41 h. After N H 4 + was < 1 uM, the N 0 3 ' disappearance rate increased to 0.57 umol L"1 h"1. Between 41 and 81 h, N0 3 " disappeared at a rate of 0.81 umol L' 1 h"1. During this period (bef. I I ) , the disappearance of urea was 0.27 ug-at N L"1 h"1. Between 33 and 65 h, very little urea was taken up and the disappearance rate was 0.06 ug-at N L' 1 h' 1. When N0 3 " was < 1 uM, the disappearance rate of urea increased to 0.14 ug-at N L"1 h"1. Between 81 and 137 h, the urea disappearance rate was 0.30 ug-at N L"1 h"1. After 56 h of N starvation, the three N sources were re-supplied. For the first 2 h immediately after the re-supply of the three N sources, all three N sources remained almost at their initial concentrations. N H 4 + disappeared from the medium within 13.5 h and the disappearance rate of N H 4 + during the first 2 h was 2.2 umol L' 1 h"1 and was followed by a faster rate of 5.5 umol L"1 h"1 (2 -5 h). The concentration of urea again decreased together with NH 4 + , but at a rate of 35 65% of that of N H 4 + during the period (aft. I ) . Between 1 -1.5 h after the re-supply of N, the disappearance rate of urea was negative (-1.3 ug-at N L' 1 h"1). NOV was not used until the concentration of NH 4* was < 1 uM Between 0 and 5 h, the disappearance rate of NOV was 1.1 umol L"1 h"1. Once NH 4 * disappeared from the culture, the disappearance of N0 3* exceeded that of urea for the following 18 h (aft. I I ) . The disappearance rates of N0 3" and urea were 2.1 umol L"1 h"1 and 0.93 ug-at N L"1 h"1 respectively during this period (aft. I I ) . Especially, between 13.5 and 25.5 h, the disappearance rate of N0 3 " was 2.5 umol L' 1 h' 1. On the other hand, urea was taken up more slowly than N0 3 " at 1.5 ug-at N L"1 h"1 between 4 and 25.5 h. Uptake rates are plotted versus the average time between sampling for the entire incubation period (Fig. 3a). The maximum N H 4 + uptake rate (0.009 ug-at lO^cells'1 h"1) by N-replete cells occurred during the 9 -17 h interval. After the re-supply of the three N sources, the maximum N H 4 + uptake rate (0.007 ug-at lO^cells'1 h'1) during the aft. I period occurred during the 3 - 4 h interval. (Fig. 3b). The maximum N0 3" uptake rate (0.003 ug-at fo^ceHs"1 h"1) occurred during the 41 - 49 h interval, which coincided with the exhaustion of NH 4*. Subsequently, the average uptake rate during the 49 - 89 h interval was 0.002 ug-at lO^cells'1 h"1. N0 3" uptake rate was 0.009 ug-at lO^cells"1 h'1 for the aft. I period and 0.002 ug-at lO^cells'1 h"1 for the aft. I I period. The maximum uptake rate of urea (0.004 ug-at lO^cells"1 h'1) by N-replete cells occurred during the 25 36 _P 0.020 -j 2 0.015 « o 0.010 a 0.005 | 0.000 a -coos Q. 3 z -0.010 -0 50 100 150 200 Time (hours) b — 0.012 -j •f 0.010 UL ^ 0.008 f ° o 0.006 % 0.004 S 0.002 | 0.000 .£ -0.002 is a- -0.004 •} z -0.006 -I , , , . . . 0 5 10 15 20 25 30 Time (hours) Figure 3 N uptake rates of triplicate, batch cultures of Prorocentrum micans grown on NH 4 + , NOV, and urea and under continuous light plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4 + , N03", and urea at t=0 to previously N-starved cultures for 56 h. (b) is an expanded time version of the re-addition phase of (a). Cultures were N-starved between 130 and 186.5 h and NH 4 + , N03", and urea were added at 186.5 h. Bars represent + 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 37 - 33 h interval during the bef. I period, and was 0.002 ug-at lO^cells"1 h"1. While NOV uptake was high during the bef. n period, the uptake rate of urea was 0.005 ug-at lO^cells'1 h'1 and 34% of N0 3" uptake. After N starvation and the re-addition of the three N sources, the maximum urea uptake (0.005 ug-at 10^cells"1 h"1) occurred during 25 - 33 h. Surge uptake of any of the N sources was not observed. The purpose of the present experiments was to determine the effect of N starvation on the preference for the three N sources, NH 4 + , N03~, and urea. Therefore, the relative disappearance rates were also calculated and expressed as ratios (Figs. 4 and 5). Before N starvation and while the three N sources (from 0 to 49 h) were present in the medium, the ratio of the disappearance and uptake rates of each N was NH4+:N03":urea = 1.9:0.6:1. Preference for urea over N H 4 + and NOV during the aft. I period increased during N starvation. Preference for N H 4 + and N 0 3 ' relative to urea decreased and the ratio was 1.5:0.5:1 after N starvation. Period II was also compared. Before N starvation (from 49 to 97 h), the ratio of the disappearance rate of N03":urea was 2.1:1 compared to 2.0:1 after N starvation and not significantly different. Amphidinium carterae grown on and re-supplied with N H 4 + , N0 3", and urea (Experiment 2) The growth curve of Amphidinium carterae is given in Fig. 6. The growth rate (u) was 0.76 d"1 calculated from 59 to 143 h. When the fluorescence value 38 NOj N sources Figure 4 Relative N preference of N H 4 + and NOV relative to urea for Prorocentrum micans during the period I before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation. before N starvation i 1 after N starvation NO, urea N sources Figure 5 Relative N preference of urea relative to N0 3" for Prorocentrum micans during the period n before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation. 39 2 i -1 -I , , 1 1 — 0 50 100 150 200 Time (hours) Figure 6 Growth of Amphidinium carterae grown on N H 4 \ N0 3", and urea under continuous light. Log plot of in vivo fluorescence versus time. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 40 was 18, the cell number was 345,600 cells L"1, which was used for the calculation for the uptake rate (i.e. 1 fluorescence value corresponds to 19,200 cells L' 1). N concentration in the medium is shown for the entire incubation period (Fig. 7a) and after the re-addition of N sources (Fig. 7b). Generally the values for separate cultures agreed well. First, NH 4* disappeared from the medium within 71 h during the bef. I period at a rate of 0.48 umol L"1 h"1. During this period, the concentration of both NOV and urea remained at almost their initial concentrations. The disappearance rate for the period (bef. I) was -0.01 umol L"1 h"1 for NOV and 0.07 ug-at N L"1 h'1 for urea. The negative value for the NOV disappearance rate may be due to the efflux of NOV. The disappearance rates for NOV and urea were 0.01 umol L"1 h'1 and 0.11 ug-at N L"1 h"1 respectively between 0 and 77 h. When the concentration of N H 4 + was < 1 uM, N 0 3 ' and urea were both taken up at the rate of 0.77 umol L' 1 h"1 and 0.46 ug-at N L' 1 h"1 respectively during bef. n period. Between 15 and 21 h, the disappearance rates of N0 3" and urea were 0.79 umol L " V 1 and 0.61 ug-at N L' 1 h"1. Finally, urea was the only N source in the medium and disappeared within the following 11 h (bef. III) at a rate of 0.48 ug-at N L"1 h'1. After the cells were N-starved for 75 h and then re-supplied with the three sources of N, N H / decreased quickly after the re-supply, and the disappearance rate was 10.1 umol L"1 h"1 during the first 2.7 h (aft. 1). Focusing only on the initial 0 to 15 min and 20 to 120 min periods, the rates were 29.8 and 9.2 1 41 a 50 -i Time (hours) Figure 7 Nitrogen concentrations (urea as ug-at N L'1) during the growth of Amphidinium carterae grown on NH 4 + , NOV, and urea under continuous light plotted against a) elapsed time and b) elapsed time since re-addition of N H 4 \ N03", and urea at t=0 to previously N-starved cultures for 75 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Cultures were N-starved between 119 and 193.5 h and then 29 uM NH 4 + , 30 uM N03", and 30 ug-at N L' 1 urea were added at 193.5 h. Bars represent + 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 42 limol L' 1 h'1 respectively. The concentration of N0 3" and urea increased during the first 5 min after the re-addition of the three N sources. N0 3" increased from 29.6 to 37.4 uM and urea increased from 30.4 to 34.5 ug-at N L"1. Between 0 and 40 min, efflux of NOV and urea occurred at rates of -12.3 umol L' 1 h"1 and -5.3 ug-at N L' 1 h'1 respectively. During the rapid decrease in N H 4 + concentration (aft. I), N0 3" and urea concentrations both remained almost constant. However, urea started to decrease slightly before N0 3 " started to decline. The decrease in urea started when N H 4 + was 18 uM, while N0 3 " started to decrease when N H 4 + was 4 uM. When NH 4* was present in the medium (aft. II), the disappearance rate of urea was 0.68 ug-at N L' 1 h"1 and -0.74 ug-at N L"1 h"1 for NOV Between 50 min and 7 h , the observed rates were 3.7 umol L"1 h'1 for N 0 3 ' and 2.7 ug-at N L"1 h"1 for urea. Once the concentration of N H 4 + in the medium reached 1.7 uM, the disappearance rate of N 0 3 ' was 6.6 umol L"1 h' 1 and that of urea was 4.9 ug-at N L"1 h"1. After the concentration of N0 3 " reached 5.6 uM, both N0 3" and urea decreased rapidly. The maximum uptake of N H 4 + (0.08 ug-at lO^cells"1 h'1) for N-replete cells occurred during the 24 - 30 h interval (Fig. 8a). After the re-addition of three N sources, the maximum NH 4* uptake rate (0.005 ug-at lO^cells"1 h"1) occurred during the initial 0 - 0.5 h interval (Fig. 8b). Uptake rates of N0 3 " during the bef. I period fluctuated widely (-0.05 - 0.03 ug-at lO^cells"1 h'1) including negative uptake, indicating efflux. Uptake rate of N0 3" during the initial 0 - 5 h interval was -0.033 ug-at lO^cells'1 h"1. The maximum N 0 3 ' uptake rates (0.027 43 0.2 <9 O If 0) I 3 0.0 -0.2 0 20 40 60 80 100 120 140 160 180 200 220 Time (hours) 10 15 Time (hours) Figure 8 N uptake rates of triplicate, batch cultures of Amphidinium carterae grown on NH 4 + , NOV, and urea and under continuous light plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4 + , NOV, and urea at t=0 to previously N-starved cultures for 75 h. (b) is an expanded version of the re-addition phase of (a). Cultures were N-starved between 119 and 193.5 h and NH 4 + , NOV and urea were added at 193.5 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 44 ug-at lO^cells"1 h'1) after the re-addition of N sources occurred during 6 - 8 h intervals when only NOV and urea were in the medium. Urea uptake during the bef. i period also fluctuated, and ranged from -0.031 to 0.055 ug-at lO^cells'1 h"1. Then, it became stable during the bef. n with an average urea uptake rate of 0.005 ug-at lO^cells"1 h"1 between 71 -113 h. Very low uptake (-0.030 ug-at lO^cells"1 h'1) occurred immediately after the re-addition of N sources (0-5 h), as the lowest NO3' uptake occurred during the same period. Urea uptake rate during the aft. 1 period was 0.004 ug-at lO^cells'1 h*1 and only 13% of N H 4 + uptake. Uptake rate of urea during aft. II (2.3 - 8 h) was 0.011 ug-at lO^cells'1 h"1 and 76% of NOV uptake and reached a maximum uptake rate (0.031 ug-at lO^cells"1 h"1) during the last interval of aft. 11 (6 - 8 h). Surge uptake of any of the N sources was not observed. The ratio of the disappearance and uptake rates was calculated to find the N preference (Figs. 9 and 10). Before N starvation, the uptake ratio of NH 4 +:N0 3 ':urea was 6.9:-0.2:1, during the period that the three N sources existed in the medium (bef. 1). After N starvation, the uptake ratio for that same period (aft. 1) was 7.5:-1.1:1. The negative ratio for NOV indicates NOV efflux. The order of preference, NH 4 + , urea, and N 0 3 ' remained the same after N deprivation. However, the relative preference of NH 4* over NOV and urea increased after N deprivation. The preference of N H 4 + relative to urea increased by 8.4%. The uptake of urea was faster than N0 3" after N starvation, which indicated that urea was preferably used by starved cells in period 1. The 45 before N starvation after N starvation NO,' N sources Figure 9 Relative N preference of NH4* and urea relative to NOV for Amphidinium carterae during the period I before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation. Figure 10 Relative N preference of urea relative to NOV for Amphidinium carterae during the period n before (•) and after (•) N starvation. The relative preference for urea is fixed at 1, both before and after N starvation. relative preference of N0 3" over urea dropped dramatically by 5-fold. During the period when only N 0 3 ' and urea were present in the medium (II ) , the ratios of disappearance rates of NOV and urea were 1.7:1 before N starvation and 1.3:1 after N starvation. Again, the order of preference was not changed. However, after N deprivation, the relative preference of N0 3" over urea decreased by 20%. N-starved Amphidinium carterae cells showed enhanced urea utilization ability over N0 3 " in periods I and I I . Heterosigma carterae grown on and re-supplied with NH 4 + , N03~, and urea (Experiment 3) The growth curve of Heterosigma carterae is given in Fig. 11. The growth rate was 0.56 d'1 calculated from 0 to 108 h. When the fluorescence values were 19.4 and 23.2, cell numbers were 457,080 and 520,640 cells L' 1 respectively. The maximum cell density was 911,040 cells L"1 at the fluorescence value of 45. The calculation of cell density for Heterosigma carterae depended on these measured values in this study. The equation for the calculation of cell density was obtained from the change in the linear regression (cell density = 21755*[fluorescence value] - 27046; except for the fluorescence values < 1.24). The N concentration in the medium is shown for the entire incubation period (Fig. 12a) and after re-supply of the three N sources (Fig. 12b). Generally, the values for separate cultures agreed well. Before N starvation, N H 4 + disappeared from the culture medium first. The disappearance rates of 47 2 i Figure 11 Growth of Heterosigma carterae grown on NH 4+ , NOV, and urea under continuous light. Log plot of in vivo fluorescence versus time. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 48 a 50 -i 20 40 60 80 Time (hours) Figure 12 Nitrogen concentrations (urea as ug-at N L"1) during the growth of Heterosigma carterae grown on NH 4 + , NOV, and urea under continuous light plotted against a) elapsed time and b) elapsed time since re-addition of NH 4 + , N0 3 ' , and urea at t=0 to previously N-starved culture for 0 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). 31 uM NH 4 + , 41 uM N0 3 ' , and 38 ug-at N L"1 were added at 175 h. Bars represent ± 1 S.D. (n=2-3). Error bars are smaller than symbols where not visible. 