Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Storage lipid metabolism during nitrogen assimilation in a marine diatom Larson, Tony Robert 1998

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

Item Metadata

Download

Media
831-ubc_1998-271846.pdf [ 9.73MB ]
Metadata
JSON: 831-1.0088751.json
JSON-LD: 831-1.0088751-ld.json
RDF/XML (Pretty): 831-1.0088751-rdf.xml
RDF/JSON: 831-1.0088751-rdf.json
Turtle: 831-1.0088751-turtle.txt
N-Triples: 831-1.0088751-rdf-ntriples.txt
Original Record: 831-1.0088751-source.json
Full Text
831-1.0088751-fulltext.txt
Citation
831-1.0088751.ris

Full Text

STORAGE LIPID METABOLISM DURING NITROGEN ASSIMILATION IN A MARINE DIATOM by TONY ROBERT LARSON B.Sc. (Botany), University of Auckland, New Zealand, 1990 M.Sc. (Hons.) (Botany), University of Auckland, New Zealand, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Botany) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA March 1998 © Tony Robert Larson, 1998  (  In  presenting  degree freely  at  this  the  available  copying  of  department publication  of  in  University  of  for  this or  thesis  this  British  reference  thesis by  partial  thesis  or for  her  of  y^77/7/1/7  T h e U n i v e r s i t y of British Vancouver, Canada  Date  DE-6  (2/88)  6  AWL  the  Columbia,  I  agree  Columbia  I  further  purposes  gain  shall  requirements that  agree  may  representatives.  financial  permission.  Department  of  study.  scholarly  for  his  and  fulfilment  It not  be is  that  the  Library  permission  granted  by  understood be  for  allowed  an  advanced  shall  that without  it  extensive  for  the  make  head  of  copying my  my or  written  11  ABSTRACT  Four major experiments were conducted to investigate the patterns of storage carbon accumulation and degradation in the marine diatom, Phaeodactylum tricornutum. When the cultures were N-starved in the light for more than 24 h, there was an accumulation of storage triacylglycerols containing primarily hexadecanoic (16:0) and cis-9hexadecenoic (16:1) fatty acids. After cultures that had been N-starved for 3 d were recovered with an addition of nitrate to the medium, carbohydrates were degraded before triacylglycerols, irrespective of the size of the intracellular triacylglycerol reserve. There was no difference in the timing or magnitude of triacylglycerol degradation during the assimilation of either nitrate or ammonium in the light. Triacylglycerol and carbohydrate degradation appeared to be metabolically linked. The activity of the glyoxylate cycle enzyme, isocitrate lyase, was 4-fold higher during nitrate assimilation compared to ammonium assimilation 24 h after adding either N source to N-starved cultures. Triacylglycerol degradation appeared to be targeted towards replacing, via gluconeogenesis, the carbohydrates that were respired during nitrate assimilation. This was confirmed with radiolabeling experiments. In dark N-recovered cultures, there was a net degradation of triacylglycerols and a net synthesis of carbohydrates irrespective of the added N source. During nitrate or ammonium assimilation in the light, some of the triacylglycerol carbon was directed towards polar lipid synthesis. However, during nitrate assimilation, more triacylglycerol carbon was directed towards low molecular weight metabolite and carbohydrate synthesis than during ammonium assimilation. Carbon stable isotope data showed that triacylglycerol carbon accumulation during N-starvation, or loss during recovery with nitrate, had little effect on whole cell 8 C values. However, the triacylglycerol pool had a 13  stable 8 C signature irrespective of the growth phase of the cells. 13  Ul  TABLE OF CONTENTS  ABSTRACT  ii  TABLE OF CONTENTS  iii  LIST OF TABLES  vi  LIST OF FIGURES  viii  LIST OF ABBREVIATIONS  xx  ACKNOWLEDGEMENTS GENERAL INTRODUCTION LIPID AND FATTY ACE) COMPOSITION AND ACCUMULATION IN MICROALGAE REGULATION OF STORAGE LIPID SYNTHESIS REGULATION OF STORAGE LIPID DEGRADATION Ecological roles for storage lipid Pathways for storage lipid degradation THESIS OUTLINE AND OBJECTIVES CHAPTER 1 - CONSTITUENT CHANGES INN-STARVED AND NITRATERECOVERED CULTURES 1.1) INTRODUCTION 1.2) MATERIALS AND METHODS 1.2.1) Culture conditions 1.2.2) Growth measurements 1.2.3) Macromolecular constituents 1.2.4) Carbon and nitrogen analyses.....". 1.2.5) Fatty acid analysis 1.2.6) Medium nitrate 1.3) RESULTS , 1.3.1) Nitrate addition to logarithmic, early stationary, and late stationary phase cultures 1.3.2) Extended monitoring of late stationary phase cultures after nitrate addition 1.4) DISCUSSION 1.4.1) Signature fatty acid pools 1.4.2) Patterns of storage carbon biosynthesis 1.4.3) Patterns of storage carbon degradation CHAPTER 2 - N ASSIMILATION AND STORED CARBON MOBILIZATION DURING RECOVERY OF STATIONARY PHASE CULTURES FROM NSTARVATION IN THE LIGHT OR DARK WITH NITRATE OR AMMONIUM  xxiv 1 3 5 8 8 10 14 16 16 18 18  19 20 22 22  24 25 25 40  28  40 42 45  50  iv 2.1) INTRODUCTION 2.2) MATERIALS AND METHODS 2.2.1) Culture conditions 2.2.2) Growth measurements 2.2.3) Macrornolecular constituents and fatty acids 2.2.4) Nitrogen measurements 2.2.5) Enzyme assays 2.2.5.1) Nitrate reductase 2.2.5.2) Glutamine synthetase 2.2.5.3) Isocitrate lyase 2.3) RESULTS 2.3.1) Nitrate or ammonium addition to 3 d N-starved cultures in continuous light 2.3.2) Nitrate or ammonium addition in darkness to 3 d N-starved cultures followed by a transition to light 2.4) DISCUSSION 2.4.1) Regulation of nitrogen assimilation 2.4.1.1) Nitrate assimilation 2.4.1.2) Ammonium assimilation 2.4.2) Carbon mobilization during nitrogen assimilation 2.4.2.1) Carbon and reductant supply for nitrogen assimilation from carbohydrates and storage lipids 2.4.2.2) Alternative roles for storage lipid carbon  50 53 53 54 54 54 55 55 56 57 57 57 59 93 93 95 97 99 99 102  CHAPTER 3 - REALLOCATION AND LOSS OF CARBON FROM C LABELED TRIACYLGLYCEROLS DURING RECOVERY OF STATIONARY PHASE CULTURES FROM N-STARVATION IN THE LIGHT WITH NITRATE OR AMMONIUM 105 3.1) INTRODUCTION 105 3.2) MATERIALS AND METHODS 108 3.2.1) Culture conditions and biomass measurements 108 3.2.2) C labeling and experimental time-course 108 3.2.3) Medium, whole cell, and intracellular fraction radioactivity measurements 110 3.2.4) Fractionation procedure Ill 3.3) RESULTS 113 3.3.1) Biomass changes during N-starvation and recovery with nitrate or ammonium.... 113 3.3.2) C labeling and reallocation patterns during N-starvation and recovery with nitrate or ammonium 114 3.4) DISCUSSION 131 3.4.1) C labeling and label distribution during N-starvation 131 3.4.2) Carbon flows during recovery from N-starvation 132 3.4.3) A model for degradative TAG carbon flows during recovery from N-starvation.. 134 14  14  14  14  CHAPTER 4 - TRIACYLGLYCEROL SYNTHESIS AND DEGRADATION AS A DETERMINANT OF CARBON STABLE ISOTOPE FRACTIONATION 4.1) INTRODUCTION 4.2) MATERIALS AND METHODS 4.2.1) Culture conditions 4.2.2) Isolation and measurement of carbon and 8 C 1 3  136 136 139 139 139  V  4.3) RESULTS 4.4) DISCUSSION 4.4.1) DIC assimilation and 8 C 4.4.2) Lipid biosynthesis and whole cell 8 C 4.4.3) Comparison of total lipid and TAG 8 C values during lipid synthesis 13  13  13  4.4.4) Lipid degradation and whole cell 8 C values 13  143 148 148 149 151 152  GENERAL CONCLUSIONS  154  ECOLOGICAL IMPLICATIONS  158  FUTURE STUDIES  160  LITERATURE CITED  161  APPENDIX A - CELL DENSITY DETERMINATION  175  APPENDIX B - PROTEIN DETERMINATION  176  APPENDIX C - CARBOHYDRATE DETERMINATION  178  APPENDIX D - CHLOROPHYLL a EXTRACTION EFFICIENCIES  179  APPENDIX E - LIPID SEPARATION  180  APPENDIX F - ULTRAVIOLET NITRATE DETERMINATION  183  APPENDIX G - AMMONIUM DETERMINATION  185  APPENDLX H - NITRATE REDUCTASE DETERMINATION  187  APPENDIX I - GLUT AMINE SYNTHETASE DETERMINATION  190  APPENDIX J - ISOCITRATE LYASE DETERMINATION  191  APPENDIX K - SCINTHLATION COUNTING  194  vi  LIST OF TABLES  Table 1.1. 16:1/16:0 fatty acid ratios in P. tricornutum. Ratios were calculated from GC derived values for total fatty acids in logarithmic, early, and late stationary phase cultures (time periods A, C, and E on Fig. 1.1). Results are means ± 1 standard error for 4 cultures.. 39 Table 2.1. Pathway models for reductant and carbon generation during the oxidation of either a 16 carbon fatty acid associated with TAG or a 16 carbon carbohydrate in P. tricornutum. e": carbon conserved = ratio of reductant produced in pathway:moles carbon not lost to CO2. For cases of anaplerotic oxidation, V* of the carbon was assumed to be mobilized for use in biosynthetic reactions, and the remainder oxidized toC0  91  Table 2.2. Equations for reductant and carbon requirements during either nitrate or ammonium assimilation in P. tricornutum. NR = nitrate reductase; NiR = nitrite reductase; GS = glutamine synthetase; GOGAT = glutamate synthase; Gin = glutamine; Glu = glutamate  92  2  Table 3.1. Label recovery efficiencies for C-labeled intracellular pools in P. tricornutum. Label recovery efficiency was calculated as the sum fraction radioactivity/whole cell radioactivity * 100% 14  117  Table D . l . Chlorophyll a extraction efficiencies for P. tricornutum. Chlorophyll a was extracted from P. tricornutum cultures growing in logarithmic phase. Whatman™ GF/F filters (25 mm) were used for all extractions. The extraction solvents were either 90% acetone or 3:1 90% acetone:DMSO (v/v). Efficiencies were calculated as percentages relative to the most chlorophyll a extracted in a single sample, which was for filters sonicated in 3:1 90% acetone:DMSO. Chlorophyll a was determined fluorometrically. Treatments: 24 h = overnight extraction at 4°C; sonication =10 min of sonication in an ice-water bath followed by overnight extraction at 4°C; grind = 30 s with a Teflon-headed pestle in a 5 mL Potter-Elvehjem grinding tube followed by overnight extraction at 4°C. Results are means ± 1 standard deviation for duplicate measurements 179  Table H . l . Effect of filter storage time on NR activity. A logarithmic phase culture of P. tricornutum was harvested and the filters stored for varying lengths of time under liquid nitrogen (-196°C). One set offilterswas not frozen, but processed immediately. The remaining filters were thawed and assayed for in vitro NR activity at different times from harvest. The results show that storing the filters for up to one week did not adversely affect NR activity. In fact, freezing the filters at -196°C for even short periods of time appeared to enhance the extraction of the NR enzyme from the cells. Routine NR measurements were therefore made on liquid nitrogen frozen and thawed cells, regardless of storage time. Results are means ± 1 standard error for triplicate determinations  viii  LIST OF FIGURES  Fig. 1.1. Changes in biomass over several dilution cycles for semicontinuous cultures of P. tricomutum. Cultures were sampled, diluted withfreshmedium containing 1 mM NaN0 , and sampled again 24 h later. The dotted lines indicate the times when dilutions were made to supply nitrate. Results show logarithmic phase cultures before (A) and 24 h after (B) dilution; early stationary phase cultures before (C) and 24 h after (D) dilution; and late stationary phase cultures before (E) and 24 h after (F) dilution. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols 3  Fig. 1.2. Changes in medium nitrate concentration over several dilution cycles for semicontinuous cultures of P. tricomutum. Samples were taken as described in Fig. 1.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols Fig. 1.3. Nitrogen containingfractionsfor P. tricomutum. Bars represent three cell cycle phases; logarithmic (A), early stationary (C), and late stationary (E), and measurements taken 24 h after nitrate addition to each of the three culture phases (B, D, and F, respectively). Graphs are shown for PON (A), chlorophyll a (B), and protein (C). Results are means ± 1 standard error for 4 cultures  31  Fig. 1.4. Carbon-containingfractionsfor P. tricomutum. Bars represent the sampling times indicated on Fig. 1.3. Graphs are shown for POC (A), carbohydrate (B), and total fatty acids (C). Results are means ± 1 standard error for 4 cultures  32  Fig. 1.5. Lipid-containingfractionsfor P. tricomutum. Bars represent the sampling times indicated on Fig. 1.3. Graphs are shown for polar lipid fatty acids (A), and TAG fatty acids (B). Results are means ± 1 standard error for 4 cultures  33  Fig. 1.6. Fatty acid components of the polar (A) and TAG (B) lipidfractionsfor P. tricomutum. Bars show changes in the concentrations for individual fatty acids before and after dilution of logarithmic phase cultures withfreshmedium containing 1 mM NaN0 (times A and B on Fig. 1.1). The fatty acid identity is given as x:y, where x is the carbon number, and y is the double bond number. Results are means ± 1 standard error for 4 cultures  34  Fig. 1.7. Fatty acid components of the polar (A) and TAG (B) lipidfractionsfor P. tricomutum. Bars show changes in the concentrations for individual fatty acids before and after dilution of early stationary phase cultures withfreshmedium containing 1 mM NaN0 (times C and D on Fig. 1.1). Results are means ± 1 standard error for 4 cultures  35  3  3  ix Fig. 1.8. Fatty acid components of the polar (A) and TAG (B) lipid fractions for P. tricomutum. Bars show changes in the concentrations for individual fatty acids before and after dilution of late stationary phase cultures withfreshmedium containing 1 mM NaNC»3 (times E and F on Fig. 1.1). Results are means ± 1 standard error for 4 cultures  36  Fig. 1.9. Polar and TAG lipidfractionsin P. tricomutum cultures N-starved for 1 week and then diluted withfreshmedium containing 1 mM NaNC»3 at t = 0 h. The nitrate in the medium was exhausted 36 h following dilution, as indicated on thefigurewith an arrow. Results are means ± 1 standard error for 4 cultures. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols 37 Fig. 1.10. Carbon-containingfractionsin P. tricomutum cultures N-starved for 1 week and then diluted withfreshmedium containing 1 mM NaNC»3 at t = 0 h. The nitrate in the medium was exhausted 36 h following dilution, as indicated on thefigurewith an arrow. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  38  Fig. 2.1. Changes in cell density for nitrate or ammonium-recovered cultures of P. tricomutum. Cultures were N-starved for 3 d in continuous light before being diluted withfreshmedium containing either 1 mM NaNC>3 or N H 4 C I (t = 0 h; dotted line). Dilutions were made at 12 to 18 h intervals following time zero to ensure NsufBciency. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  65  Fig. 2.2. Culture in vivofluorescencefor nitrate- or ammonium-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols 66 Fig. 2.3. Culture medium N concentrations for nitrate- or ammonium-recovered cultures of P. tricomutum. The initial concentrations immediately following dilution at time zero were not measured. The periodic increases in medium N concentrations following the initial dilution at time zero indicate subsequent dilutions made to ensure N-sufficiency. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  67  Fig. 2.4. Intracellular carbohydrate concentrations for nitrate- or ammonium-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  68  Fig. 2.5. Intracellular TAG fatty acid concentrations for nitrate- or ammonium-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  69  X  Fig. 2.6. Intracellular polar lipid fatty acid concentrations for nitrate- or arnrnoniumrecovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures  70  Fig. 2.7. Intracellular N assimilatory enzyme activities for nitrate- or ammoniumrecovered cultures of P. tricornutum. Nitrate reductase (NR) activity was monitored in nitrate-recovered cultures, and glutamine synthetase (GS) activity was monitored in ammonium-recovered cultures. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols 71 Fig. 2.8. Intracellular isocitrate lyase (ICL) activity for nitrate- or ammonium-recovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  72  Fig. 2.9. Changes in cell density for nitrate or ammonium dark-recovered cultures of P. tricornutum. Cultures were N-starved for 2 d in continuous light and then 1 d in darkness before being diluted with fresh medium containing either 1 mM NaN03 or NH4CI (t = 0 h; dotted line). Cultures were maintained in the dark (shaded area) for 72 h following N addition before light was made available (unshaded area). Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  73  Fig. 2.10. Culture in vivo fluorescence for nitrate or ammonium dark-recovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  74  Fig. 2.11. Culture medium N concentrations for nitrate or ammonium dark-recovered cultures of P. tricornutum. The initial concentrations immediately following dilution at time zero were not measured. Only nitrate was measured in nitrate-recovered cultures and ammonium in ammonium-recovered cultures. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  75  Fig. 2.12. Cell N concentrations for nitrate or ammonium dark-recovered cultures of P. tricornutum. The initial concentrations immediately following dilution at time zero were not measured. Only nitrate was measured in nitrate-recovered cultures and ammonium in ammonium-recovered cultures. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. 76 Fig. 2.13. Medium nitrite concentrations for nitrate dark-recovered cultures of P. tricornutum. The initial concentrations immediately following dilution at time zero were not measured. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  77  xi Fig. 2.14. Intracellular carbohydrate concentrations for nitrate or ammonium darkrecovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  78  Fig. 2.15. Intracellular TAG fatty acid concentrations for nitrate or ammonium darkrecovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  79  Fig. 2.16. Intracellular polar lipid fatty acid concentrations for nitrate or ammonium dark-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols 80 Fig. 2.17. Intracellular N assimilatory enzyme activities for nitrate or ammonium darkrecovered cultures of P. tricomutum. NR and GS activities were monitored as described in Fig. 2.7. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  81  Fig. 2.18. Intracellular ICL activity for nitrate or ammonium dark-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  82  Fig. 2.19. Model for lipid and carbohydrate Cflowslinked to N assimilation. Pathways do not represent exact stoichiometrics and not all intermediates are shown. Although succinate is the product of the glyoxylate cycle, malate is shown as the net exported C4 acid for subsequent reactions. (1) = lipase; (2) = carnitine acyltransferase; (3) = pyruvate dehydrogenase complex; (4) = decarboxylating enzymes of the TCA cycle; (5) = pyruvate kinase; (6) = malic enzyme; (7) = malate dehydrogenase, (8) = phosphoenolpyruvate carboxylase, (9) = phosphoenolpyruvate carboxykinase; (10) = 6-phosphogluconate dehydrogenase; (11) = nitrate reductase; (12) = nitrite reductase; (13) = glutamine synthetase; (14) = glutamate synthase. PEP = phosphoenolpyruvate; PYR = pyruvate; OAA = oxaloacetate; Gin = glutamine; Glu = glutamate; OPP Pathway = oxidative pentose phosphate pathway  83  Fig. 2.20. Model for carbonflowand reductant generation during respiration of TAGs assuming either mitochondrial (A) or glyoxysomal (B) 13-oxidation. Pathways do not represent exact stoichiometrics and not all intermediates are shown. (1) = 13oxidation enzymes; (2) = dehydrogenases of the TCA cycle; (3) = carnitine acyltransferase  84  xii F i g . 2.21. Model for carbon flow and reductant generation during TAG oxidation by glyoxysomal oxidation and the TCA cycle. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = 13-oxidation enzymes; (2) = dehydrogenases of the glyoxylate cycle; (3) = malic enzyme; (4) = pyruvate dehydrogenase complex; (5) = dehydrogenases of the TCA cycle  85  F i g . 2.22. Model for carbon flow and reductant generation during anaplerotic oxidation of TAGs. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = G-oxidation enzymes; (2) = dehydrogenases of the glyoxylate cycle; (3) = malic enzyme; (4) = pyruvate dehydrogenase complex; (5) = dehydrogenases of the TCA cycle.  86  F i g . 2.23. Model for carbon flow and reductant generation during gluconeogenesis. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = B-oxidation enzymes; (2) = dehydrogenases of the glyoxylate cycle; (3) = malate dehydrogenase; (4) = phosphoenolpyruvate carboxykinase; (5) = glyceraldehyde phosphate dehydrogenase. PGA = phosphoglyceraldehyde; 1,3bisPGA= 1,3-bisphosphoglyeraldehyde  87  F i g . 2.24. Model for carbon flow and reductant generation during classical (A) and malate (B) glycolysis. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = glyceraldehyde phosphate dehydrogenase; (2) = pyruvate dehydrogenase complex; (3) = dehydrogenases of the TCA cycle; (4) = phosphoenolpyruvate carboxylase; (5) = malate dehydrogenase; (6) = malic enzyme. G 3-P = glyceraldehyde 3-phosphate  88  F i g . 2.25. Model for carbonflowand reductant generation during carbohydrate degradation through the oxidative pentose phosphate pathway. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = glucose 6phosphate dehydrogenase; (2) = 6-phosphogluconate dehydrogenase; (3) = glyceraldehyde phosphate dehydrogenase; (4) = pyruvate dehydrogenase complex; (5) = dehydrogenases of the TCA cycle. 6-PG = 6-phosphogluconate  89  F i g . 2.26. Model for carbonflowand reductant generation during anaplerotic oxidation of carbohydrates. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = glyceraldehyde phosphate dehydrogenase; (2) = malate dehydrogenase; (3) = malic enzyme; (4) = pyruvate dehydrogenase complex; (5) = dehydrogenases of the TCA cycle. Dashed lines = malate glycolysis  90  F i g . 3.1. Changes in cell density for nitrate or ammonium starved and recovered cultures of P. tricornutum labeled with C sodium acetate. Cultures were N-starved in continuous light for 24 h, labeled for 2 h, and the label removed by washing the cultures at t = 0. The cultures were N-starved for a further 48 h before being spiked with 1 mM NaN03 or NH4CI (dotted line). Subsequent N additions were made at 12 h intervals following the initial spike to ensure N-sufficiency. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols 118 14  xiii Fig. 3.2. Culture in vivo fluorescence for nitrate or ammonium starved and recovered cultures of P. tricomutum labeled with C sodium acetate. Manipulations were made as described in Fig. 3.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  119  Fig. 3.3. Uptake of C by 24 h N-starved cultures of P. tricomutum previously grown on nitrate. C-acetate was pulsed to culture concentrates (2 h prior to t = 0 in Fig. 3.1) and radioactivities were measured for whole cells (A) and in several intracellular fractions (B). Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  120  Fig. 3.4. Uptake of C by 24 h N-starved cultures of P. tricomutum previously grown on ammonium. Manipulations were carried out as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  121  Fig. 3.5. Reallocation of C by 48 h N-starved cultures of P. tricomutum previously grown on nitrate. Cultures that were previously 24 h N-starved were pulsed with C-acetate (Fig. 3.3), then washed and resuspended infreshN-free medium under normal culture conditions (t = 0 in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. The legend refers to the symbols in panel B  122  Fig. 3.6. Reallocation of C by N-starved cultures of P. tricomutum previously grown on ammonium. Cultures that were previously 24 h N-starved were pulsed with C acetate (Fig. 3.4), then washed and resuspended infreshN-free medium under normal culture conditions (t = 0 in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. The legend refers to the symbols in panel B  123  Fig. 3.7. Short-term reallocation of C by nitrate-recovered cultures of previously N starved P. tricomutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mM N a N 0 at t = 0 (dotted line). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. The legend refers to the symbols in panel B  124  1 4  1 4  14  1 4  1 4  14  1 4  1 4  1 4  3  Fig. 3.8. Short-term reallocation of C by ammonium-recovered cultures of previously N-starved P. tricomutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mM N a N 0 at t = 0 (dotted line). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. The legend refers to the symbols in panel B 125 1 4  3  xiv Fig. 3.9. Reallocation of C by nitrate-recovered cultures of previously N-starved P. tricornutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mM NaN03 at t = 0 (dotted line in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. The legend refers to the symbols in panel B 126 14  Fig. 3.10. Reallocation of C by ammonium-recovered cultures of previously N-starved P. tricornutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mM NaN03 at t = 0 (dotted line in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols  127  Fig. 3.11. Reallocation of C by nitrate-recovered cultures of previously N-starved P. tricornutum amongst modified cell fractions. The data in this figure have been reprocessed from Fig. 3.7B, and show the carbohydrate and LMWM fractions combined into a single LMWM/carbohydrate fraction. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols  128  Fig. 3.12. Reallocation of C by ammonium-recovered cultures of previously N-starved P. tricornutum amongst modified cell fractions. The data in this figure have been reprocessed from Fig. 3.8B, and show the carbohydrate and LMWM fractions combined into a single LMWM/carbohydratefraction.Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols....  129  Fig. 3.13. Models for carbon flowfromTAGs during nitrate (A) or ammonium (B) assimilation. Both models represent carbon flows at the time of maximum T A G degradation in P. tricornutum; i.e. 18-48 h following N addition to N-starved cultures. Approximate carbon flow magnitudes are represented by different arrow thicknesses. The broken lines indicate carbonflowsthat occur before TAGs are mobilized. C4 acids = malate, OAA, succinate; M = mitochondria; C = chloroplast; CH 0 = carbohydrate  130  I4  14  14  2  XV  Fig. 4.1. Changes in 8 C values in P. tricomutum cellfractionsbefore, during, and after recoveryfromN-starvation. The bars represent a transitionfromlogarithmic phase cultures (white, N just depleted) to stationary phase (gray, 4 d N-starved), and recovered with nitrate (black, 2 d after recovery with NaNOa). Growth medium (M), whole cell (WC), total lipid (TL) and TAGfractionswere isolated and their 8 C values determined as described in the materials and methods. The non-lipid cell fraction (NL) 8 C value was calculated assuming mass balance and using whole cell and total lipid 8 C and pg C values. The non-TAG lipidfraction(NTL) 8 C value was similarly calculated using total lipid and TAG 8 C and pg C values. Results are means ± 1 standard error for 4 cultures. Statistical tests (repeated measures A N O V A P < 0.01, followed by Tukey's test, q < 0.01, a = 0.01) were performed to detect differences between the three growth phases for the 8 C values. Within each group of three bars, identical letters above the bars indicate no significant difference between the means. 145 13  13  13  13  13  13  13  Fig. 4.2. Carbon quotas in P. tricomutum cellfractionsbefore, during, and after recoveryfromN-starvation. The bar fills and cellfractioncodes are as described in Fig. 4.1. The carbon content of the whole cell (WC) and TAGfractionswere measured directly. The carbon content of the non-lipid (NL) and non-TAG lipid (NTL)fractionswere calculated assuming mass balance and using total lipid (TL) carbon data and datafromthe measuredfractions.Results are means ± 1 standard error for 4 cultures. Statistical tests (repeated measures A N O V A P 0.01, followed by Tukey's test, q < 0.01, a = 0.01) were performed to detect differences between the three growth phases for the cell carbon values. Within each group of three bars, identical letters above the bars indicate no significant difference between the means <  146  Fig. 4.3. Carbon isotope discrimination between cell compartments in P. tricomutum before, during, and after recoveryfromN-starvation. The bar fills and cell fraction codes are as described in Fig. 4.1. Carbon isotope discrimination (A) was calculated for the assimilation of dissolved inorganic carbon into whole cell organic carbon (M-»WC), total lipid synthesisfromnon-lipid cell carbon (NL-»TL), and T A G synthesisfromnon-lipid cell carbon (NL->TAG). Carbon discrimination values were calculatedfromabsolute C / C ratio (R) values as described in the materials and methods. Results are means ± 1 standard error for 4 cultures. Statistical tests (repeated measures A N O V A p < 0.01, followed by Tukey's test, q<0.01,a = 0.01) were performed to detect differences between the three growth phases for the assimilation step A values. Within each group of three bars, identical letters above the bars indicate no significant difference between the means 147 13  12  xvi Fig. A . l . Comparison of methods for estimating cell density in P. tricornutum. A stationary phase culture was sequentially diluted with fresh ESAW to give 20 subcultures of varying cell density. These were then subsampled for density estimates from either haemocytometer (x-axis) or Coulter Counter™ counts (y-axis). The haemocytometer counts were assumed to more closely represent actual cell densities than Coulter Counter™ counts, as the latter does not discriminate between algal and non-algal particles, and coincidence error increases as particle density increases. Thus, Coulter Counter™ counts tended to underestimate cell density as cell density increased. However, in experiments, the cell densities measured always fell between 0.3-2.0 * 10 cells-mL" ; thus this underestimation was deemed to be insignificant , 175 7  1  Fig. B.l. Typical standard curve for protein determinations. Bovine serum albumin (BSA), processed as for algal samples, was used as the standard protein. Protein was estimated in 0.1 mL of algal extract or BSA dissolved in 0.1 N NaOH 177 Fig. C l . Typical standard curve for carbohydrate determinations. Glucose, processed as for algal samples, was used as the standard carbohydrate. Carbohydrate was estimated in 0.1 mL of algal extract or glucose dissolved in 3 N H2SO4. The carbohydrates in algal extracts were assumed to contain simple sugars and polysaccharides, such as chrysolaminarin  178  Fig. E . l . Comparison of lipid separations using thin layer chromatography or solid phase extraction columns. Standard lipids were dissolved in chloroform and applied to either Baker Si250 PA silica gel thin layer chromatography (TLC) plates (J.T. Baker, Phillipsburg, New Jersey, USA) (left), or Waters Silica Light Sep-Pak™ solid phase extraction (SPE) columns (right). The standard lipids chosen were from Sigma. In increasing order of polarity, these lipids were: TAG (triacylglycerol), D A G (diacylglycerol), M A G (monoacylglycerol), and PC (phosphatidylcholine). Each lipid contained the fatty acids indicated on the left of the figure. For TLC plates, the developing solvent was 100:1 diethyl ether: acetic acid (v/v), and lipid spots were identified with iodine vapour. Duplicate runs on non-iodinated areas of the plate were scraped into tubes and transmethylated for fatty acid methyl ester (FAME) analysis by gas chromatography (GC). For SPE columns, the neutral fraction was eluted first with 10 mL 3:1 chloroform:hexane (3:1), and the polar fraction was eluted with 10 mL methanol. The eluates were dried, transmethylated, and the FAMEs analyzed by GC. Therightside of the figure shows the presence (Yes) or absence (No) for each lipid in the neutral or polar column fractions. The absence of a particular lipid in any one of the column fractions was confirmed by the absence of one of the fatty acid constituents for thatfraction(bracketed fatty acid designations). For "Yes" values, all fatty acids for a particular lipid were present in the relevant columnfraction.These tests confirmed that only TAGs were eluted in the first collectedfractionof the SPE columns 180  xvii F i g . E.2. Efficiency of solid phase extraction columns. Standard lipids (Fig. E. 1) were dissolved in chloroform and applied to Waters Silica Light Sep-Pak™ SPE columns. The neutral fraction was elutedfirstwith 10 mL 3:1 chloroform:hexane (3:1), and the polar fraction was eluted with 10 mL methanol. During elution, 1 mL sub-fractions were collected. The eluates were dried, transmethylated, and the FAMEs analyzed by GC. Total sample retrieval was at least 93% for neutral lipids (A) and 83% for polar lipids (B). Neutral lipids (A) were eluted in a bell-shaped curve, indicating good sample retrievalfromthe column. However, polar lipids (B) were eluted in a decaying curve, indicating that residual sample remained bound to the column. Thus, for any one sample, the polar lipid fatty acidfractionwas usually calculated by subtracting GC determined neutral lipid fatty acid quantitiesfromtotal fatty acids. Thorough washing of the columns with 2><10 mL methanol succeeded in removing most of the residual polar lipids, so that columns could be reused up to four times before being discarded  181  F i g . E.3. Gas chromatograph output trace for a mixture of standard fatty acid methyl esters. Sample volume injected was 1 uL. Split ratio was 100:1. Injector temperature was 250°C. Oven temperature was held constant at 180°C for the first 32 minutes of each run, and then increased to 240°C at 50°C per minute for the final 10 minutes. Total run time was 43.2 minutes. Flame ionization detector (FDD) temperature was 350°C. The numbers above each peak show the FAME identity and retention time 182 F i g . F . l . Typical standard curve for nitrate determinations using the UV method. Quartz cuvettes were used for the determinations. Readings were taken exactly 30 s after placing cuvettes in the spectrophotometer as absorbance values tended to increase slightly with measurement time. All samples were diluted by mixing 0.3 mL of sample with 0.7 mL DDW before measurement  183  F i g . F.2. Comparison of methods for nitrate determination in ESAW. The medium from a culture of P. tricomutum was sampled during the transitionfromlogarithmic phase (N-sufficient) to stationary phase (N-starved). The nitrate concentration was then determined in these samples using both the Technicon® Autoanalyzer and U V methods. As the cultures became increasingly N-starved, the U V method tended to overestimate the nitrate in solution. Thus, the UV method was only used to monitor nitrate concentrations in the medium of actively growing cultures. The Technicon® Autoanalyzer method (Wood et al. 1967) was used to determine nitrate concentrations in the mediumfromN-starved cultures and in cell extracts 184 F i g . G . l . Typical standard curve for manual ammonium determinations. All samples were diluted by mixing 0.025 mL sample with 0.475 mL DDW before adding the colour reagents and measuring  186  xviii Fig. H.l. Effect of cell number in assay homogenate on NR activity. Different volumes of a logarithmic phase culture of P. tricornutum were harvested, the filters quickfrozen in liquid nitrogen, thawed, and assayed for in vitro NR activity. Assay time was 15 minutes. Using these results, harvest volumes in routine experiments were adjusted so that 5 * 10 cells were filtered for NR assays. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible they fit inside the symbols 7  188  Fig. H.2. Effect of assay time on NR activity. A logarithmic phase culture of P. tricornutum was harvested (5 x 10 cells per filter), the filters quick-frozen in liquid nitrogen, thawed, and assayed for in vitro NR activity. Assays were conducted for varying lengths of time. The results show that NR activity increased with assay time, but not proportionally; i.e. assay efficiency decreased with increasing assay time. Therefore, in routine experiments, a 15 minute incubation time was used to obtain the best NR activities. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible they fit inside the symbols 189 7  Fig. 1.1. Comparison of methods for estimating ammonium assimilation rates. A logarithmic phase culture of P. tricornutum was sampled five times over a 36 h period. Cells were harvested for in vitro GS activity measurements, and the medium ammonium concentration was monitored. Cell density data from this experiment, and previously acquired particulate organic nitrogen (PON) data for logarithmic phase P. tricornutum cultures, were used to estimate the ammonium assimilation rate (PON x u; left bar). Assimilation rate was also estimated from ammonium disappearance from the medium (ammonium uptake; middle bar), and GS activities (GS; right bar). There were no significant differences between any of these methods (p>0.05, 1-way repeated measures ANOVA). Therefore, the GS assay was judged to accurately estimate in vivo ammonium assimilation rates for logarithmic phase P. tricornutum cultures. Results are means ± 1 standard error for four consecutive assimilation rate estimates 190 Fig. J.l. Effect of cell number in assay homogenate on ICL activity. Different volumes of a logarithmic phase culture of P. tricornutum were harvested, the filters quickfrozen in liquid nitrogen, thawed, and assayed for in vitro ICL activity. Assay time was 90 minutes. Using these results, harvest volumes in routine experiments were adjusted so that 5 x io cells were filtered for NR assays. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible they fit inside the symbols 7  191  Fig. J.2. Effect of assay time on ICL activity. A logarithmic phase culture of P. tricornutum was harvested (5 x io cells per filter), thefiltersquick-frozen in liquid nitrogen, thawed, and assayed for in vitro ICL activity. Assays were conducted for varying lengths of time. The results show that ICL activity increased with assay time. In routine experiments, a 90 minute incubation time was used to obtain ICL activities. