UDP-GLUCOSE:SINAPIC ACID GLUCOSYLTRANSFERASE IN BRASSICACEAE
by
SHAWN X. WANG
B. Sc., The Shanghai Agricultural College, 1983 M. Sc., The University of Saskatchewan, 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 PLANT SCIENCE
We accept this thesis as conforming
to therequiredstandard
THE UNIVERSITY OF BRITISH COLUMBIA
October, 1996
© Shawn X. Wang, 1996 In presenting this thesis in partial fulfilment , of the requirements: for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. .
Department of
The University of British Columbia Vancouver, Canada
Date tfew '??7
DE-6 (2/88) ABSTRACT
Sinapine is a bitter phenolic choline ester that is ubiquitous in crucifer seeds. It occupies about 1-4% of the dry matter of rapeseed. Elimination of sinapine from the seeds would improve the flavor, palatability and nutritional properties of rapeseed meal for its utilization as supplemental protein source for animal feed. UDP-glucose.sinapic acid glucosyltransferase (SGT; EC 2.4.1.120) is a key enzyme involved in biosynthesis of both sinapine in seed and sinapoylmalate in vegetative tissues of many members of
Brassicaceae. Both esters are strongly UV-absorbing but their physiological functions in plants are unknown. By a series of column chromatography techniques, SGT has been purified from 60-h-old seedlings of Brassica napus and characterized. SGT is a monomeric polypeptide with a molecular weight of 42 kDa and a pi of pH 5. The subcellular location of SGT appears to be the cytosol. SGT activity is not inducible by heat shock or UV radiation stresses. Its general characteristics are similar to those of SGT from Raphanus sativus, as well as a number of other UDP-glucose-dependent
glucosyltransferases. The Km (UDp.giUCOse) was 0.24 mM and Km (sinapic acid) was 0.16 mM.
SGT also catalyzes the reverse reaction in vitro, using UDP and sinapoylglucose to form
UDP-glucose. No cofactors are required for SGT activity, but reducing reagents and glycerol are required to stabilize the enzyme. The enzyme is strongly inhibited by p- hydroxyl-mercuribenzoic acid, UDP, TDP, Zn++, Cu++ and Hg++. Kinetic properties and substrate affinity data suggest that the catalytic mechanism of SGT is best described by a
"random bi-bi" model. An analysis of developmental profiles showed that SGT was expressed in all growth stages of B. napus plants, but was most active during the early
ii germination and seed development stages, particularly in cotyledons, juvenile leaves and young shoots. Partial amino acid sequence data were obtained from tryptic digests of the putative SGT protein. These sequences showed a very high degree of similarity to the
HSP/HSC70 protein family.
iii TABLE OF CONTENTS
ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES . x LIST OF FIGURES xii ABBREVIATIONS xvi ACKNOWLEDGMENTS xviii
Chapter I GENERAL INTRODUCTION 1 1. Sinapoyl esters and phenolics in plants 2 2. The central role of the phenylpropanoid pathway 4 3. Glucosylation and UDP-glucose-dependent glucosyltransferases 5 4. Biological significance of SGT 8 5. Economic significance of SGT 10
Chapter II TAXONOMIC DISTRIBUTION OF SGT IN
BRASSICA CEAE 14
INTRODUCTION 15
OBJECTIVE 16
MATERIALS 16 1. Plant materials 16 2. Chemicals 19 3. Equipment 19 METHODS 19 1. Determination of sinapine by HPLC 19 1.1. Sample preparation 19 1.2. HPLC analysis 20 1.3. Calculation of sinapine content • 20 2. SGT assay for dry seeds and seedlings 21 2.1. Seedling growth 21 2.2. Protein sample preparation 21 2.3. Determination of protein concentration 22 2.4. Determination of SGT activity 23 2.5. Calculation of SGT activity 23
iv RESULTS 25 1. Sinapine accumulation 25 2. SGT activity 25
DISCUSSION 29
Chapter III THE DIVERSITY OF SINAPIC ACID ESTERS, ENZYMOLOGY AND DEVELOPMENTAL EXPRESSION OF SGT IN B. NAPUS AND S. ALBA 33
INTRODUCTION 34
OBJECTIVES 39
MATERIALS 39 1. Plant materials 39 2. Chemicals 39
METHODS 40 1. Plant growth conditions and sampling 40 2. HPLC analysis of methanol-soluble phenolics 41 2.1. Sample preparation 41 2.2. HPLC analysis 41 2.3. Qualitative and quantitative determination of sinapic acid derivatives 42 3. Enzymatic assays of SE, SGT, SCT, SMT and 4CL .... 42 3.1. Sample preparation for protein extraction 42 3.2. Determination of SE activity by HPLC 42 3.3. Determination of SGT activity by HPLC 43 3.4. Determination of SMT activity by HPLC 43 3.5. Determination of SCT activity by HPLC 44 3.6. Determination of 4CL activity by HPLC 44