49 N H 4 + between 0 and 20 h and 20 and 62 h were only 0.09 umol L' 1 h' 1 and 0.57 umol L"1 h"1 respectively. While NH 4 + was primarily used (bef. I), both NOV and urea remained near their initial concentrations. During this period (bef. I), the disappearance rates of NH 4 + , NO3", and urea were 0.47 umol L"1 h'1, 0.005 umol L"1 h"1, and 0.05 ug-at N L"1h"1 respectively. Especially low disappearance rates of 0.005 umol L"1 h"1 for N0 3" and 0.05 h"1 for urea were observed between 0 and 60 h and between 0 and 67.5 h respectively. The low N0 3 " uptake was due to the increase in the N0 3" concentration during the first 12 h. Urea uptake showed less influence by NH 4* in Exp. 1 and 2, and its concentration decreased more than NOV However, once the concentration of N H 4 + reached 3.4 uM (bef. 11), N0 3 " rapidly decreased within 63 h at a rate of 0.63 umol L"1 h"1. The disappearance rate of N0 3" between 67.5 and 107.5 h increased to 0.85 umol L' 1 h' 1. During the period when N 0 3 ' and urea both existed in the medium (bef. 11), the disappearance rate of urea was 0.15 ug-at N L"1 h"1. Between 67.5 and 123 h, the rate of urea disappearance was 0.18 ug-at N L"1 h' 1. Cells took up urea very slowly even after urea became the only N source in the culture medium. The disappearance rate of urea during bef. 111 period was 0.43 ug-at N L"1 h'1 and it was 0.44 ug-at N L"1 h"1 between 123 and 163 h. Since urea did not completely disappear from the culture medium, Heterosigma carterae did not experience true N starvation. After the re-supply of the three N sources, N H 4 + was immediately used up within 6 h (aft. 1) at a disappearance rate of 5.1 umol L"1 h"1. At the end of N H 4 + 50 uptake (6-13 h), the rate slowed to 0.17 umol L' 1 h"1. During the same period (aft. i ) , NOV remained nearly constant and urea was slightly used at a disappearance rate of 0.68 ug-at N L"1 h'1. Again, in the presence of the three N sources, N H 4 + strongly influenced on N0 3" uptake. The disappearance rates of N0 3 " from 0 to 5 h, from 5 to 6 h, and from 11 to 15 h were -0.50, 3.2, and 5.1 umol L"1 h"1 respectively. Once N H 4 + concentration became < 2.4 uM, N0 3 " was quickly used up within the following 10.5 h (aft. I I ) . Urea showed different phases of uptake. From 0 to 16.5 h (when N0 3" was primarily consumed) and from 16.5 to 27.5 h (after N0 3 " was consumed), both rates were almost the same and they were 0.86 ug-at N L"1 h"1 and 0.89 ug-at N L"1 h' 1 respectively. Then, urea disappeared gradually. The maximum N H 4 + uptake rate (0.008 ug-at lO^cells"1 h'1) of N-replete cells occurred during 20 - 25 h interval, followed by a decrease in uptake before it reached a roughly constant rate of 0.008 ug-at lO^cells'1 h"1 (Fig. 13a). Uptake rate during the bef. I period was 96-fold higher than N0 3" and 10-fold higher than urea. In the N re-supply phase, the maximum uptake rate (0.003 ug-at lO^cells' 1 h'1) occurred during the initial 0 - 0.5 h interval (Fig 13b). N H 4 + uptake rate was 0.009 ug-at lO^cells"1 h"1 (4 - 5 h) from 0.007 ug-at lO^cells"1 h"1 (2 - 4 h). N0 3 " uptake rate temporally decreased from 0.007 to -0.03 ug-at lO^cells"1 h"1 during the initial 0 -20 h interval, followed by a rapid increase in uptake until it reached a roughly constant rate of 0.0008 ug-at lO^cells'1 h'1 during the bef. I period. After the re-addition of 51 i 0.10 -i = 0.08 50 uM remained for the following 5 h and resulted in this very slow N H 4 + disappearance. The disappearance rates during this period were 1.7 umol L' 1 h' 1 and 0.41 umol L"1 h"1 between 2.5 and 4.5 h and 4.5 and 7.5 h respectively. N H 4 + gradually started to disappear again at the rate of 4.1 umol L"1 h' 1 between 7.5 and 14.5 h, after the 5 h period where little uptake occurred. It took 32.5 h to consume all the added NH 4 + . The maximum NH 4* uptake rate 0.031 ug-at lO^cells"1 h"1 of N-replete cells occurred during the 14 -18 h interval (Fig. 24a). In the N re-addition 66 Figure 22 Growth of Heterosigma carterae on N H 4 + under continuous light, culture 1. Log plot of in vivo fluorescence versus time. 67 a Time (hours) b 100 -i Figure 23 N H 4 + concentrations during the growth of Heterosigma carterae grown on NH 4* under continuous light, culture 1, plotted against a) elapsed time and b) after the re-addition of N H 4 + at t=0 to N-starved cells for 33 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Culture was NH4*-starved between 65 and 98 h and then 77 uM N H 4 + was added at 98 h. 68 0.05 n | 0.04 <» r 0.03 f 0.02 2 0.01 0.00 3 *X* -0.01 20 40 60 80 100 120 Time (hours) 10 Time (hours) 15 Figure 24 N H 4 + uptake rates of a batch culture of Heterosigma carterae grown on N r V and under continuous light, culture 1, plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4* at t=0 to previously N-starved culture for 33 h. (b) is an expanded version of the re-addition phase of (a). Culture was NH4+-starved between 65 and 98 h and then N H 4 + was added at 98 h. 69 phase, maximum uptake (0.042 ug-at lO^cells"1 h'1) occurred during the 1 - 2.5 h interval, followed by a rapid decrease in uptake and very low NH 4 + uptake period (2.5 - 6.5 h) with an average rate of 0.002 ug-at lO^cells"1 h'1 (Fig. 24b). Data from the other duplicate cultures agreed well with both growth 1.02 d"1 calculated 14 to 54 h and NH 4 + disappearance rates (Figs. 25 and 26a). Before N starvation, the NH 4 + concentration continued to decline without any lag period, which matched the log period of growth. The disappearance rates of NH 4 + were 0.78 umol L'1 h"1 and 1.51 umol L'1 h'1 between 0 and 23 h and 23 and 54 h respectively. After the re-addition of NH 4 + to N-starved cells, NH 4 + concentration hardly decreased for the following 1.5 h (Fig. 26b), which also occurred in culture 1. The disappearance rate between 0 and 1.5 h was 0.96 umol L"1 h"1. Following the lag period, NH 4 + concentration rapidly dropped from 72.1 to 59.4 uM at a rate of 25.4 umol L'1 h"1. As in culture 1, the concentration of NH 4 + almost did not decrease after the rapid disappearance of NH4 +. The disappearance rate from 2 to 4.5 h was 2.8 umol L'1 h"1. From 4.5 to 5.5 h, the rate increased to 13.0 umol L"1 h"1. Then, NH 4 + started to disappear again gradually at a rate of 4.4 umol L'1 h'1 between 5.5 and 10.5 h and it took 24.5 h for the complete disappearance. NH 4 + disappearance after the re-supply of NH 4 + showed several steps. First, the concentration did not change and, then the concentration dramatically dropped followed by a slow decrease in NH4 +. This phenomenon was more distinct than that observed in the same species grown on NOV and continuous light (Exp. 4). 70 0 -I 1 1 1 1 1 1 1 . 0 20 40 60 80 100 120 140 Time (hours) Figure 25 Growth of Heterosigma carterae on NH 4* under continuous light, cultures 2 and 3. Log plot of in vivo fluorescence versus time. Bars represent ± 1 S.D. (n=1-2). Error bars are smaller than symbols where not visible. 71 a 100 n Time (hours) Figure 26 N H / concentrations during the growth of Heterosigma carterae grown on N H / under continuous light, cultures 2 and 3, plotted against a) elapsed time and b) elapsed time after the re-addition of N H / at t=0, to N-starved cells for 18 h. Thus (b) is an expanded time version of the N concentrations during the re-addition phase shown in (a). Cultures were NH/-starved between 88 and 106 h and then 74 uM N H / was added at 106 h. Bars represent ± 1 S.D. (n=1-2). Error bars are smaller than symbols where not visible. 72 A relatively high average N03* uptake (0.030 ug-at lO^cells'1 h'1) for N-replete cells occurred during the initial 0 -18 h interval (Fig. 27a). During the N-deplete phase, uptake rate fluctuated and even the NH/ concentration in the medium did not change much (2.5 - 4.5 h) (Fig. 27b). The maximum NH 4 + uptake rate (0.034 ug-at lO^cells"1 h'1) occurred during the 4.5 - 5.5 h interval. 73 Figure 27 NH/ uptake rates of duplicate, cultures of Heterosigma carterae grown on NH 4 + and under continuous light, culture 2+3, plotted against a) average time between sampling and b) average time between sampling since re-addition of NH 4 + at t=0 to previously N-starved for 18 h. (b) is an expanded version of the re-addition phase of (a). Cultures were NH4+-starved between 88 and 106 h and then NH 4 + was added at 106 h. Bars represent ± 1 S.D. (n=1-2). Error bars are smaller than symbols where not visible. 74 DISCUSSION N Preference change 1) period I In the present study, Prorocentrum micans, Amphidinium carterae, and Heterosigma carterae showed different degrees of N preference compared to the generally accepted pattern of preferring NH 4 + and then N03", over urea (Dugdale & Goering 1967; Paasche & Kristiansen 1982; Levasseur etal. 1990). Urea uptake was higher than N03" uptake in the presence of NH4 +. This result was different from previous field studies, where N utilization was 54% (NH4+), 26% (N03"), and 5% (urea) (Maguer et al. 1995) and 48% (NH4+), 32%(N03_), and 13% (urea) (L'Helguen etal. 1996). N-replete Prorocentrum micans preferred NH 4 + over N0 3" and urea with an uptake ratio of 1.9:0.6:1 respectively. This species used the three N sources most evenly among the tested species. Urea utilization was seen at > 14 uM NH4 +, which occurred in only Prorocentrum micans but not in the other species. N-depleted cells increased their urea preference relative to N03" and NH 4 + and the N0 3" preference relative to NH4*. However, the increase in the preference for urea was greater than N03". The strong preference for NH 4 + by Heterosigma carterae (Exp. 3) resulted in N preference ratios of 10 (NH4+):0.1 (N03'):1 (urea) by N-replete cells. Such a strong preference for a particular N source was found only in this species. N-75 replete cells did not use N03" when NH 4 + > 1 uM. On the other hand, urea was utilized more than N03'. These results were in good agreement with the field study on several lakes in Brazil by Mitamura et al. (1995) who showed that the contribution of N03" as N source was negligible compared to NH 4 + and urea. N-deplete cells of Heterosigma carterae decreased their preference for NH 4 + relative to both N0 3 ' and urea. The preference for NH 4 + relative to N03" showed a 59% decrease, and urea preference relative to N03" decreased 44%. This is the only species that decreased the preference for urea relative to N03* after N deprivation. Consequently, N-deplete cells used the three N sources more indiscriminately than N-replete cells. However, the preference for NH 4 + was still very high compared to the other species, before and after N deprivation. Amphidinium carterae also showed a strong preference for NH4*. Interestingly, N-replete cells did not use N03" at all when NH 4 + was available. N-deplete cells enhanced the relative preference of NH 4 + over both N0 3" and urea and the consequence was a significant decrease in the relative preference for N03". This resulted in a wider range of the preferences for N sources than N-replete cells. In general, the present study showed that influence by NH 4 + affected N0 3 ' to a greater extent than urea. For all three species, urea was more preferred than N0 3 ' in the presence of NH 4 + (bef. and aft. I). The important new finding in the present study was the interactions of the three re-added N sources. Most reports of inhibition by N sources focused on only two N sources, NH 4 + and N03". Therefore, previous reports of interactions 76 apply only partially to the present results. For example, the present results of the increased urea uptake relative to NH/ after the re-addition of three N sources (Exp. 1 and 2) were the opposite of the results by Horrigan and McCarthy (1982) and Lund (1987). This discrepancy is considered due to the interaction of N sources, culture conditions, and different species (Thalassiosira pseudonana and Skeletonema costatum). Furthermore, the previously reported increased NOV uptake by removing inhibition by NH 4 + (Maclsaac & Dugdale, 1972) was not observed in my study, perhaps due to the enhanced urea uptake, when NH 4 + was depleted in Exp. 1 and 2. However, the increased preference for urea over NH 4 + and N03" (Exp. 1 and 2) and over NH 4 + (Exp. 3) observed during aft. I periods agreed with the results by Rees and Syrett (1979). The increased urea preference by N-starved dinoflagellates in the present study is in agreement with the results of a recent study on a coccolithophore and a diatom by Waser et al. (1998a). In their study, using three N sources, NH 4 + was utilized first, followed by the simultaneous uptake of N03" and urea by N-replete Thalassiosira pseudonana and Chaetoceros debilis. N-replete Emiliania huxleyi took up N in the following order NH4+, N03', and urea. However, N-deprived Thalassiosira pseudonana and Emiliania huxleyi used reduced N forms (NH4+ and urea) immediately after the re-supply of N sources and N0 3" was used last. Interestingly, N-starved Chaetoceros debilis took up only urea after the re-addition of three N sources, and NH 4 + was not immediately utilized. These observations suggest that increasing preference for urea in the interaction of the three N sources is a general phenomenon in several groups of marine 77 phytoplankton. However, the evidence that the raphidophyte Heterosigma carterae did not increase urea preference after N deprivation in the present study also suggests increasing urea preference is group or species specific. 2) Period I I During this period ( n ) of N uptake after N addition to N-starved cells, the preference for urea relative to NOV increased 20 and 12% in Amphidinium carterae and Heterosigma carterae respectively; there was no change in the urea preference in Prorocentrum micans. Further investigation is needed to determine why Heterosigma carterae developed a preference toward NOV utilization after N starvation. Heterosigma carterae grown on three N sources could not use urea before and after N deficiency. Heterosigma carterae did not grow well on urea (Fig. 28). The growth rate of Heterosigma carterae on urea was 0.90 d'1, (calculated from 0 to 3 d) and was not the lowest compared to growth on other N sources. However, maximum fluorescence of this culture with urea was 80% of Exp. 3 (three N sources), 78% of Exp. 4 (NOV and continuous light), 79% of Exp. 5 (NOV and L D cycle), and 78% of Exp. 6 (NH 4 + and continuous light). Therefore, to compensate for N deficiency which was exacerbated by the incapability of using urea, Heterosigma carterae might have increased its capability to take up N0 3 " during N deprivation. Overall, the order of preference of N was not changed for each period. However, it was revealed that the order of N preference really depended on the 78 Figure 28 Growth of Heterosigma carterae grown on various N sources and light conditions. Log plot of in vivo fluorescence versus time. 79 combinations of N sources. In addition, the N physiological state affected the degree of N preference, which was shown by a significantly increased preference for urea relative to NH 4+ and N0 3 " by N-starved Prorocentrum micans and Amphidinium carterae, and an increased N 0 3 ' preference of N-deprived Heterosigma carterae. C e l l u l a r p h y s i o l o g i c a l s ta te 1) S u r g e u p t a k e N-starved cells did not exhibit evident of faster NH 4+ , NOV and urea uptake than non-starved cells for all three species, except for Heterosigma carterae grown on N H 4+ (culture 1). In addition, the fastest uptake of N occurred some time after the re-addition of N sources, except N H 4+ uptake by N-starved Amphidinium carterae and Heterosigma carterae grown on three N sources. Syrett (1953) and Harvey (1953) first demonstrated the effects of cellular physiological state on N uptake rates. Following these first results, a natural phytoplankton community and a batch culture were shown to exhibit rapid uptake of N H 4+ after N deprivation (Conway et al. 1976; McCarthy & Goldman 1979; Goldman etal. 1981; Goldman & Glibert 1982; Dortch etal., 1982; Wheeler et al. 1982; Harrison 1983; Suttle & Harrison 1988; Syrett & Peplinska 1988). However, this enhanced uptake rate decreased rapidly after the cells overcame the N deficiency (Fitzgerald 1968). In general, species specificity (Conway & Harrison 1977) and the duration of N deprivation (Parslow et al. 1984a) determine the magnitude of the surge uptake response. This enhanced uptake 8 0 rate of N could be an adaptive response for phytoplankton to take up regenerated N H / (McCarthy & Goldman 1979; Gilbert & Goldman 1981; Goldman & Glibert 1982). Consequently, it allows phytoplankton to maintain relatively high growth rates in oligotrophic waters (Goldman etal. 1979; Goldman & Glibert 1982). The apparent role of surge uptake for phytoplankton is to overcome the N deficiency and speed up the incorporation of N into macromolecules. Wheeler et al. (1982, 1983) suggested that growth is limited by the rate at which cellular metabolism can incorporate N into macromolecules. 2) A m m o n i u m up take in N-deplete p h a s e During the N re-supply period, the fastest uptake rate (0.07 ug-at lO^cells"1 h"1) of N H / was observed between 3 and 4 h in Prorocentrum micans. However, in Amphidinium carterae and Heterosigma carterae, the fastest rates occurred during the initial interval after the re-addition of N sources. The former species required an acclimation period to take up the re-added N H / . Waser et al. (1998a) also reported a similar phenomenon in which N H / was taken up immediately after the three N sources by N-starved Emiliania huxleyi and Thalassiosira pseudonana, but not Chaetoceros debilis. Therefore, the capacity for surge uptake of N H / must be species specific. Heterosigma carterae grown on N H / showed the maximum uptake rate of 0.042 ug-at lO^cells"1 h"1 in culture 1 and 0.33 ug-at lO^cells' 1 h'1 (culture 2+3). Following the rapid uptake of N H / , uptake rate slowed for a short period in the 81 Heterosigma carterae culture. The NH 4+ uptake rates during this period were 0.006 (2.5 - 4.5 h) in culture 1 and 0.014 ug-at lO^cells"1 h"1 (2.5 - 4.5 h) in culture 2+3. The same phenomenon also occurred in Prorocentrum micans grown on three N sources. However, such a pattern was not as distinct as the pattern that occurred in cultures grown on only NH 4+ (Exp. 6). Freshwater phytoplankton (Suttle & Harrison 1988) and a picoplankter, Micromonas pusilla (Cochlan ef al. 1991) exhibited the phenomenon of a short-term decrease in 1 5 N H 4+ uptake following surge uptake, which was similar to the results in the present study. This phenomenon could be explained by the short lag before the processing of NH 4+ into amino acids, or a sudden loss of membrane potential due to the influx of cations (Suttle & Harrison 1988). Evidence that N-starved Lemna gibba, which was re-supplied with N H 4+ suddenly lost membrane potential supports the later explanation (Ullrich ef al. 1984). Since approximately 100 ug-at N L"1 of N, with the combinations of N H 4+ (cation), NOV (anion), and uncharged urea were re-supplied to Prorocentrum micans (Exp. 1) and Heterosigma carterae (Exp. 3), membrane potential might have been lost, although the magnitude of influx of cations is less than in Heterosigma carterae supplied with only 100 uM NH 4+ . The re-supply of NH 4+ affects many cellular processes. During N deprivation, cultures continue accumulating polysaccharides by photosynthesis (Syrett 1981). Once these cultures are supplied with NH 4+ , they take up N H 4+ very rapidly and convert it to organic N compounds. Simultaneously, the accumulated polysaccharide disappears because it is converted into organic N \ 82 compounds (Syrett 1953; Hattori 1958; Reisner et al. 1960). Also, these polysaccharides are used for producing energy due to increased respiration during the rapid assimilation of NH 4+. 