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible theyfitinside the symbols 192 7  Fig. J.3. Wavelength scans of dinitrophenylhydrazone derivatives of various G4 acids. These scans were done because Haigh and Beevers (1964) suggested that the glyoxylate produced in the ICL assay reaction could not be easily distinguished from other C4 acids. Several C4 acids (100 nmol-mL" ) were converted to their dinitrophenylhydrazone derivatives with the same techniques and conditions used for ICL assay products. These were then scanned with an LKB Ultraspec II™ spectrophotometer over the visible wavelength range. A combination of all the pure C4 acids was also scanned, as were the assay productsfroman ICL activity determination in P. tricomutum. Absorbance ratios were calculated for the prominent peaks and shoulders for each C4 acid derivative, and for the combination and P. tricomutum assay derivatives. The absorbance spectrum for P. tricomutum closely matched that for glyoxylate, indicating that glyoxylate was most likely the C4 dinitrophenylhydrazone derivativefromthe ICL assay reactions being measured 1  Fig. K . l . Quench curves for various solvents used for radioactivity determinations. Unquenched radioactivities were measured in dried volumes of the source radioisotope with no added solvents. The inset shows the quench curves for the low range of radioactivities. All samples were corrected for background radiation and chemiluminescence  LIST OF ABBREVIATIONS  ct-kg  a-ketoglutarate specific growth rate  u  °/oo  parts per thousand  1,3-bisPGA  1,3-bisphosphoglyeraldehyde  6-PG  6-phosphogluconate  ACCase  acetyl-CoA carboxylase  ADP  adenosine diphosphate  AMP  adenosine monophosphate  ANOVA  analysis of variance  ATP  adenosine triphosphate  BSA  bovine serum albumin  C  chloroplast (on model diagrams only)  C3  sugars containing three carbon atoms  C4  organic acids containing four carbon atoms  CH 0  polymeric carbohydrates  CHN  carbon nitrogen hydrogen  A  isotopic discrimination  8C  standardized ratio of C/ C  DAG  diacylglycerol  DAGAT  diacylglyerol acyltransferase  DDW  deionized distilled water  2  13  13  12  DIC  dissolved inorganic carbon  DMSO  dimethylsulfoxide  DMSP  dimethylsulfoniopropionate  DPM  disintegrations per minute  ESAW  enriched artificial seawater  FAME  fatty acid methyl ester  FID  flame ionization detector  G3-P  glyceraldehyde 3-phosphate  GC  gas chromatography  Gin  glutamine  Glu  glutamate  GOGAT  glutamate synthase  GS  glutamine synthetase  HEPES  N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic  ICL  isocitrate lyase  IRGA  infra red gas analyzer  LMWM  low molecular weight metabolite  M  growth medium (Chapter 4 only)  M  mitochondrion (on model diagrams only)  MAG  monoacylglycerol  mRNA  messenger ribonucleic acid  MS  malate synthase  NAD  oxidized nicotinamide adenine dinucleotide  acid]  NADH  reduced rticotinamide adenine dinucleotide  NADPH  reduced nicotinamide adenine dinucleotide phosphate  NiR  nitrite reductase  NL  non-lipid  NR  nitrate reductase  NTL  non-triacylglycerol lipid  OAA  oxaloacetate  OPPP  oxidative pentose phosphate pathway  PC  phosphatidylcholine  PDH  pyruvate dehydrogenase  PEP  phosphoenolpyruvate  PEPCase  phosphoenolpyruvate carboxylase  PEPCK  phosphoenolpyruvate carboxykinase  PGA  phosphoglyceraldehyde  Pi  inorganic phosphate  POC  particulate organic carbon  PON  particulate organic nitrogen  ppm  parts per million  PYR  pyruvate  RUBISCO  ribulose bisphosphate carboxylase/ oxygenase  SPE  solid phase extraction  TAG  triacylglycerol  TCA  tricarboxylic acid cycle  TCAA  trichloracetic acid  TL  total lipid  TLC  thin layer chromatography  UDP  uridine diphosphate  UV  ultraviolet  WC  whole cell  xxiv  ACKNOWLEDGEMENTS  There are many people I wish to thank, who have provided valuable support in one way or another in the past 5 V2 years. First and foremost, my thanks go to Dr. Paul J. Harrison. Paul has been an excellent supervisor, and has shown me the value of collaboration and cooperation with other members of the scientific community to "get the job done." I also thank Dr. David Turpin, in whose lab I spent a good part of a year developing a culture system and conducting some preliminary experiments. I am grateful to Dr. Ljerka Kunst, who gave me" free reign on the G C , and has also inspired me to continue to work in the field of lipid metabolism. It has been an enjoyable experience working in the Harrison Lab. I am grateful to Anthony Fielding, Bente Nielsen, Dr. David Crawford, Dr. Nathalie Waser, and Dr. Rob Guy for helping me to understand the seemingly endless nuances of carbon stable isotope fractionation processes. Dr. John Berges gave valuable advice on optimizing the nitrate reductase assay. Anne Fisher helped with the method development for the carbohydrate analyses. My thanks go out to my "lab-bound" colleagues, Allen Milligan and Robert Strzepek, who provided entertaining and useful discussions on algal physiology. I also thank my other lab colleagues for providing an excellent working atmosphere - Beth Bornhold, Dr. Philip Boyd, Mingxin Guo, Dr. Catriona Hurd, Joe Needoba, Lauren Ross, Nelson Sherry, and Dr. Kedong Yin. Special thanks go to the curling team - Shannon Harris and Diana Varela, who helped me to enjoy myself in a truly "Canadian" pastime. Finally, my heartfelt thanks go out to my parents, Alan and Mary, who have supported me through seemingly endless years of university education. I dedicate this thesis to them both.  1  GENERAL INTRODUCTION  Fatty acids are important components in all living organisms. They have a vital structural role in cell membranes, some act as hormones or second messengers, and others accumulate in triacylglycerols (TAGs) to store carbon for later metabolism. From a dietary point of view, the study of the processes controlling fatty acid synthesis in autotrophs has been important, because animals are incapable of synthesizing several unique fatty acids that are essential for their normal development. Plant breeding and genetic engineering has also been applied to a number of commercial crops (e.g. varieties of Brassica napus) to improve their fatty acid composition and yield, and make them suitable for human consumption. In aquatic ecosystems, the fatty acid composition of phytoplankton has been examined to assess the suitability of various species as food sources for shellfish orfishin aquaculture. Microalgal strains with high lipid yields have also been considered as potential alternative fuel sources. Finally, the unique chemical structure of fatty acids and their derivatives has provided a useful tool for biogeochemical studies in marine ecosystems. Despite the wide interest in lipids and their component fatty acids in microalgae and higher plants, much is still unknown about the processes governing their metabolism. The reasons for this are that the biosynthetic pathways for fatty acid synthesis, desaturation, esterification to form lipids, and transacylation among lipids, occur in more than one cell compartment, are multi-branched, and differ between species. There have been two approaches to circumvent these problems. Firstly, for many years surveys have been conducted on how the fatty acid and lipid composition in microalgae and higher plants change in response to environmental variables. These studies have yielded much useful information on how variables such as light, temperature, pH, and nutrient availability affect lipid and fatty acid composition.  2  However, this approach does not provide the information required to thoroughly elucidate the metabolic events governing fatty acid synthesis and degradation. More detailed studies on the biochemistry of fatty acid and lipid biosynthesis have been carried out using model species. For higher plants, the model of choice in recent years has been Arabidopsis thaliana. Molecular genetic analyses of lipid metabolism in this species have enabled researchers to pinpoint some key regulatory steps of fatty acid synthesis and desaturation (Somerville and Browse 1991). For microalgae, the biosynthetic pathways for various fatty acids have been most thoroughly studied in the marine diatom Phaeodactylum tricornutum (Arao et al. 1987, Arao and Yamada 1994, Arao et al. 1994). The reason for this is that diatoms, particularly P. tricornutum, are known to produce significant quantities of eicosapentaenoic acid, a fatty acid thought to have an important role in preventing heart disease in humans (Alonso et al. 1994). Unfortunately, research on improving fatty acid yields and altering fatty acid composition in microalgae and higher plants has been carried out at the expense of studies on lipid degradation. From studies on higher plants, it is becoming apparent that these degradative processes need to be investigated more carefully, because plants genetically engineered to overproduce particular fatty acids may internally compensate by breaking these same fatty acids down (Eccleston et al. 1996). In higher plants, information on lipid and fatty acid degradation has comefromstudies on germinating seeds, where TAGs are mobilized to sugars by the concerted actions of lipases, B-oxidation, the glyoxylate cycle, and gluconeogenesis. In recent years, more studies have been directed at possible regulatory steps in these processes, particularly at the mechanisms whereby sugars inhibit lipid degradation (Graham et al. 1994). In algae, studies on lipid degradation have been limited to monitoring changes in lipid  composition in response to environmental variables, and the heterotrophic utilization of acetate, a lipid breakdown product. The research in this thesis will attempt to explain some of the regulatory mechanisms that govern storage lipid degradation in the marine diatom P. tricomutum. The following sections will provide background information on the processes that affect lipid and fatty acid composition and accumulation in microalgae, processes that regulate storage lipid synthesis, and processes that regulate storage lipid degradation. Finally, the objectives and chapter outline for the thesis will be presented.  LIPID A N D F A T T Y ACID C O M P O S I T I O N A N DA C C U M U L A T I O N IN MICROALGAE  There have been several major studies investigating the lipid and/or fatty acid composition in both cultured microalgae (Opute 1974, Orcutt and Patterson 1975, Shifrin and Chisholm 1981, Ben-Amotz and Tornabene 1985, Cranwell etal. 1985, Arao etal. 1987, Harvey etal. 1988, Yongmanitchai and Ward 1991, Parrish etal. 1993, Servel etal. 1994, Zhukova and Aizdaicher 1995), and natural phytoplankton populations (Kattner and Brockmann 1990, Mayzaud etal. 1990, Skerrat etal. 1995). These studies have shown that the major lipid components of microalgal species are glycoglycerolipids, phospholipids, and TAGs; however, free fatty acids, sterols and hydrocarbons may also be present. The presence of sugar groups (such as galactose) in microalgal lipids makes their structure more similar to higher plant lipids than animal lipids (Gurr and Harwood 1991).  4 The fatty acid composition of microalgal lipids is extremely variable because of the taxonomic diversity of this group. However, in general it appears that 18 carbon fatty acids are common-place in the dinoflagellates, prasinophytes, and chlorophytes (Harvey et al. 1988, Siron et al. 1989, Guckert and Cooksey 1990, Parrish et al. 1993, Reitan et al. 1994), which makes their fatty acid composition similar to higher plants (Gurr and Harwood 1991). In the diatoms, eustigmatophytes, and bangiophytes, 16 and 20 carbon fatty acids are more common (Fisher and Schwarzenbach 1978, Cohen et al. 1988, Mayzaud etal. 1990, Sukenik and Carmeli 1990, Yongmanitchai and Ward 1991, Reitan etal. 1994, Servel etal. 1994, Berge et al. 1995, Brown et al. 1996). In the prymnesiophytes and dinoflagellates, 22 carbon fatty acids are common (Harvey et al. 1988, Henderson et al. 1988, Thompson et al. 1992, Parrish et al. 1993, Alonso et al. 1994, Reitan et al. 1994, Servel et al. 1994). Of all the microalgae, fatty acid composition in diatoms has been more closely examined, because the species in this group accumulate a significant proportion of their fatty acids as eicosapentaenoic acid (20:5; using the notation x:y, where x = number of carbons and y = number of double bonds). In P. tricomutum, this fatty acid may comprise 10-29% of the total fatty acids, depending on the growth conditions (Arao etal. 1987, Siron etal. 1989, Yongmanitchai and Ward 1991, Thompson et al. 1992, Alonso et al. 1994, Reitan et al. 1994, Zhukova and Aizdaicher 1995). In microalgae the total lipid content can increase in response to a change in environmental variables, such as N- or Si-deficiency (Shifrin and Chisholm 1981, Piorreck etal. 1984, Parrish and Wangersky 1987, Suen et al. 1987, Emdadi and Berland 1989, Myklestad 1989, Parrish and Wangersky 1990, Roessler 1990), P-deficiency (Reitan et al. 1994, Siron et al. 1989), extended time in stationary phase (Fisher and Schwarzenbach 1978, Brown et al. 1996), and inhibition of growth by suboptimal pH values (Guckert and Cooksey 1990). All these conditions elicit an increase in the proportion of neutral lipids, particularly TAGs. In  contrast, conditions that alter the cell physiological state without completely inhibiting growth appear to change the fatty acid composition qualitatively rather than quantitatively. In general, the ratio of saturated to unsaturated fatty acids in several microalgae increases in response to longer daylengths (Fisher and Schwarzenbach 1978, Brown et al. 1996), higher irradiances (Thompson et al. 1990), elevated C 0 concentrations (Sato 1989), and the onset of stationary 2  phase (Brown et al. 1996). Polyunsaturated fatty acids are synthesized at lower temperatures (Baasch et al. 1984, Thompson et al. 1992), and at increased culture dilution rates (Molina Grimae/a/. 1994). The literature indicates that any condition that reduces the growth rate in microalgae appears to induce the accumulation of lipids, especially those containing saturated fatty acids. Presumably, this is because the cell division rate is slowed so that photosynthate is preferentially channeled into carbon storage compounds rather than protein. For example, microalgae from Arctic and Antarctic regions readily accumulate lipids at high light intensities because the rate of photosynthesis exceeds the rate of protein synthesis at low temperatures (Palmisano and Sullivan 1985, Palmisano etal. 1988, Priscu etal. 1988, Hawes 1990, Smith and Herman 1992). The next section will discuss some of the biochemical regulatory mechanisms that may function during storage lipid accumulation.  REGULATION OF STORAGE LIPID SYNTHESIS  TAGs comprise the storage lipids in plants and microalgae, and are synthesized in the endoplasmic reticulum from acyl-CoA precursors transported out of the plastid (Andrews and Ohlrogge 1990). Parnas and Cohen (1976) suggested that the accumulation of storage  6  materials in microalgae is not just a consequence of excess photosynthate, but a deliberate, controlled switch to the synthesis of storage material. Therefore, T A G synthesis must be enhanced relative to the synthesis of other lipids and macromolecules for storage lipids to accumulate. The primary and rate limiting step for de novo fatty acid synthesis is considered to be the activity of the plastidic enzyme, acetyl-CoA carboxylase (ACCase) (Stymne and Stobart 1987). In the oilseeds of some higher plants, the initiation of de novo storage lipid synthesis appears to be correlated with increased ACCase activity (Simcox et al. 1979, Turnham and Northcote 1983, Page etal. 1994). In the marine diatom Cyclotella cryptica, ACCase activity was enhanced during the onset of silicate starvation (Roessler 1987a). However, in the Nstarved prymnesiophyte Isochrysis galbana, ACCase activity decreased with time, but net lipid accumulation still occurred, albeit at a slower rate (Livne and Sukenik 1992). This same pattern of decreasing ACCase activity during ongoing storage lipid synthesis was also observed during seed development in oilseed rape (Kang et al. 1994). Thus, ACCase may not be entirely rate limiting in de novo lipid synthesis. Acyltransferase specificity may exert some control over TAG synthesis in microalgae and higher plants. In higher plant oilseeds, de novo TAG synthesis from recent photosynthate is enhanced by increased diacylglycerol acyltransferase (DAGAT) activity relative to other acytransferases that do not lead to the formation of TAGs (Stymne and Stobart 1987, Griffiths and Harwood 1991). Thus, non-TAG forming acytransferases must compete for the acyl-CoA pool exportedfromthe plastid. Recently, it has also been observed that TAGs can be synthesized by transacylation reactions among existing monoacyl- and diacylglycerols in developing safflower seeds (Stobart et al. 1997). Thus, TAG synthesis may be regulated at the level of cfe novo plastidic fatty acid synthesis, and/or preferential selection of acyl-CoA molecules in the endoplasmic reticulum.  One further question that needs to be addressed concerning storage carbon accumulation in microalgae is the relative importance of lipids vs. carbohydrates. In several higher plants, storage lipids are formed in the developing seeds, but starch and sucrose are the major stored carbon forms for other plant organs in the remainder of the life cycle. Green microalgae commonly accumulate starch, and marine diatoms accumulate (l-3)-linked B-Dglucans, such as chrysolaminarin, as storage polysaccharides (Myklestad 1989). In microalgae, carbohydrates accumulate when growth rates are reduced (Smith and Geider 1985, Harrison et al. 1990), whereas storage lipids accumulate when cell division ceases (Opute 1974, Thomas et al. 1984, Kattner and Brockmann 1990, Parrish and Wangersky 1990). The mass of storage lipids may easily exceed the mass of intracellular carbohydrates for stationary phase microalgae (Brown et al. 1996, Larson and Rees 1996), or exceed 50% of the dry weight (Shifrin and Chisholm 1981). Additionally, a mass comparison of stored carbohydrates vs. lipids is not valid when considering the importance of each for subsequent respiration. The calorific value for fatty acids is 9 kcalg' , whereas the value for carbohydrates is only 4 kcal-g' , so storage lipids 1  1  effectively have over double the calorific value of carbohydrates (Stryer 1988). For example, although the cell lipid content in aged cultures of the marine diatom Skeletonema costatum was reported to be only two-thirds of the value for carbohydrate (Morris 1981), the overall calorific value would have been higher for the accumulated lipid. Thus, despite their similarity as storage carbon reserves, the different energy values, structures, and storage sites for triacylglycerols and carbohydrates suggests that each has different roles during degradative carbon metabolism. These will be discussed in the following sections.  8 REGULATION OF STORAGE LIPID DEGRADATION  Ecological roles for storage lipid The accumulation of storage lipids in microalgae is species specific and unpredictable (Shifrin and Chisholm 1981). Storage carbon, whether in the form of carbohydrates or lipids, will accumulate in microalgae whenever net carbonfixationthrough photosynthesis exceeds its subsequent incorporation into protein or respiration to CO2. For microalgae, this situation may arise in the lighted portion of a day/night cycle, at low temperatures when the rate of protein synthesis is slowed, or at the end of a bloom when nutrients such as N or P are exhausted and cell division has ceased (Fernandez et al. 1992, Smith and D'Souza 1993). Thus, the stored carbon compounds may be utilized when these situations are reversed, and photosynthetic carbonfixationis temporarily limiting to growth. In situations where growth rates are not severely inhibited, carbohydrates are the predominantly stored and degraded polymers. Diel variation in B-glucans, which accumulate during the day and are degraded at night, occurs in marine microalgae (Myklestad 1989, Janse et al. 1996). This pattern of periodic carbohydrate accumulation and degradation may be a general mechanism whereby microalgae reduce their light-dependence on processes such as N assimilation (Turpin 1991). In specialized cases, lipids may have an important dynamic storage and degradative role on a day-to-day basis. MicroalgaefromArctic and Antarctic regions readily accumulate lipids at high light intensities because the rate of photosynthesis exceeds the rate of protein synthesis at low temperatures (Priscu et al. 1988, Palmisano and Sullivan 1985, Palmisano et al. 1988, Hawes 1990, Smith and Herman 1992). A portion of the accumulated lipid pool is utilized during the night to support continued N assimilation and protein synthesis (Smith etal. 1990).  9 The most likely roles for storage lipids in microalgae are to provide a concentrated energy source that can be used for maintenance respiration during severe growth inhibition, and to provide the carbon skeletons for the restoration of normal cell metabolism when growth resumes. For example, the ability of diatoms to use lipids for maintenance respiration may reduce their sinking rates in the dark. Diatoms are negatively buoyant because they possess a silicafrustule,and so must expend metabolic energy to pump heavy ions out of the cell to remain in the photic zone (Anderson and Sweeney 1977). Waite et al. (1992) found that sinking rates in the marine diatom Ditylum brightwelli increased when the cells were incubated under conditions where high energy storage compounds could not accumulate or be used, such as in low light or in the presence of respiratory inhibitors. Waite et al. (1992) suggested that the ability of diatoms to remain buoyant is a function of their energy reserves, so cells can remain in the photic zone when light is limiting during bloomsfromself-shading, or at night. Alternatively, the presence of storage lipid reserves may enable diatoms to make use of spatially separated nutrients. At the end of the light period, large D. brightwelli cells, which had presumably accumulated carbon storage productsfromphotosynthesis, sank inherently faster than small cells (Waite et al. 1992). This may enable the cells to use the higher concentrations of nitrate present near the nutricline for N assimilation at night. Perhaps the most important role for storage lipids in microalgae is in the recovery from an inhibition of growth. Lipids accumulated by the green alga Selenastrum capricornutum during cadmium inhibition of growth were used as respiratory substrates when the inhibition was removed and growth resumed (Thompson and Couture 1990). This also occurred in the marine diatom Phaeodactylum tricomutum when lipid-rich, non-dividing cells resumed growth after recoveryfromNa- or N-deficiency (Larson and Rees 1996). A suitable analogy to use here is the production and utilization of lipids in the oilseeds of some higher plants. Maturation  10 of oilseeds is essentially a process where the growth rate is slowed down and storage lipids are accumulated. The lipid reserves are then rapidly mobilized at germination to produce the carbohydrates required for rapid growth (Leegood and ap Rees 1978a, Andrews and Ohlrogge 1990). This process is necessary in germinating seeds, because photosynthetic carbon fixation cannot occur until thefirstleaves form. In microalgae resuming growth after a period of inhibition, the supply of carbonfromstorage lipids may be essential to compensate for suboptimal rates of photosynthetic carbonfixation.This is discussed in more detail in the next section.  Pathways for storage lipid degradation Photosynthesis generates reduced nicotinamide adenine dinucleotide phosphates (NADPH), adenosine triphosphate (ATP), and sugar phosphates. These can be used for a number of biosynthetic and maintenance processes. In microalgae, the transport and assimilation of nitrogenous compounds into protein are usually light-dependent processes and are coupled to photosynthesis (Pate and Layzell 1990). However, during extreme nutrient limitation that leads to a cessation of cell division, the photosynthetic rate may drop to less than 10% of the rate observed during nutrient-sufficient growth (Glover and Morris 1979, Larson and Rees 1996). When microalgae are recoveringfromN-deficiency, the photosynthetic rate is insufficient to supply the necessary reductant and carbon skeletons required to maintain rapid N assimilation rates and amino acid synthesis (Turpin 1991). The cell must compensate by increasing the rate of photosynthesis and degrading stored carbon compounds. The light reactions of photosynthesis can adapt to meet some of the requirements for rapid N assimilation by increasing the rate of cyclic photophosphorylation to provide extra ATP (Turpin and Bruce 1990, Mohanty et al. 1991). However, such a mechanism has limited value  11 for sustained N assimilation in extremely N-limited microalgae, because the photosynthetic rate may take several hours to recover to pre-limitation rates (Larson and Rees 1996). Thus, stored carbon must be respired to fulfill the energetic and carbon requirements for N assimilation. However, it is unlikely that lipids and carbohydrates are mobilized simultaneously or at the same rate. This is because the pathway for lipid degradation is longer and, in part, metabolically distinct from carbohydrate degradation. Some theoretical models for the different pathways of storage lipid and carbohydrate oxidation are presented in Chapter 2. Starch and other glucans are usually stored in the chloroplasts or plastids of higher plants (Kruger 1990) and outside the plastids in microalgae (Craigie 1974). In higher plants, sucrose is stored in the cytoplasm. Both these carbohydrates are converted to hexose phosphates, which can then enter either the glycolytic or oxidative pentose phosphate pathways (Dennis and Miernyk 1982). Ultimately, the pyruvate generated in glycolysis is respired in the tricarboxylic acid (TCA) cycle. These processes can generate various quantities of ATP and NAD(P)H, and, together withfi-carboxylationreactions, can also provide extra C4 acids for de novo amino acid synthesis. This anaplerotic replenishment of C4 acids through increased glycolytic starch breakdown and B-carboxylation reactions has been shown to be important in providing the carbon necessary for rapid ammonium assimilation in the green alga Selenastrum minutum (Elrifi and Turpin 1986). In contrast to carbohydrates, storage lipids (TAGs) are found in structures variously named oleosomes, spherosomes, or oil bodies in the cytosol of plant seed tissues (Halpin and Lord 1990). Oil bodies have also been described for P. tricornutum (Larson 1992), and Chlorella (Cooksey et al. 1987). There are several possible sinks for the carbon mobilized from TAG degradation. TAGs have been postulated to provide the carbon necessary for dark N fixationin peanut nodules (Siddique and Bai 1991), and for phospholipid synthesis in 2  12 germinating spores of the fungus Glomus versiforme (Gaspar et al. 1994) and Chlorella cells following recovery from N-starvation (BClyachko-Gurvich et al. 1973). The most likely sink for the carbon mobilized from TAGs is in carbohydrates. In germinating oilseeds, TAGs are broken down by oil-body bound lipases, and the fatty acids liberated are converted to acetyl-CoA and C4 acids by the concerted actions of B-oxidation and the glyoxylate cycle, which are located in specialized organelles called glyoxysomes (Kindl 1987). This compartmentalization allows C4 acids and acetyl-CoA to be generated without the oxidative decarboxylations associated with the TCA cycle (Kindl 1987). These C4 acids are then used for gluconeogenesis, which is an anabolic process. In animals, B-oxidation of fatty acids occurs in the mitochondria and peroxisomes, so the acetyl-CoA produced can enter directly into the TCA cycle and be respired. There are very few examples of B-oxidation occurring in plant mitochondria (ap Rees 1990), which suggests the primary role of storage lipid degradation in plants is for gluconeogenesis. Thus, compared to carbohydrate respiration, the pathway of storage lipid degradation in plants generates reductant but also conserves the mobilized carbon. The spatial separation of acetyl-CoA synthesis from storage lipids, afforded by the glyoxysomes, from the TCA cycle in the mitochondria may be advantageous for anaplerotic reactions. This is because the C4 acid products of the glyoxylate cycle may be transported into the mitochondria and used to replace intermediates removed from the TCA cycle for amino acid synthesis. Partial reactions of the glyoxylate cycle in unicellular algae may therefore function in a similar role to the anaplerotic system described for Selenastrum minutum, where the B-carboxylation of glycolytically produced phosphoenolpyruvate (PEP) by phosphoenolpyruvate carboxylase (PEPCase) supplies pyruvate and oxaloacetate to the TCA cycle (Elrifi and Turpin 1986).  13 One major problem in suggesting that the glyoxylate cycle is functional in microalgae, especially in diatoms, is the scarcity of evidence for glyoxysomes in these organisms. Part of the reason for this has been the difficulty in preparing isolated organelle suspensions from microalgae. However, Winkler and Stabenau (1995) did isolate peroxisome-like organelles from the diatoms Nitzschia laevis and Thalassiosira fluviatilis that contained the glyoxysomal marker enzyme, isocitrate lyase (ICL). Additionally, there is also conflicting evidence over whether fatty acid 13-oxidation occurs in peroxisomes/glyoxysomes, as in higher plants, or peroxisomes and mitochondria, as in animals (Stabenau etal. 1984, Gross 1989). Most of the evidence for the operation of the glyoxylate cycle in microalgae has come from the measurement of either ICL or malate synthase (MS) enzyme activities, which are considered to be markers for this cycle. Using these enzymes, the glyoxylate cycle has been documented in several bacteria, algae, ferns, gymnosperms, angiosperms, and simple worms (Cxometal. 1981, Giachetti etal. 1987). Early work with Chlorella and several other green microalgae showed that ICL activity was induced by acetate in heterotrophically grown cultures (Syrett etal. 1963, Haigh and Beevers 1964, Syrett 1966). ICL activity has also been measured in P. tricomutum (Larson 1992). Very little is known about the coordinated regulation of lipase activity, 13-oxidation, the glyoxylate cycle, and gluconeogenesis in plants. In higher plants, lipases may be under hormonal control. Gibberellic acid has been shown to stimulate the degradation of TAGs in the barley aleurone layer (Arnalte et al. 1991). An increase in TAG lipolytic activity in maize scutella tissues was also shown to be a consequence of de novo lipase protein synthesis (Wang and Huang 1987). In the yeast Candida lipolytica, regulation is at the level of fatty acid supply to either 13-oxidation or net lipid synthesis by two distinct forms of acyl-CoA synthetase (Numa and Tanabe 1984). In the bacterium Acinetobacter calcoaceticus, phosphorylation enhances  14 ICL activity but inhibits the TCA enzyme isocitrate dehydrogenase (Hoyt and Reeves 1992). Gluconeogenesis appears to be largely under the control of phosphoenolpyruvate carboxykinase activity in the cotyledons of geminating marrow (Leegood and ap Rees 1978b, Trevanion etal. 1995). Storage lipid degradation may be ultimately regulated by the demand for respiratory substrates, and be coordinated in some way with carbohydrate degradation and the availability of recent photosynthate. To this end, experiments have shown that exogenous glucose or light will inhibit ICL activity in Chlorella and Chlamydomonas (Thurston 1977, Martinez-Rivas and Vega 1993), and that expression of an element in the ICL gene in cucumber may be repressed by sucrose (Graham et al. 1994, Reynolds and Smith 1995). Nevertheless, it is clear from the literature that the relationships between lipid and carbohydrate metabolism are still far from being well understood.  THESIS O U T L I N E A N D O B J E C T I V E S  The experiments in this thesis were designed to investigate the patterns of storage lipid synthesis and degradation in response to N-starvation and recovery from it in the marine diatom Phaeodactylum tricornutum. Four major experiments were designed and carried out:  1)  Chapter 1 - Constituent changes in N-starved and nitrate-recovered cultures.  This experiment was designed to obtain baseline information for P. tricornutum cell constituents. The data describes the changes in cell constituents (protein, carbohydrate,  15 chlorophyll a, particulate organic N and C, fatty acid composition) during the processes of Nstarvation and recovery with nitrate.  2)  Chapter 2 - N assimilation and stored carbon mobilization during recovery of stationary  phase culturesfromN-starvation in the light or dark with nitrate or ammonium. This experiment investigated how TAG mobilization was coordinated with nitrate or ammonium assimilation in the light or dark following a period of N-starvation. The timing of activity changes for some key N assimilatory enzymes were compared with changes in the activity of the glyoxylate cycle enzyme, isocitrate lyase.  3)  Chapter 3 - Reallocation and loss of carbonfrom C labeled triacylglycerols during 14  recovery of stationary phase culturesfromN-starvation in the light with nitrate or ammonium. This experiment was designed to evaluate the relevance of gluconeogenesis in replenishing carbohydrates consumed during rapid N assimilation following a period of N-starvation. The experiment followed the movement of radiolabeled carbonfromTAGs into other cell constituents during N assimilation.  4)  Chapter 4 - Triacylglycerol synthesis and degradation as a determinant of carbon stable  isotopefractionation.This experiment was conducted because of the interest that biogeochemists have in the role of microalgal storage lipids in contributing to sedimentary carbon stable isotope ratios. The data show how a change in the cell lipid content over a Nstarvation and recovery cycle alters the cell carbon stable isotope signal.  16  CHAPTER 1  CONSTITUENT CHANGES IN N-STARVED AND NITRATE-RECOVERED  CULTURES  1.1)  INTRODUCTION  In marine diatoms, conditions of extreme nutrient limitation, such as at the end of a bloom, may lead to a reduction in growth rate and an accumulation of storage lipids. In general, when microalgae are N-starved there is a loss of intracellular chlorophyll a, amino acids, and protein, but an increase in biomass that is attributable to stored carbon (Healey 1973, Piorreck et al. 1984, Harrison etal. 1990). Carbohydrates accumulate in the initial stages of N-starvation (Harrison et al. 1990), whereas storage lipids accumulate during prolonged N starvation or when cell division ceases (Opute 1974, Thomas etal. 1984, Kattner and Brockmann 1990, Parrish and Wangersky 1990). Triacylglycerols (TAGs) make up the bulk of the storage lipids, and may form prominent oil bodies within the cell (Larson 1992). The ability of some algae to store prolific quantities of lipid has also spurred interest in developing alternative fuels from algal sources (Roessler 1990). Through the utilization of signature fatty acid molecules, and their ratios, diatom lipids can be traced up the food chain to zooplankton and fish (Kattner and Brockmann 1990). However, the processes governing storage lipid biosynthesis, degradation, and turnover in the source (diatoms) are poorly understood. Storage lipids have long been assumed to be important in respiratory metabolism in diatoms and other algae. However, our current  17 understanding of how lipid biosynthesis and degradation is regulated in algae is poor. This is despite the fact that several bodies of work in the literature have demonstrated that lipids can be quantitatively significant components of algal cells, especially during nutrient deficiency (Shifrin and Chisholm 1981, Piorreck et al. 1984, Parrish and Wangersky 1987, Suen etal. 1987, Emdadi and Berland 1989, Myklestad 1989, Parrish and Wangersky 1990, Roessler 1990). Amblard and Boudier (1990) suggested that cell carbon reserves in diatoms could maintain population viability during sedimentation, or allow survival in the aphotic zone. In general, stored carbon reserves may be mobilized at a later time when recent photosynthate is limiting, which occurs in darkness, or when the demand for carbon for biosynthetic processes and respiration exceeds the rate of photosynthesis. Such a demand for carbon occurs when Nstarved microalgae are resupplied with a N source (Turpin 1991). This chapter investigates how carbohydrate and storage lipid accumulation and degradation are coordinated in the marine diatom Phaeodactylum tricomutum. N-deficiency is used to induce the accumulation of storage carbon, and nitrate is provided to N-starved cultures to create a respiratory demand for this carbon. The results will describe changes in the macromolecular pools inside the cells during the transitionfromN-sufficiency to N-starvation, andfinallyto N-sufficiency again.  