RESULTS 45 1. Profiles of methanol-soluble phenolics in various tissues at different growth stages 45 1.1. Pattern of phenolic compound accumulation in young seedlings 45 1.2. Pattern of phenolic compound accumulation in various tissues of adult plants 49 1.2.1. B. napus cv. Westar 49 1.2.2. S. alba cv. Ochre 52 1.3. Profiles of phenolic compounds accumulated in developing seeds 55
V 2. Developmental pattern of expression of SE, SGT, SMT, SCT and 4CL 57 2.1. SE, SGT, SMT, SCT and 4CL activities in young seedlings 57 2.2. SE, SGT, SMT, SCT and 4CL activities in vegetative tissues of adult plant 60 2.3. SE, SGT, SMT, SCT and 4CL activities in fruit 63 3. Developmental expression of SGT in B. napus cv. Westar .66
DISCUSSION 70
Chapter IV ENZYMOLOGY OF UDP-GLUCOSE:SINAPIC ACID GLUCOSYLTRANSFERASE FROM B. NAPUS 81
INTRODUCTION 82 r OBJECTIVE 83
MATERIALS 87 1. Plant materials 87 2. Chemicals 87
METHODS 88 1. Determination of SGT activity 88 1.1. Multi-well plate method 88 1.2. HPLC method 88 1.3. Radioisotope method 89 1.4. Spectrophotometry method 90 2. Induction of SGT 92 2.1. Induction of SGT by sinapic acid 92 2.2. Induction of SGT by light 93 2.3. Induction of SGT by heat shock 94 3. Subcellular localization of SGT 94 4. Purification of SGT 96 5. Physical and chemical characterization of SGT 100 5.1. Determination of molecular weight 100 5.2. Determination of SGT pH stability and pH optimum 101 5.3. Determination of SGT thermal stability and temperature optimum 102 5.4. Requirement for metal ions 103 5.5. Requirement for reducing reagents 103 5.6. Determination of SGT substrate specificity 104
vi 5.7. Determination of the inhibitory effect of UDP-glucose analogues 105 5.8. Determination of effects of other inhibitors 105 5.9. Determination of SGT reversibility 106 5.9.1. From the forward reaction 106 5.9.2. From the reverse reaction 107 5.10. Investigation of affinity matrices binding for SGT 107 5.10.1. UDP-glucose-hydrazide membrane 107 5.10.2. UDP-glucuronic acid-agarose 109 5.10.3. Sinapic acid-Sepharose 110 5.11. Determination of SGT substrate kinetics Ill 6. Recovery of SGT activity from native PAGE 112 7. Determination of amino acid composition and sequence 113 8. Protein sequence analysis 114 9. Production of polyclonal antibodies 115 9.1. Production of polyclonal antibodies with native antigen 115 9.2. Production of polyclonal antibodies with denatured antigen 116 10. Immunodetection of SGT by Western blot 116 11. Immunoprecipitation of SGT activity 117
RESULTS 118 1. Induction of SGT 118 1.1. Substrate - sinapic acid effect on SGT 118 1.2. Light effect on SGT activity 121 1.3. Heat stress effect on SGT activity 121 2. Subcellular localization of SGT 121 3. Purification of SGT 126 4. Physical and chemical characterization of SGT 131 4.1. Molecular weight 131 4.2. pi, pH stability and pH optimum 131 4.3. Thermal stability and optimum temperature 133 4.4. Effect of metal ions 133 4.5. Effect of reducing reagents 139 4.6. Effect of other inhibitors 139 4.7. Substrate specificity 143 4.8. Inhibitory effect of substrate analogues 143 4.9. Reversibility of SGT 143 4.10. Affinity matrices , 147 4.11. Kinetic parameters 147 5. Amino acid composition and sequence 147 6. Protein sequence homology and comparison 152
vii 7. Immunological responses 152
DISCUSSION 160 1. Induction of SGT 160 2. Subcellular localization of SGT 162 3. Physical and chemical properties of SGT 164 4. Amino acid sequence homology and immunological responses 168
Chapter V ATTEMPTS TO IDENTIFY A GENE ENCODING SGT .... 172
INTRODUCTION 173
OBJECTIVE 175
MATERIALS 175 1. Plant materials 175 2. Chemicals 175
METHODS 176
1. Synthesis of BEK oligo-d(T)20 oligo adapter 176 2. Synthesis of SGT specific N-terminal primers 176 3. Isolation of total RNA from seedlings 177 4. Synthesis of single strand cDNA from total RNA by AMV-RT and M-MLV-RT 178 5. Amplification of hsp70/hsc70 and sgt by PCR 179 6. Agarose gel electrophoresis 180 7. Labeling hsplOlhsclQ probes with a-32P-dATP by random primer labeling system 180 7.1. DNA preparation 180 7.2. Random primer labeling 181 8. Northern blot analysis of seedling RNA 181 8.1. Electrophoresis of RNA 182 8.2. Capillary transfer with sodium citrate buffer 182 8.3. Pre-hybridization 183 8.4. Hybridization with probe 183 9. Southern blot analysis of PCR products 184 9.1. Electrophoresis of PCR products 184 9.2. Capillary transfer with 0.4 M NaOH 184 9.3. Pre-hybridization 184 9.4. Hybridization with probe 185