3) N i t ra te up t ake in N-deplete p h a s e NOV was not rapidly taken up by N-starved Heterosigma carterae when it was the only N source (Exp. 4 and 5), and when it was one of three N sources with NH 4+ and urea during period I (Exp. 1, 2, and 3). The concentration of N0 3" began to decline only after 3 h (Exp. 1), 6 h (Exp. 2), 11 h (Exp. 3), 6 h (Exp. 4), and 2.5 h (Exp. 5). Previous uptake rates were quite low and they were 0.001 (Exp. 1), 0.002 (Exp. 2), 0.004 (Exp. 3), 0.006 (Exp. 4), and -0.011 ug-at lO^cells'1 h"1 (Exp. 5). The difference in the period in which no N0 3 ' uptake occurs may be species dependent (Conway & Harrison 1977), and vary with the duration of N starvation (Parslow etal. 1984a), and other environmental factors. Overall, N-starved cells needed an acclimation period to start to take up N03" again. Maximum uptake rates ranged from 0.002 (Prorocentrum micans) to 0.025 ug-at lO^cells"1 h'1 (Heterosigma carterae grown on N0 3 ' under continuous light) and occurred some time after these low N0 3 ' uptake periods. In Exp. 1, 2, and 3, influence by NH 4+ and urea could be responsible for the period of no N0 3" uptake. When NH 4+ decreased to a certain concentration, then N0 3" started to decline. In Exp. 4 and 5, there must be another reason for Heterosigma carterae grown on NOV N03" was not taken up as readily by the N03'-starved culture of Heterosigma carterae compared to the reduced N form, N H 4 + by NfV-starved Heterosigma carterae. A reduction in the ability of NOV uptake after N deprivation has been commonly reported (Collos 1980; Dortch et al. 1982). N03"-starved phytoplankton usually require previous exposure to NO3" before N0 3 " uptake can occur (Dortch etal. 1982; Collos 1983; Parslow etal. 1984b). Starved cells achieve maximal N0 3" uptake after an acclimation period (Collos 1983; Harrison etal. 1989). One of reasons for a reduction in the initial N0 3 " uptake is that long term N0 3" deprivation might reduce the ability of N0 3 " uptake due to the loss of viability of cells. The decline in the ability to take up N 0 3 ' after short term N deprivation is also attributable to the loss of an active uptake system (Falkowski 1975) and to inactivation of nitrate reductase (Syrett 1981). Inactivation of nitrate reductase alone did not interfere with the initial uptake of N0 3 " since transient internal N 0 3 ' pools were observed in even N -starved cells (Dortch et al. 1984). Details will be discussed in the following "Interaction" section. 4) Urea uptake in N-deplete phase In the present study, urea uptake after N-starvation and re-supply of urea was observed in the interaction of other N sources such as N0 3 " and NH 4 + . Surge uptake of urea was not observed in any species. This agrees with the results of nb evidence of urea surge uptake by N-starved cells (Bekheet & Syrett 1979; Price & Harrison 1988). Maximum urea uptake rates were 0.044, 0.014, and 0.008 ug-at lO^cells'1 h"1 for Prorocentrum micans, Amphidinium carterae, and Heterosigma carterae respectively. However, there are many reports that 84 urea uptake is enhanced by N deprivation (Rees & Syrett 1979; Horrigan & McCarthy 1981; Price & Harrison 1988). The enhancement of urea uptake is attributed to the removal of NH 4+ originating from urea in the cell (Rees & Syrett 1979). Increased N-specific urea uptake might be due to a reduction in the cell N quota and retention of all of the urea-N by the N-starved cells. N In terac t ion The present study demonstrated that NOV uptake by Prorocentrum micans, Heterosigma carterae, and Amphidinium carterae could not proceed in the presence of NH 4+ concentrations as low as 1.6, 1.7, and 0.9 uM respectively before they experienced N deprivation. The three species started to take up N0 3" only after NH 4+ reached near zero. An interesting finding was that urea uptake rate exceeded NH 4+ uptake rate in Prorocentrum micans culture during a short-term decrease in NH 4+ uptake followed a period of rapid uptake in the N-deplete phase. Generally, NH 4+ influence on urea uptake was weaker compared to that on NOV and N03" influenced on urea uptake when only those two N sources co-existed in the medium for all three species. N-replete Prorocentrum micans (Exp. 1) showed less influence on N0 3 ' uptake by both NH 4+ and urea than Amphidinium carterae (Exp. 2) and Heterosigma carterae (Exp. 3). In addition, Prorocentrum micans used urea more favorably than Amphidinium carterae and Heterosigma carterae during the period (I), both before and after N starvation. Thus, this species could utilize the three N sources indiscriminately. In contrast, in Amphidinium carterae, NH 4 + 85 strongly influenced on both NOV and urea uptake before N starvation, and only NH 4+ was taken up preferentially. Both N03" and urea remained near their initial concentration until the NH4* concentration < 1 uM. Then, urea and N0 3" were both taken up almost simultaneously. However, after they experienced N starvation, only urea started to decline in the medium when > 18 uM NH 4+ existed. On the other hand, N03" was still not utilized at this NH 4+ concentration. Therefore after N starvation, influence of urea uptake by NH 4+ decreased but N0 3" uptake was still influenced. In Heterosigma carterae (Exp. 3), NOV was suppressed by NH 4+ but not urea, before and after N starvation because once NH 4+ disappeared from the culture medium, N0 3 ' was taken up much faster than urea. However, NH 4+ influence on N03" uptake clearly decreased after N starvation. Focusing on the interaction between N0 3 ' and urea, N0 3" influence on urea uptake during the aft. I I period was slightly strengthened in Prorocentrum micans. In contrast, such an influence was weakened in N-deplete Amphidinium carterae and Heterosigma carterae cultures. In general, species and nutritional state affect on interaction between N0 3 ' and NH 4+ (Syrett 1981; McCarthy 1981). Inhibition of N0 3 ' uptake by NH 4+ ranged from total suppression (Syrett & Morris 1963; McCarthy & Eppley 1972; Cresswell & Syrett 1979) to simultaneous and comparable rates of NH 4+ and NOV uptake in cultures (Caperon & Ziemann 1976; Conway 1977) and natural communities (Price etal. 1985, Collos etal. 1989). Higher concentrations of 86 N H 4+ inhibited N 0 3 ' uptake, but stimulation of N0 3 " uptake has been reported to occur at low N H 4+ concentrations (Yin 1988). The mechanism of the inhibition of N0 3 " utilization needs further investigation. Regulatory action at the level of N0 3 " uptake (Eppley & Rogers 1970; Tischner & Lorenzen 1979) and N 0 3 ' reduction (Syrett & Morris 1963; Hipkin et al. 1980) have been suggested to influence NOV uptake. To a certain extent, the effect of NH 4+ on both mechanisms may be independent (Ullrich 1987). It is known that N 0 3 ' uptake stops very quickly when N H 4+ is added and starts when the NH 4 * has disappeared (Syrett & Morris 1963; Eppley et al. 1969a; Conway et al. 1976). This inhibition by NH 4 + is too rapid to be accounted for by the inactivation of nitrate reductase. Probably, the primary and most rapidly acting effect of N H 4+ on N0 3* utilization is attributed to an inhibition of the N0 3 " uptake system. Then, this may be followed by the effects on N0 3 " metabolism through inhibition of nitrate reductase activity, by irreversible proteolytic breakdown (Hipkin etal. 1980), reversible inactivation (Pistorius etal. 1978), or suppression of its synthesis (Morris & Syrett 1963; Amy & Garrett 1974). The rate of N 0 3 ' uptake is modulated in response to changes in pools of some organic product of N H 4+ assimilation (Syrett 1981; Guerrero et al. 1981). Highly C-deficient cells did not show inhibition of N0 3 " uptake by N H 4+ and these cells reduced N0 3 " to NH 4+ (Syrett & Morris 1963; Thacker & Syrett 1972a; Ullrich 1987). Therefore, N H 4+ itself may not be an inhibitor and an organic product of N H 4+ assimilation is suspected to be an inhibitor. 87 NH 4+ repressed urea uptake in Thalassiosira pseudonana (Lui & Roels 1970). Lund (1987) found that NH 4+ immediately inhibited urea uptake by Skeletonema costatum, while Horrigan and McCarthy (1982) reported that urea uptake inhibition occurred after 30 min. When these two diatom species were N-depleted, NH 4+ inhibited urea uptake to an even greater extent. In general, urea suppresses NOV uptake but at a lower level than NH 4+ (Grant et al. 1967; McCarthy & Eppley 1972; Molly & Syrett 1988b). However, simultaneous uptake of urea and other N sources by phytoplankton communities (McCarthy & Eppley 1972; Price et al. 1985) and a normal uptake rate of NOV in the presence of 10 ug-at N L"1 urea by Skeletonema costatum (Lund 1987) have been reported. Urea was taken up by a natural phytoplankton assemblage simultaneously with N03" and NH 4+, with little influence on the uptake rates of NH 4+ (Robert et al. 1986). Repression of nitrate reductase by urea, which was reported in studies on Cyclotella cryptica (Liu & Hellebust 1976) and Chlorella (Smith & Thompson 1968) is a possible explanation of the inhibition of N0 3" uptake by urea. Nutritional state affects the interaction between urea and other inorganic N sources. N0 3" uptake inhibition by urea weakened after N starvation in the study of Skeletonema costatum (Lund 1987). On the other hand, Molly and Syrett (1988a) reported that urea increased the suppression of N0 3" uptake. NH 4+ suppression of urea uptake disappeared due to N starvation (Rees & Syrett 1979). Horrigan & McCarthy (1982) reported that urea and NH 4+ uptake rates were enhanced even for N03'-sufficient cells. 88 To conclude, the magnitude of the effect of the interaction of three N sources on uptake rates was species specific, and influenced by N starvation. As it was mentioned in the N preference section, the interaction among N sources was variable and depended on the N sources in the medium. Since N H / and urea are found in similar concentrations in the sea (Harrison 1992), it is important to consider urea in studies of N interactions. Increased concentrations and efflux of N An increase in the external N concentration during the initial interval after the re-addition of N sources occurred in the culture of Prorocentrum micans (three N), Amphidinium carterae (three N), Heterosigma carterae (three N), Heterosigma carterae (NOV), and Heterosigma carterae (NH/ , culture 2+3). In the Amphidinium carterae culture, an increase in the concentration of both N0 3 " and urea occurred. In Exp. 3 and 5, Heterosigma carterae excreted NOV. Urea efflux during the initial interval was -0.006 ug-at lO^cells"1 h' 1 during 0 -1 h (Prorocentrum micans), -0.03 g-at lO^cells"1 h"1 during 0 - 0.5 h (Amphidinium carterae), and -0.0006 ug-at lO^cells"1 h'1 during 0 -1 h (Heterosigma carterae) when these three N-starved species were re-supplied with three N sources. Effluxes of N0 3 " of -0.007 ^g-at lO^cells'1 h'1 (Exp. 3) and -0.022 ug-at lO^cells'1 h"1 (Exp. 5), during the initial interval of the N re-supplied phase were observed in N-starved Heterosigma carterae. The initial stage of urea uptake involves a rapid influx and probably an accumulation of an intracellular pool (Antia et al. 1991). Excretion of urea in the 89 form of dissolved organic N is not re-assimilated in the short term. Excretion into the culture medium of 14C-labelled organic products derived from urea-C could explain the lower urea uptake rates (Antia et al. 1991). Efflux of N0 3" in Exp. 3 coincided with the high uptake rate of NH 4+. Therefore, NH 4+ taken up by N-starved cells may be attributed to this efflux of N0 3 ' . The efflux of N03" of N-starved Heterosigma carterae in Exp. 5 occurred during the light period, although the N-replete culture showed several periods of efflux of N0 3 ' during dark periods. Pujo-Pay et al. (1997) reported < 10% of the N0 3" uptake was released as DON with the rate of 10.4 to 13.3 nmol N L"1 h"1. They mentioned the released DON was an important resource for some organisms. NH 4+ excretion occurred in NH4+-grown Heterosigma carterae culture, and such a phenomenon has been reported often. NH 4+ excretion was observed by axenic cultures of Thalassiosira pseudonana after the addition of 10 ug-at N L"1 of urea (Price & Harrison 1988). NH 4+ excretion also occurred in urea-grown Phaeodactylum tricornutum (Rees & Syrett 1979). Urea appears to be one of the sources of NH 4+ excretion. Ecology Amphidinium carterae showed an enhanced preference for NH 4 + relative to both N0 3" and urea during the period I and preference for urea relative to N0 3" during period I I . Prorocentrum micans also enhanced its preference for urea rather than N03" during N starvation, however, its preference for NH 4+ 90 decreased. In general, species that increase their preference for NH 4+ or urea have a better chance to take up a transit N supply of regenerated N, which is mainly NH 4+ and urea. However, the increased preference for NOV relative to NH 4+ and urea by Heterosigma carterae does not appear to offer an advantage for survival in a N-deplete environment. Nevertheless, N-deprived Heterosigma carterae still possessed a much stronger preference for NH 4+ (NH4*:N03':urea = 39:1:5.2) compared to the other two species. The reduction in the ability to take up N03* and the increase in the ability to take up NH 4+ may be an ecological acclimation by N-deprived phytoplankton in oligotrophic waters (Dortch et al. 1982). N0 3 ' is recycled over a much longer time period and it is usually supplied continuously at low rates by eddy diffusion from deeper N03'-rich waters (Dugdale 1967; Eppley et al. 1979). Sporadic N0 3" supply by upwelling, frontal mixing, and internal waves occurs in the euphotic zone at very slow rates (Walsh et al. 1978; Pingree et al. 1978; Parsons et al. 1981; Cullen etal. 1983). These physical events also dilute the phytoplankton community in the euphotic zone. Therefore, the demand for the nutrient by phytoplankton decreases and N0 3 ' remains in the euphotic zone for some time, which reduces the benefits for transient elevated uptake rates of N0 3" (Parslow et al. 1984b). Compared to NOV NH4* and urea are recycled rapidly within the euphotic zone and introduced to the system by animal excretion (Dugdale 1967) Maintaining a high uptake capability during N starvation results in metabolic costs (Raven 1986). Therefore, it is probably more beneficial for cells to have a high uptake capability for NH 4+ and urea rather than N0 3 ' because NH 4+ and 91 urea are recycled N sources. Interestingly, annual phytoplankton production expressed as N production in well mixed water (Western English Channel) was mainly based on recycling of N forms such as N H 4 + (54%), but not new N forms (Maguer et al. 1995). To conclude, phytoplankton which are capable of assimilating N H 4 + or urea rapidly after they are N starved may have an advantage for species succession and growth in the environment where N limitation is the primary stress. Therefore, Amphidinium carterae that increased the preference for N H 4 + relative to NOV (period I ) , and the preference for urea relative to N0 3 " (period I I ) , can acclimate to N-deplete water, while Prorocentrum micans and Heterosigma carterae increased their preference for N0 