18  1.2)  MATERIALS AND METHODS  1.2.1)  Culture conditions The growth medium used in all experiments was enriched artificial seawater medium  (ESAW), made up in 20 L batches as described by Harrison et al. (1980), and modified by Thompson et al. (1991). The medium was further modified by increasing the macronutrient concentrations of NaN0 or NH4CI and Na Si0 .9H 0 to 1 mM, and NaH P0 to 200 uM. 3  2  3  2  2  4  This was to ensure that these macronutrients were present in sufficient concentrations to support the growth of high-density cultures. The silicate was dissolved separately in deionized distilled water (DDW) and the pH adjusted to approximately 8.0 with 50% HC1 before it was added to the rest of the medium. The medium was buffered to pH 8.8 by adding 20 mM N-[2hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] (HEPES) and 20 mM NaOH. The freshly prepared medium was filter sterilized through a 0.22 pm Millipore™ GS membrane filter into a 20 L sterile polypropylene reservoir. The reservoir was bubbled at 1200 mL-min" with sterile, 1  humidified air supplemented with 0.5% C 0 for 24 h before the medium was used. This 2  bubbling procedure adjusted the medium pH to 7.8. Stock cultures of Phaeodactylum  tricornutum  (clone CCMP630, Provasoli-Guillard  Center for Culture of Marine Phytoplankton, Bigelow Laboratory for Ocean Sciences, West Boothbay Harbor, Maine, USA) were maintained axenically in 40 mL glass tubes in nitratecontaining ESAW. Temperature was held constant at 18°C by immersing the tubes in a temperature controlled water bath. Light was provided by four 40 W Vitalite™ fluorescent tubes to give a photosynthetic photon flux density of 200 pmol-photons-m'^s", as measured 1  with a Biospherical Instruments™ light meter fitted with a QSL-100 probe. For experiments, 1 L flat-sided glass bottles were used to grow the cultures. Prior to their first use, the bottles  19 were treated with Surfasil™ siliconizingfluid(Pierce, Rockford, Illinois, USA) to minimize cell adhesion to the glass walls. Typically, four 5 mL aliquotsfroma single stock culture were used to aseptically inoculate quadruple culture bottles for a single experiment. Each bottle was sealed with a sterile silicon stopper containing the glass tubing necessary to aerate the cultures and withdraw culture samples. The bottles were suspended by clamps in the water bath at the same temperature and light conditions as the stock cultures. Sterile silicon tubing connected each bottle to the medium reservoir. A series of valves was used to enable culture dilution, aeration, and withdrawal without removing the stoppers. Each culture was mixed and aerated by bubbling with airfromthe medium reservoir at 300 mL-min \ Cultures were periodically diluted so that the medium N source was never exhausted; this procedure typically maintained cell densities between 1 x 10 and 5 x io cells-mL" . Experimental manipulations began at least 6  6  1  one week following culture inoculation, which typically encompassed 8-10 generation times.  1.2.2) Growth measurements Biomass was approximated by monitoring in vivofluorescencewith a Turner Designs™ Model 10fluorometer.Approximately 5 mL of culture was transferred into a 13 x 100 mm glass tube and then measured directly in thefluorometer.Cultures were diluted to keep in vivo fluorescence readings between 10 and 100; this represented logarithmic phase growth under conditions of N-sufficiency. Biomass was calculated exactly by measuring cell numbers with a Coulter Counter™ model TAD" particle counterfittedwith a population accessory. For routine measurements, 100 uL of culture was diluted with 20 mLfresh3% NaCl as the electrolyte. As P. tricomutum is not spherical and coincidence readingsfromthe particle counter would sometimes reach 10% for dense cultures, counts were checked against haemocytometer estimates. Using this  20 comparison, the Coulter Counter™ returned accurate cell counts over all the experimental cell densities encountered (Appendix A).  1.2.3) Macromolecular constituents Cells were harvested for protein, carbohydrate (total carbohydrates, including simple sugars and polysaccharides, such as chrysolaminarin), and chlorophyll a measurements by filtration using either 25 mm Gelman™ A/E (nominal pore size 1.0 pm) or Whatman™ GF/F glassfiberfilters(nominal pore size 0.7 pm). Allfilterswere precombusted at 450°C for 4 h prior to use. Multiple samples werefilteredsimultaneously using a Millipore™ sampling manifold attached to a vacuum line. The applied vacuum was typically 10-15 mm Hg. All filters wererinsedwith 2><5 mLfresh3% NaCl to remove any compounds loosely adhering to the cell surface or present in the medium as cell exudates. Thefilterswere folded in half and stored in aluminum foil at -20°C for a maximum of one month before extraction and analysis. A blankfilterwasrinsedandfrozenwith each set of sample filters. For protein determinations, approximately 2.5 x i o cells werefiltered.Eachfilterwas 7  thawed and ground in solvent with a motorized Teflon-headed pestlefittedto a 5 mL PotterElvehjem grinding tube. The grinding tube was immersed in a beaker of cold water to reduce sample overheating. Filters were ground in 1 mL 10% trichloracetic acid (TCAA). The grindings were poured into a 15 mL polypropylene centrifuge tube and the grinding tube rinsed with 2 x 0.5 mL 10% TCAA. The centrifuge tubes were vortex-mixed and centrifuged for 5 min at maximum speed in an International Clinical Equipment (ICE) benchtop centrifuge. The supernatant was discarded, the pellet resuspended in 1 mL 1 N NaOH, and the tubes left to extract at room temperature for 30 min. The tubes were centrifuged as before, and triplicate 0.05 mL aliquots were transferred to disposable polystyrene semi-micro cuvettes. Protein was  21 determined in these aliquots according to Lowry et al. (1951). This involved adding 1 mL of freshly prepared reagent D to each cuvette, followed exactly 10 min later with 0.1 mL of diluted Folin-Ciocalteu phenol reagent (Appendix B). The contents of each cuvette were gently vortex-mixed, left to stabilize for 30 min, and the absorbance measured at 750 nm with a LKB Ultraspec II™ spectrophotometer against a bovine serum albumin (BSA) standard (Appendix B). For carbohydrate determinations, approximately 1 x 10 cells were filtered. Each filter 7  was thawed and placed in a 16 x 100 mm precombusted screwtop glass tube. To each tube was added 2 mL of 3 N H2SO4. The tubes were tightly capped with Teflon-lined caps, vortexmixed, and placed in a boiling water bath for 2 h. The tubes were removed from the water bath, cooled at room temperature for 5 min, uncapped, and triplicate 0.1 mL aliquots transferred into precombusted 13 x 100 mm glass tubes. Carbohydrate was determined in these aliquots according to Kochert (1978). Using this technique, the tubes were transferred to a boiling water bath and 10 uL 4:1 phenol:DDW was added to each tube. After 10 min, 1 mL of concentrated H2SO4 was added, the tubes vortex-mixed and left to cool at room temperature for 30 min. The contents of the tubes were transferred to disposable polystyrene semi-micro cuvettes, and their absorbance measured at 490 nm against a glucose standard curve (Appendix C). For chlorophyll a determinations, 2 mL of culture wasfiltered.Eachfilterwas thawed and placed in a 15 mL polypropylene centrifuge tube with 10 mL of 3:1 90% acetone.dimethylsulfoxide (DMSO). The tubes were capped, vortex-mixed, and extracted at 4°C for 24 h. This solvent extraction procedure gave better extraction efficiencies than 90% acetone alone (Appendix D). The tubes were centrifuged for 5 min, and chlorophyll a was determined in the supernatant fluorometrically (Parsons et al. 1984).  22  1.2.4) Carbon and nitrogen analyses Cells were harvested for determination of particulate organic carbon and nitrogen (POC and PON, respectively) by filtration using precombusted 13 mm Gelman™ A/E glass fiber filters. For each analysis, 2 mL of culture wasfiltered,and thefilterrinsed with 2 x 1 mL 3% NaCl. A blankfilterwas rinsed and frozen with each set of samplefilters.Thefilterswere stored for up to 1 month at -20°C in polystyrene petri dishes. Prior to analysis, thefilterswere dried at 40°C for 1 h and compressed into 8 x 5 mm tin cups. POC and PON were determined from the samefiltersusing a Carlo Erba CHN analyzer according to Verardo et al. (1990), and calibrated with sulfanilamide.  1.2.5) Fatty acid analysis Cells were harvested for quantitative fatty acid analysis byfiltrationusing precombusted 47 mm Whatman™ GF/Ffilters.Approximately 2 x io cells werefiltered,and the filters 8  rinsed with 2 x 5 mL 3% NaCl. Thefilterswere notfrozen,but were ground immediately using the same protocols described in section 1.2.3. A 2 mL volume of 2:1 methanol: chloroform (both reagents Optima™ gradefromFisher Scientific, Vancouver, BC, Canada) was used as the solvent. The grindingsfromeachfilterwere poured into a precombusted 13 * 100 mm screwtop glass tube, and the grinding tube rinsed with 2 x 1 mL of solvent. The tubes were tightly capped with Teflon-lined caps and placed in the dark at 4°C for 2 h. The tubes were then centrifuged in an ICE centrifuge at maximum setting for 5 min, and the supernatant poured intofreshtubes. The pellets were re-extracted with 2:1:0.8 methanol:chloroform:DDW, vortex-mixed, and centrifuged as before. The supernatants were pooled with thosefromthefirstextract and combined with 1 mL DDW and 1 mL chloroform.  23 A crude lipid extract was prepared using a procedure modifiedfromBligh and Dyer (1959). The tubes were capped, shaken, and centrifuged for 1 min to separate the phases. The lower phasesfromthe tubes, which contained lipid, were aspirated intofreshtubes using precombusted glass pasteur pipets. The remaining polar phase was extracted twice more with 1 mL chloroform, and the aspirated lower phases pooled. The pooled lower phases were dried under a stream of N2 at 40°C for 45 min, the tubes capped under N , and stored in the dark at 2  -20°C for up to 1 month before further analysis. For fatty acid analysis, the crude lipid extract was redissolved in exactly 100 pL 3:1 chloroform.hexane (both solvents Optima™ grade). This extract was subsampled for total fatty acid and triacylglycerol (TAG) fatty acid analyses. For total fatty acid analysis, duplicate 20 uL samples were transferred into precombusted 13 x 100 mm screwtop glass tubes. For TAG fatty acid analysis, duplicate 20 pL samples were transferred onto Waters Sep-Pak™ Silica Light solid phase extraction (SPE) columns. TAGs were elutedfromthe columns using 10 mL 3:1 chloroform:hexane as the eluent. Details concerning column efficiency and specificity are given in Appendix E. The eluate was collected under vacuum in precombusted 13 x 100 mm screwtop glass tubes. Afixedvolume of 10 uL (10 pg) heneicosanoic acid (21.0) dissolved in chloroform was added to all tubes to act as an internal standard for subsequent transmethylation and gas chromatography (this fatty acid was not present in sample extracts). The tubes containing the total fatty acid and T A G extracts were dried in a vacuum dessicator in the dark for 24 h. The extracts were redissolved in 200 pL hexane and 1 mL 0.25 N methanolic HC1 (Supelco, Toronto, ON, Canada). The tubes were tightly capped with Teflon-lined caps, vortex-mixed, and the contents transmethylated at 85°C for 1 h. The tubes were cooled for 5 min, uncapped, and 0.5 mL of 0.9% NaCl and 200 uL hexane was added to each tube. The tubes were vortex-mixed, and the top hexane layer containing fatty acid methyl  24 esters (FAMEs) carefully aspirated into polypropylene microcentrifuge tubes. A further 200 pL hexane was added to the tubes, the aspiration procedure repeated, and the FAME extracts pooled. The contents of the microcentrifuge tubes were dried under vacuum for 1 h, redissolved in 35 pL hexane, and transferred into clean 200 pL glass inserts fitted into gas chromatography vials. The vials were capped with Teflon-lined caps, and the FAMEs analyzed on a Hewlett Packard 5890A gas chromatograph fitted with a 30 m long, 0.25 mm internal diameter type DB-23 (50% cyanopropyl) column. Helium was used as the carrier gas at 30 mL-min" , and FAMEs were detected with aflameionization detector. For more complete 1  details and a sample output trace, see Appendix E. Most FAMEs were identified by comparison of retention times with mixtures of authentic standards (Sigma Chemical Co., St. Louis, Missouri, USA). The FAMEs of 16:2, 16:3, and 16:4 were commercially unavailable, and were identified by comparing their retention times with known FAME profiles from Arabidopsis  thaliana seed extracts. Fatty acid quantities were calculated by the Hewlett  Packard ChemStations™ program using peak area and response factor data obtained for sample, standard, and internal standard FAMEs.  1.2.6) Medium nitrate Medium nitrate concentrations were determined by an adaptation of the ultraviolet spectroscopy method described by Cawse (1967). Using this method, 1 mL culture samples were collected in polypropylene microcentrifuge tubes and centrifuged in an Eppendorf model 5414 microfuge at 16000 x g for 2 min. Triplicate 0.3 mL aliquots of the supernatant from each sample were vortex-mixed with 0.7 mL DDW in glass tubes. The ultraviolet absorbance of the diluted supernatants were measured in semi-micro quartz cuvettes at 200 nm against a NaNC>3 standard curve made up in N-deficient ESAW (Appendix F). For some analyses,  25 supernatant samples were diluted 30 * in DDW and the nitrate concentrations determined with a Technicon® Autoanalyzer using the procedures described by Wood et al. (1967). This was necessary for stationary phase cultures, where background ultraviolet absorbance was high and probably due to dissolved organic compounds in the medium. A comparison of the two methods is given in Appendix F.  1.3)  RESULTS  1.3.1) Nitrate addition to logarithmic, early stationary, and late stationary phase cultures P. tricomutum  cultures were set up and sampled in logarithmic phase (N-sufficient),  early stationary phase (24 h N-starved), and late stationary phase growth conditions (1 week Nstarved). Cell density and in vivo fluorescence data indicated active cell division for logarithmic phase cultures, slowed cell division for early stationary phase cultures, and no cell division for late stationary phase cultures (Fig. 1.1). In vivofluorescencevalues increased before cell division resumed after early or late stationary phase cultures were diluted withfreshmedium containing 1 mM nitrate (Fig. 1.1). These patterns of cell density and in vivo fluorescence changes were correlated with changes in the medium nitrate concentration (Fig. 1.2). Following dilution, the medium nitrate was consumed more rapidly in early and late stationary phase cultures than in logarithmic phase cultures (Fig. 1.2). This was correlated with higher cell densities observed for early and late stationary phase cultures compared to logarithmic phase cultures following dilution (Fig. 1.1). The concentrations of the N-containing pools (PON, chlorophyll a, and protein) within P. tricomutum  remained constant before and after dilution of logarithmic phase cultures (Fig.  26 1.3). All these pools decreased following 24 h of N-starvation, with little further decrease after 1 week of starvation (Fig. 1.3). For early and late stationary phase cultures, dilution with fresh medium resulted in a recovery of all pools to concentrations similar to those found in logarithmic phase cultures (Fig. 1.3). The concentration of the POC and fatty acid pools within P. tricornutum remained constant before and after dilution of logarithmic phase cultures, but carbohydrate increased slightly (Fig. 1.4). The most marked increases in the carbon-containing pools occurred during the 1 week of N-starvation between early and late stationary phase cultures (Fig. 1.4). Of the three pools, total fatty acids increased the most (Fig. 1.4C). For early stationary phase cultures, dilution had little effect on the POC (Fig. 1.4A) or total fatty acid (Fig. 1.4C) pools; however carbohydrates were consumed (Fig. 1.4B). For early and late stationary phase cultures, the POC pool actually increased following dilution (Fig. 1.4A), although both the carbohydrate and total fatty acid pools decreased (Fig. 1.4B,C). This anomaly could be explained by taking into account that a certain percentage of the protein and chlorophyll a pools, which dramatically increased following dilution of early and late stationary phase cultures (Fig. 1.3), would have contained carbon. In the lipid fraction of the carbon-containing pools, most of the fatty acids were associated with polar lipids for logarithmic and early stationary phase cultures (Fig. 1.5 A). The polar lipid fatty acid pool changed very little before or after dilution for both of these cultures. In late stationary phase cultures, there was a small increase in the polar lipid fatty acid pool, but a 16-fold increase in the T A G fatty acid pool (Fig. 1.5B). Almost half of the T A G pool fatty acids decreased 24 h after dilution of late stationary phase cultures, with no corresponding increase in the polar lipid fatty acid pool.  27 The concentrations of individual fatty acids in the polar lipid and T A G fractions were also measured before and after dilution of logarithmic, early stationary, and late stationary cultures. Figure 1.6 shows the fatty acid profile for logarithmic phase cultures. The predominant fatty acids in the polar lipid fraction were tetradecanoic (14:0), hexadecanoic (16:0), cis-9hexadecenoic (16:1), and eicosapentaenoic (20:5) (Fig. 1.6A). In contrast, the T A G fraction was almost exclusively made up of 16:0 and 16:1 fatty acids (Fig. 1.6B). In logarithmic phase cultures, most of the fatty acids were present in the polar lipidfraction.Dilution of these cultures only slightly changed the fatty acid profiles for both the polar lipid and T A G fractions. Figure 1.7 shows the fatty acid profile for early stationary phase cultures. The fatty acid profile for the polar lipidfractionwas much the same as for logarithmic phase cultures, except for an overall decrease in the concentrations of all fatty acids (Fig. 1.7A). In contrast, there was a marked increase in 16:0 and 16:1 in the TAGfractionwhen compared to logarithmic phase cultures (Fig. 1.7B). Dilution of early stationary phase cultures resulted in an increase of all fatty acids in the polarfraction,but a decrease in 16:0 and 16:1 from the T A G fraction. Figure 1.8 shows the fatty acid profile for late stationary phase cultures. The most notable trend is an extremely large increase in 16:0 and 16:1 in the TAGfractionwhen compared to logarithmic or early stationary phase cultures (Fig. 1.8B). Approximately one third of these fatty acids were degraded 24 h after dilution of the late stationary phase cultures. For all three cultures, the ratio of 16:1/16:0 in the total lipid pool was positive before and after diltution (Table 1.1). However, there was a marked decrease in this ratio as the cultures progressedfromlogarithmic to late stationary phase, indicating a preferential synthesis of 16:0 over 16:1 as a storage fatty acid. After dilution of early and late stationary phase  28 cultures, there was an increase in the ratio towards logarithmic culture values, indicating a preferential degradation of 16:0.  1.3.2) Extended monitoring of late stationary phase cultures after nitrate addition Cultures of P. tricornutum were grown and N-starved for 1 week (late stationary phase). At the end of the starvation period, the cultures were diluted with fresh medium containing 1 mMNaN0 . These cultures were monitored during and beyond the 24 h period 3  described in section 1.3.1. Under these conditions, cultures exhausted the medium nitrate in the freshly added medium 36 h after addition, and then re-entered a period of N-starvation. Figure 1.9 shows the changes in the fatty acids from the polar and T A G lipid fractions. Degradation of the fatty acids from the TAGs began 12 h after culture dilution, and continued until 36 h. During this time period, polar lipid fatty acids increased. However, T A G fatty acids decreased by 5 pg whereas polar lipid fatty acids increased by only 1 pg. When the medium nitrate was exhausted, there was a subsequent slow increase in TAG fatty acids and a decrease in polar lipid fatty acids for the remainder of the experiment. Figure 1.10 shows the changes in the carbon-containing pools for the same experiment. During the first 36 h after culture dilution, there was a net degradation of fatty acids, a decrease then an increase in carbohydrates, and a slight decrease in POC. When the cultures re-entered N-starvation, both carbohydrates and fatty acids slowly increased for the remainder of the experiment. POC decreased initially after the cultures re-entered N-starvation, and then remained constant for the remainder of the experiment.  29  0  24  48  72  96 120 144 168 192 216 240 264 288 312 336 Time from inoculation (h)  Fig. 1.1. Changes in biomass over several dilution cycles for semicontinuous cultures of P. tricornutum. Cultures were sampled, diluted with fresh medium containing 1 mM NaN03, and sampled again 24 h later. The dotted lines indicate the times when dilutions were made to supply nitrate. Results show logarithmic phase cultures before (A) and 24 h after (B) dilution; early stationary phase cultures before (C) and 24 h after (D) dilution; and late stationary phase cultures before (E) and 24 h after (F) dilution. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  30  0  24  48  72  96 120 144 168 192 216 240 264 288 312 336 Time from inoculation (h)  F i g . 1.2. Changes in medium nitrate concentration over several dilution cycles for semicontinuous cultures of P. tricornutum. Samples were taken as described in Fig. 1.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  31  0.20  TJ o  0.15  o.  ".2  0.10  S •§ U  A B log  C D E F early stat late stat. Growth Phase  &0  0.00  A B log  C D E F early stat. late stat Growth Phase  ] Before N addition I 24 h after N addition  O  3  0.05  4 h  (3 O  A  B log  C D E F early stat. late stat Growth Phase  Fig. 1.3. Nitrogen containingfractionsfor P. tricomutum. Bars represent three cell cycle phases; logarithmic (A), early stationary (C), and late stationary (E), and measurements taken 24 h after nitrate addition to each of the three culture phases (B, D, and F, respectively). Graphs are shown for PON (A), chlorophyll a (B), and protein (C). Results are means ± 1 standard error for 4 cultures.  32  B. Carbohydrate  A  B log  C D E F early stat. late stat. Growth Phase  log  early stat. late stat. Growth Phase  C. Total fatty acids  Before N addition 24 h after N addition  log  early stat. late stat. Growth Phase  Fig. 1.4. Carbon-containing fractions for P. tricomutum. Bars represent the sampling times indicated on Fig. 1.3. Graphs are shown for POC (A), carbohydrate (B), and total fatty acids (C). Results are means ± 1 standard error for 4 cultures.  33  <L> O  ao  w •O  o  Kt  8 r  A. Polar lipid fatty acids  B. T A G fatty acids  7 6 r 5 4  CM  'a, o OH  3  I  2 1 0  A B log  C D E F early stat. late stat. Growth Phase  A  B log  C D E F early stat. late stat. Growth Phase  Before N addition 24 h after N addition  Fig. 1.5. Lipid-containing fractions for P. tricornutum. Bars represent the sampling times indicated on Fig. 1.3. Graphs are shown for polar lipid fatty acids (A), and T A G fatty acids (B). Results are means ± 1 standard error for 4 cultures.  34  Logarithmic Phase A. Polar lipid fatty acids  ©  (N|  Tfr  V©  v£>  MD  MD  VO  OO  OO  00  00  OO  O  O  O  O  O  <N  <N  (N  Fatty acid identity Fig. 1.6. Fatty acid components of the polar (A) and TAG (B) lipid fractions for P. tricomutum. Bars show changes in the concentrations for individual fatty acids before and after dilution of logarithmic phase cultures withfreshmedium containing 1 mM NaN03 (times A and B on Fig. 1.1). The fatty acid identity is given as x:y, where x is the carbon number, and y is the double bond number. Results are means ± 1 standard error for 4 cultures.  35  Early Stationary Phase 300  A. Polar lipid fatty acids  250 200 150 ^  100  i  1 50 I3 500 0  £  n  n  B. TAG fatty acids  400  n I  I Before N addition 24 h after N addition  I  300 200 100 0  JjJ .1 o  ©  o  ri  o  o  >T"i  v© vb  ts| wi  MD  „ n _ TJ-  o  n  <s cn  n  n  © -7< ts  vb' <b do do do' do do o  ©  ©  o  <•/> o  ©  ri  jg  r i r i .g  Fatty acid identity Fig. 1.7. Fatty acid components of the polar (A) and TAG (B) lipid fractions for P. tricornutum. Bars show changes in the concentrations for individual fatty acids before and after dilution of early stationary phase cultures with fresh medium containing 1 mM NaN03 (times C and D on Fig. 1.1). Results are means ± 1 standard error for 4 cultures.  36  1250  Late Stationary Phase A. Polar lipid fatty acids  1000 750 500 u  i  250  Lull  ¥ 0 B. TAG fatty acids '§ 4000 X I* 3500 ^ 3000 2500 2000 1500 1000 500 0 o ©  o  CS  <Z> <Z> ~ *jr'  J5n  Before N addition 24 h after N addition  CA  •<? p  vo vb vb vb vb do do oo  T-i  CS  00 00  o  o o  CS CS  Fatty acid identity  o o  <N CS  CS  CS  CS CS  CS  cs  Fig. 1.8. Fatty acid components of the polar (A) and TAG (B) lipidfractionsfor P. tricomutum. Bars show changes in the concentrations for individual fatty acids before and after dilution of late stationary phase cultures withfreshmedium containing 1 mM NaN0 (times E and F on Fig. 1.1). Results are means ± 1 standard error for 4 cultures 3  37  0  12  24  36  48  60  72  84  96  108 120  Time from N addition (h) Fig. 1.9. Polar and TAG lipid fractions in P. tricornutum cultures N-starved for 1 week  and then diluted with fresh medium containing 1 mM NaN0 at t = 0 h. The nitrate in the medium was exhausted 36 h following dilution, as indicated on the figure with an arrow. Results are means ± 1 standard error for 4 cultures. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. 3  38  0  12  24  36  48  60  72  84  96  108 120  Time from N addition (h) Carbon-containing fractions in P. tricornutum cultures N-starved for 1 week and then diluted with fresh medium containing 1 mM NaN0 at t = 0 h. The nitrate in the medium was exhausted 36 h following dilution, as indicated on the figure with an arrow. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. Fig. 1.10.  3  39  Table 1.1. 16:1/16:0 fatty acid ratios in P. tricomutum. Ratios were calculated from GC  derived values for total fatty acids in logarithmic, early, and late stationary phase cultures (time periods A, C, and E on Fig. 1.1). Results are means ± 1 standard error for 4 cultures.  16:1/16:0 ratio (w/w)  Growth Phase  before N addition  after N addition  logarithmic  1.853 ±0.007  1.663 ±0.012  early stationary  1.3.46 ±0.002  1.806 ±0.014  late stationary  1.171 ±0.006  1.316± 0.012  40  1.4)  DISCUSSION  This study has examined how N-deficiency, and N resupply in the form of nitrate, influenced carbohydrate and lipid accumulation and degradation in the marine diatom P. tricornutum. The following sections will discuss the fatty acid profile (section 1.4.1), and how  carbohydrate and lipid accumulation may be coordinated and regulated in diatoms during the process of N-starvation (section 1.4.2). Finally, the coordination and regulation of carbohydrate and fatty acid degradation will be discussed (section 1.4.3).  1.4.1) Signature fatty acid pools  There has been interest in the fatty acid composition of microalgae for chemotaxonomy, both for the algae and their consumers. It is generally assumed that stable signature fatty acid profiles can be obtained from different species and used for classification purposes (Skerratt et al. 1995, Zhukova and Aizdaicher 1995). For diatoms, the marker fatty acids are 16:0, 16:1, 14:0,  and 20:5 (Oreutt and Patterson  1975).  These fatty acids are typically the most abundant.  The results from this study indicate that this assumption is fairly robust; these fatty acids always predominated at all growth phases before and after the addition of nitrate. In P. tricornutum, high production rates of 20:5 have been correlated to culture densities that support the highest growth rates (Chrismadha and Borowitzka 1994, Molina Grima et al.  1994,  Yongmanitchai and Ward 1991). This relationship has also been found in  the red alga Porphyridium omentum (Cohen et al.  1988).  This is consistent with the results  from this study, which show that most 20:5 is associated with polar (structural) lipids rather than TAGs. Arao et al.  (1987)  found that the 20:5 in P. tricornutum is a major component of  galactolipids and phospholipids. Thus, 20:5 would be integrated into organelle and cell  41 membranes, and not significantly accumulate when TAG synthesis increases as growth rate decreases. This would mean that 20:5 would not be a good chemotaxonomic marker, because its contribution to the total cell fatty acids would dramatically decrease as the diatom population becomes increasingly N-starved. The ratio of 16:1/16:0 was always positive in this study, and is in agreement with the pattern found for several diatoms (Skerratt etal. 1995, Zhukova and Aizdaicher 1995). The ratio of 16:1 to 16:0 has also been evaluated as a trophic level marker for diatoms as food sources for consumers (Kattner and Brockmann 1990, Skerratt et al. 1995). The results indicate that this may be inappropriate, because this ratio decreases as a consequence of Nstarvation. However, although the ratio is not stable, it does appear to change in response to phosphorus deficiency in much the same way as N-deficiency. Siron et al. (1989) found that Psufficient P. tricomutum had a 16:1/16:0 ratio of 1.86 (1.85 for N-sufficient cultures in this study) and 1.24 for 7 d P-deficient cultures (1.17 for 1 week N-starved cultures in this study). Therefore, the 16:1/16:0 ratio does appear to behave predictably according to nutrient deficiency and/or a reduction in the growth rate. However, in order to use this ratio as a tracer through the food chain, the physiological state of the food source must be known. One potentially useful outcome of thesefluctuationsin the 16:1/16:0 ratio could be in hindcasting the physiological status of a phytoplankton population. If afishspecies is known to derive its tissue fatty acidsfromultimately one or two phytoplankton species, the tissue 16:1/16:0 ratio might be used to infer at what growth stage the phytoplankton were consumed by the zooplankton; i.e. pre bloom (high ratio) or post bloom (low ratio). This may also provide useful information on the food source preference for consumers - not only at the level of classifying the producer, but also at the level of predicting producer physiological state at the time of consumptionfromthe fatty acid ratios in the consumer's tissues.  42  1.4.2) Patterns of storage carbon biosynthesis Storage lipid is unlikely to play a major metabolic role in diatom populations that are N starved in the short term. The results indicated that while there was a significant increase in TAG-specific fatty acids after one week of N-starvation, there was almost no increase after 24 h. For several diatoms, carbohydrates accumulate in the initial stages of N-deficiency (Harrison et al. 1990), whereas lipids accumulate after prolonged N-deficiency (Healey 1973, Thomas et al. 1984), or when cell division completely ceases in stationary phase (Opute 1974, Taguchi et al. 1987, Kattner and Brockmann 1990, Parrish and Wangersky 1990). In the freshwater diatom Melosira italica, sinking cells were shown to always preferentiallyfixcarbon into polysaccharides, and then lipids, regardless of the physiological state of the population (Amblard and Boudier 1990). These observations suggest that storage lipid synthesis is a consequence of excess photosynthate becoming available when triose phosphate synthesis slows down. Considering that the calorific value of fatty acids is 9 kcal-g" , whereas the value 1  for carbohydrates is only 4 kcal-g' , lipid synthesis is a highly efficient way to store carbon 1  (Stryer 1988). Therefore, a reduction in carbohydrate synthesis and an increase in storage lipid synthesis during stationary phase is energetically advantageous, because the limited volume of the cell can then be used to store the more concentrated energy source. Parnas and Cohen (1976) suggested that the accumulation of storage materials in unicellular algae is not just a consequence of an excess of photosynthate, but a deliberate, controlled switch to storage material synthesis. In order to discuss the regulation of carbohydrate vs. storage lipid synthesis, the specific identities of the relevant carbon compounds and their biosynthetic pathways needs to be known.  43  The principal storage carbohydrate in marine diatoms is a 13-1,3 linked glucan similar to starch, known as chrysolaminarin (Myklestad 1989). Chrysolaminarin has been described in P. tricornutum (Ford and Percival 1965). Chrysolaminarin synthesis appears to be regulated in diatoms by the activity of uridine diphosphate (UDP) glucose pyrophosphorylase, an enzyme which catalyzes the formation of UDPglucose from glucose. Roessler (1987b) found that the activity of this enzyme was stimulated in the marine diatom Cyclotella cryptica by the divalent cations M g  2+  and Mn at a pH of 7.8; these are the conditions which also stimulate the activity 2+  of Calvin cycle enzymes. This suggests that chrysolaminarin synthesis is linked to photosynthetic carbonfixation.However, the enzyme still remained active in the dark, and UDPglucose pyrophosphorylase activity in C. cryptica was not stimulated by 3phosphoglycerate or inhibited by inorganic orthophosphate, as ADP glucose pyrophosphorylase is in higher plants (Roessler 1987b). This suggests that chrysolaminarin may be synthesized outside the chloroplast in diatoms, which would be differentfromthe usual pattern of chloroplastic starch synthesis in higher plants. The results from this study have shown that the principal storage fatty acids in P. tricornutum are 16:0 and 16:1, and that these fatty acids are accumulated in the TAGs. This same pattern of fatty acid accumulation was observed for P. tricornutum under phosphate deficiency (Siron et al. 1989). In higher plants, TAGs accumulate in the seed cotyledons, endosperm, orfruitmesocarp (Stymne and Stobart 1987). In P. tricornutum, TAGs accumulate in oil bodies in the cytoplasm (Larson and Rees 1996). In seeds, fatty acids are synthesized in the plastid, and it is thought that acyl-CoA units are exported to the endoplasmic reticulum, where TAG synthesis occurs (Stymne and Stobart 1987). The synthesis of TAG fatty acids may be regulated at the level of primary fatty acid synthesis via acetyl-CoA carboxylase (ACCase), or at the level of TAG synthesis. From the  44 results, it is obvious that TAG synthesis must be regulated, because the quantities of 16:0 and 16:1 in the TAG pool dramatically increased relative to other fatty acids as the cells became progressively more N-starved. This would require an increase in phosphatidase activity, which is thought to be the rate limiting enzyme in the production of diacylglycerol precursors for TAG synthesis (Stymne and Stobart 1987). It is also probable that de novo fatty acid synthesis was increased, given the 16-fold increase in the size of the TAG fatty acid pool observed for stationary phase cultures in this study. In barley and maize leaves, studies where acetyl-CoA carboxylase was inhibited by herbicides revealed this enzyme to be the major flux controlling enzyme for light-stimulated lipid synthesis (Page et al. 1994). In addition, Roessler (1987a) found that ACCase activity was enhanced during the onset of silicon deficiency in C. cryptica. Enhanced de novo fatty acid synthesis may also be a consequence of enhanced plastidic glycolytic activity, with the result that substrate availability to ACCase is increased. In developing cultured canola embryos, the pyruvate/phosphoenolpyruvate ratio increases and plastidic pyruvate kinase activity increases as the embryo matures, indicating an increasing, regulated flux towards acetyl-CoA production from the pyruvate dehydrogenase (PDH) complex (Sangwan et al. 1992). However, Kang et al. (1994) found that ACCase activity was in excess of quantitative fatty acid synthesis in developing Brassica napus embryos, suggesting that some other part of the fatty acid biosynthetic pathway is regulated. At the very least, it appears that carbohydrate synthesis must occur first in diatoms (and developing oilseeds) to provide the necessary carbon substrates for lipid biosynthesis. In developing sunflower seeds, Luthra et al. (1991) found that starch synthesis increased between 5 and 15 days after flowering (DAF), whereas after this period, starch content decreased and fatty acid biosynthesis increased. Similarly, Roessler (1987c) found that the activity of  45 chrysolaminarin synthase decreased in C. cryptica after 4 h of silicon deficiency, whereas ACCase activity increased, suggesting a regulatory switchfromcarbohydrate to lipid synthesis in this diatom. In sunflower seeds, Luthra et al. (1991) suggested that the initial accumulation of starch was to provide a transient reserve material that could be metabolized for the biosynthesis of lipids. Their conclusion was based on observed increases in the activities of malic enzyme, isocitrate dehydrogenase, and succinate dehydrogenase at 15 DAF, which could provide the NADH and ATP necessary for fatty acid biosynthesis from starch degradation in the plastid. Thus, the presence of a starch reserve in the same compartment (the plastid) as the site of fatty acid synthesis would enable the initiation of lipid synthesis to be not only controlled by ACCase activity and substrate availability, but also by the availability of NAD(P)H and ATP. Similarly, in developing castor bean seeds, TAG synthesis is coordinated with ribulose bisphosphate carboxylase (RUBISCO) activity, the oxidative pentose phosphate pathway (to supply reducing equivalents), and plastidic glycolysis (Simcox et al. 1979). In conclusion, it seems likely that the control of carbonflowto carbohydrate or lipid synthesis is most likely regulated by the coordinate regulation of key enzymes such as ACCase and the PDH complex for lipid synthesis, and UDPglucose pyrophosphorylase activity for chrysolaminarin synthesis.  1.4.3) Patterns of storage carbon degradation When unicellular algae recover from N-deficiency, the photosynthetic rate is insufficient to supply the necessary reductant and carbon skeletons required to maintain rapid N assimilation and amino acid synthesis (Turpin 1991). Diatoms have a low photosynthetic rate in the stationary phase of culture; a rate 7-fold less than the rate for logarithmic phase cells has been reported for P. tricomutum (Glover and Morris 1979).  