RESULTS 185 1. RT-PCR products 185
viii 1.1. Effect of template sources, concentrations and reverse transcriptases 185 1.2. Effect of Mg++ concentration and primers 187 1.3. Effect of annealing temperature 187 2. Southern blot 190 3. Northern blot 190
DISCUSSION 193
Chapter VI CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER RESEARCH 196
1. Anti-SGT antibodies 199 2. Identification of a sgt gene 199 3. Relationship between HSP70/HSC70 and SGT 199
BIBLIOGRAPHY 201
APPENDIX A. Procedure for preparing protein extracts for SE, SGT, SMT, SCT and 4CL assay 217
ix LIST OF TABLES
Table No. Table title Page
Table 1. The names and sources of seeds of 38 genotypes from 13 species in Brassicaceae 17
Table 2. Sinapine content and SGT activities in the seeds and 2-day-old seedlings of 38 genotypes from 13 brassicaceous species 26
Table 3. Identified non-sinapine conjugates of sinapic acid in crucifer species 36
Table 4. Enzymatic reaction products and plant sources of known UDP-glucose-dependent glucosyltransferases 84
Table 5. Recovery of SGT from soluble and insoluble portions of seedling extract prepared with or without 0.2% Triton X-100 125
Table 6. SGT purification schedule and protein profile of each
step on SDS-PAGE 128
Table 7. Metal ion effects on SGT activity 138
Table 8. Inhibitory effects of Cu++, Hg++ and Zn++ on SGT activity 140
Table 9. Effects of other inhibitors on SGT activity 142
Table 10. The specificity of SGT for sugar donor and acceptor 144 Table 11. Effects of UDP-glucose analogues on SGT activity 145
Table 12. Affinity binding of SGT by immobilized UDP-glucose membrane cartridge, UDP-glucuronic acid-agarose and sinapic acid-Sepharose 148
Table 13. Amino acid composition of the putative SGT protein (42 kDa) from seedlings of B. napus cv. Westar 150
Table 14. Amino acid composition of a 33 kDa protein purified from seedlings of B. napus cv. Westar 151
x Table 15. N-terminal and internal amino acid sequences of the putative SGT protein (42 kDa) and 33 kDa polypeptides purified from seedlings of B. napus cv. Westar 153
Table 16. Alignment of amino acid sequences of the putative SGT (42 kDa) polypeptides purified from seedlings of B. napus cv. Westar and the heat shock cognate protein 70 isolated from A. thaliana 154
xi LIST OF FIGURES
Figure No. Figure title Page
Fig. 1. The central role of the phenylpropanoid pathway in biosynthesis of phenolics in plants 6
Fig. 2. The role of SGT in sinapine biosynthesis and metabolism. 11
Fig. 3. Comparisons of SGT total activities ( A ) and specific activities ( B ) in the dry seeds and 60-h-old seedlings of 38 genotypes from 13 brassicaceous species 30
Fig. 4. HPLC chromatograms of methanol extracts from young seedlings of B. napus cv. Westar and S. alba cv. Ochre 46
Fig. 5. Changes in levels of sinapoyl-conjugates in young seedlings of B. napus cv. Westar and S. alba cv. Ochre during germination 48
Fig. 6. HPLC chromatograms of methanol extracts from various tissues of B. napus cv. Westar 50
Fig. 7. HPLC chromatograms of methanol extracts from leaves of 28-day-old B. napus cv. Westar and S. alba cv. Ochre plants 51
Fig. 8. Quantitative analysis of sinapoyl-conjugates at different growth stages in different tissues of B. napus cv. Westar 53
Fig. 9. HPLC chromatograms of methanol extracts from various tissues of S. alba cv. Ochre 54
Fig. 10. HPLC chromatograms of methanol extracts from developing seeds of B. napus cv. Westar and S. alba cv. Ochre 56
Fig. 11. Changes in levels of sinapoyl-conjugates in developing seeds of B. napus cv. Westar and S. alba cv. Ochre 58
Fig. 12. Comparison of SE, SGT, SCT, SMT and 4CL specific activities in young seedlings of B. napus cv. Westar and S. alba cv. Ochre. 59
xii Fig. 13. SE, SGT, SCT, SMT and 4CL activities in various tissues of B. napus cv. Westar during plant development 61
Fig. 14. SE, SGT, SCT, SMT and 4CL activities in various tissues of S. alba cv. Ochre during plant development 62
Fig. 15. Comparison of SE, SGT, SCT, SMT and 4CL activities in silique walls of B. napus cv. Westar and S. alba cv. Ochre during seed development 64
Fig. 16. Comparison of SE, SGT, SCT, SMT and 4CL activities in developing seeds of B. napus cv. Westar and S. alba cv. Ochre 65