3" during period 1 or 11. Growth rate Growth rates were 0.30 for Prorocentrum micans (three N sources), 0.76 for Amphidinium carterae (three N sources), 0.56 for Heterosigma carterae (three N sources), 1.0 for Heterosigma carterae (N0 3 ' + continuous light), 0.72 for Heterosigma carterae (N0 3 ' + L D cycle), 1.1 for Heterosigma carterae (NH 4 +) (culture 1), and 1.02 d"1 for Heterosigma carterae (NH 4 +) (culture 2+3). Prorocentrum micans had the lowest growth rate (0.30 d"1). Prorocentrum micans grown in f/2 medium under continuous light and 19±10°C showed growth rates ranging from 0.59 to 0.99 d'1 in the study by Costas (1990). The temperature used in the present study (17°C) was lower than the 19°C in Costas' study and could partially account for the lower growth rate in 92 Prorocentrum micans. Growth rates of Prorocentrum micans were measured in a culture containing a mixture of other dinoflagellates (Fedorov & ll'yash 1991). Prorocentrum micans and other dinoflagellates such as Glenodinium foliaceum, Excuviaella cordate, Gymnodinium kovalevskii, and Olisthodiscus luteus (Heterosigma carterae) were grown in Goldberg medium (Lanskaya 1971) under natural illumination at 20 - 22 °C. The growth rate of Prorocentrum micans was 0.2 d*1. When Prorocentrum micans was grown by itself without the interaction with other species, its growth rate was also low and < 0.5 d'1, even though a relatively high temperature was used. These results are in agreement with this study, and indicate that Prorocentrum micans possesses a low growth rate. Fedorov and ll'yash (1991) concluded that Prorocentrum micans was able to increase its population density slowly but steadily by utilizing limited amounts of resources and classified it as a "patient" species. Amphidinium carterae had the highest growth rate of 0.76 d"1 among the other three species grown on three N sources (Exp. 2). This species also exhibited the same growth rate grown under a 14:10 L.D cycle at 20°C when 50 L I M NH 4 + was given to the culture at the beginning of the light period (Wheeler et al. 1983). However, if the same concentration of NH 4 + was given at the beginning of the dark period, the growth rate increased to 1.4 d"1. Amphidinium carterae has been used for many studies because it grows well in various culture conditions. The growth rates of Heterosigma carterae ranged from 0.56 to 1.1 d"1 depending on culture conditions. Since this is the first study for Heterosigma carterae grown on three combined N sources, there are no comparable data on growth rates from the literature. However, growth rates on individual N sources are available. All growth rates were < 0.8 d"1 under 14:10 L.D photoperiod of 160 umol photon m'2 s'1 at 20°C with 100 uM of each N (Hoe Chang & Page 1995). Another study by Wood & Flynn (1995) reported 0.9 d"1 and 0.6 d"1 when grown on N H 4 + and N0 3" of 75% ESAW (Harrison et al. 1980) respectively in a 12:12 L.D cycle at 18°C. The growth rates of N0 3" and NH 4* grown cells in the present study were generally higher than those of literature values. Light: Dark cycle 1) Cell division and fluorescence number An expected step-wise increase in fluorescence, which indicates synchronicity and phasing of growth to the L:D cycle, was not evident in Heterosigma carterae under the 14:10 L:D cycle (Exp. 5). Therefore, the occurrence of cell division at night could not be detected. However, the fluorescence increased much more smoothly in cultures under continuous light (Exp. 4). The growth rates for Exp. 4 and 5 were 1.00 d"1 (0 - 64 h) and 0.72 d'1 (49 -105 h) respectively. Growth rate often varies with a 24 h periodicity and the timing of division depends on experimental conditions and species (Sweeney 1983). 94 2) U p t a k e d i e l p e r i o d i c i t y Diel periodicity of NOV uptake occurred during log phase growth. N-replete Heterosigma carterae grown on NOV under a LD cycle showed high N03" uptake rate during the light period and at the beginning of the dark period. At the end of dark period, uptake rates were dramatically decreased to negative values or values close to zero. During early log phase, a large decrease in uptake rate was not observed in the culture under continuous light (Exp. 4). This observation agreed with findings by Cochlan et al. (1991) who found that NOV uptake rates were maximal during the light periods and decreased during the dark periods for Micromonas pusilla. The outdoor culture of Chaetoceros sp. which was grown on low N also exhibited cyclic variations in N0 3" uptake with the maximum at night and the minimum during the day (Malone et al. 1975). In the present study, after N starvation, N03" uptake occurred during the dark period. The uptake rate of N-deplete cells in the dark was 0.017 ug-at lO^cells"1 h"1 and higher than that of N-replete cells. This agrees with the finding that Pavlova lutheri grown on low N showed an increase in the relative importance of potential nighttime N uptake (Laws & Wong 1978). N availability has been considered a determining factor of the enhancement of nighttime uptake. When Gyrodinium aureolum was N-starved for 24 h, it exhibited an increase in nighttime N0 3 ' uptake (Paasche et al. 1984) even though it did not show nighttime N03" uptake before N starvation. N-starved dinoflagellates, Pyrocystis noctiluca and Dissodinium lunula were independent of the L:D cycle in their N0 3 ' uptake capacity (Bhovichitra & Swift 95 1977). N limitation may enhance the potential for dark N uptake more than light uptake. Consequently, diel periodicity of N uptake becomes less evident. The enhancement of light independence of N uptake might be an adaptive response to N limitation since N-starved phytoplankton could optimize their uptake capability under even low light. Cochlan et al. (1991) found that with increased N limitation, the relative dark uptake capacity increased four-fold for total (light + dark) N0 3" uptake, and two-fold for total NH 4+ uptake. A consequence of enhanced dark N uptake by N starvation is the dampening of the diel N uptake periodicity. Laboratory studies by Harrison (1976) showed dark uptake by Gonyaulax polydra increased to ca. 40% of daytime uptake rate in N-starved cultures. On the other hand, in N-sufficient cultures, N uptake during nighttime was only ca. 20% of that during daytime. The absence of diel periodicity in N03" uptake occurred in axenic cyclostat cultures (Picard 1976). Dark NO3" uptake capacity of cyclostat cultures of the prymnesiophyte, Pavlova lutheri and the chlorophyte, Dunaliella tertiolecta exceeded the supply rate of N03" at high dilution rates, which suggested that the environment had less available N for cells (Laws & Caperon 1976; Laws & Wong 1978). The coccolithophore, Emiliania huxleyi, which was grown in both N O 3 -and NH4+-limited cyclostat cultures did not exhibit diel periodicity (Eppley et al. 1971). There are two contrasting examples in field studies. In N03"-depleted water off Peru with a dinoflagellate bloom dominated by Gymnodinium sanguineum, N03" uptake during nighttime was ca. 50% of that during the 96 daytime (Dortch & Maske 1982). On the other hand, in the bloom off Baja California, which was dominated by Gonyaulax polyedra, night time N0 3 ' uptake was 10 - 20% of daytime N03" uptake (Maclsaac 1978). The later case had a higher N0 3" concentration than the former case in the surface waters where phytoplankton resided. In addition, the nitracline off Baja California was shallower than off Peru and thus a shorter downward vertical migration in order to obtain N03". This result agreed with the field study by Harrison (1976), which revealed the dampening effect on N uptake by red tide populations dominated by Gonyaulax polyedra. 3) Heterosigma carterae a n d da rk N0 3 ' up take In the present study, high N0 3 ' uptake occurred during the light and not during the dark before N0 3 ' deprivation. Low N03" uptake (81 - 90 h, 105-114 h) and even efflux of N03" (32.5 - 42 h, 57 - 66 h) occurred at the end of dark periods. However, N03" uptake rate in the dark by N-starved cells was positive and higher than by N-replete cells. During diel migration of Heterosigma carterae, the cells stay close to the surface in the day time to carry out photosynthesis and migrate downwards to deeper water at night to take up nutrients (Watanabe et al. 1983; Yamochi & Abe 1984; Nagasaki etal. 1996). Considering that maximal nitrate reductase activity (Eppley & Harrison 1975) occurs during the day time, N0 3 ' uptake during night time via vertical migration is not efficient (Raven 1986). In addition, diel migration costs are metabolically high. Therefore, night-time N0 3" uptake does 97 not contribute to the entire supply of nutrients for Heterosigma carterae. However, considering that N limitation increased the SAN ratio (Cochlan et al. 1991) and increased nitrate reductase activity during the dark period (Eppley & Harrison 1975), vertical migration at night might be more beneficial for N-starved Heterosigma carterae than N-replete cells. 98 CONCLUSIONS This dissertation examined N uptake by the dinoflagellates, Prorocentrum micans and Amphidinium carterae, and the raphidophyte, Heterosigma carterae as a function of the photoperiod and N physiological states. The specific findings of the research are summarized below. 1. A change in the relative preference of the three N sources occurred during N deprivation. N-deplete Prorocentrum micans increased its preference for urea relative to both NH 4+ and N03". N-deplete Amphidinium carterae enhanced its preference for NH 4+ relative to both N03" and urea, and its preference for urea relative to N0 3 ' under the interaction of three N sources. The preference for N0 3" was enhanced by N-deprived Heterosigma carterae. 2. Strong NH 4+ influence on N03" and urea uptake were observed in N-replete Amphidinium carterae and Heterosigma carterae. N-replete Prorocentrum micans only showed strong NH 4+ influence on N0 3" uptake, but not on urea uptake. N deprivation weakened the influence of NH 4+ on urea uptake in Heterosigma carterae. The N uptake preference depended on the interactions of each N source. 3. N0 3" uptake needed a longer acclimation period after the N0 3" pulse in N-starved Heterosigma carterae grown under a light:dark cycle than under continuous lights. A short acclimation period for N0 3 ' uptake occurred in 99 Prorocentrum micans, but not in Amphidinium carterae and Heterosigma carterae grown on three N sources. The duration of the acclimation periods appeared to be influenced by the light cycle and species specificity. 4. The uptake of N0 3" by Heterosigma carterae was phased by a light:dark cycle. N0 3 " disappearance in the culture medium accompanied by a high uptake rate occurred mostly during the light periods and only a small disappearance accompanied by a low uptake rate and an efflux occurred during the dark periods. This uptake cycle may be due to nitrate reductase activity, which usually decreases at night. Diel vertical migration to take up N0 3 " in deeper water at night seems less likely than previously thought for Heterosigma carterae. 100 FUTURE RESEARCH SUGGESTIONS The outstanding question is why N-deprivation changed the magnitude of N preference by marine phytoplankton. The impact of N deprivation on the mechanisms involved in the uptake of each N source needs further investigation. The change in the interaction among N sources during the N deprivation also needs to be examined under various conditions such as the availability of light, lower concentrations of N, and widely ranging periods of N starvation. An increase in the preference for urea as a N source during N starvation seems a general phenomenon in marine phytoplankton, which was observed in this study on dinoflagellates and other previous studies on a diatom and coccolithophores. However, the evidence for a decrease in the preference for urea by Heterosigma carterae (raphidophyte) suggests species or even group specificity. It was a surprise that N-replete Heterosigma carterae showed negative or very small N0 3 " uptake rates during dark periods because it is known to vertically migrate downwards at night, presumably to take up nutrients. N-deplete Heterosigma carterae showed dark uptake of NO3'. However, unfortunately the dark period occurred when the cells had taken up most of the re-added N0 3", and therefore it was not possible to determine the effect of N starvation on dark uptake. Therefore, it is more interesting to re-add a higher concentration of N0 3" at the beginning of the dark period, which provides a longer period of observation to determine the influence of N deprivation on N0 3 " uptake in the dark. 101 P a r t n RESULTS Isotope fract ionat ion 1) Prorocentrum micans grown on N H / , N0 3 ' , and urea In the N-sufficient phase (0-106 h), N0 3" and urea were used along with N H 4 + (Fig. 29). However, N0 3" was completely consumed before urea. During growth mainly on NH 4 + , the apparent isotope discrimination was relatively large and decreased upon exhaustion of NH 4 + (Fig. 30). At this time (31 h), N H 4 + had been completely incorporated into PN and 8 1 5 N P N was in good agreement with the expected value of -1.14%o of 6 1 5N As (Table 3). N 0 3 ' uptake was accompanied by a small change due to both a smaller e(N03") and a significant amount of preexisting PN produced during growth on NH 4 + . In the 65 -106 h time period, 8 1 5 N P N showed a slight increase due to the uptake of urea. At stationary phase, the 8 1 5 N P N was 0.24%o. This value was in good agreement with the 815Nmx of 0.26%o, which was expected when all three N sources were completely taken up. The model gave a good fit for the e(NH4) value of 12%o. However, during the drawdown of N O V and urea when the model used an e(urea) of 0%o, it systematically overestimated 8 1 5 N P N . During the N re-supply phase, again, NH 4* and urea were immediately utilized, and N 0 3 ' uptake coincided with the complete consumption of NH 4 + . Using the same variables as in the N-sufficient phase, the model gave a good fit for e(NH4+). Again, the model systematically overestimated 8 1 5 N P N during the 102 Time (hours) Figure 29 Time series of [NOV] (•), [NH4+] (O) and [urea] ( • ) during the growth of Prorocentrum micans grown on N03", NH/ , and urea under continuous light. Culture was N-starved between 105 and 186.5 h and then 36.6 ug-at N L"1 NH 4 +, 48.7 ug-at N L"1N03", and 49.4 ug-at N L"1 urea were added at 186.5 h. The best fits for the concentrations were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentration, b is a co-efficient number, and c is the time when N reached 0, and are indicated by the solid lines. 103 Figure 30 Time series of PN(A) and 815NPN(<1) during the growth of Prorocentrum micans grown on N03", NH 4+, and urea under continuous light. Culture was N-starved between 105 and 186.5 h and then 36.6 ug-at N L'1 NH 4 +, 48.7 ug-at N L"1 N0 3 ' , and 49.4 ug-at N L'1 urea were added at 186.5 h. The 8 1 5 N P N as predicted by the multiple N source uptake model is shown by the solid line. The 3 dotted lines indicate times when [NH4+], [NOV] and [urea] fits, determined by the dissolved N data to the concentration data in Fig. 29, reached 0, respectively. 104 Table 3. Listing of variables and definitions used in the multiple N source uptake model. Variable Definition Value* P N M Measured P N concentration I P N X Calculated PN concentration from dissolved N drawdown PNj Initial PN concentration 8 1 5N N S 815N of the initial N 0 3 ' source 3.04%o 8 1 5N A S 515N of the initial N H 4 + source 1.14%o 8 1 5N U S 815N of the initial urea source 0.74%o 815NMX 815N of the initial mixed N sources (calculated from mass balance) 0.26-0.46%o 8 1 5N P N i 815N of PN, e(N03) Isotope fractionation for the overall reaction: NO3" -» PN e(NH 4) Isotope fractionation for the overall reaction: N H 4+ -» PN e(urea) Isotope fractionation for the overall reaction: Urea -» PN 0%o A Apparent isotope fractionation or discrimination, i.e. 8 1 5N P N - 815Nst *: constant values are given. 105 consumption of NOV and urea. All the base variables are summarized in Table 4. A sensitivity analysis is presented later. 2) Amphidinium carterae grown on NH 4 + , N0 3", and urea In the N-sufficient phase (0 - 101 h), NH 4 + was utilized first followed by the simultaneous uptake of N 0 3 ' and urea (Fig. 31). Growth on N H 4 + produced a large initial decrease in 8 1 5 N P N to values of -8.94 to -12.1%o (Fig. 32). 