46 Therefore, reducing power and carbon must come from stored carbon compounds rather than photosynthesis during the recovery period in N-starved stationary phase cultures. The results show that 16:0 and 16:1 were the fatty acids that changed most dramatically in P. tricornutum during the onset and recovery from N-starvation. These same fatty acids were shown to increase in the light and decrease in the dark in the diatom Thalassiosira pseudonana, implying their usefulness as respiratory substrates (Fisher and Schwarzenbach 1978). However, in the marine diatom Skeletonema costatum, Handa (1969) found that chrysolaminarin was respired in the dark, whereas lipid was not respired until the stored carbohydrates had been degraded. In this study, N-starved cells showed the symptoms of Nstarvation after 24 h; namely a dramatic decrease in PON, chlorophyll a, and protein. All these N-containing pools within 24 h N-starved cells recovered after nitrate addition, without any corresponding decrease in fatty acids. These results suggest that the carbon necessary for the synthesis of new protein most likely originated from the degradation of stored carbohydrates or recent photosynthate, and not lipid. This trend has been observed in polar diatoms that are known to store large quantities of lipid; protein synthesis is fueled by either recent photosynthate or carbohydrates, while the lipid pool remains intact (Thomas and Gleitz 1993). There may be two reasons why carbohydrate degradation is activated before storage lipid degradation in diatoms. Firstly, carbohydrates may be preferentially synthesized before storage lipids. This has been discussed in section 1.4.2. The consequence of a large carbohydrate pool and a small lipid pool would be that the former would be more readily available for degradative metabolism. This may explain why, after 24 h of N-starvation, carbohydrates appear to be preferentially degraded. However, even when there is a substantial pool of storage lipid available, for example after 1 week of N-starvation, the results indicated  47 that carbohydrates were degraded before storage lipid fatty acids. Therefore, some mechanism other than substrate availability must be regulating the timing of storage lipid metabolism. Secondly, unlike carbohydrates, storage lipids may not function as a directly respired energy source. Carbohydrates may be preferentially respired to provide the reductant and ATP for biosynthetic reactions, while the storage lipid carbon may be partially metabolized for the biosynthesis of other macromolecules. One possibility is that there may be a preferential synthesis of polar, structural lipid fatty acids using TAGs as the precursors. This process could supply the membrane components for the chloroplasts, which would need to be resynthesized after a period of degradation caused by N-deficiency. The results show that polar fatty acids increased for 36 h following N addition to 1 week N-starved cultures (Fig. 1.9). The transfer of fatty acidsfromTAGs to polar lipids has been suggested for the fungus Glomus versiforme (Gaspar et al. 1994). However, this is unlikely to be wholly true in this study because of the magnitude of the degradation of storage lipid fatty acids, which decreased a factor of 3 times more than polar lipid fatty acids increased. Thus, while some of the fatty acids associated with storage lipid may be used in polar lipid synthesis, it appears at least two-thirds of the carbon is degraded and used for other processes. Another possible sink for TAGs is in carbohydrates. In higher plant seeds, the main role of TAGs is in gluconeogenesis during germination. The TAGs are degraded by lipases, the component fatty acids are oxidized to acetyl-CoA by thefi-oxidationcycle, the acetyl-CoA is oxidized to malate by the glyoxylate cycle, and this malate is used in gluconeogenesis to form glucose (Leegood and ap Rees 1978a, Andrews and Ohlrogge 1990). This glucose can then be readily translocated to the growing meristems of the seedling, providing energy by its catabolism until sufficient photosynthate is availablefromthe developing leaves. This process of gluconeogenesis requires strict regulation of the enzymes associated with storage lipid  48 degradation by carbohydrates or the intermediates in carbohydrate metabolism. The results suggest that this was occurring, because there was no net decrease in fatty acids associated with TAGs until 12 h following nitrate addition to 1 week N-starved cultures, when carbohydrates appear to have decreased to an apparent minimum cell quota of 1 pg (Figs. 1.9 and 1.10). At the level of lipase regulation, the cytoplasmic portion of glycolysis may in some way inhibit lipase activity on oil body surfaces. Siddique and Bai (1991) found that the removal of photosynthate from peanut root nodules, achieved by detopping the plants, stimulated lipase activity. Unfortunately, nothing is known about in vivo lipase activity and regulation in diatoms. It has been demonstrated that accumulated carbohydrates can inhibit B-oxidation enzymes in castor bean seedlings (Huang and Beevers 1974). Additionally, mChlorella, glucose in the medium inhibits the activity of the glyoxylate cycle enzyme, isocitrate lyase (Syrett et al. 1963). Trevanion et al. (1995) showed that phosphoenolpyruvate carboxykinase, which produces phosphoenolpyruvate from glyoxysomally produced oxaloacetate, exerts control over the activity of gluconeogenesis in Cucurbita pepo cotyledons. However, gradient centrifugation studies of the diatoms Thalassiosira fluviatilis and Nitzschia laevis have suggested that the enzymes of B-oxidation are located exclusively in the mitochondria (Winkler and Stabenau 1995). Therefore, the acetyl-CoA generated would not be available for the glyoxylate cycle and gluconeogenesis, but would be exclusively respired in the TCA cycle. In this case, perhaps the presence of glycolytically produced acetyl-CoA or pyruvate inhibits the activity of mitochondrial 6-oxidation enzymes. These regulatory mechanisms are yet to be investigated in diatoms. Bewley and Black (1985) summarized the typical lipid mobilization pattern for higher  49 plant seedlings: Lipid is used as a last resort to sustain respiration; when the adenine nucleotide pool was high, the authors postulated these camefromcarbohydrate respiration, and the carbohydrates in turn were a product of gluconeogenesis. These authors view storage lipid as a stepping stone between heterotrophy and autotrophy in germinating seedlings, and that lipid metabolism in the absence of carbohydrate is a survival measure. This would also seem to be an appropriate statement for diatoms, where an accumulation of storage lipids only occurs to any significant degree during extreme physiological stress.  50  CHAPTER 2  N ASSIMILATION AND STORED CARBON MOBILIZATION DURING RECOVERY OF STATIONARY PHASE CULTURES F R O M N-STARVATION IN T H E LIGHT OR DARK WITH NITRATE OR A M M O N I U M  2.1)  INTRODUCTION  During N assimilation, carbon must be provided to generate the ATP, reductant, and carbon skeletons necessary for the assimilatory pathways and protein synthesis to function. During steady state growth in the light, much of the ATP and reductant required for these processes can comefromthe light reactions of photosynthesis, and triose phosphates produced by the Calvin cycle can be used as carbon skeletons. However, in the light, respiration may be stimulated following N addition to N-starved algae, indicating that photosynthetic carbon fixation and/or reductant generation is insufficient for rapid N assimilation (Turpin 1991). Photosynthetic carbonfixationmay even be repressed so that reductant generated by the light reactions is utilized for N assimilation rather than by the Calvin cycle (Elrifi and Turpin 1986). The steady state photosynthetic rate in stationary phase P. tricomutum cultures is also only 10% of the rate in actively dividing N-sufficient cultures (Larson and Rees 1996), and cell chlorophyll a quotas are reduced during N-starvation (Chapter 1). Finally, protein synthesis can occur in the dark in autotrophic organisms (Geider and Osborne 1989). Thus, carbon and reductant in addition to that immediately derivedfromphotosynthesis will be essential to maintain N assimilation in the dark, or in the light following a period of N-starvation.  51 One mechanism by which extra carbon can be sequestered for N assimilation is through B-carboxylation reactions. In the green alga Selenastrum minutum, PEP carboxylase (PEPCase) activity increases during dark ammonium assimilation, which suggests that increased flux of N through the assimilatory pathways provides a carbon demand that can be met by 8carboxylation (Schuller et al. 1990). In P. tricornutum, the ratio of ribulose bisphosphate carboxylase (RUBISCO):PEPCase decreases as the cells enter stationary phase (Glover and Morris 1979), which suggests that non-photosynthetic (dark) carbon fixation becomes quantitatively more important as growth rates decrease. The main limitation of B-carboxylation reactions is that there must be a continual supply of glycolytically derived substrate, namely PEP (for PEPCase). Thus, B-carboxylation requires the ongoing oxidation of stored carbon reserves. In addition, glycolysis or other oxidative processes must be occurring to provide the reductant necessary for N assimilation. Therefore, in the dark or during carbon-limited (as a consequence of insufficient recent photosynthate) N assimilation, carbohydrates or storage lipids must be degraded. There is good evidence to suggest that carbohydrates are utilized for N assimilation in N-starved microalgae. Ammonium assimilation in N-replete Chlamydomonas reinhardtii (Menancho and Vega 1989) and Selenastrum minutum (Arnory et al. 1991), which are both starch-storing green microalgae, was dependent on light and C0 . However, when these algae 2  were N-starved, ammonium was assimilated independent of light or C 0 , suggesting that the 2  endogenous starch reserves were being utilized for N assimilation. In the green microalga Selenastrum minutum, glycolytic breakdown of starch reserves and accompanying Bcarboxylation reactions have been proposed as an anaplerotic mechanism by which TCA cycle intermediates removed during protein synthesis can be replaced (Turpin 1991).  52  Given that lipids, particularly triacylglycerols (TAGs) accumulate in N-starved P. tricomutum (Chapter 1), it seems reasonable to propose that TAGs have a role in supplying the carbon for anaplerotic reactions in this diatom. This could happen via the glyoxylate cycle, where net C4 acid synthesis (primarily malate and succinate) occurs when acetyl-CoA from fatty acid B-oxidation combines with isocitrate in glyoxysomes (Kornberg 1966). One key marker enzyme that indicates the presence of the glyoxylate cycle is isocitrate lyase (ICL) (Giachetti et al. 1987). Glyoxysomal ICL has been identified in a number of higher plant seeds, fern spores, and algae (Cioni etal. 1981), and its activity has been measured in P. tricomutum (Larson and Rees 1996). In higher plants the role of the glyoxylate cycle is in gluconeogenesis, to provide a mobile carbon source (sucrose) that can be translocated through the phloem to the developing shoot. Storage oils are converted to carbohydrates necessary for growth before photosynthetic C 0 fixationis sufficient (Leegood and ap Rees 1978a, Gerhardt 1986, 2  Andrews and Ohlrogge 1990). However, in single celled organisms such as algae, there is no requirement for translocatable carbon sources; rather carbohydrates may need to be replenished within the cell, or the glyoxylate cycle may have an entirely different function. This chapter investigates the interactions between carbohydrate and storage lipid degradation during nitrate or ammonium assimilation in N-starved P. tricomutum. Nitrate assimilation has a higher energetic cost than ammonium assimilation, but this has not led to a readily discernible disadvantage, at least in terms of growth rates, for nitrate assimilation vs. ammonium assimilation when light is limiting (Thompson et al. 1989, Levasseur et al. 1993). Therefore, it may be possible that processes such as carbohydrate and lipid mobilization are regulated, according to the N source, to provide adequate or even excess carbon and reductant for the processes of N assimilation.  53  2.2)  MATERIALS AND METHODS  2.2.1) Culture conditions The growth medium (ES AW) was prepared with either 1 mM NaN03 or NH4CI as the N source, and stock cultures of Phaeodactylum tricomutum were inoculated into 1 L glass bottles as described in Chapter 1. For experiments conducted in continuous light, cultures were grown for 8-10 generation times in logarithmic phase (approximately 1 week) with either 1 mM NaN03 or NH4CI  as the N source. The cultures were then allowed to enter stationary phase, which was  defined as the point at which the medium N became undetectable by the analytical methods used. The point of N exhaustionfromthe medium, whether the cultures were grown on nitrate or ammonium, was typically equivalent to an in vivofluorescencereading of 100-120. Thus, in vivofluorescencewas used to routinely determine the onset of stationary phase in the cultures. The cultures were maintained in this N-starved state for 3 d before being diluted with approximately 3 volumes offreshmedium containing the same N source on which the cultures had originally been grown. The time zero samples were taken just prior to this dilution. Two or three subsequent dilutions were made for the remainder of the experiment to ensure that the medium N was not exhausted. For experiments conducted in the dark, cultures were grown for 8-10 generation times following the same protocols as for experiments done in continuous light. However, 24 h following the exhaustion of N from the medium, the culture bottles were wrapped in aluminum foil to exclude light. The cultures were maintained in this darkened state for 2 d (3 dfromthe exhaustion of N from the medium) before being diluted, in the dark, with approximately 3  54 volumes of fresh medium containing the same N source on which the cultures had originally been grown. Samples were taken as described for continuous light experiments, except that the vessels and apparatus used for sample handling were shielded from light with aluminum foil. Light was made available to these cultures 72 h after dilution by removing the aluminum foil wrapping. Using the same protocols as the continuous light experiments, the time zero samples were taken just prior to dilution. No further dilutions were made. The 72 h samples were taken just prior to making light available to the cultures.  2.2.2)  Growth  measurements  Biomass was measured by monitoring in vivo fluorescence and measuring cell numbers as described in Chapter 1.  2.2.3)  Macromolecular  constituents  andfatty  acids  Cells were harvested for carbohydrate and fatty acid measurements as described in Chapter 1.  2.2.4)  Nitrogen  measurements  Medium nitrate concentrations were determined as described in Chapter 1. For the dark experiments, only the method described by Wood et al. (1967) using the Technicon® Autoanalyzer was used for nitrate determinations. For cultures grown on nitrate, medium nitrite concentrations were also determined by running samples through the Technicon® Autoanalyzer without the cadmium reducing column in place. Medium ammonium concentrations were determined using a manual adaptation of the Technicon® Autoanalyzer procedure described by Slawyk and Maclssac (1972). Briefly, 3 x 0.5 mL aliquots of culture sample were obtained using the same technique that was used for nitrate measurements  55 (Chapter 1). To each aliquot was added 0.25 mL of reagent A followed by 0.25 mL of reagent B in a semi-micro polystyrene cuvette. The contents were vortex-mixed and the ammonium concentration was determined spectrophotometrically (Appendix G). Cell nitrate or ammonium concentrations were determined in dark N-recovered cultures from boiling water extracts as outlined by Thoresen et al. (1982). One mL of cells were collected on precombusted 25 mm Gelman™ A/Efiltersusing a Millipore™ sampling manifold attached to a vacuum line. Thefilterswere rinsed with 3 x 5 mL 3% NaCl before 10 mL of boiling DDW was pipetted onto eachfilterin their sample holders. Thefiltrateswere collected in precombusted 16 x 100 mm glass tubes. Thefilterswere rinsed with 5 mL of cold DDW, and thefiltratespooled with the boiling water extracts. The extracts were stored in capped 20 mL polypropylene vials at 4°C before analyses of either nitrate (Wood et al. 1967) or ammonium (Slawyk and Maclsaac 1972) using the Technicon® Autoanalyzer.  2.2.5) Enzyme assays 2.2.5.1) Nitrate reductase Cells were harvested for determination of in vitro nitrate reductase (NR) activity using precombusted 25 mm Gelman™ A/Efilters.Whatman™ GF/Ffilterswere found to inhibit NR activity by 30-50% compared to Gelman™ A/Efilters,so the former were not used. Triplicate aliquots containing approximately 5 x io cells werefilteredfor each assay using a Millipore™ 7  sampling manifold attached to a vacuum line. Thefilterswere rinsed with 2 x 5 mL 3% NaCl, folded twice, and inserted into 1.5 mL polypropylene microcentrifuge tubes. The tubes were capped and immediately plunged into liquid nitrogen. The samples were stored in liquid nitrogen in the dark for up to 7 d before being assayed; this storage time did not significantly reduce NR activity (Appendix H).  56 The filters were removed from the microcentrifuge tubes and ground, unthawed, in 1 mL N R buffer (Berges and Harrison 1995) with a motorized Teflon-headed pestle fitted to a 2 mL Potter-Elvehjem grinding tube. The grinding tube was immersed in a beaker containing an ice slurry to reduce sample overheating. The cell homogenates were pipetted into 1.5 mL polypropylene microcentrifuge tubes and stored on ice before further analysis. When 12 filters had been ground, the homogenates were centrifuged in an Eppendorf model 5414 microfuge at 16000 x g for 5 s, and 0.3 mL of the supernatant was assayed for N R activity according to Berges and Harrison (1995). The assay was carried out at 18°C in the light (200 pmol-photons-m'^s" ) for 15 min. N R activity assayed over 15 min was linear between 1.25 x 1  10 and 7.5 x io cells harvested (Appendix H). However, N R activity did not increase 7  7  proportionally with incubation time (Appendix H). Therefore, a 15 min incubation time was routinely used because it was the shortest practical time in which the assay could be simultaneously performed on multiple samples.  2.2.5.2) Glutamine synthetase Cells were harvested for determination of in vitro glutamine synthetase (GS) activity using the same methodology as was used for NR. Filters were stored and processed as for NR, except the buffer system described by Rees et al. (1995) was used, with the addition of 1% Triton X-100. GS activity was assayed in 0.2 mL of supernatant from the centrifuged homogenate at 18°C in the light (200 pmol-photons-nrV ) for 15 min. Linearity of GS activity 1  with cell concentration and incubation time were not checked, because this has already been tested for P. tricornutum (Rees et al. 1995). However, ammonium assimilation rates in vivo were found to match in vitro GS activities (Appendix I).  57  2.2.5.3) Isocitrate lyase Cells were harvested for determination of in vitro isocitrate lyase (ICL) activity using the same methodology as was used for NR. Filters were stored and processed as for NR, except the buffer system described by Larson and Rees (1996) was used with the addition of 1% Triton X-100. ICL activity was assayed in 0.25 mL of supernatantfromthe centrifuged homogenate at 30°C for 90 min. The activity was normalized to 18°C by conducting a simultaneous test assay with subsamplesfromthe same homogenate assayed at 18 and 30°C. By separately assaying four cultures, a conversion factor of 30°C values x 0.3125 was obtained and used to derive 18°C ICL activities. ICL activity assayed was linear between 1.25 x 10 and 7  7.5 x 10 cells harvested during 90 min incubation times (Appendix J). ICL activity was linear 7  over a range of incubation timesfrom30 to 120 min when 5 x 10 cells were assayed 7  (Appendix J). The absorption spectrum of the assay product, glyoxylate, was also found to closely match the absorption spectrum of commercially purchased glyoxylic acid (Sigma Chemical Co., St. Louis, Missouri, USA), and not other keto acids, indicating that ICL activity was in fact being assayed (Appendix J).  2.3)  RESULTS  2.3.1) Nitrate or ammonium addition to 3 dN-starved cultures in continuous tight P. tricomutum cultures grown on either 1 mM NaN0 or NH4CI were N-starved for 3 3  d in continuous light and then recovered with the same N source that the cultures had been grown on. Figures 2.1 and 2.2 show that both cell density and in vivofluorescencefor N -  58 starved cultures prior to dilution were both high and stable regardless of the N source on which the cultures had been conditioned. After either nitrate or ammonium addition to stationary phase cultures, there was no apparent cell division until 24 h following the initial dilution (Fig. 2.1), although in vivo fluorescence did increase after 12 h for both cultures (Fig. 2.2). The draw-down rate for nitrate or ammonium was similar between 6 and 18 h after N addition to either culture (Fig. 2.3). There was a difference between cultures in the amount of carbohydrate accumulated during the period of N-starvation. For cultures conditioned on nitrate, carbohydrate accumulated to a maximum of 7 pg-cell' during the starvation period, whereas the maximum 1  for ammonium-conditioned cultures was 4 pg-cell" during the same period (Fig. 2.4). The 1  carbohydrate in nitrate-recovered cells was rapidly degraded during the first 18 h following initial culture dilution, whereas there was no apparent decrease during the same period for ammonium-recovered cultures (Fig. 2.4). From 18 to 48 h following dilution, carbohydrate content in the nitrate-recovered cells steadily increased from an 18 h minimum of 2 pg-cell' to a 1  48 h maximum of 4 pg-cell" (Fig. 2.4). This pattern was not evident for ammonium-recovered 1  cultures. In contrast to the N-starved carbohydrate content of the cells, T A G fatty acids accumulated to the same quota of 6.5 pg-cell" after 3 d of N-starvation for cultures conditioned 1  on either nitrate or ammonium (Fig. 2.5). TAG fatty acids were not mobilized for the first 12 h following either nitrate or ammonium addition to the cultures. At 12 h following dilution, TAG mobilization began for nitrate-recovered cultures, and at 18 h for ammonium-recovered cultures (Fig. 2.5). For both cultures, the rate of TAG fatty acid mobilization appeared to stabilize 24 h following initial dilution, so that TAG fatty acids were reduced to a final cell quota of 1 pg-cell" by 48 h (Fig. 2.5). Polar lipid fatty acids remained at less than 1.5 pg-cell' 1  1  59 during the N-starvation period for cultures conditioned on either nitrate or ammonium (Fig. 2.6). The polar lipid fatty acid content began to increase 12 h after the initial dilution for both cultures, and peaked at 3 pg-cell" for ammonium-recovered cultures and 2 pg-cell* for nitrate1  1  recovered cultures (Fig. 2.6). The in vitro activity of GS was constitutively high for ammonium-recovered cultures before and after dilution (Fig. 2.7). In contrast, NR activity was not detected during the Nstarvation period for nitrate-recovered cultures. NR activity rapidly increased during thefirst6 h following dilution in these cultures, and thereafter steadily declined. The in vitro activity of ICL was almost undetectable during the N-starvation period for both cultures (Fig. 2.8). For both cultures, activity increased between 6 and 24 h after the initial dilution, and then decreased for the remainder of the experiment. Although ICL activity peaked at 24 h for both nitrate- and ammonium-recovered cultures, the activity in cells assimilating nitrate was 4-fold higher than the activity in cells assimilating ammonium (Fig. 2.8).  2.3.2)  Nitrate  transition  or ammonium  addition  in darkness  to 3 d N-starved  cultures followed  by a  to light  P. tricornutum cultures grown on either 1 mM NaNC>3 or NH4CI were N-starved for 2 d in continuous light, placed in the dark for a further 1 d, and then recovered with the same N source on which the cultures had been grown. After 72 h of dark exposure to the appropriate N source, light was made available to the cultures. There was no cell division in the dark period following either nitrate or ammonium addition to N-starved cultures (Fig. 2.9). For both cultures, cell division resumed 12 h after light was resupplied (Fig. 2.9). In vivo fluorescence decreased sharply within 6 h of diluting the cultures with either nitrate or ammonium in the dark (Fig. 2.10). For both cultures, in vivo  60 fluorescence values gradually increased to their previous N-starvation values within 24 h of dilution, and thereafter remained stable for the remainder of the dark period (Fig. 2.10). When light was resupplied to either culture, in vivofluorescencebegan to decrease after 12 h of light exposure (Fig. 2.10). Although there was no cell division in the dark, the cells did assimilate some of the added nitrate or ammonium (Fig. 2.11). Most of this assimilation occurred within 24 h of diluting the cultures, with a total measured draw-down during the dark period of 174 pM nitrate and 303 uM ammonium for nitrate- and ammonium-recovered cultures, respectively (Fig. 2.11). The concentrations for medium nitrate or ammonium immediately following dilution were probably higher than the values shown for 6 h following dilution. A single determination of the immediate post-dilution concentration for ammonium in the ammoniumrecovered cultures returned a value of 676 uM (data not shown). Thus, there was a high rate of ammonium assimilation in the dark immediately following dilution for ammonium-recovered cultures. Although an immediate post-dilution medium nitrate measurement was not made for nitrate-recovered cultures, the pattern was probably similar to that observed for ammonium assimilation, given the similarity in the shape of the curves shown in Fig. 2.11. For both cultures, the remaining medium N was rapidly assimilated within 12 h of making light available to the cultures (Fig. 2.11). Intracellular nitrate or ammonium concentrations increasedfromvery low values immediately after diluting the cultures with the appropriate N source (Fig. 2.12). The intracellular nitrate concentration subsequently decreased by 1.5frnol-ceU' during the dark 1  period in nitrate-recovered cultures, whereas the intracellular ammonium concentration decreased by 4.3frnol-ceU" in ammonium-recovered cultures (Fig. 2.12). When light was made 1  available to nitrate-recovered cultures, there was no decrease in intracellular nitrate until 6 h  61 had elapsed. In contrast, intracellular ammonium in ammonium-recovered cultures decreased immediately when light was resupplied to these cultures. For both cultures, the intracellular nitrate or ammonium concentrations decreased to dark-starvation values after 18 h of light exposure (Fig. 2.12). A comparison of Figs. 2.11 and 2.12 shows that the medium N in both cultures was exhausted before the intracellular nitrate or ammonium concentrations decreased to Nstarvation values. For nitrate-recovered cultures, medium nitrite was also measured as an indicator of nitrite excretionfromthe cells. Medium nitrite in these cultures was undetectable before dilution, and peaked at 18 uM 12 h following dilution with medium containing nitrate (Fig. 2.13). Thereafter, the medium nitrite concentration decreased to between 11-13 uM for the remainder of the dark period. When light was resupplied to the cultures, medium nitrite decreased to undetectable concentrations within 12 h (Fig. 2.13), which was identical to the pattern observed for medium nitrate (Fig. 2.11). Intracellular nitrite concentrations were always measured at less than 0.5 frnol-ceU" for nitrate-recovered cultures (data not shown). 1  Throughout the dark N-starvation period, cell carbohydrate concentrations for cultures previously conditioned on medium containing either nitrate or ammonium were maintained between 3-4 pg-cell" (Fig. 2.14). These concentrations were similar to the values obtained for 1  carbohydrates in cultures conditioned on ammonium and N-starved in continuous light (Fig. 2.4). After dilution of the dark N-starved cultures with either nitrate or ammonium, carbohydrate decreased to a minimum cell quota of 0.9 (ammonium-recovered) and 1.6 (nitrate-recovered) pg-cell' within 18 h (Fig. 2.14). However, between 18 and 72 h after 1  dilution in the dark period, there was a net increase in cell carbohydrate for nitrate- or ammonium-recovered cultures (Fig. 2.14). At the 72 h sampling time, the intracellular carbohydrate concentration had increased by 1.3 pg-cell" fromthe minimum quota for nitrate1  62 recovered cultures, and 0.7 pg-cell" fromthe minimum quota for ammonium-recovered 1  cultures (Fig. 2.14). These increases in cell carbohydrate in the dark between 18 and 72 h after nitrate or ammonium addition were statistically significant (1-tailed paired sample Student's ttest, P<0.05). Cell carbohydrate increased for thefirst18 h after light was resupplied to both cultures, and continued increasing for 24 h in nitrate-recovered cultures. Carbohydrate peaked at 12.1 pg-cell" 24 h following light resupply to nitrate-recovered cultures, and at 6.5 pg-cell" 1  1  18 h following light resupply to ammonium-recovered cultures (Fig. 2.14). During the dark N-starvation period, cell TAG fatty acids decreased for cultures conditioned on ammonium, and remained at the same concentration for cultures conditioned on nitrate (Fig. 2.15). At the time of dilution, both cultures had intracellular TAG fatty acid concentrations close to 3 pg-cell" , which was half the concentration measured for N-starved 1  cells in continuous light (Fig. 2.5). Following dilution in the dark with medium containing either nitrate or ammonium, TAG fatty acids decreased rapidly for thefirst12 h, and then more gradually for the remainder of the dark period (Fig. 2.15). When light was made available to either culture, TAG fatty acids increased by approximately 1 pg-cell" in the following 24 h 1  period (Fig. 2.15). Polar lipid fatty acids remained constant throughout the N-starvation and dark recovery period for either culture, although the quota for cultures conditioned on ammonium was higher (Fig. 2.16). When light was made available to either culture, the cell polar lipid fatty acid quota increased, then stabilized, or decreased in the case of nitraterecovered cultures (Fig. 2.16). The activity of the N assimilatory enzymes, NR for nitrate conditioned cultures, and GS for ammonium conditioned cultures, is shown in Fig, 2.17. NR activity was undetectable during the dark starvation period. When nitrate was added to these cultures in the dark, NR activity increased to a maximum of 3.2finolnitrite-cell' -!!* 6 h after dilution, decreased sharply 1  1  63 over the next 18 h, and was very low for the remainder of the dark period. The period of maximum activity in the dark period coincided with the decrease in medium nitrate (Fig. 2.11); however NR activity was only half of that observed during nitrate assimilation in continuous light (Fig. 2.7). NR activity increased a second time within 6 h of light being made available to these cultures (Fig. 2.17), with a slightly greater activity than was observed in the dark. NR activity tailed off for the remainder of the light period, and was coincident with the exhaustion of nitrate from the medium (Fig. 2.11). For cultures conditioned on ammonium, GS activity was constitutively high during the N-starvation period and after ammonium addition in the dark (Fig. 2.17). There was no noticeable change in GS activity when the cultures were diluted. However, when light was resupplied to these cultures, GS activity increased to 18 fmol phosphate-cell'^h' within 18 h 1  (Fig. 2.17), which was double the maximum activity observed during ammonium assimilation in continuous light (Fig. 2.7). GS activity remained high even when ammonium had been exhausted from the medium (Fig. 2.11). ICL activity was low during the dark N-starvation period for cultures conditioned on either nitrate or ammonium (Fig. 2.18), and was similar to the activity observed for N-starved cells in continuous light (Fig. 2.8). When nitrate or ammonium was added to the cultures in the dark, ICL activity increased over a period of 24 h for both cultures, to a maximum of approximately 0.26 fmol glyoxylate-ceir -h" . This was close to the maximum ICL activity 1  1  observed for ammonium-recovered cultures in continuous light (Fig. 2.8). Activity stayed high for the remainder of the dark period, and then increased for thefirst12 h after the transition into the light period (Fig. 2.18). For both cultures, ICL activity peaked 12 h after light was made available, and then slowly decreased for the remainder of the light period. For cultures recovered with ammonium, this 12 h peak (Fig. 2.18) was greater than the maximum activity  64 recorded for recovery during continuous light (Fig. 2.8). The opposite was true for cultures recovered with nitrate; the 12 h ICL activity peak (Fig. 2.18) was less than the maximum activity recorded for recovery during continuous light (Fig. 2.8).  Fig. 2.1. Changes in cell density for nitrate or ammonium-recovered cultures of P. tricornutum. Cultures were N-starved for 3 d in continuous light before being diluted with fresh medium containing either 1 mM NaN0 or NH4CI (t = 0 h; dotted line). Dilutions were made at 12 to 18 h intervals following time zero to ensure N-sufficiency. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. 3  66  250  0 I  ;  J  -12  •  :  L  1  0  12  i  —I  24  1  :  36  • 1 —  48  Time after N addition to stationary phase cultures (h) Fig. 2.2. Culture in vivo fluorescence for nitrate- or ammonium-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  67  Time after N addition to stationary phase cultures (h)  Fig. 2.3. Culture medium N concentrations for nitrate- or ammonium-recovered cultures of P. tricornutum. The initial concentrations immediately following dilution at time zero were not measured. The periodic increases in medium N concentrations following the initial dilution at time zero indicate subsequent dilutions made to ensure N-sufficiency. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  10'  1  :  1  -12  0  1  12  :  1  1  24  36  48  Time after N addition to stationary phase cultures (h) Fig. 2.4. Intracellular carbohydrate concentrations for nitrate- or ammonium-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  69  -12  0  12  24  36  48  Time after N addition to stationary phase cultures (h) Fig. 2.5. Intracellular TAG fatty acid concentrations for nitrate- or ammonium-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  o.o  1  l J  -12  1  ±  0  1  1  12  1  24  i  36  48  Time after N addition to stationary phase cultures (h)  Fig. 2.6. Intracellular polar lipid fatty acid concentrations for nitrate- or ammoniumrecovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures.  71  10  Time after N addition to stationary phase cultures (h) Fig. 2.7. Intracellular N assimilatory enzyme activities for nitrate- or ammoniumrecovered cultures of P. tricomutum. Nitrate reductase (NR.) activity was monitored in nitrate-recovered cultures, and glutamine synthetase (GS) activity was monitored in ammonium-recovered cultures. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  72  T-  1  r-  1  1  r  -12  0  12  24  36  48  Time after N addition to stationary phase cultures (h)  Fig. 2.8. Intracellular isocitrate lyase (ICL) activity for nitrate- or ammonium-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  73  j  0 o L~  *  0  *——  12  «  24  36  *•  *  48  60  1—I  1  72  84  96  Time after N addition to stationary phase cultures (h) Fig. 2.9. Changes in cell density for nitrate or ammonium dark-recovered cultures of P. tricornutum. Cultures were N-starved for 2 d in continuous light and then 1 d in darkness before being diluted with fresh medium containing either 1 mM NaN0 or NH4CI (t = 0 h; dotted line). Cultures were maintained in the dark (shaded area) for 72 h following N addition before light was made available (unshaded area). Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. 3  74  0  12  24  36  48  60  72  84  96  Time after N addition to stationary phase cultures (h)  Fig. 2.10. Culture in vivo fluorescence for nitrate or ammonium dark-recovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  75  Time after N addition to stationary phase cultures (h)  Fig. 2.11. Culture medium N concentrations for nitrate or ammonium dark-recovered cultures of P. tricomutum. The initial concentrations immediately following dilution at time zero were not measured. Only nitrate was measured in nitrate-recovered cultures and ammonium in ammonium-recovered cultures. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols.  76  0  12  24  36  48  60  72  84  96  Time after N addition to stationary phase cultures (h)  F i g . 2.12. Cell N concentrations for nitrate or ammonium dark-recovered cultures of P. tricornutum. The initial concentrations immediately following dilution at time zero were not measured. Only nitrate was measured in nitrate-recovered cultures and ammonium in ammonium-recovered cultures. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  77  2 5 |  +  •  i  1  *  T  ^  Time after N addition to stationary phase cultures (h)  Fig. 2.13. Medium nitrite concentrations for nitrate dark-recovered cultures of P. tricomutum. The initial concentrations immediately following dilution at time zero were not measured. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols.  78  T  1  0  12  j  24  j  1  r  36  48  60  72  84  96  Time after N addition to stationary phase cultures (h)  Fig. 2.14. Intracellular carbohydrate concentrations for nitrate or ammonium darkrecovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  0  79  0  12  24  36  48  60  72  84  96  Time after N addition to stationary phase cultures (h)  Fig. 2.15. Intracellular TAG fatty acid concentrations for nitrate or ammonium darkrecovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  80  3.5  5  Nitrate Ammonium!  o bo  8  •3 OH  0.0 0  12  24  36  48  60  72  84  96  Time after N addition to stationary phase cultures (h)  Fig. 2.16. Intracellular polar lipid fatty acid concentrations for nitrate or ammonium dark-recovered cultures of P. tricomutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  81  Time after N addition to stationary phase cultures (h)  Fig. 2.17. Intracellular N assimilatory enzyme activities for nitrate or ammonium darkrecovered cultures of P. tricornutum. NR and GS activities were monitored as described in Fig. 2.7. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  Time after N addition to stationary phase cultures (h)  Fig. 2.18. Intracellular IGL activity for nitrate or ammonium dark-recovered cultures of P. tricornutum. Manipulations were made as described in Fig. 2.9. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols.  83  Mitochondrion  Chloroplast (13)  •NH,  Glyoxysome  (2)  ^ ( 6 ) ^ malate—*-»-PYR  N0 " 2  4  malate C)  (5)  7  • acetyl CoA-  HCO, . J '  ^-r  (11)  N0 "^3  •N0 3  ^PEP  (9)\ CO.  •  Carbohydrate Pool (10)  CO,  OPP Pathway  Fig. 2.19. Model for lipid and carbohydrate C flows linked to N assimilation. Pathways do not represent exact stoichiometrics and not all intermediates are shown. Although succinate is the product of the glyoxylate cycle, malate is shown as the net exported C4 acid for subsequent reactions. (1) = lipase; (2) = carnitine acyltransferase; (3) = pyruvate dehydrogenase complex; (4) = decarboxylating enzymes of the TCA cycle; (5) = pyruvate kinase; (6) = malic enzyme; (7) = malate dehydrogenase, (8) = phosphoenolpyruvate carboxylase, (9) = phosphoenolpyruvate carboxykinase; (10) = 6-phosphogluconate dehydrogenase; (11) = nitrate reductase; (12) = nitrite reductase; (13) = glutamine synthetase; (14) = glutamate synthase. PEP = phosphoenolpyruvate; PYR = pyruvate; OAA = oxaloacetate; Gin = glutamine; Glu = glutamate; a-kg = a-ketoglutarate; OPP Pathway = oxidative pentose phosphate pathway.  84  A. Mitochondrial B-oxidation  B. Glyoxysomal B-oxidation  -rS Mitochondrion  Mitochondrion  NADH NAD  NADH NAD  ( T C A cycle 1  acetyl  r v r  NADH  NAD  >  CoA-*-J  \  (3)  Glyoxysome acetyl CoA—I  O i l Body TAG  y  Oil Body] TAG  Fig. 2.20. assuming represent enzymes;  Model for carbon flow and reductant generation during respiration of TAGs either mitochondrial (A) or glyoxysomal (B) B-oxidation. Pathways do not exact stoichiometries and not all intermediates are shown. (1) = B-oxidation (2) = dehydrogenases of the T C A cycle; (3) = carnitine acyltransferase.  Mitochondrion NADH  Fig. 2.21.  NAD  M o d e l for carbon flow and reductant generation during T A G oxidation by  glyoxysomal oxidation and the T C A cycle. Pathways do not represent exact stoichiometrics and not all intermediates are shown. (1) = B-oxidation enzymes; (2) dehydrogenases o f the glyoxylate cycle; (3) = malic enzyme; (4) = pyruvate dehydrogenase complex; (5) = dehydrogenases of the T C A cycle.  =  86  Fig. 2.22. Model for carbon flow and reductant generation during anaplerotic oxidation of TAGs. Pathways do not represent exact stoichiometrics and not all intermediates are shown. (1) = B-oxidation enzymes; (2) = dehydrogenases of the glyoxylate cycle; (3) = malic enzyme; (4) = pyruvate dehydrogenase complex; (5) = dehydrogenases of the TCA cycle.  87  NAD NADH  Glyoxysome  malate  malate •  ^  (3)  ^ •-OAA (4) PEP  NADH  1,3-bisPGA NADH (5) NAD V i PGA  1  NADH NAD  A.  Carbohydrate Pool!  Fig. 2.23. Model for carbonflowand reductant generation during gluconeogenesis. Pathways do not represent exact stoichiometrics and not all intermediates are shown. (1) = B-oxidation enzymes; (2) = dehydrogenases of the glyoxylate cycle; (3) = malate dehydrogenase; (4) = phosphoenolpyruvate carboxykinase; (5) = glyceraldehyde phosphate dehydrogenase. PGA = phosphoglyceraldehyde; l,3-bisPGA= 1,3bisphosphoglyeraldehyde.  88  A. Glycolysis  B. Malate glycolysis  Mitochondrion  Mitochondrion NADH N A D  NADH NAD (3)  j  I TCA cycle  V  I T C A cycle L  acetyl  x e t y l  J CoA  NADH-—J(2) NAD """pYR  NADH1(2) NAD "pYR s  ;  PYR  t  PEP  t  1,3-bisPGA NADH NAD  NAD  ^  t NNAADD| H  NADH-^ & OAA^  G3-P  L Carbohydrate Pool  ±->  (6) ' malate-^-^PYR UcUC S\ P  (4)  PEP  t  1,3-bisPGA NADH ^ NAD - \  ( 1 )  G 3-P  t_ Carbohydrate Pool  Fig. 2.24. Model for carbonflowand reductant generation during classical (A) and malate (B) glycolysis. Pathways do not represent exact stoichiometrics and not all intermediates are shown. (1) = glyceraldehyde phosphate dehydrogenase; (2) = pyruvate dehydrogenase complex; (3) = dehydrogenases of the TCA cycle; (4) = phosphoenolpyruvate carboxylase; (5) = malate dehydrogenase; (6) = malic enzyme. G 3-P = glyceraldehyde 3-phosphate.  Mitochondrion NADH NAD  Y  (5  TCA cycle  V>  v_y  •  co Aetyl Carbohydrate Pool  t  NADH—-J(4) NAD"py  H R  PYR  t PEP t  1,3-bisPGA NADH^i NAD G3-P  glucose 6-P NADP NADPH"  (1) 6-PG  NADP NADPH"  (2)  ribulose 5-P  Fig. 2.25. Model for carbon flow and reductant generation during carbohydrate degradation through the oxidative pentose phosphate pathway. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = glucose 6phosphate dehydrogenase; (2) = 6-phosphogluconate dehydrogenase; (3) = glyceraldehyde phosphate dehydrogenase; (4) = pyruvate dehydrogenase complex; (5) dehydrogenases of the TCA cycle. 6-PG = 6-phosphogluconate.  90  Mitochondrion NADH NAD X " g a  k  jacetyl CoA  protein synthesis  NADH-—J( ) 4  N A D ^ PYR  ma ate  malate-  t  NAD^v)  J(2)  PYR NAD j NADH  NADH-^|  I OA A -  PEP  t  1,3-bisPGA NADH ^>t NAD Carbohydrate Pool  + G3-P  Fig. 2.26. Model for carbon flow and reductant generation during anaplerotic oxidation of carbohydrates. Pathways do not represent exact stoichiometries and not all intermediates are shown. (1) = glyceraldehyde phosphate dehydrogenase; (2) = malate dehydrogenase; (3) = malic enzyme; (4) = pyruvate dehydrogenase complex; (5) = dehydrogenases of the TCA cycle. Dashed lines = malate glycolysis.  91  Table 2.1. Pathway models for reductant and carbon generation during the oxidation of either a 16 carbon fatty acid associated with TAG or a 16 carbon carbohydrate in P. tricomutum. e": carbon conserved = ratio of reductant produced in pathway:moles carbon not lost to C0 . For cases of anaplerotic oxidation, V* of the carbon was assumed to be mobilized for use in biosynthetic reactions, and the remainder oxidized to C0 . TCA = tricarboxylic acid cycle; OPPP = oxidative pentose phosphate pathway. 2  2  Substrate  Pathway  e" equivalent Carbon conserved  e": carbon  Figure  conserved  reference  (%) Fatty acid  TCA cycle  92  0  -  2.20  glyoxysomal + TCA  84  0  -  2.21  1/4 anaplerotic oxidation  68  31.2  13.6  2.22  gluconeogenesis  36  75  9  2.23  64  0  -  2.24  OPPP + glycolysis + TCA  42.7  33.3  4*  2.25  1/4 anaplerotic oxidation  42.7  31.2  7.9  2.26  Carbohydrate glycolysis + TCA  * Carbon retained as erythrose 4-P  92  Table 2.2. Equations for reductant and carbon requirements during either nitrate or ammonium assimilation in P. tricornutum. N R = nitrate reductase; NiR = nitrite reductase; GS = glutamine synthetase; G O G A T = glutamate synthase; Gin = glutamine; Glu = glutamate.  Reaction NR  N0 " 3  M  V  e" equivalent  > N0 -  2  2  ^  K  +  M  V  N H Z - ^ G l n Gin  e': carbon*  GOGAT  6  °  2  0.4  NO3" assimilation  10  2.0  NFL* assimilation  2  0.4  > Glu  * Assumes 5 C incorporated/N assimilated. ATP costs were not included in e" calculations.  93 2.4)  DISCUSSION  The resultsfromChapter 1 showed that storage lipids, particularly TAG fatty acids, were quantitatively the most important carbon storage compounds in stationary phase cultures of P. tricomutum. This chapter has examined the patterns of storage carbon degradation in cultures of this diatom during a transitionfromstationary phase to N-sufficiency. This transition was initiated by diluting stationary phase cultures with medium containing either nitrate or ammonium, and these dilutions were carried out in either the light or the dark. The following sections will discuss the factors regulating N assimilation over this transition period, and how the requirements for carbon during N assimilation may be met by the oxidation of carbohydrates and storage lipids.  2.4.1) Regulation of nitrogen assimilation  Nitrogen assimilation in unicellular algae is much more rapid following a period of Nstarvation than during steady state, logarithmic phase growth (Syrett 1988). This could be a consequence of enhanced nitrate or ammonium transport across the cell membrane, or a combination of enhanced transport and increased assimilatory enzyme activities within the cell. In unicellular algae, the transport systems for nitrate and ammonium have not been characterized, but both are postulated to require ATP (Falkowski and Rivkin 1976, Larson and Rees 1994). Assuming that either N source would require at most two protons for the transport of one nitrate or ammonium molecule across the cell membrane (Syrett 1988), then a maximum of two ATP, (or approximately 1.3 e') would be required for transport, which is well below the energetic costs for the remainder of the assimilatory pathways (Table 2.1). The data in this chapter indicates that, in the dark, nitrate or ammonium transport were not limiting  94 because of the presence of large intracellular pools of both these ions (Fig. 2.12). Additionally, the rate of N transport exceeds assimilation in N-starved P. tricomutum (Syrett 1988). Ultimately N assimilation must be coupled to protein synthesis, and in the long term, steady state growth. In this study, nitrate and ammonium assimilation in the dark (Fig. 2.11) were probably indirectly limited by the cell volume; i.e. in the absence of cell division, the cells reached their maximum quotas for protein, and thus repressed any further N assimilation. Larson and Rees (1996) found that P. tricomutum was able to assimilate ammonium and synthesize new protein in the absence of cell division, but only if the cell protein quotas were low before ammonium addition. Therefore, the reduction in the rate of nitrate or ammonium assimilation after 12 h in darkness in this study (Fig. 2.11) was most likely a consequence of cell division ceasing (Fig. 2.9) rather than insufficient reductant or carbon supply to the assimilatory enzymes. However, the experiments in this study also showed that carbohydrates and storage lipids were degraded in the light or the dark following N addition to N-starved cultures. This indicates that storage carbon may play an important role during N assimilation in N-starved algae. There is evidence to suggest that a decrease in the rate of amino acid synthesis removes any inhibition of ammonium or nitrate assimilation in microalgae. Lara etal. (1987) found that ammonium inhibition of nitrate transport in the cyanobacterium Anacystis nidulans was relieved when GS was inhibited. Similarly, Reyes and Florencio (1994) found that a mutant of Synechocystis lacking GS activity did not show a repression of NR or NiR activity by ammonium. They found glutamine to be present at almost undetectable levels in the mutant. In Zea mays roots, Sivasankar and Oaks (1995) found that NR activity was inhibited 70% by 1 mM concentrations of glutamine or asparagine. All these results suggest that nitrate or ammonium assimilation will only continue when intracellular glutamine concentrations are low  95 and protein synthesis is in demand. Tischner and Schmidt (1982) also found that reduced thioredoxin increased GS activity 8-fold and GOGAT 4-fold when added to Chlorella sorokiniana cultures in the dark. Thus, ammonium assimilation, whether the source is extracellular or from the nitrate assimilatory pathway, may be ultimately controlled by a lightmediated thioredoxin system. Together, these observations suggest that N transport is not carbon- or energy-limited in N-starved P. tricornutum, and that N assimilation will continue as long as glutamine concentrations are low and protein synthesis is occurring. These are the biochemical characteristics of an algal culture during N-starvation. Therefore, the major control points for both nitrate and ammonium assimilation in P. tricornutum most likely involve the regulation of the assimilatory enzymes within the cell.  2.4.1.1) Nitrate assimilation Nitrate assimilation in unicellular algae is limited by the activity of NR and/or nitrite reductase (NiR). Although NiR has received little attention in the literature, it is worth investigating further, since the excretion of nitrite from the cell under some conditions suggests it is a rate-limiting enzyme (e.g. Fig. 2.13). NR is a cytosolic enzyme that catalyzes the reduction of nitrate to nitrite in the presence of NADH (Pate and Layzell 1990). Enzyme activity regulation appears to primarily involve de novo protein synthesis through the translation of NR mRNA (Crawford et al. 1992), which is induced by nitrate (Pate and Layzell 1990). This substrate inducible property of NR explains why in vitro activity was undetectable during the N-starvation period for P. tricornutum cells in the light or dark (Figs. 2.7, 2.17). In the light it is important that NR activity is tightly regulated, because it consumes up to 20% of the energy produced from the light reactions of  96 photosynthesis that would otherwise be used for carbon fixation (Pate and Layzell 1990). In this study, NR activity increased in the light or the dark in response to nitrate addition, and after a dark-light transition in the presence of nitrate (Figs. 2.7, 2.17). However, NR activity after nitrate addition to N-starved cells in the dark was only about half that observed for nitrate addition in the light. This has also been observed for the green microalga Dunaliella salina (Jimenez de Rio et al. 1994). Thus, there appears to be a complex regulation of NR occurring, involving light, nitrate, and the nutritive history of the cell. In addition to NR regulation by de novo protein synthesis, an inactive form of the enzyme may be activated by covalent modification of the protein. In higher plants, Kaiser and Huber (1994) distinguished between the processes of NR activation and synthesis of new protein. Events such as light transitions, sugar feeding, and CO2 or O2 availability alter NR activity on a time-scale of minutes. They proposed that dephosphorylation of the NR protein by NR kinase activates the enzyme, and phosphorylation by NR phosphatase inactivates the enzyme. Kaiser and Huber (1994) found that NR activation was stimulated by AMP and Pi, and inactivation was stimulated by Mg  2+  and P-ester sugars.  The results from this chapter suggest that the increases in NR activity observed after nitrate addition to N-starved cultures in the light or dark in Figs. 2.7 and 2.17 were a consequence of de novo protein synthesis induced by nitrate, the substrate. Conversely, the secondary increase in activity in dark nitrate-recovered cells after the lights were switched on (Fig. 2.17) was probably a consequence of light activation of existing NR protein. However, NR can only function if there is sufficient reductant available to reduce nitrate to nitrite. In heterotrophically grown algae (or autotrophically grown algae in the dark), the NADH necessary to maintain NR activity in vivo may come from glycolysis, the oxidative pentose phosphate pathway (OPPP), or the mitochondrial dehydrogenases of the TCA cycle. Pate and  97 Layzell (1990) suggested that light activation of NR may depend on readily available photoreductant and minimal competition for this photoreductantfromthe mitochondrial dehydrogenases. In the dark, NR activity would be inhibited by increased competition from mitochondrial NADH oxidation. Therefore, under some conditions NR activity may be dependent on the supply of carbon and reductantfromcarbohydrate or storage lipid degradation. This will be discussed in section 2.4.2.1.  2.4.1.2) Ammonium assimilation The high constitutive GS activity observed for N-starved cultures in continuous light (Fig. 2.7) and during the dark (Fig. 2.17) is typical for P. tricomutum. GS is not a substrate inducible enzyme (as NR is), but rather is synthesized during N-starvation. Rees et al. (1995) observed that GS activity increased during thefirst12 h of N-starvation in this diatom, and then remained stable and high for at least a further 36 h of N-starvation. In the green microalga Dunaliella salina, GS activity was enhanced 4-fold within 4 h of N-starvation, and thereafter activity remained high (Jimenez de Rio et al. 1994). It is interesting to note that assayable GS activity in cultures maintained in the dark remained high 24 h following ammonium addition (Fig. 2.17), even though ammonium assimilation had almost ceased (Fig. 2.11). This suggests that either the enzyme was present but inactive in vivo, and activated in the in vitro assay procedure, or that GS was active in vivo, but another reaction in the assimilatory pathway, possibly GOGAT, was inhibited in the dark. Thus, in vitro GS activity measurements are not good estimators of in vivo ammonium assimilation under all conditions. It is probable that activation of existing GS enzyme protein, rather than de novo synthesis, is the main regulatory mechanism for ammonium assimilation in P. tricomutum. In bacterial systems, GS is regulated by a system of adenylylation (addition of  98 AMP units to a tyrosine residue on the enzyme) and deadenylylation reactions, with the end result being an activation of GS in response to N-starvation and an increase in the ratio of aketoglutarate:glutamine (Magasanik 1988). However, such a regulatory system for GS has yet to be demonstrated in eukaryotes. Unlike nitrate assimilation, ammonium assimilation has a much lower requirement for reductant, with only 2e" being necessary for GOGAT activity (Table 2.1). Thus, reductant supply is much less likely to be a limiting factor for ammonium assimilation than it is for nitrate assimilation. Therefore, changes in the rates of ammonium assimilation in P. tricornutum are more likely to be a function of carbon availability for amino acid synthesis rather than reductant supply. This is discussed in more detail in section 2.4.2.1.  99 2.4.2)  Carbon  mobilization  during  nitrogen  assimilation  2.4.2.1) Carbon and reductant supplyfor nitrogen assimilation from carbohydrates and storage lipids Storage lipid and carbohydrate degradation can take place via a multitude of different r  i  routes. These are outlined in Table 2.1 and in more detail in Figures 2.19-2.26. Fatty acid oxidation (Figs. 2.20, 2.21) yields more reductant than carbohydrate oxidation (Figs. 2.24, 2.25) on a per carbon basis when complete oxidation to CO2 occurs. In anaplerotic oxidation, where some of the TCA cycle intermediates are drawn off for protein synthesis, fatty acid degradation (Fig. 2.22) can still provide more reductant on a per carbon basis than carbohydrate oxidation (Fig. 2.26). Additionally, even the process of gluconeogenesis (Fig. 2.23), which synthesizes carbohydrates from lipids, provides an excess of reductant that could be used for other processes such as N assimilation. The control of the carbon flux through each of these respiratory pathways is likely to be dependent on the carbon and reductant requirements that the cell has that cannot be met directly by photosynthesis. It is well established that carbohydrate degradation, particularly that of starch and sucrose, is stimulated during N assimilation in N-starved algae. It would be logical to assume that storage lipid, being an alternative carbon storage product, acts in a similar role to carbohydrate. For example, storage lipid breakdown was linked to an enhanced respiration rate in Selenastrum capricomutum cultures recovering from cadmium-induced growth inhibition (Thompson and Couture 1990). The assimilation of nitrate or ammonium have markedly different effects on the carbon flow into the cell through the Calvin cycle and B-carboxylation, and respiratory flows through the TCA cycle. Elrifi and Turpin (1986) found that nitrate assimilation suppressed photosynthetic carbon fixation more than ammonium assimilation in Selenastrum minutum, but  100 that ammonium assimilation also stimulated much more dark respiration and dark carbon fixation than nitrate assimilation. This pattern of carbon usage is likely to be a consequence of the lower e'xarbon ratio that ammonium assimilation requires compared to nitrate assimilation (Table 2.1). Therefore, the process of storage lipid degradation via 13-oxidation and the glyoxylate cycle, which generates a higher e'xarbon ratio than equivalent anaplerotic carbohydrate mobilization (Table 2.2), will more closely match the carbon and reductant requirements for nitrate assimilation than ammonium assimilation. In fact, lipid oxidation in general appears to produce more reductant than carbohydrate oxidation, whether the carbon is respired to CO2 or used for subsequent biosynthetic processes (Table 2.2). Enhanced glyoxylate cycle activity during nitrate assimilation could provide extra reductant within the glyoxysome, which could be exported to NR in the cytosol via a glyoxysomal membrane NADH-shuttle (Bowditch and Donaldson 1990). Despite the theoretical advantage that storage lipid oxidation might have in supplying the carbon and reductant necessary for nitrate assimilation, the results show that carbohydrates were preferentially mobilized during the initial recovery period. For cultures recovered with nitrate, carbohydrate was the primary carbon storage compound mobilized following nitrate addition to cultures in the light or the dark (Figs. 2.4, 2.14), whereas TAG fatty acids were mobilized 18 h later in the light, and simultaneously in the dark (Figs. 2.5, 2.15). The same pattern was observed for ammonium recovered cultures, except there was no noticeable carbohydrate decrease in these cultures recovered in continuous light (Fig. 2.4). It appears that storage lipid mobilization was repressed during the initial 24 h period following recovery. Assuming that a portion of the storage lipid carbon was mobilized through the glyoxylate cycle, the delay in storage lipid mobilization could be explained by the 18 to 24 h delay in the increase in ICL activity following nitrate or ammonium addition to N-starved  101 cultures (Figs. 2.8, 2.18). The increase in ICL activity in both cultures may have been a consequence of increased substrate supply. ICL is similar to NR in that it is a substrate inducible enzyme. Acetate induces ICL activity in Chlorella (Goulding and Merrett 1970, Thurston 1977), Scenedesmus obliquus (Combres et al. 1994), and Chlorogonium (Schmidt and Zetsche 1990). ICL synthesis is also known to be repressed by sugars. Glucose inhibits ICL activity in Chlorella (Syrett 1966) and in the fungus Aspergillus nidulans (De Lucas et al. 1994). Giachetti et al. (1987) also described the inhibition of pea and lupin ICL by H P O 4 ' ions and 2  PEP. This may explain why ICL activity did not increase in nitrate recovered cultures in the light (Fig. 2.8) until the carbohydrates had reached their minimum cell quota (Fig. 2.4). It would also explain the similar maximum activities for ICL from dark nitrate or ammonium recovered cultures (Fig. 2.18), which both had low carbohydrate quotas (Fig. 2.14). However, carbohydrate inhibition of ICL cannot fully explain the differences between the activities measured for nitrate and ammonium recovered cultures in the light. ICL activity was 4-fold greater following nitrate addition in the light when compared to ammonium addition (Fig. 2.8). Although the carbohydrate quota was low in ammonium recovered cultures (Fig. 2.4), it is known that the glycolytic carbon flow during ammonium assimilation is greater than that during nitrate assimilation, at least in Selenastrum minutum (Elrifi and Turpin 1986). It is also known that PEPCase activity is enhanced by glycolytic substrates, particularly phosphorylated hexoses, in Selenastrum minutum (Schuller et al. 1990). Graham et al. (1994) suggested that, in addition to the substrates of glycolysis inhibiting ICL activity, the glycolytic flux may also inhibit activity. They suggested that hexose phosphorylation events associated with glycolysis inhibit ICL activity. Thus, sugar phosphorylation may serve as a signal to prevent lipid mobilization through the glyoxylate cycle and gluconeogenesis when glycolysis or  102 anaplerotic carbonfixationis occurring coupled to carbohydrate oxidation. PEPCase has also been suggested to have an important role in supplying the malate necessary for lipid biosynthesis via glycolysis (Sangwan et al. 1992). Therefore, it is only logical that PEPCase activation and lipid degradation through the glyoxylate cycle do not occur simultaneously, or a futile cycle may result.  2.4.2.2) Alternative roles for storage lipid carbon TAG fatty acids were neither degraded during the initial stages of nitrate or ammonium assimilation in the light (Fig. 2.5), nor was ICL activity strictly coupled to N assimilation. In the dark, ICL activity was actually increasing (Fig. 2.18) when nitrate and ammonium assimilation were decreasing (Figs. 2.11, 2.17). Therefore, apartfromsupplying the carbon and reductant necessary for protein synthesis during N assimilation, storage lipid degradation may fulfill more indirect roles during recoveryfromN-starvation. These include the restoration of cell membranes through fatty acid transacylation and unsaturation, and the replenishment of carbohydrate reserves through gluconeogenesis. One function of TAG fatty acid mobilization may be to restore membrane structures degraded during N-starvation. In P. tricomutum, acetate, which is a product of fatty acid 13oxidation, is incorporated into new lipids in the light, but oxidized to C O 2 in the dark (Cooksey 1974). Klyachko-Gurvich et al. (1973) suggested that the mobilization of storage lipids in Chlorella was primarily to restore photosynthetic membranes degraded during N-starvation. Similarly, mobilization of TAG fatty acids to membrane phospholipids has been described for the fungus Glomus versiforme (Gaspar et al. 1994). However, in the light, polar lipid fatty acids increased by a maximum of 1.5 pg-cell" after nitrate or ammonium addition to N-starved 1  P. tricomutum cultures (Fig. 2.6), whereas TAG fatty acids decreased by more than 5 pg-cell"  1  103 (Fig. 2.5). In the dark, polar lipid fatty acids actually decreased for ammonium recovered cultures, and only increased marginally for nitrate recovered cultures (Fig. 2.16), although T A G fatty acids decreased by at least 2 pg-cell" for both cultures (Fig. 2.15). Therefore, the majority 1  of the T A G fatty acid carbon was utilized for processes other than membrane restoration during recoveryfromN-starvation. During nitrate assimilation in the light and after the initial nitrate or ammonium assimilation period in the dark, the cell carbohydrate content increased (Figs. 2.1, 2.14), coincident with ongoing T A G degradation. This suggests that gluconeogenesis was occurring; in nitrate recovered cultures in the light in response to the prior degradation of cell carbohydrate reserves, and in both cultures in the dark in response to the metabolism of carbohydrate reserves for maintenance respiration. This may also explain why ICL activity did not taper off in the dark period following nitrate or ammonium addition to N-starved cultures (Fig. 2.18). Gluconeogenesis appears to be regulated by PEP carboxykinase (PEPCK) activity (Leegood and ap Rees 1978b, Trevanion et al. 1995), which maintains a high PEP:OAA ratio to increase the flow towards carbohydrate synthesis. Trevanion et al. (1995) noted that inhibiting PEPCK with 3-mercaptopicolinic acid increased the flow of carbonfromlabeled acetate into CO2 and decreased theflowinto OAA. Thus, it appears that regulation of gluconeogenesis occurs further down the gluconeogenic pathway than ICL, and that some of the storage lipid mobilized through the glyoxylate cycle may be respired via the TCA cycle rather than converted to carbohydrates. Ultimately, it appears that not all the storage lipid carbon mobilized could be accounted for by the glyoxylate cycle. Assuming the in vitro determination of ICL activity was realistic, then an in vivo activity of 0.3 fmol-glyoxylate-cell'^h' over 48 h (Fig. 2.18) would only support 1  a T A G fatty acid decrease of 0.9 pg, when in fact a decrease of over 2 pg was observed (Fig.  104 2.15). Some of these TAG fatty acids may have been relocated into structural lipids, but these can only account for a maximum of 0.5 pg in the dark (Fig. 2.16). Thus, there may be a case for some of the storage lipid carbon to bypass the glyoxylate cycle and be directly respired in the TCA cycle (Fig. 2.20). In conclusion, it appears that the TAG fatty acids were being oxidized in P. tricornutum by a variety of pathways, including but not limited to the glyoxylate cycle. Some of this carbon was then used for N assimilation, to supply reductant and/or carbon, whereas the remainder may have been respired or used for gluconeogenesis. Further investigation on how these various mobilization processes are regulated during N assimilation is needed to elucidate the importance of storage lipid metabolism in unicellular algae.  105  CHAPTER 3  REALLOCATION AND LOSS OF CARBON F R O M C L A B E L E D 1 4  TRIACYLGLYCEROLS DURING RECOVERY OF STATIONARY PHASE CULTURES F R O M N-STARVATION IN THE LIGHT WITH NITRATE OR AMMONIUM  3.1)  INTRODUCTION  The close relationship between carbon assimilation and biomass production in cultured microalgae and natural phytoplankton populations is fairly well understood. However, the relationship between biomass production and carbon loss rates from aquatic ecosystems is difficult to predict, partly because the processes that regulate the degradation and loss of carbon-containing compounds from microalgae are poorly understood (Keller and Riebesell 1989). In phytoplankton populations, the respiration of intracellular carbon appears to be second in importance only to sedimentation when considering carbon loss processes (Keller and Riebesell 1989). The data from Chapters 1 and 2 indicate that lipids, particularly TAGs, are quantitatively important carbon storage products in N-starved P. tricornutum cultures. The phenomenon of lipid accumulation during nutrient deficiency in diatoms and other microalgae is well documented (Shifrin and Chisholm 1981, Piorreck et al. 1984, Parrish and Wangersky 1987, Suen et al. 1987, Emdadi and Berland 1989, Myklestad 1989, Parrish and Wangersky 1990, Roessler 1990). However, the fate of storage lipid carbon in phytoplankton metabolism,  106 and its contribution to respiratory carbon loss, has not been adequately investigated. The data from Chapters 1 and 2 showed that most of the lipids accumulated during N-starvation were triacylglycerols (TAGs), and that these were degraded during subsequent nitrate or ammonium assimilation. The datafromChapter 2 also suggested that the fate of the degraded T A G carbon depends on the N source, because carbohydrate replacement via the glyoxylate cycle seemed to be occurring during nitrate assimilation, but not ammonium assimilation. One way to investigate the possible sinks for the carbonfromdegraded TAGs is to use radiolabeling techniques. These techniques have been used extensively to measure intracellular carbon allocation in polar diatoms (Terry et al. 1983, McConville et al. 1985, Palmisano et al. 1988, Tillman etal. 1989, Smith etal. 1990, Smith and Herman 1992, Smith and D'Souza 1993, Thomas and Gleitz 1993). Studies have been done with cultures and natural populations to examine changes in physiological state, where enhanced C incorporation into 14  carbohydrates vs. proteins is taken to indicate nutrient limitation (McConville et al. 1985, Palmisano et al. 1988, Smith and D'Souza 1993). It is intuitive that whether the final carbon storage products are lipids or polysaccharides, these carbon-containing compounds must be mobilized and degraded at some point in the life-history of the diatom. In studies with polar diatoms, it has been suggested that polysaccharides are utilized for night metabolism, whereas storage lipids are not, because the 14  C label decreases in the polysaccharidefractionduring the dark period, but remains invariant  in the lipidfraction(Li and Piatt 1982, Tillman etal. 1989). Similarly, C labeling experiments 14  done with P. tricomutum revealed that polysaccharides and lipids were synthesized in the light, but only polysaccharides were consumed in the dark to support protein synthesis (Terry et al. 1983). However, past C reallocation studies in diatoms have sufferedfromtwo flaws. 14  Firstly, these studies are typically designed to measure photosynthate allocation patterns rather  107 than carbon degradation and loss from the cells, so the cultures or population samples are continually incubated with the C label throughout light and dark periods in the experiments. 14  Therefore, dark C incorporation processes, such as B-carboxylation reactions that may 14  supplement glycolysis, are ignored. Secondly, a dark period may not provide a sufficient, prolonged carbon demand to stimulate lipid degradation. The data from Chapters 1 and 2 indicate that the onset of significant storage lipid mobilization lags 12 to 18 h behind N addition to N-starved cultures. This chapter examines the patterns of radiolabeled carbon fixation and reallocation during the transition period between N-starvation and N-sufficiency in nitrate or ammonium recovered cultures of the marine diatom P. tricornutum. In Chapter 2, it was apparent that the timing and quantity of triacylglycerol (TAG) degradation was similar for nitrate or ammonium recovered cultures in the light or dark. However, in the light, isocitrate lyase activity was much higher for nitrate recovered cultures. This suggested that some of the T A G carbon was being utilized for gluconeogenesis rather than being respired to CO2 during nitrate assimilation. Therefore, the experiments in this chapter were designed to investigate possible differences in the sinks for degraded TAG carbon during nitrate or ammonium assimilation.  108  3.2)  MATERIALS AND METHODS  3.2.1) Culture conditions and biomass measurements Growth medium (ES AW) was prepared without added N (see Chapter 1 for details); this medium contained sub-uM concentrations of nitrate or ammonium, and for convenience was labeled "N-free". Stock cultures of Phaeodactylum tricomutum were inoculated into 1 L glass bottles as described in Chapter 1, and diluted with N-free ESAW. Small volumes (< 1 mL) of sterile 1 M stock solutions of either NaN03 or NH4CI were added to the cultures to give afinalconcentration of 1 mM. For each N source, quadruple cultures were grown in continuous light for 8-10 generation times in logarithmic phase (approximately 1 week). The cultures were then allowed to enter stationary phase (as defined in Chapter 2). Biomass was measured throughout the experiments by monitoring in vivofluorescenceand measuring cell numbers as described in Chapter 1.  3.2.2) C labeling and experimental time-course 14  Cultures were maintained in stationary phase for 24 h before they were harvested for radiolabeling. For each culture, 250 mL was aseptically transferred to sterile polycarbonate centrifuge bottles, which were then capped and centrifuged in an International Clinical Equipment model PRJ refrigerated centrifuge for 10 min at 2410 x g at 4°C. The cell pellets were transferred to sterile 40 mL polyethylene centrifuge tubes and diluted to 20 mL with fresh N-free medium. This concentrated the cultures 12-fold, and gave afinalcell density of approximately 3 x io cells-mL" . The tubes were loosely capped and transferred to the same 8  1  water bath in which the cultures were grown, and allowed to equilibrate for 15 min before  109 further manipulations. During the equilibration period, 20 pL aliquots were withdrawn and diluted to 20 mL with 3% NaCl for cell count measurements as described in Chapter 1. Following the equilibration period, 40 uL of 1- C labeled sodium acetate dissolved in 14  ethanol was added to each tube (2 GBq-mmol' , 37 MBq-mL' , New England Nuclear). This 1  1  gave afinalacetate specific activity of 37 kBqrnL" . The tubes were briefly vortex-mixed 1  immediately following label addition, and aliquots were withdrawn for medium, whole cell (section 3.2.3), and cell fraction radioactivity estimates (section 3.2.4). The pulse period was continued for 2 h and samples were withdrawn for analysis every 15 min. After the 2 h pulse period, the tubes were centrifuged as described previously, and the radioactive supernatant discarded. The cell pellets were rinsed three times withfreshN-free medium, centrifuged, and the supernatant discarded. Finally, the cells were transferred to sterile 1 L culture vessels, diluted to 800 mL withfreshN-free medium to give afinalcell density of approximately 6 x io cells-mL' , and placed under normal culture conditions 6  1  (Chapter 1). Aliquots were withdrawn for medium, whole cell, and cellfractionradioactivity estimates, immediately following culture dilution, 4 h later, and then every 6 h. The cultures were maintained in these conditions (stationary phase) for 46 h to allow preferential C label 14  redistribution into TAGs. Nitrate or ammonium was added to the cultures 72 h following the onset of Nstarvation (48 h following label addition) as 0.6 mL 1 MNaN03 or NFLjCl, respectively, to give final concentrations of approximately 1 mM N. The cultures were spiked with the appropriate N sources at 12, 24, and 36 h intervals to ensure N-sufficiency. Medium nitrate was measured by UV spectroscopy (Chapter 1), and medium ammonium was measured as described in Chapter 2. Aliquots were withdrawn for medium, whole cell, and cellfractionradioactivity estimates, immediately following culture dilution, and then every 6 h. The cultures were  110 maintained in these conditions (recovery from stationary phase) for 48 h to monitor C label 14  redistribution into other intracellular pools and loss to CO2 through respiration.  