Fig. 17. Distribution of SGT in 68-h-old seedlings of B. napus cv. Westar 67
Fig. 18. Distribution of SGT in leaves from different positions on 50-day-old plants of B. napus cv. Westar 68
Fig. 19. Distribution of SGT in different tissues of flowers of B. napus cv. Westar 69
Fig. 20. Changes in SGT activity and protein content in developing seeds of B. napus cv. Westar 71
Fig. 21. Developmental expression of SGT throughout the life
cycle of a B. napus cv. Westar plant 72
Fig. 22. Structures of three affinity matrices for binding SGT 108
Fig. 23. Effect of sinapic acid concentration in growth medium on seedling growth 119 Fig. 24. Effect of sinapic acid in growth medium on SGT activity in germinating seeds of B. napus cv. Westar 120
Fig. 25. Dynamics of sinapoylglucose accumulation in young seedlings of B. napus cv. Westar grown in medium containing various concentrations of sinapic acid 122
Fig. 26. Effect of different light conditions on SGT activity during the seed germination 123
"xiii Fig. 27. Effect of heat stress on SGT activity during B. napus seed germination 124
Fig. 28. Distribution profiles of SGT activity in 2-day-old B. napus seedling extracts centrifuged through in situ self-generated Percoll gradients 127
Fig. 29. Chromatograms of systematic purification of SGT from the seedlings of B. napus cv. Westar by various column chromatography 129
Fig. 30. Chromatogram, SGT activity pattern and protein profile on SDS-PAGE for the SGT preparation from the last step of purification 130
Fig. 31. Determination of molecular weight of the putative
SGT protein 132
Fig. 32. The stability of SGT during storage at different buffer pH 134
Fig. 33. Determination of the optimum buffer pH for SGT activity 135
Fig. 34. Determination of thermal stability of SGT during storage 136
Fig. 35. Determination of the optimum temperature for the SGT
reaction 137
Fig. 36. Effects of reducing reagents on SGT activity 141
Fig. 37. Reversibility of the SGT reaction 146
Fig. 38. Lineweaver-Burk plots of SGT reaction velocity at four different concentrations of the second substrate 149 Fig. .39. Immunotitration of SGT activity by anti-pea HSP70 serum 156 Fig. 40. Comparison of Western blots generated using anti-pea HSP70 serum and electrophoretically-separated proteins from 2-day-old seedlings and immature green seeds of B. napus cv. Westar 157
Fig. 41. Comparison of Western blots generated using anti-pea HSP70 serum and proteins from heat-treated seedling extracts 158
xiv Fig. 42. Comparison of Western blots generated using anti-human HSP70 (StressGen 810) monoclonal antibody and proteins from heat-treated seedling extracts 159
Fig. 43. Analysis of RT-PCR products using RNA isolated from seedlings at different ages 186
Fig. 44. Analysis of RT-PCR products with single primer control
and addition of Mg++ 188
Fig. 45. Effect of annealing temperature on RT-PCR products 189
Fig. 46. Southern blot analysis of specific RT-PCR products separated by 1% agarose gel electrophoresis 191 Fig. 47. Northern blot analysis of HSP70 and SGT specific transcripts 192
XV ABBREVIATIONS
ADPG adenosine 5'-diphosphoglucose AMV-RT avian myeloblastosis virus reverse transcriptase AUFS absorbing unit full scale BCIP 5-bromo-4-chloro-3-indolyl phosphate CAPS 3-[cyclohexylamino]-l-propanesulfonic acid CDPG cytidine 5'-diphosphoglucose 4CL 4-coumaroyl CoA ligase CoA Coenzyme A DCC A^A^-dicyclohexylcarbodiimide DEPC diethyl pyrocarbonate DMF dimethyl formamide DTT dithiothreitol EDTA ethylenediaminetetraacetic acid FPLC Fast protein liquid chromatography GDPG guanosine 5'-diphosphoglucose HCA hydroxycinnamic acids HPLC high-performance liquid chromatography HPRNI human placenta RNase inhibitor HSP70 70 kD heat shock protein HSC70 70 kD heat shock cognate protein HTP hydroxyapatite KPi potassium phosphate 2-ME (3-mercaptoethanol MES 2-|7V-rnorpholino] ethane sulfonic acid M-MLV-RT moloney murine leukemia virus reverse transcriptase MOPS 3-[Af-morpholino]-propanesulfonic acid MS macro nutrient NBT nitroblue tetrazolium PAL phenylalanine ammonia-lyase PBS phosphate buffered saline PCR polymerase chain reaction PHMB jp-hydroxyl-mercuribenzoic acid PMSF phenylmethanesulfonyl fluoride PVDF polyvinylidene fluoride PVPP polyvinylpolypyrrolidone RT retention time SCT l-0-sinapoyl-(3-Z)-glucose:choline sinapoyltransferase SDS-PAGE sodium dodecylsulfate polyacrylamide gel electrophoresis SE sinapine esterase SGT Uridine-5'-diphosphoglucose:sihapic acid glucosyltransferase Sin sinapine SinA sinapic acid