5 1 5 N P N then increased to about -0.6 %o during the simultaneous growth on urea and N0 3 " (60 - 94 h) and finally reached 0.73%o at the stationary phase. The model gave a relatively good fit for the NH 4+ uptake phase with e(NH 4+) values of 20%o. However, the model systematically underestimated 8 1 5 N P N during the simultaneous growth on urea and NOV 8 1 5 N S T of 0.76%o showed a relatively good agreement with the initial 8 1 5 N M x (0.27%o). In the N re-supply phase, NH 4+ was taken up immediately, followed by simultaneous uptake of urea and NOV 5 1 SN was -0.24%o (193 h) and, then, increased to 0.33 and 0.04%o at 30 min and 1 h respectively after the re-addition of the three N sources. Positive 8 1 5 N P N values indicate more incorporation of 1 5 N into the cells than 1 4 N . Using the same variables as in the N-sufficient phase, the model systematically underestimated the 8 1 5 N P N during the growth on any N. A sensitivity study for Amphidinium carterae is also provided later. 106 Tab le 4 Var iables and their values used in the multiple N source uptake model . Prorocentrum micans, Amphidinium carterae, and Heterosigma carterae were grown on multiple N sources ( N O V , N H 4 \ and urea) under N-sufficient condit ions. After 0 - 82 h of starvation, all three N sources were added simultaneously to the culture (i.e. N re-supply phase). Heterosigma carterae was grown on N O V under N0 3 ' -suf f ic ient condit ion. After 52 h of N starvation, N 0 3 " was added to the culture (i.e. N re-supply phase). Symbols are def ined in Table 3. Species e ( N H 4+ ) s ( N 0 3 ) PNi 81 5 N j * (%o) (%o) G*M) (%o) N-suff icient phase - 0.42 Prorocentrum micans 12 9 12 Amphidinium carterae 20 0 3.04 0.27 Heterosigma carterae 24 9 3.87 - 1 . 3 6 Heterosigma carterae — » 5 5.1 1.08 N re-supply phase - 0 . 5 4 Prorocentrum micans 18 9 107.4 Amphidinium carterae 4 0 77.1 - 0 . 2 4 Heterosigma carterae 20 5 100 - 0 . 0 1 Heterosigma carterae 9 92.9 0.78 *: 8 1 5 Ni in the initial N source 107 5 0 i 0 2 0 4 0 6 0 80 1 0 0 1 8 0 185 1 9 0 1 9 5 2 0 0 Time (hours) Figure 31 Time series of [N0 3 " ] ( • ) , [NH 4+ ] (O) and [urea] ( T ) dur ing the growth ol Amphidinium carterae grown on N0 3 " , N H 4+ , and urea under cont inuous light. Culture was N-starved between 119 and 193.5 h and then 29.6 ug-at N L'1 N H 4 + , 34.3 ug-at N L'1 N 0 3 " , and 31.2 ug-at N L"1 urea were added at 193.5 h. The best f its for the concentrat ions were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentrat ion, b is a co-efficient number, and c is the t ime w h e n N reached 0, and are indicated by the solid lines. 108 Figure 32 Time series of PN ( • ) and 8 1 5 N P N ( • ) during the growth of Amphidinium carterae grown on N0 3", NH 4+ , and urea under continuous light. Culture was N-starved between 119 and 193.5 h and then 29.6 ug-at N L'1 N H 4 \ 34.3 ug-at N L'1 N0 3", and 31.2 ug-at N L"1 urea were added at 193.5 h. The 8 1 5 N P N as predicted by the multiple N source uptake model is shown by the solid line. The 3 dotted lines indicate times when [NH 4+], [ N 0 3 ] and [urea] fits, determined by the dissolved N data to the concentration data in Fig. 31, reached 0, respectively. 109 3) Heterosigma carterae grown on NH 4 + , NOV, and urea In the N-sufficient phase ( 0 - 1 7 5 h), the three N sources were utilized in sequence (Fig. 33). NO3" was taken up only when N H 4 + decreased to < 1 uM. Large 8 1 5 N P N values such as -8.16, -13.3, and -10.6%owere observed during the growth on N H 4+ at 11, 36, and 49 h respectively (Fig. 34). The isotope discrimination decreased to a small value upon exhaustion of NH 4 + . At stationary phase, the 8 1 5 N P N was -0.41 %o. This is a little lower than the value of 0.46%o for 815NMX- The model gave a relatively good fit. When the exhaustion of N H 4+ occurred (68 h), 8 1 5 N P N (-2.32%o) was in good agreement with the expected values of -1.14%o for 8 1 5 N A S -Since urea was hardly consumed in the N-replete phase, cells were not N-starved completely. NH 4+ was used first and followed by N0 3 " and urea in the N re-supply phase. The 8 1 5 N P N showed a relatively large decrease which was well predicted by the model using the same variables as in the N-sufficient phase. An s(NH 4+) value of 20%o was used for the fit in N-deplete phase. 4) Heterosigma carterae grown on N0 3 " Figure 35 shows the time series data for the nutrient fit. N0 3 " was taken up without showing a distinct lag phase. In the N-sufficient phase (0 - 6 3 h), the model gave a relatively good fit with e(N03") values of 5%o (Fig. 36). However, 8 1 5 N P N at the beginning of the experiment was lower (1.08%o) than the expected value (3.04%o). During the log phase, the 8 1 5 N P N decreased to -1.04%o at 48 h 110 50 i 0 50 100 150 200 250 Time (hours) Figure 33 Time series of [N0 3"] ( • ) , [NH 4+] (O) and [urea] ( T ) during the growth of Heterosigma carterae grown on N0 3", NH 4+ , and urea under continuous light. Culture was added 32.2 ug-at N L'1 NH 4 + , 41.2 ug-at N L'1 NOV, and 37 ug-at N L'1 urea at 175 h. The best fits for the concentrations were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentration, b is a co-efficient number, and c is the time when N reached 0, and are indicated by the solid lines. 111 Figure 34 Time series of PN (A) and 8 1 5 N P N (•) during the growth of Heterosigma carterae grown on NOV, NH 4 + , and urea under continuous light. Nitrogen additions at 175 h to the culture were 32.2 ug-at N L"1 NH 4 + , 41.2 ug-at N L"1 N0 3 ' , and 37 ug-at N L"1 urea. The 8 1 5 N P N as predicted by the multiple N source uptake model is shown by the solid line. The 3 dotted lines indicate times when [NH4 +], [N0 3 ] and [urea] fits, determined by the dissolved N data to the concentration data in Fig. 33, reached 0, respectively. 112 100 i Time (hours) Figure 35 Time series of [NOV] during the growth of Heterosigma carterae grown on NOV under continuous light. Culture was N-starved between 64 and 115.5 h and then 85.6 ug-at N L"1 N0 3 "was added at 115.5 h. The best fits for the concentrations were determined by the dissolved N data using the equation: y = a*(1-EXP(b*(x-c))) where a is an initial N concentration, b is a co-efficient number, and c is the time when N reached 0, and are indicated by the solid lines. 113 Figure 36 Time series of PN (A) and 6 1 5 N P N ( • ) during the growth of Heterosigma carterae grown on N0 3 " under continuous light. Culture was N-starved between 64 and 115.5 h and then 85.6 ug-at N L"1 NOV was added at 115.5 h. The 8 1 5 N P N as predicted by the N source uptake model is shown by the solid line. The dotted line indicates times when [N0 3 ' ] fits, determined by the dissolved N data to the concentration data in Fig. 35, reached 0. 114 from the expected value for 3.04%o for the N O V source. During N starvation (100 h), the 8 1 5 N p N was 0.90%o, which was lower than the expected value of 8 1 5 N N S ( 3 . 0 4 % O ) . In the N re-supply phase, NO3" was not taken up immediately. Positive 8 1 5 N P N values (0.78 and 0.86%o) similar to the one in the stationary phase occurred when re-added N O V was not utilized ( 1 1 5 . 5 - 1 1 8 h). At the end of this period of N 0 3 " re-supply (119 h), 8 1 5 N decreased again to -0.23%o. Then, negative values (-0.42 and -0.02%o) followed at 121 and 124 h respectively. During the exhaustion of N O V (128 h), 8 1 5 N P N became positive (1.26%o) again. An e ( N H 4 + ) of 9%o was used for the model fit. 115 SENSITIVITY ANALYSIS Changes in various variables were made to study their effect on the 8 1 5 N P N in N-replete and N re-addition conditions. The value of s(urea) was taken as a constant of 0%o. Changes in the values of e(NH 4+), s(N0 3"), and PN, are presented in Table 3. 1) Prorocentrum micans grown on NH 4 + , N0 3", and urea In N-replete phase, variations in the values of e(NH 4+) produced large changes in the 8 1 5 N P N (Fig. 37a). The e(NH 4+) that gave the best fit was 12%o. During this N-replete period, urea which has a lower s value than NH 4+ , was also taken up with NH 4+ . The values of e(N0 3 ') from various fits ranged from 5 to 9%o (Fig. 37b). Since 8 1 5 N P N data during the drawdown of NOV were lacking, it was impossible to obtain the best e(N03"). Changes in PN f were made since PN, concentration seemed large in this culture (11.1 ug-at N L"1). PN, values used for the fit ranged from 11 to 13 ug-at N L'1 (Fig. 37c). A higher PN, value tended to improve the fit. Again, variations in the values of e(NH 4+) produced a large change during the N-deplete phase (Fig. 38). Interestingly in the N-replete phase, the s(NH 4+) that gave the best fit, increased to 18%o from 12%o. 116 e(NH4*) = 12 e(NH44) = 16 - e(NH4*) = 20 40 80 120 160 Time (hours) 200 Figure 37 Response of the model to changes in a) e(NH 4+), b) e(N03"), and c) PNj in the N-sufficient condition for Prorocentrum micans grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4. 117 Figure 38 Response of the model to changes in s(NH 4+) in the N re-supply phase for Prorocentrum micans grown on three N sources. The solid line represents the response of the model using the base variables indicated in Table 4. 118 2) Amphidinium carterae grown on NH 4 + , N 0 3 \ and urea Changing s (NH 4 + ) values made a large difference in predicting 5 1 5 N P N , however, it was not significant in the beginning (0 - 6 h) and the end (59 - 61 h) of N H 4 + uptake (Fig. 39a). The model predicted the data well for growth on N H 4 + for E(NH 4 + ) values of 18 - 22%o. e(N03') decreased to a minimum value of 0%o (Fig. 39b). Lower s(N03") values tended to improve the fit. For the N-deplete phase, e(NH4+) ranged from 4 to 20%o (Fig. 40). None of the e(NH4+) values could fit the data during N H 4 + drawdown. The model underestimated 5 1 5 N P N in the entire N-deplete phase. 3) Heterosigma carterae grown on NH 4 + , N 0 3 \ and urea During the N-replete phase, the model predicted the data well for growth on N H 4 + for e(NH4+) values of 22 - 26%o (Fig. 41 a). The s(NH4 +) value that gave the best fit was about 24%o. PNj had a large effect on the response of the model during the initial growth on N H 4 + (Fig. 41b). Variations in the values of s(NH4 +) and PNj produced large changes in the 8 1 5 N P N at low PN concentrations. The values of s(N03") ranged from 5 to 9%o (Fig. 41 c). The e(N03') value that gave the best fit was about 9%o. After the re-addition of three N sources, the model predicted well the data for values of e(NH4+) of 20%o (Fig. 42). This value was smaller in magnitude than the one derived in the N-sufficient phase. 119 Figure 39 Response of the model to changes in a) e (NH 4+ ) and b) e(N0 3 " ) in the N-sufficient condit ion for Amphidinium carterae g rown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4. 120 r -1 CO -2 \ 7 s(NH 4*) = 4 E ( N H / ) = 12 s ( N H / ) = 20 180 182 184 186 Time (hours) 188 190 Figure 40 Response of the model to changes in e (NH 4+ ) in the N-sufficient condit ion for Amphidinium carterae g rown on three N sources. The solid line represents the response of the model using the base values indicated in Tab le 4. 121 0 50 100 150 200 Time (hours) Figure 41 Response of the model to changes in a) e(NH4+) and b) s(N03") in the N re-supply phase for Heterosigma carterae grown on three N sources. The solid lines represent the response of the model using the base variables indicated in Table 4. 122 1 i _ 0 o u> CO 170 180 • s(NH4*) = 4 — s(NH4*) = 12 — s(NH4*) = 20 190 200 Time (hours) 210 220 Figure 42 Response of the model to changes in e(NH 4+ ) in the N re-supply phase for Heterosigma carterae grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4. 123 4) Heterosigma carterae grown on N 0 3 ' Variations in the values of e(N0 3") produced large changes in the 8 1 5 N P N at low PN concentrations in N-replete phase (Fig. 43). The values of e ( N 0 3 ) that fitted the data best was in the range of 5 - 7%o. The best fit value of e(N0 3 ' ) in the N-deplete phase clearly increased (Fig. 44). An e(N0 3") of 9 % o was used for the best fit model, although 8 1 5 N P N values at 119 and 121 h were lower than the predicted 8 1 5 N P N of e(N0 3 ' ) of 11 % o . The changes of e(N0 3") had a large effect on the fit for the 121 -125 h time period, but a small effect on the fit immediately after the re-addition of N0 3 " (115.5 -116 h) and near the completion of NOV utilization (126-127 h). 124 Figure 43 Response of the model to changes in s(N03") in the N-sufficient condition for Heterosigma carterae grown on three N sources. The solid line represents the response of the model using the base values indicated in Table 4. 125 _2 -I - i 1 1 1 1 ' < — — 1 114 116 118 120 122 124 126 128 130 Time (hours) Figure 44 Response of the model to changes in e(N0 3") in the N re-supply phase for Heterosigma carterae grown on three N sources. The solid lines represent the response of the model using the base variables indicated in Table 4. 126 DISCUSSION The present study attempted to show that the 8 1 5 N P N resulting from the growth of two dinoflagellates and a raphidophyte in a medium containing N 0 3 ' or three N sources was predicted by the model. The model assumed 8 1 5 N P N was described as the weighted sum of the changes in 8 1 5 N P N resulted from the incorporation of each individual N source. Each N incorporation was treated as a unidirectional reaction and the total N incorporation was described as the sum of these reactions. The models fitted the data relatively well. However, during the simultaneous growth on two N sources, especially on N H 4 + and urea, the prediction was difficult. For example, Prorocentrum micans had a much lower s(NH 4 +) than the expected value of 20%o. The data for Amphidinium carterae and Heterosigma carterae were relatively well predicted by using the expected e value of each N. e(N0 3 ) The e(N03') value for Amphidinium carterae in N-replete condition that gave the best fit was 0%o. This value was lower than e(N03") of 1.9+0.3%o obtained by Needoba (1997) for Amphidinium carterae grown on N03*. s(N0 3 ') of 0%o is also lower than previous estimates for diatoms, a coccolithophore and even dinoflagellates. e(N03") for the diatoms, Chaetoceros debilis, Chaetoceros simplex, Skeletonema costatum, Thalassiosira weissflogii, and Ditylum brightwellii were 4 - 5, 3.0±0.5, 2.2+0.4, 6.2±0.3, and 2.8+0.3%o respectively 127 (Needoba 1997; Waser et al. 1998a). The coccolithophore, Emiliania huxleyi had a N fractionation of 3.9±0.3%o (Needoba 1997). Dinoflagellates such as Prorocentrum minimum had e(N03') of 2.4±0.3%0 (Needoba 1997). Only Heterosigma carterae grown on N0 3 ' whose best fit e(N03") was 5%o, showed a good agreement with these values. On the other hand, s(N03") for Heterosigma carterae grown on three N sources appeared to be higher than 9%o. A high e(N03") was reported by Montoya and McCarthy (1995). These authors reported that diatoms such as Thalassiosira weissflogii and Skeletonema costatum had an e of 9.7 -15.4 and 6.4 -11.9%o respectively for a continuous culture experiment. Interestingly, 9.0%o for Skeletonema costatum reported by Pennock (1996) was very close to the e(N03") for Heterosigma carterae grown on three N sources in the present study. The discrepancy may be due to some process such as recycling of DON or NH 4+ (Rees & Syrett 1979; Price & Harrison 1988; Pujo-Pay et al. 1997). This process was not accounted for the model and may produce discrepancies. At the moment, it is impossible to explain the reason of the discrepancy. The difference in isotope fractionation between dinoflagellates and flagellates on the one hand (i.e. low fractionation) (Montoya & McCarthy 1995) and coccolithophores and diatoms on the other hand (e.g. high fractionation) (Montoya & McCarthy 1995; Pennock et al. 1996) was not seen in this study. It has been hypothesized that the ability of dinoflagellates and flagellates to swim, decreases the zone of diffusion around the cell and results in the reduction of the potential of the fractionating step during the uptake of N03" across the outer cell membrane (Montoya & McCarthy 1995). This process could explain the low 128 s value observed. The silica shell of diatoms is considered to provide a barrier to NOV uptake and results in higher discrimination of isotopes (Montoya & McCarthy 1995). These morphological effects on isotope fractionation remain unclear. E(NH 4 + ) Growth on N H 4 + in N-replete conditions was accompanied by very large values of