3.2.3) Medium,  whole cell, and intracellular  fraction  radioactivity  measurements  Medium samples were withdrawn during the labeling (100 pL) and post-labeling periods (5 mL) and pipetted into 20 mL glass scintillation vials. Immediately following sampling, 100 pL 0.1 N KOH was added to each vial to kill the cells and prevent any C 0 loss 2  from solution. After 10 min, 10 mL of Beckman Readyvalue™ scintillation cocktail was added to each vial. The vials were capped, shaken, and allowed to sit for at least 48 h before counting on a Beckman LS 6000IC scintillation counter. The counter was set at 1% efficiency, corrected for chemiluminescence, and returned radioactivity values quench corrected by the external standard ratios method. In addition, quench curves were constructed for the different sample volumes and constituents used throughout the experiment to check for linearity with increasing sample radioactivity. All these curves were linear, so a manual quench correction was not used (Appendix K). For whole cell and intracellular fraction radioactivity measurements, cells were harvested on precombusted (450°C, 4 h) 25 mm Gelman™ A/E filters set up on a Millipore™ vacuum manifold. The filters were prewetted with 1 mL of 3% NaCI containing 50 pM sodium acetate, and rinsed 3 * with 5 mL of this same solution following sample filtering to remove any residual C labeled sodium acetate. During the labeling period, 100 pL of culture was 14  sampled, and during the post-labeling period, 5 mL of culture was sampled. For whole cell radioactivity measurements, thefilterswere immediately transferred to scintillation vials after filtering, and 100 pL 0.1N KOH was added to kill the cells and trap any CO2 released. After  Ill  10 min, 1 mL of DDW was added to each vial followed by 10 mL of scintillation cocktail. The vials were capped, shaken, and counted after a waiting period of at least 48 h. For intracellular fraction radioactivity measurements, the filters were immediately folded, wrapped in aluminum foil and flash-frozen in liquid N to halt cell metabolism. These 2  were then stored for less than a week at -20°C before further processing.  3.2.4) Fractionation procedure The radioactivities in the cell intracellular fractions were determined based on the method described by Li et al. (1980). The frozen filters were transferred to a 5 mL PotterElvehjem grinding tube, and ground for 10 strokes with a Teflon-headed pestle in 2 mL of 2:1 (v/v) methanol.chloroform. The grinding tube was cooled during the procedure with cold tap water. The grindings were poured into a 13 x 100 mm glass screwtop tube, the tube rinsed with 2 * 1 mL solvent, and the rinses pooled with the grindings. The tubes were capped with Teflon-lined caps, extracted in the dark at 4°C for 1 h, and centrifuged in an ICE benchtop centrifuge at maximum speed for 5 min. The supernatants were transferred to fresh tubes, the pellets resuspended in 2:1:0.8 methanol:chloroform:DDW, vortex-mixed, recentrifuged, and the supernatants pooled. One mL of 1% NaCI and 1 mL of chloroform was added to the supernatants, the tubes capped, shaken, and centrifuged for 5 min. The upper phases (methanol fraction, containing low molecular weight metabolites (LMWM) comprising monosaccharides, sugar polyols, organic- and amino acids (Thomas and Gleitz 1993)) were transferred to scintillation vials, loosely capped, and dried in a vacuum desiccator for 24 h. One mL of DDW was added to the vials, followed by 10 mL of scintillation cocktail. The vials were shaken and counted after at least 48 h as described in section 3.2.3.  112 Polar lipid radioactivities were calculated by subtracting TAG radioactivities from total lipid radioactivities. The lower phases (chloroform fraction, containing lipids) were dried under a stream of N2 gas for 30 min. The dried lipids were redissolved in 100 pL 3:1 (v/v) chloroform: hexane and 40 uL aliquots were fractionated into TAGs and total lipids according to the procedures described in Chapter 2. For TAG elutions, the eluate in the collecting tubes wasrinsedwith 2 x 1 mL of running solvent into scintillation vials. The scintillation vials containing the total lipid and T A G fractions were dried in a vacuum desiccator for 24 h. Ten mL of scintillation cocktail was added to each vial, which were then capped, shaken, and counted after at least 48 h as described in section 3.2.3. One mL of 5% TCAA was added to each of the filter + cell pellets left over from the methanol:chloroform extractions. The tubes were capped, vortex-mixed and placed in a boiling water bath for 30 min. The tubes were centrifuged for 5 min and the supernatants (containing carbohydrates, primarily polysaccharides (Thomas and Gleitz 1993)) transferred to scintillation vials. The pellets were resuspended in 1 mL of 5% TCAA, vortex-mixed, recentrifuged, and the supernatants pooled. Ten mL of scintillation cocktail was added to each vial, which was then capped, shaken, and counted after at least 48 h as described in section 3.2.3. One mL of 0.5 N NaOH was added to each of the filter + cell pellets left over from the TCAA extractions. The tubes were processed exactly as for the TCAA extracts, except 0.5 N NaOH was used as the solvent. This fraction contained protein. Following the 0.5 N NaOH extraction, the leftover filters + cell residues were rinsed into scintillation vials with 4 x 1 mL DDW. Ten mL of scintillation cocktail was added to each of the vials, which were then capped, shaken, and counted after at least 48 h as described in section 3.2.3. The radioactivity from this fraction was determined to evaluate the completeness  113 of the extraction procedures for the intracellular fractions. All the extraction procedures were optimized to minimize the residual fraction radioactivity count, and sum as close as possible to the whole cell radioactivity count. The fractionation procedure typically recovered between 85 and 95% of the radioactivities measured in the whole cells, regardless of the N source used or stage in the experiments (Table 3.1).  3.3)  R E S U L T S  3.3.1) Biomass changes during N-starvation and recovery with nitrate or ammonium During the N-starvation period for cultures previously grown on either nitrate or ammonium as the N source, cell density was maintained near 4 * 10 cells-mL" (Fig. 3.1). Cell 6  1  division did not resume until 18 h following nitrate or ammonium addition to these cultures; thereafter cell density steadily increased for the remainder of the experiment (Fig. 3.1). The rate of cell division appeared to be more rapid for ammonium recovered cultures than for nitrate recovered cultures in the 12-18 h period after cell division resumed (Fig. 3.1). Culture in vivo fluorescence was stable and low during the N-starvation period for both cultures (Fig. 3.2). Following nitrate or ammonium addition to the cultures, there was a slight depression in in vivo fluorescence, followed by a steady increasefrom12 h after N addition until the end of the experiment (Fig. 3.2).  114 3.3.2) or  C labeling  14  and reallocation  patterns during N-starvation  and recovery with  nitrate  ammonium During the 2 h C sodium acetate pulse, whole cell label incorporation was linear for 14  N-starved cultures grown previously on nitrate (Fig. 3.3A) or ammonium (Fig. 3.4A). For both cultures, the polar lipid and LMWM fractions accumulated most of the label over the pulse period (Figs. 3.3B, 3.4B). The TAG fraction accumulated one-third to one-half of the label accumulated in total lipids, whereas the protein and carbohydrate fractions only accumulated low amounts of label. The reallocation patterns for the intracellular C following external label removal and 14  resuspension of the cells in N-free medium is shown in Figs. 3.5 and 3.6. For cultures grown previously on nitrate (Fig. 3.5A) or ammonium (Fig. 3.6A), whole cell radioactivities decreased steadily over the 46 h N-starvation period following label removal. N-starved cultures previously grown on nitrate lost a total of 1146 DPM-mL culture" (20%) during this period, 1  whereas cultures previously grown on ammonium lost a total of 801 DPM-mL culture" (16%). 1  Presumably, these losses were a result of respiration to CO2. For both cultures, the C label in 14  polar lipids decreased over the remainder of the N-starvation period; however, TAGs became the dominantly labeled fraction (Figs. 3.5B, 3.6B). These data indicated that the C label 14  increased in the TAG fraction relative to the polar lipid and other fractions as N-starvation progressed. The short term reallocation patterns for intracellular C following nitrate or ammonium 14  addition to 48 h N-starved cultures is shown in Figs. 3.7 and 3.8, respectively. Over the 24 h time period shown, there was no cell division (Fig. 3.1). For both cultures, whole cell radioactivities decreased steadily before and after nitrate (Fig.3.7A) or ammonium (Fig. 3.8A) addition, with no discernible perturbation immediately following N addition. There was also no  115 appreciable change in the radioactivities for any of the intracellular fractions during this transition period (Figs. 3.7B, 3.8B). The long term reallocation patterns for intracellular C following nitrate or ammonium 14  addition to 48 h N-starved cultures is shown in Figs. 3.9 and 3.10, respectively. In these figures, the reallocation patterns for intracellular C are shown for the 48 h period following N 14  addition to the N-starved cultures. N-starved cultures previously grown on nitrate lost a total of 1042 DPM-mL culture" (22%) during this period, whereas cultures previously grown on 1  ammonium lost a total of 633 DPM-mL culture" (15%). Thus, the carbon loss to C 0 during 1  2  N-recovery was comparable to the loss during the previous N-starvation period. This is despite the observation that cell division resumed between 18 and 24 h after nitrate or ammonium were resupplied to N-starved cultures (Fig. 3.1). For cultures recovered with nitrate, there was a loss of labelfromTAGs in the 18-48 h period following N addition (Fig. 3.9B). Corresponding with this depletion in TAG label, there was an increase in the label in the polar lipid and LMWMfractions(Fig. 3.9B). There was a small increase in the label in the carbohydratefractionnear the end of the 48 h recovery period (Fig. 3.9B). For cultures recovered with ammonium, the labeling patterns for the TAG, polar lipid, and LMWM fractions were similar to those observed during recovery with nitrate, although the magnitude of the changes was smaller (Fig. 3.10B). There was no discernible increase in the label in the carbohydratefraction(Fig. 3.1 OB). Figures 3.11 and 3.12 show modified datafromFigs. 3.9 and 3.10, respectively, where the carbohydrate and LMWMfractionshave been combined for reasons outlined in the discussion. When the datafromthe LMWM and carbohydratefractionswere combined into a single LMWM/carbohydratefraction,the radioactivities in thisfractionwere increased during nitrate assimilation (Fig. 3.11). In contrast, a similar combination offractionsfor the  ammonium recovered cultures did not result in such a large increase for the radioactivities the LMWM/carbohydratefraction(Fig. 3.12).  117  Table 3.1. Label recovery efficiencies for C-labeled intracellular pools in P. tricornutum. Label recovery efficiency was calculated as the sum fraction radioactivity/whole cell radioactivity x 100%. 14  N-source  Experimental Phase  Label Recovery Efficiency (mean ± s.e.)  Nitrate  Ammonium  -N pulse  9 3 . 2 ± 9 . 7 ( n = 9)  post-labeling N-starvation  86.6 ± 2.0 (n = 9)  N recovery  95.7 ± 2.8 (n = 9)  total experiment  91.8 ± 3 . 3 (n = 27)  -N pulse  86.3 ± 2 . 5 (n = 9)  post-labeling N-starvation  8 3 . 7 ± 0 . 9 ( n = 9)  N recovery  8 9 . 5 ± 2 . 2 ( n = 9)  total experiment  86.5 ± 1.2(n = 27)  118  0.0 ^ 0  12  24  36  48  60  72  84  Time from C-acetate label removal (h) 14  Fig. 3.1. Changes in cell density for nitrate or ammonium starved and recovered cultures of P. tricomutum labeled with C sodium acetate. Cultures were N-starved in continuous light for 24 h, labeled for 2 h, and the label removed by washing the cultures at t = 0. The cultures were N-starved for a further 48 h before being spiked with 1 mM NaN03 or NH4CI (dotted line). Subsequent N additions were made at 12 h intervals following the initial spike to ensure N-sufficiency. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. 14  119  Fig. 3.2. Culture in vivo fluorescence for nitrate or ammonium starved and recovered cultures of P. tricornutum labeled with C sodium acetate. Manipulations were made as described in Fig. 3.1. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. 1 4  120  Time from C-acetate pulse (h) 14  3.3. Uptake of C by 24 h N-starved cultures of P. tricomutum previously grown on nitrate. C-acetate was pulsed to culture concentrates (2 h prior to t = 0 in Fig. 3.1) and radioactivities were measured for whole cells (A) and in several intracellular fractions (B). Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. 14  Fig.  14  Fig. 3.4. Uptake of C by 24 h N-starved cultures of P. tricornutum previously grown on ammonium. Manipulations were carried out as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. 14  122  0  '  ;  0  1  1  12  24  36  Time from C label removal (h) 14  Fig. 3.5. Reallocation of C by 48 h N-starved cultures of P. tricomutum previously 14  grown on nitrate. Cultures that were previously 24 h N-starved were pulsed with C acetate (Fig. 3.3), then washed and resuspended infreshN-free medium under normal culture conditions (t = 0 in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. The legend refers to the symbols in panel B. 14  123  0 ^  ' 24  1  0  12  36  Time from C label removal (h) 14  Fig. 3.6. Reallocation of C by N-starved cultures of P. tricomutum previously grown on ammonium. Cultures that were previously 24 h N-starved were pulsed with C acetate (Fig. 3.4), then washed and resuspended infreshN-free medium under normal culture conditions (t = 0 in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. The legend refers to the symbols in panel B. 14  14  124  -12  -10  -8  j -6  i -4  i -2  i  L  0  2  4  6  8  10  12  Time from N addition (h)  Fig. 3.7. Short-term reallocation of C by nitrate-recovered cultures of previously Nstarved P. tricornutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mM NaN0 at t = 0 (dotted line). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. The legend refers to the symbols in panel B. 14  3  125  -12 -10 -8  -6  -4  -2  0  2  4  6  8  10 12  Time from N addition (h)  Fig. 3.8. Short-term reallocation of C by ammonium-recovered cultures of previously N-starved P. tricomutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mM N a N 0 3 at t = 0 (dotted line). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. The legend refers to the symbols in panel 14  B.  126  0  l_L  0  .  •  •  12  24  36  48  Time from N addition (h)  Fig. 3.9. Reallocation of C by nitrate-recovered cultures of previously N-starved P. tricornutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mM NaN0 at t = 0 (dotted line in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. The legend refers to the symbols in panel B. 14  3  127  Fig. 3.10. Reallocation of C by ammonium-recovered cultures of previously N-starved P. tricomutum. Cultures that had been N-starved for a total of 72 h were spiked with 1 mMNaN0 at t = 0 (dotted line in Fig. 3.1). Radioactivities were measured as described in Fig. 3.3. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible theyfitinside the symbols. 14  3  128  0  12  24  36  48  Time from N addition (h)  Fig. 3.11. Reallocation of C by nitrate-recovered cultures of previously N-starved P. tricornutum amongst modified cell fractions. The data in this figure have been reprocessed from Fig. 3.7B, and show the carbohydrate and LMWM fractions combined into a single LMWM/carbohydrate fraction. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. 14  2000 r—r  o  —'  j  0  12  —— L  1  --  24  36  48  Time from N addition (h) Fig. 3.12. Reallocation of C by ammonium-recovered cultures of previously N-starved P. tricomutum amongst modified cell fractions. The data in this figure have been reprocessed from Fig. 3.8B, and show the carbohydrate and L M W M fractions combined into a single LMWM/carbohydrate fraction. Results are means ± 1 standard error for 4 cultures. Where error bars are not visible they fit inside the symbols. 1 4  130  A. Nitrate  B. Arnmonium  Fig. 3.13. Models for carbon flow from TAGs during nitrate (A) or ammonium (B) assimilation. Both models represent carbon flows at the time of maximum TAG degradation in P. tricornutum, i.e. 18-48 h following N addition to N-starved cultures. Approximate carbonflowmagnitudes are represented by different arrow thicknesses. The broken lines indicate carbonflowsthat occur before TAGs are mobilized. C4 acids = malate, OAA, succinate; M = mitochondria; C = chloroplast; CH 0 = carbohydrate. 2  131 3.4)  D I S C U S S I O N  3.4.1)  14  C labeling and label distribution during N-starvation  The results from this chapter showed that polar lipids and low molecular weight metabolites (LMWM) were preferentially labeled by C sodium acetate during the pulse 14  period. The T A G fraction was also extensively labeled. Thomas and Gleitz (1993) found that 14  C from bicarbonate was preferentially assimilated into LMWM in the Antarctic diatoms  Nitzschia curta and Chaetoceros sp. during label incubation under N-sufficient conditions. However, in the present study the N-starved condition of the P. tricornutum cells at the time of C-acetate addition contributed to lipid labeling. This was expected, as any monomeric sugar  14  components or organic acids in the LMWM fraction would eventually be directed towards lipid synthesis during N-starvation. In P. tricornutum, lipid synthesis increases during N-starvation (Chapter 1), which would explain why C was maintained in the TAG fraction but decreased in the other fractions 14  after the external label was removed and the cultures N-starved for a further 46 h (Figs. 3.5, 3.6). In an Antarctic sea ice diatom community, carbon incorporation rates from NaH CC>3 14  into neutral lipid were greatest close to the peak bloom period (Palmisano et al. 1988). However, these authors observed a decrease in C incorporation into neutral lipids near the 14  end of the bloom period. This could be interpreted as a decrease in neutral lipid synthesis as the population reached late senescent phase. TAG synthesis also eventually decreases after several days of N-starvation in P. tricornutum (Chapter 1). In this study, the C labeled cells were N-starved for almost 2 d in the absence of 14  external C before nitrate or ammonium was added to the cultures. This allowed C to 14  14  stabilize amongst the intracellular pools (Figs. 3.5, 3.6) before the cultures were perturbed by N  addition. Historically, label distribution and reallocation in diatom intracellular fractions has been measured in the presence of external label (McConville et al. 1985, Palmisano et al. 1988, Smith and D'Souza 1993, Thomas and Gleitz 1993). This technique is useful for determining photosynthate allocation patterns, but cannot be used to accurately determine the flows between intracellular metabolite pools and respiration. It is doubtful that the fractionation procedure used was accurate in separating out the carbohydrate and LMWM fractions, which is why these have been combined in Figs. 3.11 and 3.12. When the Li et al. (1980) method is applied to diatom samples, it appears that much of the cell polysaccharide is extracted into the methanol-water (LMWM) fraction (McConville et al. 1985). The storage polysaccharide in diatoms is chrysolaminarin, a B 1-3 linked glucan (Myklestad 1989), and has been described in P. tricomutum (Ford and Percival 1965). McConville et al. (1985) found that approximately 50% of this glucan and glucose were extracted into the LMWM fraction during a Li et al. (1980) -type fractionation. Although this is problematic, a combined carbohydrate and LMWM fraction may be especially useful when considering the products of lipid degradation via the glyoxylate cycle, as these may include organic acids such as malate and succinate, and glucose as the immediate product of gluconeogenesis. The implications of this are further considered in section 3.4.2.  3.4.2) Carbonflowsduring recoveryfrom N-starvation The results suggest that during nitrate or ammonium assimilation in previously Nstarved P. tricomutum cultures, TAG carbon was largely redistributed to polar lipids, low molecular weight metabolites, and carbohydrates. Very little TAG carbon was respired to CO2. However, the flow of TAG carbon to low molecular weight metabolites and carbohydrates seemed to be much greater during nitrate assimilation compared to ammonium assimilation.  133 This suggests that, in the case of nitrate assimilation, TAG carbon was utilized for carbohydrate synthesis via 13-oxidation, the glyoxylate cycle, and gluconeogenesis. There are two additional observations that suggest that gluconeogenesis occurred during nitrate assimilation, and either not at all or at very low levels during ammonium assimilation. Firstly, the activity of the glyoxylate cycle enzyme, isocitrate lyase (ICL) was higher during nitrate assimilation than during ammonium assimilation in previously N-starved P. tricomutum (Chapter 2). In the higher plant literature, an increase in ICL activity marks the onset of gluconeogenesis in the cotyledon or endosperm tissues of germinating seeds (Leegood and ap Rees 1978a, Gerhardt 1986, Andrews and Ohlrogge 1990). Secondly, TAG mobilization in nitrate-recovered cultures only began when cell carbohydrates had reached a minimum quota, and thereafter cell carbohydrate quotas increased (Chapter 2). A similar observation for N-starved Chlorella pyrenoidosa was made by Klyachko-Gurvich et al. (1973). These researchers observed that nitrate addition to 5 d N-starved cultures resulted in lipid degradation and carbohydrate accumulation after a 8 h lag period. There are two alternative explanations for the C reallocation patterns observed after 14  nitrate or ammonium addition to N-starved cultures. One possibility is that the increase in the 14  C in the LMWM/carbohydratefractionin nitrate recovered cultures (Fig. 3.11) was a  consequence of photosynthetic refixation of recently respired C 0 . However, this was 14  2  unlikely to have occurred because there was no increase in the C in the proteinfractionduring 14  the recovery period. Some increase in thisfractionwould be expected if respired C 0 was 1 4  2  being refixed, since the intracellular protein quota is known to at least doublefromde novo synthesis within 24 h of nitrate addition to N-starved P. tricomutum cultures (Chapter 1). Secondly, the increase in the proportion of C in the LMWM/carbohydratefractionmay have 14  been a consequence of organic acid synthesis for anaplerotic metabolism rather than  134 gluconeogenesis. In other words, malate and oxaloacetate derived from the glyoxylate cycle may have been used to replace TCA cycle intermediates drawn off for protein synthesis. However, as noted previously, there was no increase in C in the protein fraction, which would 14  be expected if anaplerotic protein synthesis was occurring. Additionally, ammonium assimilation in the green alga Selenastrum minutum appears to stimulate a greater anaplerotic demand for carbon than does nitrate assimilation (Schuller et al. 1990). However, in the present study, there was a negligible increase in the proportion of C in the L M W M pool 14  observed during ammonium assimilation in the short term (Fig. 3.8) or long term (Fig. 3.10), compared with the greater increase observed during nitrate assimilation in the long term (Fig. 3.9). Thus, it is unlikely that TAG degradation supplies carbon via anaplerotic metabolism for protein synthesis during nitrate or ammonium assimilation in P. tricornutum.  3.4.3) A modelfor degradative TAG carbon flows during recovery from N-starvation The data from this study cannot unequivocally clarify the pathways for carbon degradation from TAGs during nitrate or ammonium assimilation. In order to accomplish this, 14  C acetate pulse-chase studies need to be carried out over short time intervals throughout the  18-48 h period of TAG degradation following nitrate or ammonium addition to N-starved cultures. If this is done, and the C in the gluconeogenic intermediates (malate, OAA, glucose) 14  is determined, then the sinks for the mobilized TAG carbon could be accurately determined. However, based on the data from this chapter and the data in Chapter 2, a model has been prepared to demonstrate the possible sinks for TAG carbon during nitrate or ammonium assimilation in P. tricornutum (Fig. 3.13). In this model, during nitrate assimilation, the supply of intracellular carbohydrate is rapidly exhausted by respiration through glycolysis and the TCA cycle to provide the reductant needed for nitrate reduction via nitrate reductase (Chapter 2).  135 The respired carbohydrate is replaced by gluconeogenesis. During ammonium assimilation, storage carbohydrates are not readily demanded or utilized, so the TAGs are mobilized at a lower rate. In neither case is there an anaplerotic flowfromTAGs to protein synthesis in the chloroplast. During both nitrate or ammonium assimilation, the resumption of cell division will also provide a demand for new cell and organelle membranes. The lipid components of these membranes may be supplied by transacylation and desaturation of T A G lipids. The role of degradative lipid metabolism during recoveryfromnutrient or other stresses that reduce phytoplankton growth rates is still unclear. Together with observations that it is difficult to measure any advantage in terms of growth rates when comparing nitrate and ammonium as N-sources for microalgae (Thompson et al. 1989, Levasseur et al. 1993), the datafromthese experiments suggests that diatoms are rarely energy limited. Rather, it appears that storage carbohydrate is rapidly turned over as a cell carbon and energy source, and TAG degradation is one mechanism by which recently degraded carbohydrate can be replenished. It also appears that morefieldstudies are warranted on the enhancement of storage lipid and carbohydrateflowsduring recoveryfromnutrient deficiencies. These transitions seem to have a much greater stimulating effect on storage carbon mobilization than light/dark transitions.  136  CHAPTER 4  T R I A C Y L G L Y C E R O L SYNTHESIS AND DEGRADATION AS A DETERMINANT OF CARBON STABLE ISOTOPE FRACTIONATION  4.1)  INTRODUCTION  One of the more controversial subjects in biological oceanography and paleoceanography in recent years has been the utility of stable isotope ratios in marine sediments to hindcast prehistoric atmospheric C 0 concentrations. More specifically, these 2  predictions are made assuming that the standardized ratio of C / C (8 C) in the particulate 13  12  13  organic carbon (POC) of marine phytoplankton is influenced by the concentration of dissolved inorganic carbon (DIC) in the oceans, which in turn is in equilibrium with atmospheric CO2. As the concentration of CO2 decreases, more C isfixedinto POC by the action of the primary 13  CO2fixingenzyme, ribulose bisphosphate carboxylase/oxygenase (RUBISCO). This enzyme discriminates against C at approximately 29%o during photosynthetic CO2fixation(Roeske i 3  and O'Leary 1984). When CO2 becomes limiting, relatively more C isfixedand 8 C values 13  13  for the POC approach those of the source CO2. Several difficulties arisefromthe assumptions made about the C / C discrimination 13  12  process. These are all related to processes that may change the isotopic composition and concentration of the CO2 carbon at the site of RUBISCO fixation. Physical processes may limit the supply of CO2 to RUBISCO. These include increased surface ocean temperatures, which decrease the concentration of C 0 that can remain in solution (Rau et al. 1997), the chemical 2  137 speciation of DIC amongst HC0 ", C0 " and C 0 (Mook et al. 1974), and changes in the 2  3  3  2  permeability of the cell membrane to C 0 (Marcus et al. 1982). There is also increasing 2  evidence that phytoplankton cell physiology may be altered by the induction of carbon concentrating mechanisms, which increase the concentration of C 0 at the active site of 2  RUBISCO, preventing the expression of discrimination against C because C 0 efflux is 13  2  minimized (Fielding et al. in press). Lipid accumulation has been proposed as a possible important variable in contributing to changes in phytoplankton 8 C values (Sackett et al. 1965). The study of the 8 C signature 13  13  in lipids is particularly important in marine biogeochemistry, because all sedimentary biomarkers except porphyrins are derived from lipids (Hayes 1993). Storage lipid biosynthesis is one process that occurs under the right physiological conditions to facilitate discrimination against C . Lipid biosynthesis utilizes carbon that originates from a glycolytically generated 13  pool of pyruvate in the chloroplast (Andrews and Ohlrogge 1990). The preferential selection of pyruvate depleted in C is thought to be the mechanism that leads to lipids having more 13  negative 8 C values than the source pyruvate, or other cell carbon (DeNiro and Epstein 1977, 13  Melzer and Schmidt 1987). During lipid biosynthesis in Escherichia coli, the carbon in the lipid pool is C-depleted 7%o relative to the source pyruvate (DeNiro and Epstein 1977). 13  These observations suggest that the high lipid content in cold water diatoms may be a causative factor for these populations having more negative whole cell 8 C values compared to 13  their lipid-poor temperate or tropical counterparts (Sackett etal. 1965). However, the 8 C 13  values for a number of plant and algal lipids have been shown to be negatively correlated with cell lipid content (Park and Epstein 1961). The most likely reason for this is the retention of C-enriched pyruvate not utilized for lipid synthesis. Unless this pyruvate is respired to C 0 ,  13  2  an overall depletion of whole cell C cannot be attributed to lipid synthesis. Thus, the 13  138 correlation of cooler sea surface temperatures with more negative POC 8 C values (Sackett et 13  al. 1965, Rau et al. 1997) is more likely to be a consequence of increased C 0 supply to 2  RUBISCO, facilitated by low light levels, low temperatures, or low growth rates (DescolasGros and Fontugne 1990, Johnston 1996). Although the quantity of storage lipid is unlikely to influence whole cell 8 C values in 13  phytoplankton, the relationships between carbon isotopic discrimination and the processes of lipid accumulation and degradation in these organisms have not been investigated. Notably, none of the studies that have examined phytoplankton 8 C values have compared lipid 8 C 13  13  values independently of temperature changes. Additionally, the metabolism of lipid carbon during the resumption of growth following a period of inhibition may influence the 8 C values 13  of other carbon components in the cell. This might be a potential mechanism for the whole cell 8 C values to be maintained at negative values (relative to rapidly growing populations) 13  following storage lipid degradation in phytoplankton recoveringfromnutrient limitation. In the marine diatom P. tricomutum, it is expected that discrimination against C 1 3  should increase during N-starvation because the cells do not divide (Chapter 2); thus CO2 supply to RUBISCO will be increased. In this chapter, 8 C values are reported for the 13  triacylglycerols (TAGs), total lipid, and whole cells of the marine diatom P. tricomutum. The results show the relative importance of total and storage lipid (TAG) biosynthesis and degradation in contributing to the whole cell 8 C values before, during, and after recovery 13  from a period of N-starvation.  139 4.2)  MATERIALS AND METHODS  4.2.1)  Culture  conditions  Artificial seawater (ESAW) containing 1 mM NaN0 was prepared and used to grow 3  P. tricornutum cultures as described in Chapter 1. DIC concentrations and 5 C were closely 13  controlled by scrubbing CO2 out of the laboratory air supply with two in-line soda lime columns, and aerating the cultures with controlled quantities (0.5% v/v) of bottled C 0 mixed 2  with the scrubbed air. Quadruple cultures were grown in continuous light at 18°C for 8-10 generation times in logarithmic phase (approximately 1 week). The cultures were then allowed to enter stationary phase (as defined in Chapter 2). After 4 d of N-starvation, the cultures were diluted with 4 volumes of ESAW containing 1 mM NaN0 to resupply N. Samples were taken 3  for measurements just before the cultures entered stationary phase, 4 d after the onset of Nstarvation, and 2 d following the recovery of stationary phase cultures. Biomass was measured throughout the experiments by monitoring in vivo fluorescence and measuring cell numbers as described in Chapter 1. Medium nitrate concentrations were monitored as described in Chapter 2.  4.2.2) Isolation  and measurement  of carbon and S C 13  The concentration of DIC in the growth medium was determined by acidifying a 2.8 mL medium sample and measuring the concentration of the released C 0 with an infrared gas 2  analyzer (IRGA) (Analytical Development Company, Hoddesdon, England). The medium sample was pipetted into a N -flushed 44 mL glass vial. The vial was capped with a septum2  lined cap. Using a syringe and needle, 0.2 mL of 85% phosphoric acid was injected into the vial. The vial was vortex-mixed, and after 2 min a 1 mL gas sample was withdrawn. Triplicate  140 0.2 mL volumes were injected into the IRGA, which was set up with N as a carrier gas and 2  calibrated with a 450 ppm C 0 in air standard. DIC concentration was calculated according to 2  equation 1:  ppmC02 x V C DIC (mM) = *f. TT 7T~~ '  v  (1)  VI x V A x V M  where Vc = gas volume collected (mL) Vi = gas volume injected (mL) VA = aqueous sample volume (mL) V M = ideal gas volume, 22414 mL-mol'  1  The 5 C of the medium was measured in purified C 0 samples. These were obtained 13  2  by withdrawing 6-8 mL medium samples into evacuated glass ampoules containing 2 mL of 85% phosphoric acid. The C 0 released was trapped cryogenically using liquid N , sealed into 2  2  evacuated glass tubes, and the 8 C determined with a Prism™ triple collector mass 13  spectrometer (VG Isotech, Manchester, UK). All 8 C determinations were made relative to 13  Pee Dee Belemnite standard, and calculated according to equation 2 (Boutton 1991):  13/-.  8"C  _  Rsample - Rstandaid Rstandard  where R = the absolute molar C / C ratio 13  Rstandarf = 0.0112372  12  x  10  3  (2)  141  The 8 C values for whole cells, total lipids, and TAGs were determined in particulate 13  samples. For whole cell determinations, 1 * 10 cells were vacuum filtered onto precombusted 7  (450°C, 4 h) 13 mm Gelman™ A/E glass fibre filters, rinsed with 2 x 1 mL 3% NaCI, dried at 55°C for 24 h, and stored in a desiccator at -20°C. The filters were compressed in tin pellets and combusted in a Fisons™ automated CFfN analyzer online with the mass spectrometer for carbon isotope and mass analyses. Total lipids were extracted and concentrated from filterharvested cells as described previously (Chapter 1). For logarithmic phase and N-recovered cultures, 5 x 10 cells were harvested because of the low cell lipid content during these growth 8  phases. For stationary phase cultures, 3 x 10 cells were harvested. TAGs were isolated from 7  the total lipid fraction as described in Chapter 1. Liquid samples of the total lipid and T A G extracts were transferred into tin cups, dried in a vacuum dessicator overnight, and analyzed in the mass spectrometer as for whole cells. Lipid fractionation during the extraction procedures and T A G separation was checked using a tripalmitin standard. Relative to the 8 C value of 13  this standard (-28.67%o), the lipid extraction procedure changed the 8 C by 0.49%o, whereas 13  TAG separation changed the 8 C value by 0.53%o. 13  Fractional abundances (F) were calculated according to equation 3 (Boutton 1991):  142 Molar C quantities were calculated according to equation 4: 13  mol C = F x total moles C  (4)  13  Molar C quantities were calculated by subtracting mol Cfromtotal molar carbon. These 12  13  values were used to calculate R, F, and 8 C values for the cell components not directly 13  measured (non-lipids (NL) and non-TAG lipids (NTL)) by assuming mass balance and rearranging equation 5: (5)  ricFc = niFi + n F . . . 2  2  where nc = whole cell carbon (moles) F = whole cellfractionalabundance c  and the subscripts 1,2... represent n and F values for subcellular fractions  Finally, carbon isotope discrimination (A) expressed during the movement of carbon between pools was calculated according to equation 6 (Farquhar et al. 1989):  A (%>) =  Rsource  Rproduct  -1  xlO  3  (6)  143  4.3)  RESULTS  Culture DIC concentrations did not varyfroman average of 7.6 ± 0.2 mM throughout all growth phases. Cultures at the end of logarithmic phase growth had an average whole cell 8 C value of-47%o, a difference of 15%o compared to the source DIC of-32%o (Fig. 4.1). 