xvi SinG sinapoylglucose SinM sinapoylmalate SMT 1 -0-sinapoyl-|3-D-glucose:L-malate O-sinapoy transferase SSC sodium citrate buffer SSPE sodium phosphate-EDTA buffer TAE Tris-sodium acetate-EDTA buffer TBS Tris buffered saline TDP thymidine 5'-diphosphate TDPG thymidine 5'-diphosphoglucose TE Tris/HCl-EDTA buffer UDP uridine 5'-diphosphate UDPG uridine 5'-diphosphoglucose
xvii ACKNOWLEDGMENTS
Much of the work presented in this thesis would not have been possible without the advice, help and support of my supervisor, Dr. Brian Ellis. It is he guided me to the kingdom of plant secondary metabolism. His comprehensive and precise knowledge in science helped to shape the way I think and write, and his tireless efforts and patience in editing the manuscript of this thesis made it possible today.
I would also like to thank the members of my advisory committee: Drs. Brian Holl, John Carlson and D'Ann Rochon for their helpful ideas, critical comments and advice throughout my Ph.D. program.
The kind and constructive assistance of my colleagues, Seong-Hwan Kim, Bjorn Orvar, Stefanie Butland, Grant McKegney and Amrita Singh, in the Plant Biochemistry Laboratory of the Department of Plant Science, UBC was very much appreciated.
Thanks were also extended to Sandra Allina and Jennifer Norton for their valuable proof reading of this thesis.
Most importantly, I would like to express my deepest gratitude to my wife Yan and daughter Jini for their patience and understanding, and to my Dad and Mom who always had faith in my ability.
xviii Chapter I
GENERAL INTRODUCTION
l Uridine-5'-diphosphoglucose (UDPG):sinapic acid glucosyltransf erase (SGT; EC
2.4.1.120) is an enzyme that catalyzes the glucosylation of sinapic acid (SinA) to form a sinapic acid glucose ester, l-0-sinapoyl-|3-/>glucose (SinG). Sinapoylglucose is a precursor for the formation of sinapine (Sin) and sinapoylmalate (SinM), which are the major phenolic compounds accumulated in the seeds or the vegetative tissues of many members of the Brassicqceae family (Gadamer, 1896; Schultz and Gmelin, 1952 and
1953; Hegnauer, 1964; Strack, 1977; Kerber and Buchloh, 1980 and 1981; Voskerusa and
Kolovrat, 1989; Bouchereau et al, 1991). SGT was first identified in red radish seedlings
(Raphanus sativus) (Strack, 1980), and is of interest both biologically and economically.
1. Sinapoyl esters and other phenolics in plants
Sinapoyl esters are one class of phenolic compounds in plants. Phenolics occur in very great abundance throughout the plant kingdom. Not only are they quantitatively extremely important, but they also demonstrate a tremendous diversity of structure
(Goodwin and Mercer, 1983). Their backbone carbon skeletons can be as simple as C6, e.g. phenol, or can be as complicated as heteropolymers, e.g. polyphenols or lignin
(Haslam, 1993; Boudet et al., 1995). Some phenolics are distributed ubiquitously, while others are confined to a small number of species.
The structural diversity of phenolics in plants is matched by their extensive range of proposed functions. The fact that phenolic compounds are important plant structural components has been known for a century and a half (Lewis and Yamamoto, 1990). It has
2 been estimated that 15-20% of the carbon fixed in the biosphere each year is eventually incorporated into the durable cell wall polymer, lignin. The lignin polymer reinforces and waterproofs the walls of specialized cells and plays a fundamental role in the strategies of mechanical support, solute conductance and prevention of insect feeding and pathogen attack of higher plants (Hahlbrock and Scheel, 1989).
Flavonoids have a C6-C3-C6 carbon skeleton, and form another large class of phenolic compounds in plants. Their extended resonance structure is very efficient at absorbing
UV-B radiation, thus allowing them to serve as a screen to the harmful radiation
(Nicholson and Hammerschmidt, 1992; Jorgensen, 1994). Anthocyanidins are produced by oxidizing and dehydrating the heterocyclic central ring of the flavonoid skeleton in plants. These phenolics absorb the visible region of the radiation, resulting in a range of pigments from red-orange to blue color (Markham, 1982; Mazza and Miniati, 1993).