e(NH4 +) ranging from 12 to 24%o. e(NH4+) for Heterosigma carterae (18 - 22%o) and Amphidinium carterae (20%o) were in good agreement with previously reported values such as 25 - 26%o for Skeletonema costatum (Pennock et al. 1996) and 18 - 20%o for Thalassiosira pseudonana (Waser et al. 1998b). However, Prorocentrum micans had a very low e(NH4+) of 12%o. This value is close to the estimates (6.5 - 9%o) of phytoplankton grown on N H 4 + in coastal environments (Cifuentes etal. 1989; Montoya etal. 1991). Because of a very low e(NH4+) and a linear uptake of N H 4 + and urea in N-replete conditions, it is suspected that the cell might have experienced N starvation at the beginning of the experiment. An e(NH4+) of 12%o is very close to the values of N-starved Emiliania huxleyi and a diatom assemblage (Waser et al. unpublished results). However, NOV concentration of the stock culture was 5.25 uM, thus Prorocentrum micans was not N-starved. One clue of this low s(NH4 +) was that cells had a high PN. The calculated PN was 11.1 ug-at N L' 1 . This was the highest among other species in the present study. PNj of Amphidinium carterae (grown on three N sources), Heterosigma carterae (grown on three N sources), 129 and Heterosigma carterae (grown on N0 3 ') were 4.7, 3.9, and 5.1 ug-at N L"1 respectively. If the true PNj is higher than the calculated PNj, some of the discrepancy is explained. The PN of the first few points of Prorocentrum micans was 29.6 at 9 h and 45.2 ug-at N L' 1 at 17 h. The calculated PNjS from these two measured PNs were 14.7 and 16.4 ug-at N L' 1 respectively. These numbers are higher than 11.1 ug-at N L"1 of PNj that was used for the model and the true PNj is suspected higher than the calculated value. Another possibility for low e(NH4+), again, may be the excretion of N H 4 + (Rees & Syret 1979; Price & Harrison 1988). The hypothesis that mobile cells may have a lower e value compared to non-mobile cells (Montoya & McCarthy 1995) could be another possibility for the low e(NH4+). This species has the ability to swim as fast as 0.85 m h"1 (Edler & Olson 1985). In contrast, it is known that high e(NH4+) values are found in chain-forming diatoms (Pennock et al. 1996; Waser et al. 1998a). Chaetoceros debilis and Skeletonema costatum had an e(NH4+) of 25 and 26%o respectively. These chain-forming diatoms discriminated against 1 5 N during N H 4 + incorporation than mobile cells. The mechanism of isotope fractionation for N H 4 + and N0 3 " incorporation is still unclear, and so is the different e values obtained (Waser et al. 1998a). Culture conditions such as N H 4 + concentration, pH, and bubbling may affect e values of N H 4 + relative to N0 3" because of efflux and influx of N H 3 (Hoch et al. 1992) and the loss of gaseous NH 3 from the culture medium at high pH (Waser etal. 1998b). 130 Effect of N deprivation Physiological state of dinoflagellates and raphidophytes affected the 815NpN. After the re-addition of N sources, the positive values of 8 1 5 N P N were observed in the cultures of Amphidinium carterae grown on three N sources and Heterosigma carterae grown on N0 3". This phenomenon was observed in the study of coastal diatoms and the coccolithophore (Waser et al. 1998a). A positive number indicates that 1 5 N was incorporated more quickly than 1 4 N by the N-starved cells. One of the reasons for this inverse discrimination is the excretion of a 15N-depleted compound by the N-starved cells. An example of such an excreted compound is NH 3 which is excreted by cells when it is incorporated into urea (Price & Harrison 1988). Amino acids can become enriched in 1 5 N for a while when cells accumulate 15N-replete amino acids before the cells excrete some of these amino acids into the medium. Accumulating compounds indicate that the enzyme to aid the transformation to another compound may be rate limiting. Usually, intrinsic isotope fractionation ( P ) for the enzyme activity results in 15N-replete substrate and 15N-deplete products. However, the e value for membrane diffusion of NH 3 is quite large (39%o) (Hermes et al. 1985), and results in excretion of 15N-deplete NH 3 . In fact, the NH 4* concentration did not increase initially in Amphidinium carterae culture. Therefore, it is difficult to explain the inverse isotope discrimination by efflux of N H 3 for this species. Another possibility is the excretion of isotopically light amino acids as explained above. It is not possible to determine what 15N-deplete compound was excreted in this study. 131 The e (NH/) of Prorocentrum micans grown on three N sources and the E(N03") of Heterosigma carterae grown on N0 3" increased after the cells became N-starved. Prorocentrum micans and Heterosigma carterae changed their best fit e values from 12 to 20%o and from 5 to 9%o respectively. One possible explanation for an increased e(NH/) has to do with a change in the transport system (active transport versus passive diffusion) during N starvation. The increase in e(N0 3 ) may be for a different reason than for e(NH/). Perhaps, the efflux of N0 3 " increased after N starvation due to a slowed assimilation, which allowed for more 15N-replete N0 3" to efflux and resulted in high e values. Since Heterosigma carterae grown on N0 3" in the N-replete phase had much lower 5 1 5 N of PNj (1.08%o) than the expected values of 3.04%o, the increased e(N03') value in the N-deplete phase was of a concern. It is not clear at present why there is such a discrepancy between the measured 8 1 5N and the fit by the model. It is possible that DON may have been released into the medium (Rees & Syrett 1979; Price & Harrison 1988; Pujo-pay 1997). The PN concentration (90.6 ug-at N L"1) at steady-state was in good agreement with the initial N0 3 " concentration (78.5 ug-at N L'1) suggesting that the release of DON, if it occurred, was relatively small. Nonetheless, it is interesting that N-deprivation enhanced isotope fractionation and incorporated more 1 4 N than 1 5 N , which was the opposite of what was expected. Again, the hypothesis that mobile cells have lower e values (Montoya & McCarthy 1995) could be one possibility of higher e(NH/) of N-starved Prorocentrum micans and e(N03") of Heterosigma carterae. 132 During N starvation, cells tended to sink and stay at the bottom of the flask even though the flask was swirled by hand and after the re-addition of N sources, cells gradually dispersed in the medium again. The temporary loss of ability to move might result in the higher e values for these species. On the other hand, Amphidinium carterae decreased its e(NH4+) from 18 - 20%o to < 4%o. Previously, no change or decreased e values were reported to be due to N deprivation in the study of the open ocean clone of Emiliania huxleyi and the coastal clone of the same species grown on three N sources respectively (Waser et al. 1998a). Diatoms grown on three N sources also showed decreased e values after N deprivation (Waser et al. 1998a). All species changed the mechanism of isotope fractionation during N deprivation. This may be due to the rate limiting transport across the membrane and diffusion to the cell that occurred during N deprivation (Wada & Hattori 1978; Mariotti etal. 1982; Handily & Raven 1992; Evens etal. 1996). Otherwise, the rate-limiting step was the reduction of NO3" to NO2' by nitrate reductase which was affected by N deprivation. At this time, it is impossible to tell which step, either at the cell membrane or at the nitrate reductase level, is responsible for the isotope fractionation. If the cell did not change the mechanism of isotope fractionation, it indicates that the cell has recovered quickly from the impact of N deprivation. Such a species is able to live in the oligotrophic ocean because it takes up episodically supplied N H 4 + (Dugdale 1967) with the same mechanism that occurs under N-replete conditions. 133 Difficulties in predicting 8 1 5 N P N occurred during simultaneous uptake of two or three N sources. Since the purpose of this type of laboratory experiment involves the application for understanding N isotope fractionation in the ocean, interaction of each N is important for interpreting the 8 1 5 N P N in the ocean. For example, if NH 3 is released and recycled during the simultaneous uptake of N H 4+ and urea, 8 1 5 N P N was much lower than 81 5 N P N grown on only NH 4+ . Therefore, not only e for each N source, but also the interaction of N sources needs to be considered in the model fit. N availability affected N isotope fractionation. It is interesting to compare the results of acclimation to the N-deplete condition obtained in Part. I of this study. Prorocentrum micans and Amphidinium carterae showed a response to N depletion by enhancing their ability to take up the reduced N form and episodically supplied urea (Dugdale 1967) rather than N0 3 ' , although they decreased their ability to take up NH 4*. However, none of species were able to keep the same isotope fractionation mechanism during the N deprivation. These observations indicate that marine phytoplankton have different methods and magnitudes of adaptation to N-deplete conditions. 134 C O N C L U S I O N S N availability had a great impact on 815N incorporation by marine phytoplankton. e (isotope fractionation) changed before and after N deprivation and following N re-supply. e(N0 3 ) of Heterosigma carterae grown on N0 3 " increased during N starvation probably due to the increased efflux of 1 5 N -deplete D O N . For cultures grown on three N sources, E (NH 4 + ) decreased in Amphidinium carterae and Heterosigma carterae and increased in Prorocentrum micans. The magnitudes of isotope fractionation were species specific. The multiple N source uptake model predicted 6 1 5N P N well. 135 FUTURE STUDY Species specificity on N isotope fractionation to the N availability needs further investigation since each species showed a different response to N deprivation. Differences in morphology of marine phytoplankton may influence the isotope fractionation. Therefore, the correlation of morphological features such as chain-formation, and motility of flagellates and low fractionation, and the existence of silica frustules of diatoms and high fractionation, needs further analysis as well as the physiological state. Simultaneous uptake of two N sources have a different impact on the 8 1 5N of the PN depending on the N forms and the species, which is critical for interpreting the 8 1 5N of the PN in the field. Thus, understanding of interactions between different N sources is required. 136 REFERENCES Altabet, M. A. and Francois, R. 1994. Sedimentary nitrogen isotopic ratio as a recorder for surface ocean N 0 3 ' utilization. Global Biogeochem. Cycles 8: 103-116. Amy, N. K. and Garrett, R. H. 1974. Purification and characterization of the nitrate reductase from the diatom Thalassiosira pseudonana. Plant Physiol. 54: 629-637 Antia, N. J., Harrison, P. J., and Oliveira, L. 1991. The role of dissolved organic nitrogen in phytoplankton nutrition, cell biology and ecology. Phycologia 29: 1 -89 Avaria, P. S. 1982. Fenomenos de marea roja en el mar de CI. Cienc. Tecnol. Mar. (Valparaiso ) 6: 117-127. Bekheet, I. A. and Syrett, P. J. 1979. The uptake of urea by Chlorella. New Phytol. 82: 179-186. Berman, T. 1974. Urea in the waters of Lake Kinneret (Sea of Galilee). Limnol. Oceanogr. 19: 977-980. Bhovichitra, M. and Swift. E. 1977. Light and dark uptake of nitrate and ammonium by large oceanic dinoflagellates: Pyrocystis noctiluca, Pyrocystis fusiformis, and Dissodinium lunula. Limnol. Oceanogr. 22: 73-83. Buijsman, E., Jonker, P. J., Asman, W. A. H., and Ridder, T. B. 1991. Chemical composition of precipitation collected on a weather ship on the North Atlantic. Atmos. Environ. 25: 873-883. Calvert, S. E., Nielsen, B., and Fontugne, M. R. 1992. Evidence from nitrogen isotope ratios for enhanced productivity during formation of eastern Mediterranean sapropels. Nature 359: 223-225. Caperon, J. and Ziemann, D. A. 1976. Synergistic effects of nitrate and ammonium ion on the growth and uptake kinetics of Monochrysis lutheri in continuous culture. Mar. Biol. 36: 73-84. Cassie, V. 1981. Non-toxic blooms of Prorocentrum micans (Dinophyceae) in the Karamea Bight. N. Z. J. Mar. Fresh Res. 15: 181-184. 137 Chang, F. H., Pridmore, R., and Boustead, N. 1993. Occuurence and distribution of Heterosigma cf. Akashiwo (Raphidophyceae) in a 1989 bloom in Big Blory Bay, New Zealand. In: Toxic phytoplankton blooms in the sea (eds.) T. Smayda, and Y. Shimizu. Proceedings of the Fifth International Conference on Toxic Marine Phytoplankton, Newport, Rhode Island, October, 1991, Elsevier, p. 675-680. Cifuentes, L A., Fogel, M. L , Pennock, J. R., and Sharp, J. H. 1989. Biogeochemical factors that influence the stable nitrogen isotope ratio of dissolved ammonium in the Delaware Estuary. Geochim Cosmochim Acta. 53: 2713-2721. Cleveland, J. S., Perry, M. J., Kiefer, D. A., and Talbot, M. C. 1989 Maximal quantum yield of photosynthesis in the northwestern Sargasso sea. J. Mar. Res. 47: 869-86. Cochlan, W. P. and Harrison, P. J. 1991. Uptake of nitrate, ammonium, and urea by nitrogen-starved cultures of Micromonas pusilla (Prasinophyceae): transient responses. J. Phycol. 27: 673-679 Cochlan, W. P., Harrison, P. J., and Denman, K. L. 1991. Diel periodicity of nitrogen uptake by marine phytoplankton in nitrate-rich environments. Limnol. Oceanogr. 36: 1689-1700 Collos, Y. 1980. Transient situations in nitrate assimilation by marine diatoms. 1. Changes in uptake parameters during nitrogen starvation. Limnol. Oceanogr. 25: 1075-1081. Collos, Y. 1983. Transient situations in nitrate assimilation by marine diatoms. 4. Non-linear phenomena and the estimation of the maximum uptake rate. J. Plankton Res. 5: 677-691. Collos, Y., Maestrini, S. Y., and Robert, J.-M. 1989. High long-term nitrate uptake by oyster-pond microalgae in the presence of high ammonium concentrations. Limnol. Oceanogr. 34: 957-964. Conover, S. A. M. 1975. Nitrogen utilization during spring blooms of marine phytoplankton in Bedford Basin, Nova Scotia, Canada. Mar. Biol. 32: 247-261. Conway, H. L. 1977. Interactions of inorganic nitrogen in the uptake and assimilation by marine phytoplankton. Mar. Biol. 39: 221-232. 138 Conway, H. L , Harrison, P. J., and Davis, C. O. 1976. Marine diatoms grown in chemostats under silicate or ammonium limitation. II. Transient response of Skeletonema costatum to a single addition of a limiting nutrient. Mar. Biol. 35: 187-199. Conway, H. L. and Harrison, P. J. 1977. Marine diatoms grown in chemostats under silicate or ammonium limitation. IV. Transient response of Chaetoceros debilis, Skeletonema costatum, and Thalassiosira gravida to a single addition of the limiting nutrient. Mar. Biol. 43: 33-43. Conway, H. L. and Whitledge, T. E. R. 1979. Distribution of fluxes and biological utilization of inorganic nitrogen during a spring bloom in the New York Bight. J. Mar. Res. 37: 657-668. Costas, E. 1990. Genetic variability in growth rates of marine dinoflagellates. Genetica 82: 99-102 Cresswell, R. C. and Syrett, P. J. 1979. Ammonium inhibition of nitrate uptake by the diatom, Phaeodactylum thcornutum. Plant Sci. Lett. 14: 321-325. Cullen, J. J., Stewart, E., Renger, E., Eppley, R. W., and Winant, C. D. 1983. Vertical motion of the thermocline, nitracline and chlorophyll maximum layers in relation to currents on the Southern California Shelf. J. Mar. Res. 41: 239-262. D'elia, C. F. and DeBoer, J. A. 1978. Nutritional studies of two red algae II. Kinetics of ammonium and NOV uptake. J. Phycol. 14: 266-272. Demanche, J. M. and Curl, H. C. 1979. The rapid response of the marine diatom Skeletonema costatum to changes in external and internal nutrient concentration. Mar. Biol. 53: 323-333. Dortch, Q. and Maske, H. 1982. Dark uptake of nitrate and nitrate reductase activity of a red tide population off Peru. Mar. Ecol. Prog. Ser. 9: 299 -303. Dortch, Q., Clayton, J. R. Jr., Thoresen, S. S., Bressler, S. L , and Ahmed, S. I. 1982. Response of marine phytoplankton to nitrogen deficiency: decreased nitrate uptake vs enchanced ammonium uptake. Mar. Biol. 70: 13-19. 139 Dortch, Q., Clayton J. R. Jr., Thoresen, S. S., and Ahmed, S. I. 1984. Species differences in accumulation of nitrogen pools in phytoplankton. Mar. Biol. 81: 237-250. Dugdale, R. C. 1967. Nutrient limitation in the sea: dynamics, identification, and significance. Limnol. Oceanogr. 12: 685-695. Dugdale, R. C. and Goering, J. J. 1967. Uptake of new and regenerated forms of nitrogen in primary productivity. Limnol. Oceanogr. 