13  Although the medium DIC 8 C values were significantly different between the three growth 13  phases, they did not vary by more than 1.5%o (Fig. 4.1). The 5 C value for whole cells became 13  more negative by 1 l%o during the transitionfromlogarithmic phase to stationary phase. At the end of the logarithmic phase growth, the most negative 8 C values were for the total lipid, 13  non-TAG lipid, and TAG lipidfractions(Fig. 4.1). In stationary phase cultures, the 8 C values 13  for all cellfractionsbecame more negative, although the change was insignificant for TAGs (Fig. 4.1). When nitrate was added back to the cultures, the 8 C values for the whole cells and 13  the cellfractionsreturned to the values observed for logarithmic phase cultures (Fig. 4.1). During the transitionfromlogarithmic to stationary phase, whole cell POC increased significantly by 3.9 pg-cell' (Fig. 4.2). Most of this increase was a consequence of lipid 1  synthesis; total lipids increased by 3.7 pg-cell' , and the TAG component of the total lipids 1  increased by 3.0 pg-cell' (Fig. 4.2). However, although there was no increase in the non-lipid 1  carbonfraction,this carbon represented over half of the whole cell carbon in both logarithmic and stationary phase cultures (Fig. 4.2). In the recovery phase, total lipid carbon decreased relative to logarithmic phase concentrations as a consequence of TAG degradation, but the non-lipid carbonfractionincreased (Fig. 4.2). Isotopic discrimination (A%o) was calculated for carbon transferfromDIC into the cell, and for carbon transfer among the various cellfractions(Fig. 4.3). The within cell transfers were calculated by assuming that total and TAG lipid synthesis both use non-lipid cell carbon  144 components as biosynthetic precursors. At the end of logarithmic phase growth, A for DIC assimilation into POC was 15.8%o (Fig. 4.3). The A values for total lipid and TAG synthesis were somewhat lower, being 8.7 and 13.5%o, respectively (Fig. 4.3). In stationary phase cultures, the A for DIC assimilation into whole cell POC increased significantly to 25.0%o (Fig. 4.3). However, the A for total lipid and TAG synthesis decreased to 4.0%o and 2.7%o, respectively (Fig. 4.3). When the N-starved cells recovered after nitrate addition, all A values returned to values similar to those observed for logarithmic phase cultures (Fig. 4.3).  145  :SSS  H  Stationary I Recovered  Cell fraction  Fig. 4.1. Changes in 8 C values in P. tricornutum cell fractions before, during, and after recoveryfromN-starvation. The bars represent a transitionfromlogarithmic phase cultures (white, N just depleted) to stationary phase (gray, 4 d N-starved), and recovered with nitrate (black, 2 d after recovery with NaNOs). Growth medium (M), whole cell (WC), total lipid (TL) and TAGfractionswere isolated and their 6 C values determined as described in the materials and methods. The non-lipid cellfraction(NL) 8 C value was calculated assuming mass balance and using whole cell and total lipid 8 C and pg C values. The non-TAG lipidfraction(NTL) 8 C value was similarly calculated using total lipid and TAG 8 C and pg C values. Results are means ± 1 standard error for 4 cultures. Statistical tests (repeated measures ANOVA, p < 0.01, followed by Tukey's test, q < 0.01, a = 0.01) were performed to detect differences between the three growth phases for the 8 C values. Within each group of three bars, identical letters above the bars indicate no significant difference between the means. 13  13  13  13  13  13  13  WC  NL  TL  NTL  TAG  Cell fraction  Fig. 4.2. Carbon quotas in P. tricomutum cell fractions before, during, and after recoveryfromN-starvation. The barfillsand cell fraction codes are as described in Fig. 4.1. The carbon content of the whole cell (WC) and TAGfractionswere measured directly. The carbon content of the non-lipid (NL) and non-TAG lipid (NTL) fractions were calculated assuming mass balance and using total lipid (TL) carbon data and data from the measuredfractions.Results are means ± 1 standard error for 4 cultures. Statistical tests (repeated measures ANOVA P < 0.01, followed by Tukey's test, q < 0.01, a = 0.01) were performed to detect differences between the three growth phases for the cell carbon values. Within each group of three bars, identical letters above the bars indicate no significant difference between the means.  147  30  Assimilation step  Fig. 4.3. Carbon isotope discrimination between cell compartments in P. tricornutum before, during, and after recovery from N-starvation. The barfillsand cell fraction codes are as described in Fig. 4.1. Carbon isotope discrimination (A) was calculated for the assimilation of dissolved inorganic carbon into whole cell organic carbon (M-*>WC), total lipid synthesis from non-lipid cell carbon (NL->TL), and TAG synthesis from non-lipid cell carbon (NL->TAG). Carbon discrimination values were calculated from absolute C / C ratio (R) values as described in the materials and methods. Results are means ± 1 standard error for 4 cultures. Statistical tests (repeated measures ANOVA, p < 0.01, followed by Tukey's test, q < 0.01, a = 0.01) were performed to detect differences between the three growth phases for the assimilation step A values. Within each group of three bars, identical letters above the bars indicate no significant difference between the means. 13  12  148 4.4)  DISCUSSION  4.4.1) DIC assimilation and <5"C In phytoplankton, the observed worldwide 8 C values for whole cell POC range 13  between -14 and -35%o (Descolas-Gros and Fontugne 1990). This range in isotopic composition has largely been attributed to variations in sea surface temperature (Sackett et al. 1965) or phytoplankton growth rates relative to C 0 supply (Laws et al. 1995). Low 2  temperatures or growth rates lead to a greater expressed discrimination against C , and more 13  negative 6 C values. Thus, the mechanism by which temperature and growth rates affect 6 C 13  13  is thought to ultimately rest with a change in the supply of C 0 to RUBISCO. 2  The resultsfromthis chapter show that the whole cell POC 8 C in P. tricomutum can l3  become more negative during the process of N-starvation. This appears to be a consequence of reducing the net growth rate to zero. Laws et al. (1995) demonstrated that 8 C in the POC of 13  P. tricomutum became increasingly more negative as the growth rate slowed in nutrient-limited chemostat cultures. Their values for isotopic discrimination (e « A) calculated for the p  maximum and minimum ratios of growth rates .ambient aqueous C 0 concentrations were 15 2  and 25%o, respectively. These were very similar to the A values calculated for DIC assimilation into POC during logarithmic and stationary phase growth, respectively, in this study (Fig. 4.4). Laws et al. (1995) suggested that the relationship between isotopic discrimination and growth rate would hold as long as carbon concentrating mechanisms do not operate, and discrimination is largely controlled by the flow of C 0 through RUBISCO. A similar study was done by 2  Johnston (1996), who showed that P. tricomutum 8 C became more negative if the culture 13  temperature was decreased, which suggests that difiusional C 0 limitation was also decreased. 2  149 Alternatively, one assimilatory mechanism that may contribute to a less negative 8 C 13  signal in the POC is a switch from RUBISCO mediated carboxylation, which utilizes C 0 , to 132  carboxylation reactions, which utilize HCO3". In P. tricornutum, Descolas-Gros and Fontugne (1985) showed that some carbon fixation occurred via the 13-carboxylating enzyme phosphoenolpyruvate carboxykinase (PEPCKase), and they suggested that this process was partly responsible for the observed 8 C of -14.9%o. However, these authors could not 13  demonstrate a significant correlation between PEPCKase activity and the 8 C signal, and have 13  since suggested that 13-carboxylation is not the mechanism of primary carbon fixation in diatoms because the initial photosynthetic products are C3 sugars (Descolas-Gros and Fontugne 1990). Moreover, Glover and Morris (1979) demonstrated that 6-carboxylating activity increased relative to RUBISCO in P. tricornutum cultures entering stationary phase. This should result in a less negative 8 C value if RUBISCO decreases in significance as the primary C 0 fixing l3  2  enzyme; however, this was not seen in this study. Thus, the assimilation of C 0 via RUBISCO 2  will be the main determinant for discrimination against C during the process of DIC 13  assimilation.  4.4.2) Lipid biosynthesis and whole cell Lipids are known to accumulate in diatoms at low temperatures (Smith and Morris 1980), so it seems plausible that a link might exist between lipid accumulation and discrimination against C . However, studies on the role of phytoplankton lipids in determining 13  whole cell 8 C have been inconclusive (Descolas-Gros and Fontugne 1990). One major 13  shortfall of these studies is that they have failed to separate the effects of lipid synthesis from the effects of temperature changes on whole cell 8 C. This separation is conceptually 13  important, because whereas whole cell 8 C values become more negative by 0.28-0.3 5%o with 13  150 every 1°C decrease in temperature (Descolas-Gros and Fontugne 1990), the expression of discrimination against C during the metabolic process of lipid synthesisfromcarbohydrate 13  precursors has been shown to decrease as temperatures decrease (DeNiro and Epstein 1977). In their study on the yeast pyruvate decarboxylase, DeNiro and Epstein (1977) found that discrimination against C during the transformationfrompyruvate to acetaldehyde decreased 13  approximately 0.18%o for every 1°C decrease. Thus, temperature and lipid accumulation may have contradictory effects on the final 8 C signal expressed in the lipid pool. 13  The data indicate that lipid biosynthesis was unlikely to be a contributing factor to more negative whole cell 8 C values as P. tricomutum progressedfromlogarithmic to stationary 13  phase. This is because the 8 C value of the non-lipid (NL)fractionalso became more negative 13  as the cultures entered stationary phase (Fig. 4.1). However, although RUBISCO-mediated discrimination against C can be invoked as an explanation for the variable 8 C values 13  13  observed in phytoplankton, this view may be too simplistic to completely explain a change in the whole cell POC 8 C. Isotope effects during DIC assimilation into POC is one of four 13  determinants that may affect a 8 C signal, the others being the carbon source, isotope effects 13  during carbon metabolism, and intracellular carbon budgets (Hayes 1993). An important consideration is that intracellular metabolism will only alter whole cell 8 C values if there is a 13  disproportionate loss of one isotope (or one of its metabolic derivatives)fromthe cell into the surrounding medium. In stationary phase cultures of P. tricomutum, mannose and glucoronic acid may be excreted into the surrounding medium (Hoagland et al. 1993). Additionally, many unicellular algae are known to synthesize and excrete dimethylsulfoniopropionate (DMSP) during the onset of stationary phase (Matrai and Keller 1994). DMSP is synthesizedfromthe sulfur-containing amino acid, methionine (Gage et al. 1997), which may in turn be indirectly synthesizedfrompyruvate (Ireland 1990). Thus, it is possible that the carbonfrom C 1 3  151 enriched pyruvate left behind during the process of lipid synthesis may be excreted from the cell. However, this link is tenuous at best because no data exist on the carbon isotopic composition of DMSP or excreted polysaccharides.  4.4.3) Comparison of total lipid and TA G b* C values during lipid synthesis 3  In Escherichia coli, DeNiro and Epstein (1977) demonstrated that the selection of pyruvate was the main determinant in the depletion of C in the lipid fraction. Melzer and 13  Schmidt (1987) showed that the step responsible for the discrimination against C during lipid 13  synthesis in E. coli and yeast was the pyruvate dehydrogenase complex (PDH), which catalyzes the decarboxylation and oxidation of pyruvate to acetyl CoA. The discrimination between carbohydrate and lipid during this step was 4-6%o (Melzer and Schmidt), and is in accordance with the resultsfromFig. 4.3. Another possible branch point in the synthesis of lipids where the final lipid 8 C may be altered, is the selection of PDH-derived acetyl CoA for respiration in 13  the tricarboxylic acid cycle, rather than lipid synthesis. However, given that lipid synthesis in autotrophs occurs in the plastids and uses acetyl CoA derivedfroma plastidic PDH complex (Reid et al. 1977), it is unlikely that this acetyl CoA would be available for mitochondrial respiration. The 8 C data in Fig. 4.1 show that the lipid pool in P. tricornutum, particularly TAGs, 13  were the most C-depleted components of all the cell carbon pools during all growth phases. 13  As the lipid content increasedfromlogarithmic phase to stationary phase, the 8 C values for 13  the T A G pool did not vary (Fig. 4.1), although the carbon in this pool increased significantly (Fig. 4.2).  If the 8 C values for the TAG carbon are viewed in isolation of the 8 C values for 13  13  the rest of the cell carbon, the data in Fig. 4.1 seems to suggest that pyruvate was never limiting for T A G synthesis during any of the three growth phases. However, the isotopic discrimination  152 values in Fig.4.3 show the reason for the apparently equivalent TAG 8 C values in Fig. 4.1. 13  During the transitionfromlogarithmic to stationary phase, the expressed discrimination during DIC assimilation into whole cell carbon increased by approximately 10%o (Fig. 4.3). Conversely, the expressed discrimination during TAG synthesisfromnon-lipid intracellular precursors decreased by approximately 10%o (Fig. 4.3). This suggests that the precursors (e.g. pyruvate) for TAG synthesis did become more limiting in stationary phase, as might be expected if the rate of lipid synthesis was increased. This would have led to the stationary phase TAG 8 C values approaching those of the intracellular source carbon. 13  In contrast to the TAG carbon, the 8 C values for the non-TAG lipids (NTL) became 13  significantly more negative during the transitionfromlogarithmic to stationary phase (Fig. 4.1), although the carbon increased very little in this pool (Fig. 4.2). This was most likely a consequence of the difference in the carbon composition between the TAG and non-TAG lipid pools. All TAG carbon is derived directlyfromfatty acid synthesis, except for the three glycerol carbons. In contrast, non TAG lipids in P. tricomutum include monoacylglycerols and diacylglycerols containing galactose and digalactose headgroups (Yongmanitchai and Ward 1992). The carbon in these headgroups will reflect the 8 C values of their biosynthetic 13  sources, which would be carbon derivedfromrecent photosynthate or glycolysis.  4.4.4) Lipid degradation and whole celltf Cvalues 3  Following nitrate addition to stationary phase cultures of P. tricomutum, TAG carbon was degraded (Fig. 4.2). The datafromChapter 3 indicated that most of the TAG carbon was reallocated to polar lipids, low molecular weight metabolites, and carbohydrates within 48 h of recovering N-starved cultures with nitrate. However, the 8 C values for the whole cell and 13  component pools returned to their logarithmic phase values (Fig. 4.1). This is not consistent  153 with the transfer of TAG carbon to other cell pools (Chapter 3). However, the fixation of new carbon by photosynthesis will also contribute to less negative whole cell 8 C values, as 13  relatively more C 0 is made available to RUBISCO as a consequence of decreased C 0 efflux 13  2  2  rates. Thus, if the contribution to other cell carbon from TAGs is discounted, perhaps the whole cell 8 C values for nitrate-recovered cells would have been closer to the source DIC 13  values than shown in Fig. 4.1. The data also suggests that TAG synthesis was also no longer limited by the supply of pyruvate or other substrates, as TAG 8 C values remained very 13  negative, whereas the whole cell values became closer to the source DIC (Fig. 4.1). Thus, lipid synthesis or degradation in natural diatom populations is unlikely to be a causative mechanism for altering whole cell 8 C values. Utilizing 8 C values from 13  13  phytoplankton-derived lipid products in oceanic sediments for hindcasting global C 0  2  concentrations should therefore be done with caution. If phytoplankton T A G breakdown products are deposited in the sediments, then the 8 C value of these products will be an 13  underestimate of the C content in the original total cell carbon pool. Additionally, the 8 C 13  13  values for the lipid fraction are more likely to closely resemble those for the rest of the cell carbon in N-starved cells than in N-sufficient cells. This is because discrimination against C 13  during lipid synthesis is only likely to be important in determining lipid 8 C values when the 13  rate of lipid synthesis is low. In stationary phase, when the rate of lipid synthesis is increased, the lipid 8 C values will more closely resemble those measured for the substrate carbon. 13  154  GENERAL CONCLUSIONS  This thesis has examined the patterns of storage carbon accumulation, particularly in triacylglycerol fatty acids, in the marine diatom, Phaeodactylum tricomutum. Although storage lipid accumulation has been well documented in the past for several phytoplankton species, both in the laboratory and thefield,this is thefirststudy that has specifically investigated the accumulation of carbon in triacylglycerol fatty acids over the transition period from N-sufficiency to N-starvation. To date, the role of accumulated triacylglycerols in phytoplankton has been unclear. This work also presents thefirstevidence for the utilization of triacylglycerols to provide the carbon and energy necessary for nitrate and ammonium assimilation in lipid-accumulating phytoplankton. Chapter 1 investigated the timing of storage carbon accumulation during the process of N-starvation in P. tricomutum. The results showed that lipids did not accumulate unless the cultures were N-starved for more than 24 h. This confirms the importance of lipids as longterm energy and carbon storage compounds. When lipids did accumulate, they accumulated as triacylglycerols. Almost all the triacylglycerol carbon accumulated in hexadecanoic (16:0), and c/.s-9-hexadecenoie (16:1) fatty acids. Although both fatty acids accumulated in the triacylglycerols, 16:0 was preferentially synthesized relative to 16:1. When these fatty acids were subsequently degraded following nitrate addition to the cultures, 16:0 also appeared to be preferentially degraded. Thus, the 16:0/16:1 ratio is not stable in marine diatoms, and the use of this ratio in chemotaxonomic studies should be viewed with caution. The datafromChapter 1 also showed that carbohydrates were degraded before triacylglycerols after the cultures were recovered with the addition of nitrate. This was not a function of storage carbon availability, because carbohydrates were degradedfirstirrespective  155  of the size of the triacylglycerol pool. This suggests that there is a biochemical preference for the utilization of carbohydrate carbon over storage lipid carbon for respiration or other metabolic processes during N assimilation. Chapter 2 investigated the effect that different N sources had on the mobilization of carbohydrates and triacylglycerols following recovery from N-starvation. Although nitrate assimilation theoretically requires more reductant per unit carbon than ammonium assimilation, there was no difference in the timing or magnitude of triacylglycerol degradation during the assimilation of either of these two N sources. Theoretically, fatty acid oxidation can yield more reductant and conserve more carbon per unit reductant produced, compared to carbohydrate oxidation. However, when either fatty acid or carbohydrate oxidation proceeds through anaplerotic pathways, the calculated ratios of reductant produced:carbon conserved are always in excess of the ratios required for protein synthesis from nitrate or ammonium. Thus, although carbohydrate degradation occurred before triacylglycerol degradation, even if the demand for carbon and reductant was high following recovery from N-starvation, there appears to be no energetic advantage to degrade lipids rather than carbohydrates. Nevertheless, triacylglycerol and carbohydrate degradation do appear to be metabolically linked. In Chapter 2, the induction of higher isocitrate lyase activities following nitrate addition, compared to ammonium addition, to N-starved cultures in the light suggested that there is a greater carbon demand during nitrate assimilation. This demand appears to be met by carbohydrate oxidation, but some triacylglycerol degradation is targeted towards replacing the respired carbohydrates through gluconeogenesis. This process is probably not linked to the N source being assimilated, but the carbohydrate status of the cell. This conclusion is further reinforced by the observation that there was a net synthesis of carbohydrate in dark-recovered cultures, irrespective of the added N source. In this case,  156 carbohydrate quotas that were lowered during the processes of dark nitrate or ammonium assimilation and maintenance respiration could have been ameliorated by gluconeogenesis from triacylglycerols. The radiolabeling experiments in Chapter 3 showed that carbon that was lost from triacylglycerols accumulated in N-starved cultures between 18 and 48 h following recovery with nitrate or ammonium. This confirmed the data collected in Chapter 2, where the triacylglycerol fatty acids levels were shown to decline under the same conditions. The data in Chapter 3 showed that the sink for triacylglycerol carbon was in low molecular weight metabolites, carbohydrates, and polar lipids. During nitrate assimilation, more triacylglycerol carbon was directed towards low molecular weight metabolite and carbohydrate synthesis than during ammonium assimilation. This suggests that gluconeogenesis was more important during nitrate assimilation than during ammonium assimilation. This makes sense in light of the increased isocitrate lyase activities measured during nitrate assimilation in Chapter 2, and the increased particulate organic carbon values following recovery with nitrate (Chapter 1). Thus, in Chapter 3, new carbohydrate, protein, and lipid synthesis would have been occurring following nitrate addition, with some carbon being provided de novofromphotosynthetic CO2 fixation. Triacylglycerol carbon mobilization is most likely limited to the initial period following N recovery, when carbohydrate quotas are low and the photosynthetic rate is low. Once photosynthetic CO2fixationhas recovered and is balanced with the growth rate, any remaining triacylglycerol carbon might be respired and lostfromthe cell. Two different conclusions can be drawnfromthe carbon stable isotope data in Chapter 4. Firstly,froma biochemical point of view, the 8 C data suggested that triacylglycerol 13  synthesis was not limited by its substrates in N-sufficient cells. This makes sense in light of the low rate of storage lipid synthesis and accumulation in rapidly growing phytoplankton cells.  However, in N-starved cells, the triacylglycerol 8 C values became closer to the values for the 13  other cell carbon, which would be expected if the rate of triacylglycerol synthesis increases. Despite these differences, the 6 C valuesfromtriacylgfycerols seemed independent of the 13  physiological state of the cultures. Therefore, the second conclusion is that triacylglycerol carbon may be a stable indicator of dissolved inorganic carbon 6 C values, irrespective of 13  phytoplankton physiological state. This suggests that triacylglycerol carbon, and its derivatives, should make excellent biogeochemical markers. However, a caveat must be added to this conclusion - triacylglycerol 6 C values will always be more negative than whole cell 8 C 13  13  values. Thus, any phytoplankton degradation products derivedfromtriacylglycerols are likely to have more negative 8 C values relative to the original whole cell carbon. I3  158  ECOLOGICAL IMPLICATIONS  Although Phaeodactylum tricomutum is considered to be an atypical marine diatom because its cell wall is very weakly silicified (Darley 1974), it has a biochemical composition comparable to other diatoms with respect to carbohydrates (Craigie 1974) and lipids (Wood 1974). This typical composition, combined with the ease of culturing P. tricomutum in the laboratory, has resulted in the use of this diatom as a representative model for important physiological processes such as photosynthesis, respiration, and nutrient assimilation. In the oceans, physical factors such as light and nutrient levels, wind mixing, and sinking rates, combined with biotic factors such as grazing by zooplankton, are some of the major determinants of phytoplankton community structure. Thus, it is a difficult task to use physiological and biochemical observations from laboratory studies on a single species to make predictions about diatom communities in the oceans. Nevertheless, some key conclusions from the chapters in this thesis may be relevant to diatom communities in situ. The degradation of storage lipids in diatoms may serve to maintain other metabolic • processes when the carbon and energy availablefromphotosynthesis are diminished. For example, at the nutricline, where light levels are low and nutrient levels are high relative to the surface of the ocean, storage lipid degradation may allow some nitrate assimilation to continue. This would effectively lower the C:N ratio of the nutricline diatom community with time. Additionally, the results suggested that lipid degradation can mobilize carbon and generate reductant in excess of the requirements for energetically expensive processes such as nitrate assimilation. Thus, lipid degradation may supply the carbon necessary for sustained nitrate assimilation when photosynthetic rates are low, such as following wind mixing mediated recoveryfromnitrate limitation. However, storage lipids are probably not important substrates  159 for dark respiratory metabolism. This is because triacylglycerol degradation did not begin until at least 12 h after a respiratory demand was provided by nitrate or ammonium addition to Nstarved cultures. Thus, during the normal course of diel light fluctuations at the surface of the oceans, carbohydrates are more likely to be respired during the hours of darkness.  160  FUTURE STUDIES  The conclusions made in this thesis warrant further investigation in several areas. From an ecological perspective, triacylglycerol degradation should be studied in a number of different species of lipid-storing diatoms and other phytoplankton. It would also be valuable to follow the development of triacylglycerol synthesis during the post-bloom period in a natural diatom population. Of special interest would be the analysis of the change in lipid composition and content over time in a post-bloom diatom population that has settled at a deep nutricline. In this thesis, triacylglycerol degradation was measured in the light and dark, but not under low light conditions. When light is limiting to growth, but not absent, storage lipid carbon may be more relevant in sustaining a low but continuous rate of nitrate assimilation at the nutricline. From a biochemical perspective, the compartmentation of triacylglycerol degradation in diatoms, and microalgae in general, needs to be clarified. At present, the pathway for triacylglycerol degradation is highly speculative; the site of 13-oxidation is unknown, as is the existence of glyoxysomes. Great advances could be made in understanding the degradative processes of lipid metabolism in diatoms if clean organelle preparations could be isolated from these organisms. It would also be useful to develop immunocytochemical techniques to locate the intracellular sites of triacylglycerol degradation.  161  LITERATURE CITED  Alonso, D.L., Segura del Castillo, C L , Garcia Sanchez, J.L., Sanchez Perez, J.A,. Garcia Camacho, F. (1994) Quantitative genetics of fatty acid variation in Isochrysis galbana (Prymnesiophyceae) and Phaeodactylum tricomutum (Bacillariophyceae). J. Phycol. 30, 553-558 Amblard, C , Boudier, G. (1990) The spring bloom of the diatom Melosira italica subsp. subarctica in Lake Pavin: biochemical, energetic and metabolic aspects during sedimentation. J. Plankton. Res. 12, 645-660 Amory, A.M., Vanlerberghe, G.C., Turpin, D.H. (1991) Demonstration of both a photosynthetic and nonphotosynthetic CO2 requirement for NIL* assimilation in the green alga Selenastrum minutum. Plant Physiol. 95, 192-196 Anderson, L.W.J., Sweeney, B.M. (1977) Diel changes in sedimentation characteristics of Ditylum brightwelli: changes in cellular lipid and effects of respiratory inhibitors and ion-transport modifiers. Limnol. Oceanogr. 22, 539-552 Andrews, J.E., Ohlrogge, J. (1990) Fatty acid and lipid biosynthesis and degradation. In: Plant physiology, biochemistry and molecular biology, pp. 339-352, Dennis, D.T., Turpin, D.H., eds, Longman Scientific and Technical, Essex ap Rees, T. (1990) Carbon metabolism in mitochondria. In: Plant physiology, biochemistry and molecular biology, pp. 106-123, Dennis, D.T., Turpin, D.H., eds, Longman Scientific and Technical, Essex Arao, T., Kawaguchi, A., Yamada, M. (1987) Positional distribution of fatty acids in lipids of the marine diatom Phaeodactylum tricomutum. Phytochemistry 26, 2573-2576 Arao, T., Sakaki, T., Yamada, M. (1994) Biosynthesis of polyunsaturated lipids in the diatom, Phaeodactylum tricomutum. Phytochemistry 36, 629-635 Arao, T., Yamada, M. (1994) Biosynthesis of polyunsaturated fatty acids in the marine diatom, Phaeodactylum tricomutum. Phytochemistry 35, 1177-1181 Arnalte, M., Cornejo, M., Bush, D.S., Jones, R.L. (1991) Gibberellic acid stimulates lipid metabolism in barley aleurone protoplasts. Plant Science 77, 223-232 Baasch, K., Kohlhase, M., Pohl, P. (1984) Axenic mass culture of prokaryotic and eukaryotic microalgae and lipid production: influence of the temperature and of the nitrate concentrations in the growth medium. In: Structure, function and metabolism of plant lipids, pp. 587-590, Siegenthaler, P.A, Eichenberger, W., eds, Elsevier Science Publishers, Amsterdam  162 Ben-Amotz, A., Tornabene, T.G. (1985) Chemical profile of selected species of microalgae with emphasis on lipids. J. Phycol. 21, 72-81 Berge, J., Gouygou, J., Dubacq, J., Durand, P. (1995) Reassessment of lipid composition of the diatom, Skeletonema costatum. Phytochemistry 39, 1017-1021 Berges, J.A., Harrison, P.J. (1995) Nitrate reductase activity quantitatively predicts the rate of nitrate incorporation under steady-state light limitation: A revised assay and characterization of the enzyme in three species of marine phytoplankton. Limnol. Oceanogr. 40, 82-93 Bewley, J.D., Black, M . (1985) Seed physiology of development and germination. Plenum, New York Bligh, E.G., Dyer, W.J. (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911-917 Boutton, T.W. (1991) Stable carbon isotope ratios of natural materials: I. Sample preparation and mass spectrometric analysis. In: carbon isotope techniques, pp. 155-171, Coleman, D.C., Fry, B., eds, Academic Press, San Diego Bowditch, M.L, Donaldson, R.P. (1990) Ascorbate free-radical reduction by glyoxysomal membranes. Plant Physiol. 94, 531-537 Brown, M.R., Dunstan, G.A., Norwood, S.J.N., Miller, K A . (1996) Effects of harvest stage and light on the biochemical composition of the diatom Thalassiosira pseudonana. J. Phycol. 32, 64-73 Cawse, P.A. (1967) The determination of nitrate in soil solutions by ultraviolet spectroscopy. Analyst 92, 311-315 Chrismadha, T., Borowitzka, M.A. (1994) Effect of cell density and irradiance on growth, proximate composition and eicosapentaenoic acid production of Phaeodactylum tricornutum grown in a tubular photobioreactor. J. Appl. Phycol. 6, 67-74 Cioni, M., Pinzauti, G., Vanni, P. (1981) Comparative biochemistry of the glyoxylate cycle. Comp. Biochem. Physiol. 70B, 1-26 Cohen, Z., Vonshak, A., Richmond, A. (1988) Effect of environmental conditions on fatty acid composition of the red alga Porphyridium cruentum: correlation to growth rate. J. Phycol. 24, 328-332 Combres, C , Laliberte, J.S.R., de laNoue, J. (1994) Effect of acetate on growth and ammonium uptake in the microalga Scenedesmus obliquus. Physiol. Plant. 91, 729-734 Cooksey, K.E. (1974) Acetate metabolism by whole cells of Phaeodactylum tricornutum Bohlin. J. Phycol. 10, 253-257  163 Cooksey, K.E., Guckert, J.B., Williams, S.A., Callis, P.R. (1987) Fluorometric determination of the neutral lipid content of microalgal cells using Nile Red. J. Microbiol. Methods 6, 333-345 Craigie, J.S. (1974) Storage products. In: Algal physiology and biochemistry, pp. 207-235, Stewart, W.D.P., ed, University of California Press, Los Angeles Cranwell, P.A., Robinson, N., Eglinton, G. (1985) Esterified lipids of the freshwater dinoflagellate Peridinium lomnickii. Lipids 20, 645-651 Crawford, N.M., Wilkinson, J.Q., LaBrie, S.T. (1992) Control of nitrate reduction in plants. Aust. J. Plant Physiol. 19, 377-385 Darley, W.M. (1974) Silicification and calcification. In: Algal physiology and biochemistry, pp. 655-675, Stewart, W.D.P., ed, University of California Press, Los Angeles De Lucas, J.R., Valenciano, S., Laborda, F., Turner, G. (1994) Glucose-induced inactivation of isocitrate lyase in Aspergillus nidulans. Arch. Microbiol. 162, 409-413 DeNiro, M.J., Epstein, S. (1977) Mechanism of carbon isotope fractionation associated with lipid synthesis. Science 197, 261-263 Dennis, D.T., Miernyk, J.A. (1982) Compartmentation of nonphotosynthetic carbohydrate metabolism. Ann. Rev. Plant Physiol. 33, 27-50 Descolas-Gros, C , Fontugne, M.R. (1985) Carbonfixationin marine phytoplankton: carboxylase activities and stable carbon-isotope ratios: physiological and paleoclimatological aspects. Mar. Biol. 87, 1-6 Descolas-Gros, C , Fontugne, M.R. (1990) Stable isotope fractionation by marine phytoplankton during photosynthesis. Plant Cell Environ. 13, 207-218 Eccleston, V.S., Cranmer, A M . , Voelker, T.A., Ohlrogge, J.B. (1996) Medium-chain fatty acid biosynthesis and utilization in Brassica napus plants expressing lauroyl-acyl carrier protein thioesterase. Planta 198, 46-53 Elrifi, I.R., Turpin, D.H. (1986) Nitrate and ammonium induced photosynthetic repression in N-limited Selenastrum minutum. Plant Physiol. 81, 273-279 Emdadi, D., Berland, B. (1989) Variation in lipid class composition during batch growth of Nannochloropsis salina m& Pavlova lutheri. Mar. Chem. 26, 215-225 Falkowski, PIG., Rivkin, R.B. (1976) The role of glutamine synthetase in the incorporation of ammonia in Skeletonema costatum (Bacillariophyceae). J. Phycol. 12, 448-450 Farquhar, G.D., Ehleringer, J.R., Hubick, K.T. (1989) Carbon isotope discrimination and photosynthesis. Ann. Rev. Plant Physiol. Mol. Biol. 40, 503-537  164 Fernandez, E., Serret, P., de Madariaga, I., Harbour, D.S. (1992) Photosynthetic carbon metabolism and biochemical composition of spring phytoplankton assemblages enclosed in microcosms: the diatom-Phaeocystis sp. succession. Mar. Ecol. Prog. Ser. 90, 89102 Fielding, A S , , Crawford, D., Turpin, D.H., Harrison, P.J., Calvert, S.E., Guy, R.D. Influence of the carbon concentrating mechanism on carbon stable isotope fractionation by a marine phytoplankter. Can. J. Bot. In press Fisher, N.S., Schwarzenbach, R.P. (1978) Fatty acid dynamics in Thalassiosirapseudonana (Bacillariophyceae): implications for physiological ecology. J. Phycol. 14, 143-150 Ford, C.W., Percival, E. (1965) The carbohydrates of Phaeodactylum tricomutum. Part I. Preliminary examination of the organism, and characterization of low molecular weight material and a glucan. J. Chem. Soc. 7035-7041 Gage, D.A., Rhodes, D., Nolte, K.D., Hicks, W.A., Leustek, T., Cooper, A.J.L., Hanson, A D . (1997) A new route for synthesis of dimethylsulphoniopropionate in marine algae. Nature 387, 891-894 Gaspar, M.L., Pollero, R.J., Cabello, M. (1994) Triacylglycerol consumption during spore germination of vesicular-arbuscular mycorrbizal fungi. J. Am. Oil Chem. Soc. 71, 449452 Geider, R.J., Osborne, B A . (1989) Respiration and microalgal growth: a review of the quantitative relationship between dark respiration and growth. New Phytol. 112, 327341 Gerhardt, B. (1986) Basic metabolic function of the higher plant peroxisome. Physiol. Veg. 24,397-410. Giachetti, E., Pinzauti, G., Bonaccorsi, R., Vincenzini, M.T., Vanni, P. (1987) Isocitrate lyase from higher plants. Phytochemistry 26, 2439-2446 Glover, H.E., Morris, I. (1979) Photosynthetic carboxylating enzymes in marine phytoplankton. Limnol. Oceanogr. 24, 510-519 Goulding, K.H., Merrett, M J . (1970) Experiments on enzyme induction in Chlorella pyrenoidosa. J. Biol. Educ. 4, 43-52 Graham, I.A., Denby, K.J., Leaver, C.J. (1994) Carbon catabolite repression requires glyoxylate cycle gene expression in cucumber. The Plant Cell 6, 761-772 Griffiths, G., Harwood, J.L. (1991) The regulation of triacylglycerol biosynthesis in cocoa (Theoroma cacao L.) Planta 184, 279-284 Gross, W. (1989) Intracellular localization of enzymes of fatty acid-B-oxidation in the alga Cyanidium caldarium. Plant Physiol. 91, 1476-1480  165 Guckert, J.B., Cooksey, K.E. (1990) Triglyceride accumulation and fatty acid profile changes in Chlorella (Chlorophyta) during high pH-induced cell cycle inhibition. J. Phycol. 26, 72-79 Gurr, M.I., Harwood, J.L. (1991) Lipid biochemistry. Chapman and Hall, London Haigh, W.G., Beevers, H. (1964) The occurrence and assay of isocitrate lyase in algae. Arch. Biochem. Biophys. 107, 147-151 Halpin, C , Lord, J.M. (1990) The structure and formation of microbodies. In: Plant physiology, biochemistry and molecular biology, pp. 329-338, Dennis, D.T., Turpin, D.H:, eds, Longman Scientific and Technical, Essex Handa,N. (1969) Carbohydrate metabolism in the marine diatom Skeletonema costatum. Mar. Biol. 4, 208-214 Harrison, P.J., Thompson, P.A., CalderwOod, G.S. (1990) Effects of nutrient and light limitation on the biochemical composition of phytoplankton. J. Appl. Phycol. 