Recently, some phenolics have been found to be signal compounds, and to play a role in plant growth and in the interaction between plants and other organisms, e.g. strigol and acetosyringone (Smith et al., 1989; Lynn and Chang, 1990). However, many of the physiological roles of specific phenolic compounds are unknown, including the functions of the sinapoyl esters.
A common structural property of phenolic compounds is that they contain an aromatic ring with one or more hydroxyl groups attached. Virtually, all of these compounds share a
3 common biosynthetic intermediate (phenylalanine or tyrosine), derived from the precursor, shikimic acid.
2. The central role of the phenylpropanoid pathway
The linkage between plant primary metabolism and the tremendous diversity of phenolic compounds is the phenylpropanoid pathway. The key reaction leading to the structures of these secondary metabolites is catalyzed by phenylalanine ammonia-lyase (PAL; EC
4.3.1.5) (Koukol and Conn, 1961). This unique enzyme acts upon L-phenylalanine to remove both the a-amino group and the pro-S hydrogen from the (3-position of the side- chain, yielding f-cinnamic acid. This non-oxidative deamination initiates the general phenylpropanoid pathway. Cinnamic acid undergoes a sequence of reactions that involves hydroxylation at the 4-position of the ring of cinnamic acid to form />-coumaric acid
(Gabriac et al., 1991), and addition of a second hydroxyl group at the 3-position to yield caffeic acid (Bolwell and Butt, 1983). These hydroxylations are catalyzed by cinnamate-
4-hydroxylase (C4H; EC 1.14.13.11) and coumarate-3-hydroxylase (C3H; EC 1.14.18.1), respectively. An O-methylation of the 3-hydroxyl in caffeic acid, catalyzed by caffeic acid O-methyltransferase (COMT; EC 2.1.1.6), leads to the formation of ferulic acid
(Kuroda et al., 1981; Edwards and Dixon, 1991). The ferulic acid undergoes hydroxylation at the 5-position of the ring to form 5-hydroxyl-ferulic acid, catalyzed by ferulate-5-hydroxylase (F5H) (Grand, 1984; Ohashi et al., 1987), and further O- methylation of the 5-hydroxyl to yield sinapic acid (Bugos etal., 1991).
4 The free hydroxycinnamic acids are seldom found to be accumulated in significant amounts in plant tissues, since they are rapidly converted to Coenzyme A (CoA) esters, a reaction catalyzed by 4-coumaroyl CoA ligase (4CL; EC 6.2.1.12) (Knobloch and y
Hahlbrock, 1975; Wallis and Rhodes, 1977), or to glucose esters, in a reaction catalyzed by UDP-glucose:hydroxycinnamic acid glucosyltransferase (HGT; EC 2.4.1.-) (Fleuriet etal., 1980; Baumker etal., 1987; Mock and Strack, 1993). These activated derivatives form a crucial branch point in phenylpropanoid metabolism, and lead to a wide range of specific reactions, since they serve as the precursors for all major classes of phenolic compounds which accumulate in plant tissues, e.g. flavonoids, lignins, stilbenes, phenolamides, benzoic acid derivatives, hydroxycinnamic acid esters (Fig. 1).