12: 196-206. Edler, L. and Olsson, P. 1985. Observations on diel migration of Ceratium furca and Prorocentrum micans in a stratified bay in the Swedish west coast. In: Toxic Dinoflagellates (eds.) D. M. Anderson, A. W. White, and D. G. Baden, Elsevier, New York. p. 195-200. Eppley, R. W. and Coatsworth, J. L. 1968. Uptake of nitrate and nitrite by Ditylum brightwellii - kinetics and mechanisms. J. Phycol. 4: 151-156. Eppley, R. W., Coatsworth, J. L., and Solorzano, L. 1969. Studies of nitrate reductase in marine phytoplankton. Limnol. Oceanogr. 14: 194-205. Eppley, R. W. and Rogers, J. N. 1970. Inorganic nitrogen assimilation of Ditylum brightwellii, a marine plankton diatom. J. Phycol. 6: 344-351. Eppley, R. W., Rogers, J. N., McCarthy, J. J., and Sournia, A. 1971. Light/dark periodicity in nitrogen assimilation of the marine phytoplankters Skeletonema costatum and Coccolithus huxleyi in N-limited chemostat cultures. J. Phycol. 7: 150-154. Eppley, R. W., Renger, E. H., Venrick, E. L , and Mullin, M. M. 1973. A study of plankton dynamics and nutrient cycling in the central gyre of the North Pacific Ocean. Limnol. Oceanogr. 18: 534-551. Eppley, R. W. and Harrison, W. G. 1975. Physiological ecology Gonyaulax polyedra, a red water dinoflagellate of Southern California. In: Proceedings of the First International Conference on Toxic Dinoflagellate Blooms, (ed.) V. R. LoCicero. Massachusetts Science & Technology Foundation, Wakefield, Massachusetts, p. 11-22. Eppley, R. W., Rogers, E. H., and Harrison, W. G. 1979. Nitrate and phytoplankton production in southern California coastal waters. Limnol. Oceanogr. 24: 483-494. 140 Evans, R. D., Bloom, A. J., Sukrapanna, S. S., Ehleinger, J. R. 1996. Nitrogen isotope composition of tomato (Lycopersion esculentum Mill. Cv. T-5) grown under ammonium or nitrate nutrition. Plant Cell Environ. 19: 1317-1323. Falkowski, P. G. 1975. Nitrate uptake in marine phytoplankton: (nitrate, chloride) - activated adenosine triphosphatase from Skeletonema costatum (Bacillariophyceae). J. Phycol. 11: 323-326. Falkowski, P. G. and Rivkin, R. B. 1976. The role of glutamine synthetase in the incorporation of ammonium in Skeletonema costatum (Bacillariophyceae). J. Phycol. 12: 448-450. Farrell, J. W., Pedersen, T. F., Calvert, S. E., and Nielsen, B. 1995. Sedimentary 1 5 N / 1 4 N ratios reveal nutrient utilization history in the equatorial Pacific. Nature 377: 514-517. Fedorov, V. D., and H'yash, V. 1992. Adaptive mechanisms in life-history strategies of microalgae. Hydrobiol. J. 28: 38-45. Fisher, T. R., Carlson, P. R., and Barber, R. T. 1982. Carbon and nitrogen primary productivity in three North Carolina estuaries. Est. Coastal Shelf Sci. 15: 621-644. Fitzgerald, G. P. 1968. Detection of limiting or surplus nitrogen in algae and aquatic weeds. J. Phycol. 4: 121-126. Florencio, F. J. and Vega, J. M. 1983. Utilization of nitrate, nitrite and ammonium by Chlamydomonas reinhardii. Planta 158: 288-293. Francois, R., Altabet, A., and Burckle, L. H. 1992. Glacial to interglacial changes in surface nitrate utilization in the Indian sector of the Southern Ocean as recorded by sediment 8 1 5N. Paleoceanography 7: 589-606. French, D. P., Smayda, T. J., Lassus, P., Arzul, G., Erard-Le Denn, E., Gentien, P., and Marcailou-Le Baut, C. 1995. Temperature regulated responses of nitrogen limited Heterosigma akashiwo, with relevance to its blooms. Harmful marine algal blooms. In: Proliferation d'algues marines nuisibles. p. 585-590. Glivert, P. M. and Goldman, J. C. 1981. Rapid ammonium uptake by marine phytoplankton. Mar. Biol. Lett. 2: 25-31. 141 Goericke, R., Montoya, J. P., and Fry, B. 1994. Physiology of isotope fractionation in algae and cyanobactera. In: Stable isotope in ecology and environmental science, (eds.) K. Lajtha and R. H. Michener. Blackwell Scientific Publications, p. 187-221. Goldman, J. C , McCarthy, J. J., and Peavey, D. G. 1979. Growth rate influence on the chemical composition of phytoplankton in oceanic waters. Nature (Lond.) 279: 210-215. Goldman, J. C , Taylor, C. D., and Glibert, P. M. 1981. Non-linear time-course uptake of carbon and ammonium by marine phytoplankton. Mar. Ecol. Prog. Ser. 6: 137-148. Goldman, J. C. and Glibert, P. M. 1982. Comparative rapid ammonium uptake by four species of marine phytoplankton. Limnol. Oceanogr. 27: 814-827. Grant, B. R., Madgwick, J., and Dalpont, G. 1967. Growth of Cylindrotheca closterium var. californica (Mereschk) Rieman & Lewin on nitrate, ammonia, and urea. Aust. J. Mar. Freshwater Res. 18: 129-135. Guerrero, M. G., Vega, J. M., and Losada, M. 1981. The assimilatory nitrate-reducing system and its regulation. Ann. Rev. Plant Physiol. 32: 169-204. Hallegraeff, G. M. 1991. Aquaculturists' guide to harmful Australian microalgae. CSIRO, Hobart. Halstead, B. W. 1965. Poisonous and Venomous Marine Animals of the World, vol. 1 U. S. Govt. Printing Office, Washington, D. C , p. 159. Handily, L. L. and Raven, J. A. 1992. The use of natural abundance of nitrogen isotopes in plant physiology and ecology. Plant Cell Enrivon. 15: 965-985. Hara, T. 1990. Heterosigma akashiwo (HAD A) HAD A. In: Nihon no akashiwo seibutsu. (eds.) Y. Fukuyo, H. Takano, M. Chihara, and K. Matsuoka. Uchida Rokkakueho, Tokyo, p. 346-347. Harrison, P. J., Waters, R. E., and Taylor, F. J. R. 1980. A broad spectrum artificial seawater medium for coastal and open ocean phytoplankton. J. Phycol. 16: 28-35. Harrison, P. J. and Turpin D. H. 1982. The manipulation of physical, chemical and biological factors to select species from natural phytoplankton 142 communities. In: Marine Mesocosms: Biological and chemical research in experimental ecosystems, (eds.) G. D. Grice and M. R. Reeve. Springer-Verlag, New York. p. 275-289. Harrison, P. J., Druehl, L. D., Lloyd, K. E., and Thompson, P. A. 1986. Nitrogen uptake kinetics in three-year classes of Laminaria groenlandica (Laminariales: Phaeophyta). Mar. Biol. 93: 29-35. Harrison, P. J., Parslow, J. S., and Conway, H. L. 1989. Determination of nutrient uptake kinetic parameters: a comparison of methods. Mar. Ecol. Prog. Ser. 52: 301-312. Harrison, W. G. 1976. Nitrate metabolism of the red tide dinoflagellate Gonyaulax polyedra Stein. J. Exp. Mar. Biol. Ecol. 21: 199-209. Harrison, W. G. 1983. The time-course of uptake of inorganic and organic nitrogen compounds by phytoplankton from the eastern Canadian Arctic: A comparison with temperate and tropical populations. Limnol. Oceanogr. 28: 1231-1236. Harrison, W. G., Piatt, T., and Lewis, M. R. 1987. F-Ratio and its relationship to ambient nitrate concentration in coastal waters. J. Plankton Res. 9: 235-248. Harrison, W. G. 1992. Regeneration of nutrients. In: Primary productivity and biogeochemical cycles in the sea. (eds.) P. G. Falkowski, and A. D. Woodhead. Plenum Press, New York. p. 385-407. Harvey, H. W. 1953. Synthesis of organic nitrogen and chlorophyll by Nitzschia closterium. J. Mar. Biol. Ass. U. K. 31: 477-487. Harvey, G. W. 1966. Microlayer collection from the sea surface by new method and initial results. Limnol. Oceanogr. 11: 608-612. Hata, M., Abe, S., and Hata, M. 1982. Occurrence of peridinin-chlorophyll a-protein complex in red tide dinoflagellate Prorocentrum micans Bull. Jap. Soc. Scient. Fish. 48: 459-461. Hattori, A. 1958. Studies on the metabolism of urea and other nitrogenous compounds in Chlorelfa ellipsoidea. II. Changes in levels of amino acids and amides during the assimilation of ammonia and urea by nitrogen-starved cells. J. Biochem. (Tokyo) 45: 57-64. 143 Hattori, H., Yuki, K., Zaitsev, Y. P., and Motoda, S. 1983. A prelimary observation on the neuston in Suruga Bay. La Mer (B Soc. Franco-jap. Oceanogr., Tokyo) 21: 11 -20. Hersey, R. L. and Swift, E. 1976. Nitrate reductase activity of Amphidinium carteh and Cachnina hei (Dinophyceae) in batch culture: Diel periodicity and effects of light intensity and ammonia. J. Phycol. 12: 36-44. Hedrich, R. and Schroeder, J. I. 1989. The physiology of ion channels and electrogenic pumps in higher plants. Annu. Rev. Plant Physiol. 40: 539- 569. Herdman, E. C. 1924a. Notes on dinoflagellates and other organisms causing discoloration of the sand at Port Erin. III. Liverpool Biol. Soc. Proc. Trans. 38: 58-63. Herdman, E. C. 1924b. Notes on dinoflagellates and other organisms causing discoloration of the sand at Port Erin. IV. Liverpool Biol. Soc. Proc. Trans. 38: 75-84. Hermes, J. D., Weiss, P. M., and Cleland, W. W. 1985. Use of nitrogen-15 and deuterium isotope effects to determine the chemical mechanism of phenylalanine ammonia-lyase. Biochemistry 24: 2959-2967. Hipkin, C. R., Al-Bassam, B. A., and Syrett, P. J. 1980. The roles of nitrate and ammonium in the regulation of the development of nitrate reductase in Chlamydomonas reinhardii. Planta 150: 13-18. Hoch, M. P., Foge, M. L., and Kirchman, D. L. 1994. Isotope fractionation associated with ammonium uptake by a marine bacterium. Limnol. Oceanogr. 37: 1447-1459. Hoe Chang, F. 1993. Occurrence and distribution of Heterosigma akashiwo (Raphidophyceae) in a 1989 bloom in Big Glory Bay, New Zealand. In: Toxic Phytoplankton Blooms in the Sea. (eds.) T. J. Smayda and Y. Shimizu. Elsevier, Amsterdam, p. 675-679. Hoe Chang, F and Page, M. 1997. Influence of light and three nitrogen sources on growth of Heterosigma carterae (Raphidophyceae). N. Z. J. Mar. Freshwat. Res. 29: 299-304. Honjo, T. 1992. Harmful red tides of Heterosigma akashiwo. NOAA Tech. Rep. NMFS 111: 27-32 144 Honjo, T. 1993. Overview on bloom dynamics and physiological ecology of Heterosigma akashiwo. In: Toxic Phytoplankton Blooms in the sea. (eds.) T. J. Smayda and Y. Shimizu. Elsevier, New York. p. 33-41. Honjo, T. and Tabata, K. 1985. Growth dynamics of Olisthodiscus luteus in outdoor tanks with flowing coastal water and in small vessels. Limnol. Oceanogr. 30: 653-664. Horiguchi, T. 1990. Prorocentrum micans Ehrenberg. In: Nihon no akashiwo seibutsu. (eds.) Y. Fukuyo, H. Takano, M. Chihara, and K. Matsuoka. Uchida Rokkakueho, Tokyo, p. 28-29. Horrigan, S. G. and McCarthy, J. J. 1.981. Urea uptake by phytoplankton at various stages of nutrient depletion. J. Plankton Res. 3: 403-414. Horrigan, S. G. and McCarthy, J. J. 1982. Phytoplankton uptake of ammonium and urea during growth on oxidized forms of nitrogen. J. Plankton Res. 4: 379-389. Hulbert, E. M. 1957. The taxonomy of unarmored Dinophyceae of shallow embayments on Cape Cod, Massachusetts. Biol. Bull. 112: 196-219. Ignatiades L. 1986. Annual variability of [14C] urea utilization by natural marine phytoplankton. British Phycol. J. 21: 209-215. Ikawa, M. and Sasner, J. J. Jr. 1975. Chemical and physiological studies on the marine dinoflagellate Amphidinium carterae. In: Proceedings of the first international conference on toxic dinoflagellate blooms (ed.) V. R. LoCicero. The Massachusetts Science and Technology Foundation, Wakefield, Massachusetts, p. 323- 332. Jones, M. N. 1984. Nitrate reduction by shaking with cadmium, alternative to cadmium columns. Water Res. 18: 643-646. Khan, S., Arakawa, O. and Onoue, Y. 1997. Neurotoxins in a toxic red tide of Heterosigma akashiwo (Raphidophyceae) in Kagoshima Bay, Japan. Aqua. Res. 28: 9-14. Lanskaya, L. A. 1971. Cultivation of algae. In: Ecological physiology of marine planktonic algae. Naukova Dumka Press, Kiev. p. 5-21. Larsen, J. and Moestrup, O. 1989. Guide to Toxic and Potentially Toxic Marine Algae. Fish Inspection Service, Ministry of Fisheries, Copenhagen. 145 Larsen, J. and Moestrup. 0. 1992. Potentially toxic phytoplankton. 1. Haptophyceae (Prymnesiophyceae). Ices Identif. Leafl. Plankton. 179: 11. Laws, E. and Caperon, J. 1976. Carbon and nitrogen metabolism by Monochrysis lutheri: measurement of growth-rate-dependent respiration rates. Mar. Biol. 36: 85-97. Laws, E. A. and Wong, D. C. L. 1978. Studies of carbon and nitrogen metabolism by three marine phytoplankton species in nitrate-limited continuous culture. J. Phycol. 14: 406-416. Lee, R. B. 1980. Sources of reductant for nitrate assimilation in non-photosynthetic tissue: a review. Plant Cell Environ. 3: 65-90. Lehman, J. T. and Scavia, D. 1982. Microscale patchiness of nutrients in plankton communities. Science 216: 729-730. Lesser, M. P. 1996. Acclimation of phytoplankton to UV-B radiation: oxidative stress and photoinhibition of phytosynthesis are not prevented by UV-absorbing compounds in the dinoflagellate Prorocentrum micans. Mar. Ecol. Prog. Ser. 132: 287-297. Levasseur, M. E., Harrison, P. J., Heimdal, B. R., and Therriault, J-C. 1990. Simultaneous nitrogen and silicate deficiency of a phytoplankton community in a coastal jet-front. Mar. Biol. (Berl.) 104: 329-338. L'Helguen, S., Made, C , and Le Corre, P. 1996. Nitrogen uptake in permanently well-mixed temperate coastal waters. Estuar. Coast. Shelf Sci. 42: 803-818. Liebig, J. 1846. Chemistry in its application to agriculture and physiology. 4th ed. Taylor and Walter, London. Liu, M. S. and Hellebust, J. A. 1976. Effects of salinity and osmolarity of the medium on amino acid metabolism in Cyclotella cryptica. Can. J. Bot. 54: 938-948. Lobban, C. S., Harrison, P. J., and Duncan, M. J. 1985. The physiological ecology of seaweeds. Cambridge University Press, Cambridge, England, p. 242. Lobban, C. S. and Harrison, P. J. 1994. Seaweed ecology and physiology. Cambridge University Press. New York, New York. 146 Lui, N. S. T. and Roels, 0. A. 1970. Nitrogen metabolism of aquatic organisms. I. The assimilation and formation of urea in Ochromonas malhamensis. Arch. Biochem. Biophys. 139: 269-277. Lund, B. A. 1987. Mutual interference of ammonium, nitrate, and urea on uptake of 15N sources by the marine diatom Skeletonema costatum (Grev.) Cleve. J. Exp. Mar. Biol. Ecol. 113: 167-180. Maclsaac, J. J. 1978. Diel cycles of inorganic nitrogen uptake in a natural phytoplankton population dominated by Gonyaulax polyedra. Limnol. Oceanogr. 23: 1-9. Maclsaac, J. J. and Dugdale, R. C. 1972. Interactions of light and inorganic nitrogen in controlling nitrogen uptake in the sea. Deep-Sea Res. 19: 209-232. Maguer, J. F., Le Corre, P., and L'Helguen, S. 1995. Nitrogen sources for phytoplankton in shallow well-mixed waters of the western English Channel: Importance of the regenerated production. J. Rech. Oceanogr. 21: 103-108. Malone, T. C , Garside, C , Haines, K. C , and Roels, O. A. 1975. Nitrate uptake and growth of Chaetoceros sp. in large outdoor cultures. Limnol. Oceanogr. 20: 9-19. Mariotti, A., Germon, J. C , Hubert, P., Kaiser, P., Letolle, R., Tardieux, A., and Tarcieux, P. 1981. Experimental determination of nitrogen kinetic isotope fractionation: some principles; illustration for the denitrification and nitrification processes. Plant Soil 62: 413-430. Marriotti, A., Mariotti, F., Champigny, M., Amarger, N., and Moyse, A. 1982. Nitrogen isotope fractionation associated with nitrate reductase acitivity and uptake of N 0 3 by Pearl Millet. Plant Physiol. 69: 880-884. Martin, G. W. 1929. Dinoflagellates from marine and brackish waters of New Jersey. Univ. Iowa Stud. Nat. Hist. 12: 3-32. McCarthy, J. J. 1980. Nitrogen. In : The physiological ecology of phytoplankton. (ed.) I. Morris. Blackwell Sci. Publ., Oxford, England, p. 191-233 McCarthy, J. J. 1981. The kinetics of nutrient utilization, p. 211-233. In: Physiological bases of phytoplankton ecology, (ed.) T. Piatt. Can. Bull. Fish. Aquat. Sci. 210. p. 211-233. 147 McCarthy, J. J. and Eppley, R. W. 1972. A comparison of chemical, isotopic, and enzymatic methods for measuring nitrogen assimilation of marine phytoplankton. Limnol. Oceanogr. 17: 371-382. McCarthy, J. J. and Goldman, J. C. 1979. Nitrogenous nutrition of marine phytoplankton in nutrient depleted waters. Science 203: 670-672. McCarthy, J. J., Taylor, W. R., and Taft, J. L. 1977. Nitrogenous nutrition of the plankton in the Chesapeake Bay. 1. Nutrient availability and phytoplankton preferences. Limnol. Oceanogr. 22: 996-1011. Mitamura, O., Saijo, Y., Hino, K., and Barbosa, F. A. R. 1995. The significance of regenerated nitrogen for phytoplankton productivity in the Rio Doce Valley lakes Brazil. Arch. Hydrobiol. 