2, 45^56 Harrison, P.J.., Waters, R.E., Taylor, F.J.R. (1980) Abroad spectrum artificial seawater medium for coastal and open ocean phytoplankton. J. Phycol. 16, 28-35 Harvey, H.R., Bradshaw, S.A., O'Hara, S.C.M., Eglinton, G., Corner, E.D.S. (1988) Lipid composition of the marine dinoflagellate Scrippsiella trochoidea. Phytochemistry 27, 1723-1729 Hawes, I. (1990) The effects of light and temperature on photosynthate partitioning in Antarctic freshwater phytoplankton. J. Plankton. Res. 12, 513-518 Hayes, J.M. (1993) Factors controlling C contents of sedimentary organic compounds: principles and evidence. Mar. Geol. 113, 111-125 13  Healey, F.P. (1973) Inorganic nutrient uptake and deficiency in algae. CRC Crit. Rev. Microbiol. 3, 69-113 Henderson, R.J., Leftley, J.W., Sargent, J.R. (1988) Lipid composition and biosynthesis in the marine dinoflagellate Crypthecodinium cohnii. Phytochemistry 27, 1679-1683 Hoagland, K.D., Rosowski, J.R., Gretz, M.R., Roemer, S.C. (1993) Diatom extracellular polymeric substances: function, fine structure, chemistry, and physiology. J. Phycol: 29, 537-566 Hoyt, J.C, Reeves, H.C. (1992) Phosphorylation olAcinetobacter isocitrate lyase. Biochem. Biophys. Res. Comm. 182, 367-371 Huang, A.H.C., Beevers, H. (1974) Developmental changes of germinating castor bean independent of embryonic axis. Plant Physiol. 54, 364-368  166 Ireland, R. (1990) Amino acid and ureide biosynthesis. In: Plant physiology, biochemistry and molecular biology, pp. 407-421, Dennis, D.T., Turpin, D.H., eds, Longman Scientific and Technical, Essex Janse, I., van Rijssel, M., van Hall, P., Gerwig, G.J., Gottschal, J.C, Prins, R.A. (1996) The storage glucan of Phaeocystis globosa (Prymnesiophyceae) cells. J. Phycol. 32, 382387 Jimenez de Rio, M., Ramazanov, Z., Garcia-Reina, G. (1994) Dark induction of nitrate reductase in the halophilic alga Dimaliella salina. Planta 192, 40-45 Johnston, A.M. (1996) The effect of environmental variables on C discrimination by two marine phytoplankton. Mar. Ecol. Prog. Ser. 132, 257-263 13  Kaiser, W.M., Huber, SC. (1994) Posttranslational regulation of nitrate reductase in higher plants. Plant Physiol. 106, 817-821 Kang, F., Ridout, C.J., Morgan, C.L., Rawsthorne, S. (1994) The activity of acetyl-CoA carboxylase is not correlated with the rate of lipid synthesis during development of oilseed rape (Brassica napusL.) embryos. Planta 193, 320-325 Kattner, G., Brockmann, U.H. (1990) Particulate and dissolved fatty acids in an enclosure containing unialgal Skeletonema costatum (Greve) Cleve culture. J. Exp Mar. Biol. Ecol. 141, 1-13 Keller, A. A., Riebesell, U. (1989) Phytoplankton carbon dynamics during a winter-spring diatom bloom in an enclosed marine ecosystem: primary production, biomass and loss rates. Mar. Biol. 103, 131-142 Kindl, H. (1987) B-oxidation of fatty acids by specific organelles. In: The biochemistry of plants, Vol 9. pp. 31-90, Stumpf, P.K., ed, Academic Press, Florida Klyachko-Gurvich, G.L., Rudova, T.S., Kuznetsova, G.P. (1973) The role played by lipid fatty acids oi Chlorella in recovery of cells after nitrogen starvation. Sov. Plant. Physiol. 20, 92-99 Kochert, G. (1978) Carbohydrate determination by the phenol-sulphuric acid method. In: Handbook of phycological methods, pp. 95-97, Hellebust, J.A., Craigie, J.S., eds, Cambridge University Press, New York Kornberg, H.L. (1966) Anaplerotic sequences and their role in metabolism. In: Essays in biochemistry, Vol 2. pp. 131, Campbell, P.N., Greville, G.D., eds, Academic Press, New York Kruger, N.J. (1990) Carbohydrate synthesis and degradation. In: Plant physiology, biochemistry and molecular biology, pp. 59-76, Dennis, D.T., Turpin, D.H., eds, Longman Scientific and Technical, Essex  167 Lara, C , Romero, J.M., Guerrero, M G . (1987) Regulated nitrate transport in the cymobacterium Anacystisnidulans. J. Bacteriol. 169,4376-4378 Larson, T.R. (1992) The role of sodium in growth and metabolism of Phaeodactylum tricornutum. M.Sc. thesis, University of Auckland, Auckland, New Zealand Larson, T.R., Rees, T. A V . (1994) Arrest of cell division but not protein synthesis in sodium deficient cells of the marine diatom Phaeodactylum tricornutum Bohlin. Planta 195, 195-200 Larson, T.R., Rees, T.A.V. (1996) Changes in cell composition and lipid metabolism mediated by sodium and nitrogen availability in the marine diatom Phaeodactylum tricornutum (Bacillariophyceae). J. Phycol. 32, 388-393 Laws, E.A., Popp, B.N., Bidigare, R.R., Kennicutt, M . C , Macko, S.A. (1995) Dependence of phytoplankton carbon isotope composition on growth rate and [C02]aq: theoretical considerations and experimental results. Geochim. Cosmochim. Acta 59, 1131-1138 Leegood, R C , ap Rees, T. (1978a) Identification of the regulatory steps in gluconeogenesis in cotyledons of Cucurbita pepo. Biochim. Biophys. Acta 542, 1-11 Leegood, R C , ap Rees, T. (1978b) Phosphoenolpyruvate carboxykinase and gluconeogenesis in cotyledons of Cucurbita pepo. Biochim. Biophys. Acta 524, 207-218 Levasseur, M., Thompson, P.A., Harrison, P.J. (1993) Physiological acclimation of marine phytoplankton to different nitrogen sources. J. Phycol. 29, 587-595 Li, W.K.W., Glover, H.E., Morris, I. (1980) Physiology of carbon photoassimilation by Oscillatoria thiebautii in the Caribbean Sea. Limnol. Oceanogr. 25, 447-456 Li, W.K.W., Piatt, T. (1982) Distribution of carbon among photosynthetic end-products in phytoplankton of the Eastern Canadian arctic. J. Phycol. 18, 466-471 Livne, A., Sukenik, A. (1992) Lipid synthesis and abundance of acetyl CoA carboxylase in Isochrysis galbana (Prymnesiophyceae) following nitrogen starvation. Plant Cell. Physiol. 33, 1175-1181 Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J. (1951) Protein measurement with Folin-phenol reagent. J. Biol. Chem. 193, 265-275 Luthra, R., Munchi, S.K., Sukhija, P. S. (1991) Relationship of carbohydrate metabolism with lipid biosynthesis in developing sunflower (Helianthus annuus L.). J. Plant Physiol. 137,312-318 Magasanik, B. (1988) Reversible phosphorylation of an enhancer binding protein regulates the transcription of bacterial nitrogen utilization genes. Trends Biochem. 12, 475-479  168 Marcus, Y., Zenvirth, D., Hard, E., Kaplan, A. (1982) Induction of HCO3" transporting capability and high photosynthetic affinity to inorganic carbon by low concentration of C 0 in Anabaena variabilis. Plant Physiol. 69, 1008-1012 2  Martinez-Rivas, J.M., Vega, J.M. (1993) Effect of culture conditions on the isocitrate dehydrogenase and isocitrate lyase activities in Chlamydomonas reinhardtii. Physiol. Plant. 88, 599-603 Matrai, P. A., Keller, M.D. (1994) Total organic sulfur and dimethylsulfoniopropionate in marine phytoplankton: intracellular variations. Mar. Biol. 119, 61-68 Mayzaud, P., Claustre, H , Augier, P. (1990) Effect of variable nutrient supply on fatty acid composition of phytoplankton grown in an enclosed experimental ecosystem. Mar. Ecol. Prog. Ser. 60, 123-140 McConville, M.J., Mitchell, C„ Wetherbee, R. (1985) Patterns of carbon assimilation in a microalgal communityfromannual sea ice, East Antarctica. Polar Biol. 4, 135-141 Melzer, E., Schmidt, H.L. (1987) Carbon isotope effects on the pyruvate dehydrogenase reaction and their importance for relative carbon-13 depletion in lipids. J. Biol. Chem. 262, 8159-8164 Menancho, A., Vega, J.M. (1989) Effect of nitrogen starvation on ammonium assimilation by Chlamydomonas reinhardtii. Physiol. Plant. 75, 285-289 Mohanty, N., Bruce, D., Turpin, D.H. (1991) Dark ammonium assimilation reduces the plastoquinone pool of photosystem II in the green alga Selenastrum minutum. Plant Physiol. 96, 513-517 Molina Grima, E., Sanchez Perez, J.A, Garcia Camacho, F., Acien Fernandez, F.G., Fernandez Sevilla, J.M., Valdes Sanz, F. (1994) Effect of dilution rate on eicosapentaenoic acid productivity of Phaeodactylum tricomutum UTEX 640 in outdoor chemostat culture. Biotech. Lett. 16, 1035-1040 Mook, W.G., Bommerson, J.C, Staverman, W.H. (1974) Carbon isotope fractionation between dissolved bicarbonate and gaseous carbon dioxide. Earth Planet. Sci. Lett. 22, 169-176 Morris, I. (1981) Photosynthesis products, physiological state and phytoplankton growth. In: Physiological bases of phytoplankton ecology, Vol 210. pp. 83-102, Piatt, T., ed, Can. Bull. Fish. Aquat. Sci. Myklestad, S.M. (1989) Production, chemical structure, metabolism, and biological function of the (l-3)-linked, B-D-glucans in diatoms. Biol. Oceanogr. 6, 313-326 Numa, S., Tanabe, T. (1984) Acetyl-coenzyme A carboxylase and its regulation. In: Fatty acid metabolism and its regulation, pp. 1-27, Numa, S., ed, Elsevier, Amsterdam Opute, F.I. (1974) Lipid and fatty acid composition of diatoms. J. Exp. Bot. 25, 823-835  169 Orcutt, D.M., Patterson, G.W. (1975) Sterol, fatty acid and elemental composition of diatoms grown in chemically defined media. Comp. Biochem. Physiol. 50, 579-583 Page, R.A., Okada, S., Harwood, J.L. (1994) Acetyl-CoA carboxylase exerts strong flux control over lipid synthesis in plants. Biochim. Biophys. Acta 1210, 369-372 Palmisano, A C , Lizotte, M.P., Smith, G.A, Nichols, P.D., White, D . C , Sullivan, C W . (1988) Changes in photosynthetic carbon assimilation in Antarctic sea-ice diatoms during the spring bloom: variation in synthesis of lipid classes. J. Exp. Mar. Biol. Ecol. 116, 1-13 Palmisano, A.C., Sullivan, CW. (1985) Pathways of photosynthetic carbon assimilation in sea ice microalgaefromMcMurdo Sound, Antarctica. Limnol. Oceanogr. 30, 674-678 Park, R., Epstein, S. (1961) Metabolicfractionationof C 133-138  13  and C  1 2  in plants. Plant Physiol. 36,  Parnas, H., Cohen, D. (1976) An optimal policy for the metabolism of reserve materials in unicellular algae. J. Theor. Biol. 56, 1-18 Parrish, C . C , Bodennec, G., Sebedio, J., Gentien, P. (1993) Intra- and extracellular lipids in cultures of the toxic dinoflagellate, Gyrodinium aureolum. Phytochemistry 32, 291-295 Parrish, C . C , Wangersky, P.J. (1987) Particulate and dissolved lipid classes in cultures of Phaeodactylum tricomutum grown in cage culture turbidostats with a range of nitrogen supply rates. Mar. Ecol. Prog. Ser. 35, 119-128 Parrish, C . C , Wangersky, P.J. (1990) Growth and lipid class composition of the marine diatom Chaetoceros gracilis, in laboratory and mass culture turbidostats. J. Plankton Res. 12, 1011-1021 Parsons, T.R., Maita, Y., Lalli, C M . (1984) A manual of chemical and biological methods for seawater analysis. Pergamom Press, Oxford Pate, J.S., Layzell, D.B. (1990) Energetics and biological costs of nitrogen assimilation. In: The biochemistry of plants, Vol 16. pp. 1-42, Stumpf, P.K., ed, Academic Press, Florida Piorreck, M., Baasch, K., Pohl, P. (1984) Biomass production, total protein, chlorophylls, lipids and fatty acids, offreshwatergreen and blue-green algae under different nitrogen regimes. Phytochemistry 23, 207-216 Priscu, J.C, Priscu, L.R., Howard-Williams, C , Vincent, W.F. (1988) Diel patterns of photosynthate biosynthesis by phytoplankton in permanently ice-covered Antarctic lakes under continuous sunlight. J. Plankton Res. 10, 333-340 Rau, G.H., Riebesell, U., Wolf-Gladrow, D. (1997) C0 aq-dependent photosynthetic C fractionation in the ocean: A model versus measurements. Global Biogeochem. Cycles 11, 267-278 1 3  2  170 Rees, T.A.V., Larson, T.R., Heldens, J.W.G., Huning, F.G.J. (1995) In situ glutamine synthetase activity in a marine unicellular alga. Development of a sensitive colorimetric assay and the effects of nitrogen status on enzyme activity. Plant Physiol. 109, 14051410 Reid, E E . , Thompson, P., Lyttle, C.R., Dennis, D T . (1977) Pyruvate dehydrogenase complex from higher plant mitochondria and proplastids. Plant Physiol. 59, 842-848 Reitan, K.I., Rainuzzo, J.R., Olsen, Y. (1994) Effect of nutrient limitation on fatty acid and lipid content of marine microalgae. J. Phycol. 30, 972-979 Reyes, J.C, Florencio, F.J. (1994) A mutant lacking the glutamine synthetase gene (glnA) is impaired in the regulation of the nitrate assimilation in the cyanobacterium Synechocystis sp. strain PCC 6803. J. Bacterid. 176, 7516-7523 Reynolds, S.J., Smith, S.M. (1995) Regulation of expression of the cucumber isocitrate lyase gene in cotyledons upon seed germination and by sucrose. Plant Mol. Biol. 29, 885896 Roeske, C. A , OLeary, M.H.O. (1984) Carbon isotope effects on the enzyme-catalyzed carboxylation of ribulose bisphosphate. Biochemistry 23, 6275-6284 Roessler, P.G. (1987a) Lipid accumulation in silicon-deficient diatoms. In: The metabolism, structure and function of plant lipids, pp. 649-651, Stumpf, P.K., Mudd, J.B., Nes, W.B., eds, Plenum, New York Roessler, P.G. (1987b) UDPglucose pyrophosphorylase activity in the diatom Cyclotella cryptica. Pathway of chrysolaminarin synthesis. J. Phycol. 23, 494-498 Roessler, P.G. (1987c) The effects of silicon deficiency on lipid and carbohydrate metabolism in the diatom Cyclotella cryptica. Ph.D. thesis, University of Colorado, Boulder, CO, USA Roessler, P.G. (1990) Environmental control of glycerolipid metabolism in microalgae, commercial implications and future research directions. J. Phycol. 26, 393-399 Sackett, W.M., Eckelmann, W.R., Bender, M.L., Be, A.W.H. (1965) Temperature dependence of carbon isotope composition in marine phytoplankton and sediments. Science 148, 235-237 Sangwan, R.S., Singh, N., Plaxton, W.C. (1992) Phosphoenolpyruvate carboxylase activity and concentration in the endosperm of developing and germinating castor oil seeds. Plant Physiol. 99, 445-449 Sato, N. (1989) Modulation of lipid and fatty acid content by carbon dioxide in Chlamydomonas reinhardtii. Plant Sci. 61, 17-21 Schmidt, A , Zetsche, K. (1990) Regulation of the synthesis of isocitrate lyase and the corresponding mRNA in the green alga Chlorogonium. Bot. Acta 103, 48-53  171 Schuller, K.A., Plaxton, W.C., Turpin, D.H. (1990) Regulation of phosphoewo/pyruvate carboxylase from the green alga Selenastrum minutum. Properties associated with replenishment of tricarboxylic acid cycle intermediates during ammonia assimilation. Plant Physiol. 93, 1303-1311 Servel, M., Claire, C , Derrien, A., Coiffard, L., de Roeck-Holtzhauer, Y. (1994) Fatty acid composition of some marine microalgae. Phytochemistry 36, 691-693 Shifrin, N.S., Chisholm, S.W. (1981) Phytoplankton lipids, interspecific differences and effects of nitrate, silicate and light-dark cycles. J. Phycol. 17, 374-384 Siddique, A.M., Bai, A.K. (1991) Nitrogen fixation in peanut nodules during dark periods and detopped conditions with special reference to lipid bodies! Plant Physiol. 95, 896-899 Simcox, P.D, Garland, W., DeLuca, V., Canvin, D.T., Dennis, D.T. (1979) Respiratory pathways and fat synthesis in the developing castor oil seed. Can. J. Bot. 57, 10081014 Siron, R., Giusti, G., Berland, B. (1989) Changes in the fatty acid composition of Phaeodactylum tricornutum and Dunaliella tertiolecta during growth and under phosphorus deficiency. Mar. Ecol. Prog. Ser. 55, 95-100 Sivasankar, S., Oaks, A. (1995) Regulation of nitrate reductase during early seedling growth. A role for asparagine and glutamine. Plant Physiol. 107, 1225-1231 Skerrat, J.H., Nichols, P.D., McMeekin, T.A., Burton, H. (1995) Seasonal and inter-annual changes in planktonic biomass and community structure in eastern Antarctica using signature lipids. Mar. Chem. 51, 93-115 Slawyk, G., Maclssac, J.J. (1972) Comparison of two automated ammonium methods in a region of coastal upwelling. Deep Sea Res. 19, 521-524 Smith, A.E., Morris, I. (1980) Synthesis of lipids during photosynthesis of phytoplankton in the Southern Ocean. Science 207, 197-199 Smith, R.E.H., Clement, P., Head, E.J. (1990) Night metabolism of recent photosynthate by sea ice algae in the high Arctic. Mar. Biol. 107, 255-266 Smith, R.E.H., D'Souza, M.L. (1993) Macromolecular labeling patterns and inorganic nutrient limitation of a North Atlantic spring bloom. Mar. Ecol. Prog. Ser. 92, 111-118 Smith, R.E.H., Geider, R.J. (1985) Kinetics of intracellular carbon allocation in a marine diatom. J. Exp. Mar. Biol. Ecol. 93, 191-210 Smith, R.E.H., Herman, A. W. (1992) In situ patterns of intracellular photosynthate allocation by sea ice algae in the Canadian high Arctic. Polar Biol. 12, 545-551 Somerville, C , Browse, J. (1991) Plant lipids: metabolism, mutants, and membranes. Science 252, 80-86  172 Stabenau, H., Winkler, U., Saftel, W (1984) Enzymes of B-oxidation in different types of algal microbodies. Plant Physiol. 75, 531-533 Stobart, K., Mancha, M., Lenman, M., Dahlqvist, A., Stymne, S. (1997) Triacylglycerols are synthesized and utilized by transacylation reactions in microsomal preparations of developing safflower (Carthamus tinctoriusL.) seeds. Planta203, 58-66 Stryer, L. (1988) Biochemistry. W.H. Freeman and Company, New York Stymne, S., Stobart, A K . (1987) Triacylglycerol biosynthesis. In: The biochemistry of plants, Vol 9, pp. 175-214, Stumpf, P.K., Conn, E.E., eds, Academic Press, Toronto Suen, Y., Hubbard, J.S., Holzer, G , Tornabene, T.G. (1987) Total lipid production of the green algaNannochloropsissp. Ii. under different nitrogen regimes. J. Phycol. 23, 281296 Sukenik, A., Carmeli, Y. (1990) Lipid synthesis and fatty acid composition in Nannochloropsis sp. (Eustigmatophyceae) grown in a light-dark cycle. J. Phycol. 26, 463-469 Syrett, P.J. (1966) The kinetics of isocitrate lyase formation in Chlorella: evidence for the promotion of enzyme synthesis by photophosphorylation. J. Exp. Bot. 17, 641-654 Syrett, P.J. (1988) Uptake and utilization of nitrogen compounds. In: Biochemistry of the algae and cyanobacteria. pp. 23-39, Rogers, L.J., Gallon, J.R., eds, Clarendon Press, Oxford Syrett, P.J. Merrett, M.J., Bocks, S.M. (1963) Enzymes of the glyoxylate cycle in Chlorella vulgaris. J. Exp. Bot. 14, 249-264 Taguchi, S., Hirata, JA., Laws, E.A. (1987) Silicate deficiency and lipid synthesis of marine diatoms. J. Phycol. 23, 260-267 Terry, K.L., Hirata, J., Laws, E.A. (1983) Light-limited growth of two strains of the marine diatom Phaeodactylum tricomutum Bohlin: chemical composition, carbon partitioning and the diel periodicity of physiological processes. J. Exp. Mar. Biol. Ecol. 68, 209227 Thomas, D.N., Gleitz, M. (1993) Allocation of photoassimilated carbon into major algal metabolite fractions: variation between two diatom species isolated from the Weddell Sea (Antarctica). Polar Biol. 13, 281-286 Thomas, W.H., Seibert, D.L.R., Alden, M., Neori, A , Eldridge, P. (1984) Yields, photosynthetic efficiencies and proximate composition of dense marine microalgal cultures. I. Introduction and Phaeodactylum tricomutum experiments. Biomass. 5, 181-209 Thompson, P., Couture, P. (1990) Aspects of carbon metabolism in the recovery of Selenastrum capricomutum populations exposed to cadmium. Aquatic Toxicology 17, 1-14  173 Thompson, P.A., Guo, M., Harrison, P.J. (1992) Effects of variation in temperature. II. On the biochemical composition of eight species of marine phytoplankton. J. Phycol. 28, 481488 Thompson, P. A., Harrison, P.J., Parslow, J.S. (1991) Influence of irradiance on cell volume and carbon quota for ten species of marine phytoplankton. J. Phycol 27, 351-360 Thompson, P.A., Harrison, P.J., Whyte, J.N.C (1990) Influence of irradiance on the fatty acid composition of phytoplankton. J. Phycol. 26, 278-288 Thompson, P.A., Levasseur, M.E., Harrison, P.J. (1989) Light limited growth on ammonium vs. nitrate: what is the advantage for marine phytoplankton? Limnol. Oceanogr. 34, 1014-1024 Thoresen, S.S., Dortch, Q., Ahmed, S.I. (1982) Comparison of methods for extracting intracellular pools of inorganic nitrogen from marine phytoplankton. J. Plankton Res. 4, 695-704 Thurston, C F . (1977) Control of isocitrate lyase synthesis in Chlorella fusca var. vacuolata. The basal activity of the enzyme and the kinetics of induction. Biochem. J. 164, 147151 Tillman, U., Baumann, M.E.M., Aletsee, L. (1989) Distribution of carbon forming photosynthetic end products in the bloom-forming Arctic diatom Thalassiosira antarcticaCQMSEK. Polar Biol. 10,231-238 Tischner, R , Schmidt, A. (1982) A thioredoxin-mediated activation of glutamine synthetase and glutamate synthase in synchronous Chlorella sorokiniana. Plant Physiol. 70, 113116 Trevanion, S.J., Brooks, A.L., Leegood, R C (1995) Control of gluconeogenesis by phosphoewo/pyruvate carboxykinase in cotyledons of Cucurbita pepo L. Planta 196, 653-658 Turnham, E., Northcote, D.H. (1983) Changes in the activity of acetyl-CoA carboxylase during rape-seed formation. Biochem. J. 212, 223-229 Turpin, D.H. (1991) Effects of inorganic N availability on algal photosynthesis and carbon metabolism. J. Phycol. 27, 14-20 Turpin, D.H., Bruce, D. (1990) Regulation of photosynthetic light harvesting by nitrogen assimilation in the green alga Selenastrum minutum. FEBS Lett. 263, 99-103 Verardo, D.J., Froelich, P.N., Mclntyre, A. (1990) Determination of organic carbon and nitrogen in marine sediments using the Carlo Erba NA-1500 Analyzer. Deep Sea Res. 37, 157-165 Waite, A.M., Thompson, P.A., Harrison, P.J. (1992) Does energy control the sinking rate of marine diatoms? Limnol. Oceanogr. 37, 468-477  174 Wang, S.M., Huang, A.H.C. (1987) Biosynthesis of lipase in the scutellum of maize kernel. J. Biol. Chem. 262, 2270-2274 Winkler, U., Stabenau, H. (1995) Isolation and characterization of peroxisomes from diatoms. Planta 195, 403-407 Wood, B.J.B. (1974) Fatty acids and saponifiable lipids. In: Algal physiology and biochemistry, pp. 236-265, Stewart, W.D.P., ed, University of California Press, Los Angeles Wood, E.D., Armstrong, F.A.J., Richards, F.A. (1967) Determination of nitrate in seawater by cadmium-copper reduction to nitrite. J. Mar. Biol. Assoc. U.K. 47, 23-31 Yongmanitchai, W., Ward, OP. (1991) Growth of and omega-3 fatty acid production by Phaeodactylum tricornutum under different culture conditions. Appl. Environ. Microbiol. 57, 419-425 Yongmanitchai, W., Ward, OP. (1992) Separation of lipid classes from Phaeodactylum tricornutum using silica cartridges. Phytochemistry 31, 3405-3408 Zhukova, N.V., Aizdaicher, N.A. (1995) Fatty acid composition of 15 species of marine microalgae. Phytochemistry 39, 3 51-3 56  175  APPENDIX A  C E L L DENSITY DETERMINATION  7  1  Haemocytometer determined cell density (10 cells-mL" )  Fig. A . l . Comparison of methods for estimating cell density in P. tricornutum. A stationary phase culture was sequentially diluted with fresh ESAW to give 20 subcultures of varying cell density. These were then subsampled for density estimates from either haemocytometer (x-axis) or Coulter Counter™ counts (y-axis). The haemocytometer counts were assumed to more closely represent actual cell densities than Coulter Counter™ counts, as the latter does not discriminate between algal and non-algal particles, and coincidence error increases as particle density increases. Thus, Coulter Counter™ counts tended to underestimate cell density as cell density increased. However, in experiments, the cell densities measured always fell between 0.3-2.0 * 10 cells-mL" ; thus this underestimation was deemed to be insignificant. 7  1  176  APPENDIX B  PROTEIN DETERMINATION  Reagents: Reagent A  20 g Na C0 dissolved in 1000 mL DDW (fresh weekly)  Reagent B  1 g KNa tartrate + 0.5 g CuS0 dissolved in 100 mL DDW  Reagent D  50 mL Reagent A + 1 mL Reagent B (fresh daily)  Folin Reagent  5 mL Folin-Ciocalteu's Reagent (Sigma F-9252) + 10 mL  2  3  4  DDW  Protocol: 1. )  Pipet 0.1 mL of standards and samples into polystyrene semi-micro cuvettes.  2. )  Add 1 mL Reagent D and gently vortex-mix.  3. )  After exactly 10 min, add 0.1 mL Folin Reagent and gently vortex-mix.  4. )  After 30 min, measure absorbance at 750 nm vs. a 1 N NaOH blank processed as for the standards and samples.  0.25  0.20 fi c  °> r- 0.15 13  0.05  0.00  0  20  40  60  80  100  120  140  BSA (ug-mL" extract) 1  Fig. B.l. Typical standard curve for protein determinations. Bovine serum albumin (BSA), processed as for algal samples, was used as the standard protein. Protein was estimated in 0.1 mL of algal extract or BSA dissolved in 0.1 N NaOH.  178 APPENDIX G  CARBOHYDRATE DETERMINATION  0.20  0.15 B c  ©  0.10 h o o  I/)  3 0.05  0.00  10  20  30  40  Glucose (ug-mL") 1  Fig. C l . Typical standard curve for carbohydrate determinations. Glucose, processed as for algal samples, was used as the standard carbohydrate. Carbohydrate was estimated in 0.1 mL of algal extract or glucose dissolved in 3 N H S0 . The carbohydrates in algal extracts were assumed to contain simple sugars and polysaccharides, such as chrysolaminarin. 2  4  179 APPENDIX D  CHLOROPHYLL a EXTRACTION EFFICIENCIES  Table D . l . Chlorophyll a extraction efficiencies for P. tricomutum. Chlorophyll a was extractedfromP. tricomutum cultures growing in logarithmic phase. Whatman™ GF/F filters  (25 mm) were used for all extractions. The extraction solvents were either 90% acetone or 3:1 90% acetone:DMSO (v/v). Efficiencies were calculated as percentages relative to the most chlorophyll a extracted in a single sample, which was forfilterssonicated in 3:1 90% acetone:DMSO. Chlorophyll a was determined fluorometrically. Treatments: 24 h = overnight extraction at 4°C; sonication =10 min of sonication in an ice-water bath followed by overnight extraction at 4°C; grind = 30 s with a Teflon-headed pestle in a 5 mL Potter-Elvehjem grinding tube followed by overnight extraction at 4°C. Results are means ± 1 standard deviation for duplicate measurements.  Extraction Efficiency (%)  Treatment  90% acetone  3:1 90% acetone:DMSO  24 h  79.4 ± 2.4  98.5 ±0.7  sonication  86.6 ± 0.7  99.1 ± 1.3  grind  95.9 ±0.3  96.1 ± 1.0  180 APPENDIX E  LIPID SEPARATION Column Fractions Neutral  Polar  •  TAG 16:0, 18:0,20:0, 22:0, 22:1  Yes  No (20:0, 22:0, 22:1)  £  DAG 16:0, 18:0, 18:1  No (18:1)  Yes  •  MAG 16:0, 18:0, 18:1  No (18:1)  Yes  No (18:2)  Yes  _ ^ P C _  16:0, 18:2  Fig. E.l. Comparison of lipid separations using thin layer chromatography or solid phase extraction columns. Standard lipids were dissolved in chloroform and applied to either Baker Si250 PA silica gel thin layer chromatography (TLC) plates (J.T. Baker, Phillipsburg, New Jersey, USA) (left), or Waters Silica Light Sep-Pak™ solid phase extraction (SPE) columns (right). The standard lipids chosen were from Sigma. In increasing order of polarity, these lipids were: TAG (triacylglycerol), DAG (diacylglycerol), MAG (monoacylglycerol), and PC (phosphatidylcholine). Each lipid contained the fatty acids indicated on the left of the figure. For TLC plates, the developing solvent was 100:1 diethyl ether:acetic acid (v/v), and lipid spots were identified with iodine vapour. Duplicate runs on non-iodinated areas of the plate were scraped into tubes and transmethylated for fatty acid methyl ester (FAME) analysis by gas chromatography (GC). For SPE columns, the neutral fraction was eluted first with 10 mL 3:1 chloroform:hexane (3:1), and the polar fraction was eluted with 10 mL methanol. The eluates were dried, transmethylated, and the FAMEs analyzed by GC. The right side of thefigureshows the presence (Yes) or absence (No) for each lipid in the neutral or polar column fractions. The absence of a particular lipid in any one of the column fractions was confirmed by the absence of one of the fatty acid constituents for that fraction (bracketed fatty acid designations). For "Yes" values, all fatty acids for a particular lipid were present in the relevant columnfraction.These tests confirmed that only TAGs were eluted in thefirstcollectedfractionof the SPE columns.  181  TD  §  -o .«  1  2  4  6  8  10 total  Cumulative fraction (mL)  100  -i o  o <5 TO 2  TO ^ to so  1  2  4  6  8  10 total  Cumulative fraction (mL)  Fig. E.2. Efficiency of solid phase extraction columns. Standard lipids (Fig. E. 1) were dissolved in chloroform and applied to Waters Silica Light Sep-Pak™ SPE columns. The neutral fraction was eluted first with 10 mL 3:1 chlorofornrhexane (3:1), and the polar fraction was eluted with 10 mL methanol. During elution, 1 mL sub-fractions were collected. The eluates were dried, transmethylated, and the FAMEs analyzed by GC. Total sample retrieval was at least 93% for neutral lipids (A) and 83% for polar lipids (B). Neutral lipids (A) were eluted in a bell-shaped curve, indicating good sample retrievalfromthe column. However, polar lipids (B) were eluted in a decaying curve, indicating that residual sample remained bound to the column. Thus, for any one sample, the polar lipid fatty acidfractionwas usually calculated by subtracting GC determined neutral lipid fatty acid quantitiesfromtotal fatty acids. Thorough washing of the columns with 2x10 mL methanol succeeded in removing most of the residual polar lipids, so that columns could be reused up to four times before being discarded.  182  **g'0 3  3 - 2 T J - in  S B t 222jS2S2  t- O N O R RR  g te>2cri 3«e  U S  7 -  6 -  85  -  0 &  :  3 JUL  2 O  IO  20  30  40  Time (min.)  Fig. E.3. Gas chromatograph output trace for a mixture of standard fatty acid methyl esters. Sample volume injected was 1 uL. Split ratio was 100:1. Injector temperature was 250°C. Oven temperature was held constant at 180°C for the first 32 minutes of each run, and then increased to 240°C at 50°C per minute for the final 10 minutes. Total run time was 43.2 minutes. Flame ionization detector (FID) temperature was 350°C. The numbers above each peak show the FAME identity and retention time.  183 APPENDIX F  ULTRAVIOLET NITRATE DETERMINATION  0  100  200  300  400  500  600  700  800  900  1000  Nitrate (uM)  Fig. F.l. Typical standard curve for nitrate determinations using the UV method. Quartz cuvettes were used for the determinations. Readings were taken exactly 30 s after placing cuvettes in the spectrophotometer as absorbance values tended to increase slightly with measurement time. All samples were diluted by mixing 0.3 mL of sample with 0.7 mL DDW before measurement.  184  UV determined nitrate (uM)  Fig. F.2. Comparison of methods for nitrate determination in ESAW. The medium from a culture of P. tricomutum was sampled during the transitionfromlogarithmic phase (Nsufficient) to stationary phase (N-starved). The nitrate concentration was then determined in these samples using both the Technicon® Autoanalyzer and UV methods. As the cultures became increasingly N-starved, the UV method tended to overestimate the nitrate in solution. Thus, the UV method was only used to monitor nitrate concentrations in the medium of actively growing cultures. The Technicon® Autoanalyzer method (Wood et al. 1967) was used to determine nitrate concentrations in the mediumfromN-starved cultures and in cell extracts.  185 APPENDIX G  AMMONIUM DETERMINATION  Reagents:  Reagent A  35 g phenol + 0.4 g sodium nitroprusside dissolved in 100 mL DDW (store at 4°C)  Reagent B  280 g sodium citrate + 15 g NaOH dissolved in 830 mL DDW + 3 5 mL domestic bleach (store at 4°C)  Protocol:  1. )  Pipet 0.025 mL of standards and samples into polystyrene semi-micro cuvettes. Add 0.475 mL DDW.  2. )  Add 0.25 mL Reagent A and gently vortex-mix.  3. )  Add 0.25 mL Reagent B and gently vortex-mix.  4. )  After 30 min, measure absorbance at 630 nm vs. a DDW blank processed as for the standards and samples.  186  0  10  20  30  40  50  Ammonium (uM)  Fig. G . l . Typical standard curve for manual ammonium determinations. All samples were diluted by mixing 0.025 mL sample with 0.475 mL DDW before adding the colour reagents and measuring.  187 APPENDIX H  NITRATE REDUCTASE DETERMINATION  Table H . l . Effect of filter storage time on NR activity. A logarithmic phase culture of P. tricornutum was harvested and the filters stored for varying lengths of time under liquid  nitrogen (-196°C). One set of filters was not frozen, but processed immediately. The remainingfilterswere thawed and assayed for in vitro NR activity at different times from harvest. The results show that storing thefiltersfor up to one week did not adversely affect NR activity. In fact, freezing thefiltersat -196°C for even short periods of time appeared to enhance the extraction of the NR enzyme from the cells. Routine NR measurements were therefore made on liquid nitrogen frozen and thawed cells, regardless of storage time. Results are means ± 1 standard error for triplicate determinations.  NR activity  Time from harvest  Liquid N  (d)  treatment  (fmol-nitrite-ceir^h")  0  No  2.4 ±0.2  0  Yes  7.7 ±0.2  1  Yes  8.2 ±0.4  7  Yes  8.6 ±0.1  2  1  188  700  0  1  L  0  2  —  1  4  1  :  6  —  1  8  Cells in homogenate (x 10) 7  Fig. H.l. Effect of cell number in assay homogenate on NR activity. Different volumes of a logarithmic phase culture of P. tricornutum were harvested, the filters quick-frozen in liquid nitrogen, thawed, and assayed for in vitro NR activity. Assay time was 15 minutes. Using these results, harvest volumes in routine experiments were adjusted so that 5 x 10 cells were filtered for NR assays. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible theyfitinside the symbols. 7  189  8  7r  2 I  1  15  •  I  •  30  1  45  :  1—  60  Assay time (min) Fig. H.2. Effect of assay time on NR activity. A logarithmic phase culture of P. tricomutum was harvested (5 x 10 cells perfilter),thefiltersquick-frozen in liquid nitrogen, thawed, and assayed for in vitro NR activity. Assays were conducted for varying lengths of time. The results show that NR activity increased with assay time, but not proportionally; i.e. assay efficiency decreased with increasing assay time. Therefore, in routine experiments, a 15 minute incubation time was used to obtain the best NR activities. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible theyfitinside the symbols. 7  190  APPENDIX I  GLUTAMINE SYNTHETASE DETERMINATION  12 i  *  '  .  .  .  PON x u  ammonium uptake  GS  M r  Assimilation rate estimator  Fig. 1.1. Comparison of methods for estimating ammonium assimilation rates. A logarithmic phase culture of P. tricornutum was sampled five times over a 36 h period. Cells were harvested for in vitro GS activity measurements, and the medium ammonium concentration was monitored. Cell density datafromthis experiment, and previously acquired particulate organic nitrogen (PON) data for logarithmic phase P. tricornutum cultures, were used to estimate the ammonium assimilation rate (PON x u; left bar). Assimilation rate was also estimatedfromammonium disappearancefromthe medium (ammonium uptake; middle bar), and GS activities (GS;rightbar). There were no significant differences between any of these methods (p>0.05, 1-way repeated measures ANOVA). Therefore, the GS assay was judged to accurately estimate in vivo ammonium assimilation rates for logarithmic phase P. tricornutum cultures. Results are means ± 1 standard error for four consecutive assimilation rate estimates.  APPENDIX J  ISOCITRATE LYASE DETERMINATION  300  0  i  0  :  ,  1  . _  2  4  6  8  Cells in homogenate (x 10 ) 7  Fig. J . l . Effect of cell number in assay homogenate on ICL activity. Different volumes of a logarithmic phase culture of P. tricomutum were harvested, thefiltersquick-frozen in liquid nitrogen, thawed, and assayed for in vitro ICL activity. Assay time was 90 minutes. Using these results, harvest volumes in routine experiments were adjusted so that 5 x io cells werefilteredfor NR assays. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible theyfitinside the symbols. 7  192  o  I  1  >-> "Tab "o  M > CO  O  60  80  120  Assay time (min)  Fig. 5.2. Effect of assay time on ICL activity. A logarithmic phase culture of P. tricomutum was harvested (5 x 10 cells perfilter),thefiltersquick-frozen in liquid nitrogen, thawed, and assayed for in vitro ICL activity. Assays were conducted for varying lengths of time. The results show that ICL activity increased with assay time. In routine experiments, a 90 minute incubation time was used to obtain ICL activities. Results are means ± 1 standard error for triplicate determinations. Where error bars are not visible theyfitinside the symbols. 7  193  1.50  300  350  400  450  500  550  600  Wavelength (nm) o • A O  •  445:520 =1.965 445:520 =1.907 445:520 =1.499 445:520 =1.482 445:520 =1.684 445:520 =2.016 Phaeodactylum a-kg glyoxylate OAA pyruvate combination  400:520 =2.660 400:520 =0.833 400:520 =1.012 400:520 =0.775 400:520 =1.146 400:520 =1.026  Fig. J.3. Wavelength scans of dinitrophenylhydrazone derivatives of various C4 acids. These scans were done because Haigh and Beevers (1964) suggested that the glyoxylate produced in the ICL assay reaction could not be easily distinguishedfromother C4 acids Several C4 acids (100 nmol-mL') were converted to their dinitrophenylhydrazone derivatives with the same techniques and conditions used for ICL assay products. These were then scanned with an LKB Ultraspec II™ spectrophotometer over the visible wavelength range. A combination of all the pure C4 acids was also scanned, as were the assay productsfroman ICL activity determination in P. tricomutum. Absorbance ratios were calculated for the prominent peaks and shoulders for each C4 acid derivative, and for the combination and P. tricomutum assay derivatives. The absorbance spectrum for P. tricomutum closely matched that for glyoxylate, indicating that glyoxylate was most likely the C4 dinitrophenylhydrazone derivativefromthe ICL assay reactions being measured. 1  194  APPENDIX K  SCINTILLATION COUNTING  Unquenched DPM (* 10 ) 3  Fig. K . l . Quench curves for various solvents used for radioactivity determinations. Unquenched radioactivities were measured in dried volumes of the source radioisotope with no added solvents. The inset shows the quench curves for the low range of radioactivities. All samples were corrected for background radiation and chemiluminescence.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088751/manifest

Comment

Related Items