3. Glucosylation and UDP-glucose-dependent glucosyltransferases
Although glucosylation of sinapic acid is specifically important in sinapine biosynthesis and metabolism, the glucosylation/glycosylation of plant secondary products occurs widely in higher plants. The physiological functions of glucosylation/glycosylation are thought to be: (1) to increase the solubility of secondary products for compartmentation and storage; (2) to detoxify exogenous compounds and (3) to convert plant secondary products to conjugated compounds as intermediates for the formation of other physiologically important metabolites. Plant glycosides include simple phenolics, hydroxycinnamic esters, coumarins, flavonoids, glucosinolates, cyanogenic glucosides, steryl glucosides and alkaloids (Hosel, 1981). By definition, when the sugar being transferred is glucose, the process is defined as glucosylation and the enzyme catalyzing
5 Primary metabolism
H2NvCOOH COOH HOOC Carbohydrates HO^Y^OH (fll DH Lj_l Arogenic acid
Shikimic acid
Tyrosine OH Phenylalanine
Phenylpropanoid pathway
COOH COOH COOH COOH
HsCO-^pOCHs HO^^OCH3 -OCH3 >H OH OH
Sinapic acid 5-OH-Ferulic acid Ferulic acid Caffeicacid p-Coumaric acid Cinnamicacid
Specific phenolic branch pathway O COOH H-o-x GO,
Coumarins Xanthones Benzoic acid derivatives Hydroxycinnamic acid esters
OH OH Lignin or cr" Fig. 1. The central role of the phenylpropanoid pathway in biosynthesis of phenolics in plants. R = H, OH, or OCH3, X = CoA, glucose or other conjugates. 6 the reaction is referred to as a glucosyltransferase. When other types of sugars are being transferred, the process is designated as glycosylation, and the transferases are referred to as glycosyltransferases. UDP-glucose-dependent transferases are the most important class of glucosyltransferase responsible for the glucosylation (Hosel, 1981). Depending on the mode of attachment of the sugar to the aglycone, the resulting sugar conjugates can be classified as glucosides, i.e. glucose attached through its anomeric carbon to hydroxyl or phenolic groups; or glucose esters, i.e. glucose attached through its anomeric carbon to a carboxylic acid group. Despite acting upon the same substrates, the enzymes catalyzing formation of these two types of glucose conjugates in Tulipa were recently shown to be unrelated proteins (Baumker et al, 1987). Most glucosylation of simple phenols (Pridham and Saltmarsh, 1963; Miles and Hagen, 1968; Yalpani et al, 1992), flavonoids (Sun and Hrazdina, 1991; Cheng et al, 1994; Ishikura and Yang, 1994), cyanogenic glucosides (Hosel and Schiel, 1984), glucosinolates (Guo and Poulton, 1994; GrootWassink et al, 1994), steryl glucosides (Ury et al, 1989; Warnecke and Heinz, 1994) and alkaloids (Stapleton et al, 1992) is catalyzed by the type of glucosyltransferases that yields glucosides. However, when hydroxycinnamic acids (HCA) are conjugated with sugars, the products can be either glucose esters, glucosides or both (Harborne and Corner, 1961; Imperato, 1976; Fleuriet et al, 1980). 7 Synthesis of glucose esters of HCA by UDP-glucose-dependent transferases in vitro has been reported in number of species, including the formation of 1-^-coumaroyl, 1- caffeoyl, 1-feruloyl and 1-sinapoyl glucose in Geranium zonale (Corner and Swain, 1965), 1-sinapoyl glucose in B. oleracea (Corner and Swain, 1965) and R. sativus (Strack, 1980), 1-feruloyl glucose in young unripe apples (Macheix, 1977), 1-p- coumaroyl, 1-feruloyl and 1-sinapoyl glucose in tomato fruits (Fleuriet et al, 1980), \-p- coumaroyl glucose in leaves of Coleus, Pilea, Cistus and Cestrum (Nagels et al, 1981), 1-cinnamoyl, 1-jO-coumaroyl, 1-caffeoyl and 1-feruloyl glucose in root of sweet potato (Shimizu and Kojima, 1984), l-/>-coumaroyl, 1-caffeoyl, 1-feruloyl and 1-sinapoyl glucose in Tulipa anthers (Baumker et al, 1987), and l-/?-coumaroyl and 1-feruloyl glucose in flowers of Ruschieae (Strack et al, 1990). 4. Biological significance of SGT In many members of the Brassicaceae, especially Brassica and its closely-related genera, one of the remarkable chemotaxonomic characteristics is the accumulation of sinapic acid esters in their tissues. In seeds, sinapine, an ester of sinapic acid and choline, can reach levels of 40 (j,mol/g seed (Bouchereau et al, 1991), which is equivalent to 1-4% of the dry weight of the seed (Blair and Reichert, 1984; Uppstrom and Johansson, 1985; Wang, 1992). In contrast, the predominant phenolic compound in the vegetative tissue is usually O-sinapoylmalate, an ester of sinapic acid and malic acid (Strack, 1982; Strack et al, 1990). In red radish seedlings, the content of sinapoylmalate in one pair of cotyledons could reach a level as high as 110 nmoles (Strack et al, 1984). It has been suggested that 8 the sinapine accumulated in seeds might serve as a ready-to-use pool of sinapic acid and choline for the development of young seedlings (Strack, 1981), and that sinapoylmalate in vegetative tissues might play a role as a light filter by absorbing damaging UV-B radiation in order to protect the guard-cell chloroplasts (Harborne, 1980). However, the physiological functions of these two compounds are not yet clear. The discovery of sinapine in the Brassicaceae can be traced back to 1825 when Henry and Garot first isolated this phenolic ester from seeds of S. alba (cited from Remsen and Coale, 1884). It was not until 72 years later that a chemical structure was proposed for sinapine (Gadamer, 1897). Although this ester is widely distributed