134: 179-194. Molloy, C. J. and Syrett. P. J. 1988a. Interrelationships between uptake of urea and uptake of ammonium by microalgae. J. Exp. Mar. Biol. Ecol. 118: 85-95. Molloy, C. J. and Syrett. P. J. 1988b. Effect of light and N deprivation on inhibition of nitrate uptake by urea in microalgae. J. Exp. Mar. Biol. Ecol. 118: 97-101. Montaya, J. P. and McCarthy, J. J. 1995. Isotopic fractionation during nitrate uptake by phytoplankton grown in continuous culture. J. Plankton Res. 17: 439-464. Montaya, J. P., Horrigan, S. G., and MacCarthy, J. J. 1991. Rapid, storm-induced changes in the natural abundance of 1 5 N in a planktonic ecosystem. Geochim. Cosmochim. Acta. 55: 3267-3638. Morris, I. and Syrett, P. J. 1963. The development of nitrate reductase in Chlorella and its repression by ammonium. Arch. Mikrobiol. 47: 32-41. Nagasaki, K., Itakura, I., Imai, I., Nakagiri, S., and Yamaguchi, M. 1996. The disintegration process of a Heterosigma akashiwo (Raphidophyceae) red tide in northern Hiroshima Bay, during the summer of 1994. In Proceedings of the 7th International Conference on Toxic Phytoplankton. Nagasaki, K and Yamaguchi, M. 1997. Isolation of a virus infectious to the harmful bloom causing microalga Heterosigma akashiwo (Raphidophyceae). Aquat. Microb. Ecol. 13: 135-140. 148 Needoba, J. 1997. Nitrogen isotope fractionation by four groups of phytoplankton grown on nitrate. BSc dissertation, University of British Columbia, Vancouver. Ohmori, M., Ohmori, K., and Strotmann, H. 1977. Inhibition of nitrate uptake in a blue-green alga, Anabaena cylindrica. Arch. Microbiol. 114: 225-229. Paasche, E. 1971. Effect of ammonia and nitrate on growth, photosynthesis, and ribulosediphosphate carboxylase content of Dunaliella tertiolecta. Physiol. Plant. 24: 294-299. Paasche, E. and Kristiansen, S. 1982. Nitrogen nutrition of the phytoplankton of the Oslofjord. Estuarine Coastal Shelf Sci. 14: 237-249. Paasche, E., Bryceson, I., and Tange, K. 1984. Interspecific variation in dark nitrogen uptake by dinoflagellates. J. Phycol. 20: 394-401. Pachard, T. T. and Blasco, D. 1974. Nitrate reductase activity in upwelling regions. 2. Ammonia and light dependence. Tethys 6: 269-280. Park, J. S. 1989. Studies on red tide phenomena in Korean coastal waters. In: Red Tides (eds.) T. Okaichi, D. M. Anderson, and T. Nemoto. Elsevier, New York. p. 37-40. Parslow, J. S., Harrison, P. J., and Thompson, P. A. 1984a. Development of rapid ammonium uptake during starvation of batch and chemostat cultures of the marine diatom Thalassiosira pseudonana. Mar. Biol. 83: 43-50. Parslow, J. S., Harrison, P. J., and Thompson, P. A. 1984b. Saturated uptake kinetics: transient response of the marine diatom Thalassiosira psseudonana to ammonium, nitrate, silicate or phosphate starvation. Mar. Biol. 83: 51-59. Parsons, T. R., Stronach, J., Borstad, G. A., Louttit, G., and Perry, R. I. 1981. Biological fronts in the Strait of Georgia, British Columbia and their relation to recent measurements of primary productivity. Mar. Ecol. Prog. Ser. 6: 237-242. Parsons, T. R., and Harrison, P. J. 1983. Nutrient cycling in marine ecosystems. In: physiological plant ecology IV. Ecosystem Processes : Mineral Cycling, Productivity and Pollution, (eds.) B. Richards and J. L. Charley. Springer-Verlag, Heidelberg, p. 85-115. 149 Pennock, J. R., Velinsky, D. J., Ludlam, J. M., and Sharp, J. H. 1996. Isotopic fractionation of ammonium and nitrate during uptake by Skeletonema costatum: implications for 8 1 5N dynamics under bloom conditions. Limnol. Oceanogr. 41: 451-459. Picard, G. A. 1976. Effects of light and dark cycles on the relationship between nitrate uptake and cell growth rates of Chaetoceros sp. (STX-105) in continuous culture. Ph.D. Thesis, Dept. Biology, City Univ. of New York, N. Y. p. 220 Pingree, R. A., Holligan, P. M., and Mardell, G. T. 1978. The effects of vertical stability on phytoplankton distributions in the summer on the northwest European Shelf. Deep-Sea Res. 25: 1011-1028. Price, N. M., Cochlan, W. P., and Harrison, P. J. 1985. Time course of uptake of inorganic and organic nitrogen by phytoplankton in the Strait of Georgia: comparison of frontal and stratified communities. Mar. Ecol. Prog. Ser. 27: 39-53. Price, N. M. and Harrison P. J. 1987. Comparison of methods for the analysis of dissolved urea in seawater. Mar. Biol. 94: 307-317. Price, N. M. and Harrison, P. J. 1988. Uptake of urea C and urea N by the coastal marine diatom Thalassiosira pseudonana. Limol. Oceanogr. 33: 528-537. Probyn, T. A. and Chapman, A. R. 0. 1982. Nitrogen uptake characteristics of Chordaria flagelliformis (Phaeophyta) in batch mode and continuous mode experiments. Mar. Biol. 71: 129-133. Pujo-Pay, M., Conan, P., and Raimbault, P. 1997. Excretion of dissolved organic nitrogen by phytoplankton assessed by wet oxidation and super(15)N tracer procedures. Mar. Ecol. Prog. Ser. 155: 99-111. Pybus, C. 1984. Unusual concentrations of the dinoflagellate Ceratium tripos in Galway Bay. Ir. Nat. J. 21: 317-320. Raven, J. A. 1986. Physiological consequences of extremely small size for autotrophic organisms in the sea. In: Photosynthetic picoplankton. (eds.) T. Piatt, and W. K. W. Li. Can. Bull. Fish. Aquat. Sci. 214. p. 1-70. 150 Ree, T. A. V. and Syrett, P. J. 1979. Mechanisms for urea uptake by the diatom Phaeodactylum thcornutum: the uptake of thiourea. New Phytol. 83: 37-48. Reed, R. H. 1990. Solute accumulation and osmotic adjustment. In: Biology of Red Algae, (eds.) K. M. Cole and R. G. Sheath. Cambridge University Press, p. 147-170. Rees, T. A. V. and Syrett, P. J. 1979. The uptake of urea by the diatom Phaeodactylum. New Phytol. 82: 169-178. Reid, F. M., Lange, C. B., and White, M. M. 1985. Microplankton species assemblages at the Scripps Pier from March to November 1983 during the 1982 -1984 El Nino event. Botanica Mar. 28: 443- 452. Reisner, G. S., Gering, R. K., and Thompson, J. F. 1960. The metabolism of nitrate and ammonia by Chlorella. Plant Physiol. 35: 48-52. Remsen C. C. 1971. The distribution of urea in coastal and oceanic waters. Limonol. Oceanogr. 16: 732-740. Remsen, C. C , Carpenter, E. J., and Schroeder, B. W. 1972. Competition for urea among estuarine microorganisms. Ecology 53: 921-926. Rhee G-Y. and Lederman, T. C. 1983. Effects of nitrogen sources on P-limited growth of Anabaena flos-aquae. J. Phycol. 19: 179-185. Ricketts, T. R. 1988. Homeostatis in nitrogenous uptake/assimilation by the green alga Platymonas (Tetraselmis) striata (Prasinophyceae). Annals. Bot. 61: 451-458. Robert, J-M. and Maestrini, S. Y. 1986. Absorptions simultanees des ions N 0 3 et N H 4 par trois diatomees de claires a huitres, en culture axenique. Phycologica 25: 152-159. Rosenberg, G., Probyn, T. A., and Mann, K. H. 1984. Nutrient uptake and growth kinetics in brown seaweeds: response to continuous and single additions of ammonium. J. Exp. Mar. Biol. Ecol. 80: 125- 146. Ryther, J. H. and Dunstan, W. M. 1971. Nitrogen, phyosphorus and eutrophication in the coastal marine environment. Science 171: 1008-1013. Sampayo, M. A. de M. 1985. Encystment and excystment of a Portuguese isolate of Amphidinium carterae in cultures. In: Toxic Dinoflagellates. 151 (eds.) D. M. Anderson, A. White, and D. G. Baden. Elsevier, New York. p. 125-130. Satoh, Y. and Hanya, T. 1981. The distribution of urea in Lake Yunoko. Jap. J. Limnol. 42: 254-258. Serra, J. L., Llama, M. J., and Codenas, E. 1978. Nitrate utilization by the diatom Skeletonema costatum. 2. Regulation of nitrate uptake. PI. Physiol. 62: 991-994. Sharp, J. H. 1983. The distributions of inorganic nitrogen and dissolved and particulate organic nitrogen in the sea. In: Nitrogen in the marine environment, (eds.) E. J. Carpenter and D. G. Capone. Academic Press, New York, N.Y. p. 1-35. Smith, F. W. and Thompson, J. F. 1968. Regulation of nitrate reductase in Chlorella. Plant Physiology 43: 58. Slawyk, G. and Maclsaac, J. J. 1972. Comparison of two automated methods in a region of coastal upwelling. Deep-Sea Res. 19: 521-524. Solomonson, L. P. and Barver, M. J. 1990. Assimilatory nitrate reductase: functional properties and regulation. Annu. Rev. Plant Physiol. Plant Molec. Biol. 4: 225-253. Steinmann, J. 1976. Untersuchungen uber den bacteriellen Abbau von Harnstoff and Harnsaure in der Westlichen Ostsee. Botanica Mar. 19: 47-58. Suttle, C. A. and Harrison, P. J. 1988. Rapid ammonium uptake by freshwater phytoplankton. J. Phycol. 24: 13-16. Suttle, C. A., Chan, A. M., and Cottrell, M. T. 1990. Infection of phytoplankton by viruses and reduction of primary productivity. Nature 347: 467-469. Sweeney, B. M. 1975. Red tides I have known. In: Proceedings of the first International Conference on Toxic Dinoflagellate Blooms, (ed.) LoCicero, V. R. Massachusetts Science & Technology Foundation. Wakefield, Massachusetts, p. 225-234. Sweeney, B. M. 1983. Circadian time-keeping in eukaryotic cells, models and hypotheses. Prog. Phycol. Res. 2: 189-225. 152 Syrett, P. J. 1953. The assimilation of ammonia by nitrogen-starved cells of Chlorella vulgaris. I. The correlation of assimilation with respiration. Ann. Bot. 17: 1-19. Syrett, P. J. 1981. Nitrogen metabolism of microalgae. In: Physiological bases of phytoplankton ecology, (ed.) T. Piatt. Can. Bull. Fish. Aquat. Sci. 210. p. 182-210. Syrett, P. J. 1988. Uptake and utilization of nitrogenous compounds. In: Biochemistry of the Algae and Cyanobacteria. (eds.) L. J. Rogers and J. R. Gallon. Oxford: Clarendon Press, p. 23-39. Syrett, P. J. and Morris, I. 1963. The inhibition of nitrate assimilation by ammonium in Chlorella. Biochem. Biophys. Acta. 67: 566-575. Syrett, P. J. and Peplinska, A. M. 1988. Effects of nitrogen-deprivation, and recovery from it, on the metabolism of microalgae. New Phytol. 109: 289-296. Taylor, F. J. R. 1992. The taxonomy of harmful marine phytoplankton, G. Bot. Ital. 126: 209-219. Taylor, F. J. R. and Haigh, R. 1993. The ecology offish-kills blooms of the chloromonad flagellate Heterosigma in the Strait of Georgia and adjacent waters. In: Toxic phytoplankton Blooms in the Sea (ed.) T. J. Smayda and T. Shimizu. Elsevier, Amsterdam, p. 705-710. Thacker, A. and Syrett, P. J. 1972a. Disappearance of nitrate reductase activity from Chlamydomonas reinhardi. New Phytol. 71: 435-441. Thacker, A. and Syrett, P. J. 1972b. Assimilation of nitrate and ammonium by Chlamydomonas reinhardi. New Phytol. 71: 423-434. Thurberg, E. P. and Sasner, J. J. Jr. 1973. Biological activity of a cell extract from the dinoflagellate Amphidinium carteri, Chesapeake Sci. 14: 48-51. Timperley, M. H., Vigor-Brown, R. J., Kawashima, M., and Ishigami, M. 1985. Organic nitrogen compounds in atmospheric precipitation: their chemistry and availability to phytoplankton. Can. J. Fish. Aquat. Sci. 42: 1171-1177. Tischner, R. and Lorenzen, H. 1979. Nitrate uptake and nitrate reduction in synchronous Chlorella. Plant 146: 287-292. 153 Turley, C. M. 1986. Urea uptake by phytoplankton at different fronts and associated stratified and mixed waters on the European Shelf. British Phycol. J. 21: 225-238. Turpin, D. H. and Harrison, P. J. 1978. Fluctuations in free amino acid pools of Gymnodinium simplex (Dinophyceae) in response to ammonia pertubation. Evidence for glutamine synthetase pathway. J. Phycol. 14: 461-464. Turpin, D. H. and Bruce, D. 1990. Regulation of phytosynthetic light harvesting by nitrogen assimilation in the green alga Selenastrum minutum. FEBS 263: 99-103. Thomas, T. E. and Harrison, P. J. 1987. Rapid ammonium uptake and nitrogen interactions in five intertidal seaweeds grown under field conditions. J. Exp. Mar. Biol. Ecol. 107: 1-8. Thompson, P. A., Levasseur, M. E., and Harrison, P. J. 1989. Light-limited growth on ammonium vs nitrate: What is the advantage for marine phytoplankton? Limnol Oceanogr. 34: 1014-1024. Tischner, R., Ward, M. R., and Huffaker, R. C. 1989. Evidence for a plasma-membrane-bound nitrate reductase involved in nitrate uptake of Chlorella sorokiniana. Planta 178: 19-24. Tomas, C. R. 1978. Olisthodiscus luteus (Crysophyceae) I. Effect of salinity and temperature on growth, motility and survival. J. Phycol. 14: 309-313. Tomas, C. R. 1982. Olisthodiscus luteus (Chrysophyceae). V. Its occurrence, abundance and dynamics in Narragansett Bay. Rhode Island. J. Phycol. 16: 157-166. Uchida, T. 1976. Excretion of ammonia by Prorocentrum micans Ehrenberg in urea-grown culture. Jap. J. Ecol. 26: 42-44. Ullrich, W. R. 1987. Nitrate and ammonium uptake in green algae and higher plants: mechanism and relationship with nitrate metabolism. In: Inorganic nitrogen metabolism, (eds.) W. R. Ullrich, P. J. Aparicio, P. J. Syrett, and F. Castillo. Springer-Verlag, Berlin, Heidelberg, New York. p. 32-38. Ullrich, W. R., Larsson, M., Larsson, C-M., Lesch, S., and Novacky, A. 1984. Ammonium uptake in Lemna gibba G1, related membrane potential changes, and inhibition of anion uptake. Physiol. Plant. 61: 369-376. 154 Vernet, M, Neori, A., and Haxo, F. T. 1989. Spectral properties and photosynthetic action in red-tide populations of Prorocentrum micans and Gonyaulax polyedra. Mar. Biol. 103: 365-371 Wada, E. and Hattori, A. 1978. Nitrogen isotope effects in the assimilation of inorganic nitrogenous compounds. Geomicrobiol. J. 1: 85-101. Wada, E. 1980. Nitrogen isotope fractionation and its significance in biogeochemical processes occurring in marine environments. In: Isotope Marine Chemistry, (eds.) E. Goldberg, Y. Horibe, and K. Saruhashi. Uchida Rokakukuho, Tokyo, p. 375-398. Walsh, J. J., Whitledge, T. E., Barvenik, F. W., Wirick, C. D., and Howe, S.O. 1978. Wind events and food chain dynamics within the New York Bight. Limnol. Oceanogr. 23: 659-683. Wang, Z., Zhang, Q., Lu, Y., and Lu, H. 1996. The effects of nutrients, vitamins and trace metals on the growth of the red tide organism Prorocentrum micans. Donghai Mar. Sci. 14: 33-38. Wangersky, P. J. and Guillard, R. L. 1960. Low molecular weight organic base from the dinoflagellate Amphidinium carteri. Nature (London) 185: 689-690. Ward, A. K. and Wetzel, R. G. 1980. Interactions of light and nitrogen source among planktonic blue-green algae. Arch. Hydrobiol. 90: 1-25. Waser, N. A. D., Yin, K., Yu, Z., Tada, K., Harrison, P. J., Turpin, D. H., and Calvert, S. E. 1998a. Nitrogen isotope fractionation associated with growth of marine diatoms and coccolithophores on the combined nitrate, ammonium and urea sources. Mar. Ecol. Prog. Ser. 169: 29-41. Waser, N. A. D. ( Harrison, P. J., Nielsen, B., and Calvert, S. E. 1998b. Nitrogen isotope fractionation during the uptake and assimilation of nitrate, nitrite, ammonium, and urea by a marine diatom. Limnol. Oceanor. 43: 215-224. Watanabe, M., Nakamura, M., and Kohata, K. 1983. Diurnal vertical migration and dark uptake of nitrate and phosphate of the red tide flagellates, Heterosigma akashiwo HADA and Chattonella antiqua (HADA) ONO (Raphidophyceae). Jpn. J. Phycol. 31: 161-166. 155 Wheeler, P. A., Gilbert, P. M., and McCarthy, J. J. 1982. Ammonium uptake and incorporation by Chesapeake Bay phytoplankton: Short-term uptake kinetics. Limnol. Oceanogr. 27: 1113-1128. Wheeler, P. A., Olson, R. J., and Chisholm, S. W. 1983. Effects of photocycles and periodic ammonium supply on three marine phytoplankton species. II. Ammonium uptake and assimilation. J. Phycol. 19: 528-533. Wood, G. J. and Flynn, K. J. 1995. Growth of Heterosigma carterae (Raphidophyceae) on nitrate and ammonium at three photon flux densities: Evidence for N stress in nitrate-growing cells. J. Phycol. 31: 859-867. Yamochi, S. 1983. Mechanisms for outbreak of Heterosigma akashiwo red tide in Osaka Bay, Japan. Part 1. Nutrient factors involved in controlling the growth of Heterosigma akashiwo Hada. J. Oceanogr. Soc. Jap. 39: 310-316. Yamochi, S. and Abe, T. 1984. Mechanisms to initiate a Heterosigma akashiwo red tide in Osaka Bay. Part 2. Diel vertical migration. Mar. Biol. 83: 255-261. Yamochi, S. 1989. Mechanisms for outbreak of Heterosigma akashiwo red tide in Osaka Bay, Japan, In: Red Tides (eds.) T. Okaichi, D. M. Anderson, and T. Nemoto. Elsevier, New York, p 253-256. Yin, K. 1988. Short-term interaction between nitrate and ammonium uptake for cells of a marine diatom grown under different degrees of light limitation. M.Sc. Thesis, University of British Columbia, Vancouver. Yokote, M., Honjo, T., and Asakawa, M. 1985. Histochemical demonstration of a glycocalyx on the cell surface of Heterosigma akashiwo. Mar. Biol. (Berl.)88: 295-299. 156 APPENDIX 1 Equations used to calculate growth rates Growth rates (d"1) were calculated from fluorescence values between the beginning and the end of log phase according to the following equation: H = In (F2/F1) / (t2-t1) where F2 and F1 are the fluorescence values at time 2 (t2) and time 1 (t1), respectively and are reported as the mean (n = 3) ±1 S. D. for triplicate cultures. In vivo fluorescence was measured with a Turner Designs model 10 fluorometer.