in many members of Brassicaceae family (Schultz and Gmelin, 1952 and 1953; Kerber and Buchloh, 1980 and 1981), its biosynthetic pathway, and the underlying enzymology, were not clarified until the early 1980's (Bopp and Ludicke, 1980; Strack, 1980; Strack etal, 1983). Sinapine biosynthesis occurs in the embryo tissue during brassicaceous seed development. Like other hydroxycinnamic acids, the carbon skeleton of sinapic acid is supplied by L-phenylalanine via the phenylpropanoid pathway (Gross, 1981). Sinapic acid is first activated by SGT to form the glucose ester, and this sinapoylglucose then provides the acyl donor for formation of sinapine. The latter reaction is catalyzed by l-O- sinapoyl-|3-D-glucose:choline sinapoyltransferase (SCT; EC 2.3.1.91) (Strack et al, 1983; Grawe and Strack, 1986; Vogtetal, 1993). 9 During seed germination, sinapic acid released from sinapine through the activity of sinapine esterase (SE; EC 3.1.1.49) (Tzagoloff, 1963a; Nurmann and Strack, 1979) is re- esterified to sinapoylglucose by SGT. The sinapoylglucose is then used as a substrate for the synthesis of 0-sinapoylmalate catalyzed by l-0-sinapoyl-[3-Z)-glucose:Z--malate sinapoyltransferase (SMT; EC 2.3.1.92) (Tkotz and Strack, 1980; Strack et al, 1990). SGT is thus a key enzyme involved in both sinapine biosynthesis during seed development and sinapoylmalate biosynthesis during vegetative growth of the plant (Fig. 2). The temporal and spatial patterns of its expression are crucial factors in establishing the pattern of sinapic acid derivatives formed in Brassicaceae. 5. Economic significance of SGT Species of Brassica and allied genera have been used intensively in agriculture since ancient times. Particularly, the seed of Brassica, known as rapeseed, is one of the most important sources of edible vegetable oil. The seed contains approximately 40% oil and produces a meal with up to 40% protein after oil extraction. As the second largest oil-seed crop, rapeseed provides not only huge quantities of edible oil for human consumption, but also yields a valuable protein source as animal feed (Downey and Robbelen, 1989). However, the high content of sinapine in the seed limits the utilization of this protein-rich resource since the bitter taste produced by sinapine can result in poor palatability and other nutritional defects (Schwarze, 1949; Clandinin, 1961; Sosulski, 1979; Ismail et al, 1981). Unpleasant flavors in meat and milk have been reported from cows fed with 10 11 canola or rapeseed meal (Larsene? al, 1983; Bille et al, 1983; Andersen, 1985; Andersen and Sorensen, 1985). A "fishy" taint in the yolk of brown-shelled eggs has also been associated with the presence of sinapine in the feed (Hobson-Frohock et al, 1973; Hobson-Frohock etal, 1977; Butler and Fenwick, 1984). Elimination of sinapine from the seed would improve the flavor, palatability and nutritional properties of rapeseed meal, and thereby enlarge the markets for rapeseed and its by-products. The sinapine content in the meal can be reduced through various processing techniques (Fenwick et al, 1979; Goh et al, 1982; Dabrowski and Siemieniak, 1987; Dabrowski et al, 1989; Tayaranian and Henkel, 1991), but so far none of these have proven to be economical. Conventional plant breeding would be a more efficient long-term means of lowering or eliminating sinapine. However, the success over the past decades in plant breeding of rapeseed for improvement in both oil and meal quality has not reduced the sinapine content in the seeds (Uppstrom and Johansson, 1985). Studies of genetic variability for sinapine level in Brassica and its allies indicate that a lack of "low sinapine" germplasm is the major limitation for incorporating this trait into new varieties through conventional plant breeding (Clausen et al, 1985; Kraling et al, 1991; Wang, 1992). Biotechnology provides powerful tools for modifying plant genomes. The introduction of appropriate sense/antisense gene constructs into Brassica may make it possible to block the sinapine biosynthetic pathway and thus create a novel "low sinapine" genotype. 12 Before attempting to modify metabolism through this approach, however, it is essential to have a thorough understanding of the metabolic patterns involved. Taxonomic analysis of metabolite distribution can provide insights into the range of variability that exists naturally within Brassicaceae, while developmental analysis within a species can reveal the variation associated with tissue type, age and environmental influence. In the present study, the taxonomic distribution of SGT was investigated in order to establish whether a correlation exists between the level of enzyme activity and the level of sinapine in the seeds. The developmental pattern of SGT expression in both B. napus and Sinapis alba was then studied in detail, since S. alba was found to have a unique sinapoyl ester profile. The enzymology of sinapine biosynthesis/metabolism, and the diversity of phenolic metabolites, were also explored in both species. Finally, SGT was purified from seedlings of B. napus cv. Westar and fully characterized in order to gain access to the amino acid sequence, and thus to the corresponding gene. A preliminary attempt to identify and clone the putative SGT gene was also undertaken. The results of this research provide essential baseline information in support of our ultimate objective to genetically engineer a new genotype of B. napus with reduced sinapine content in the seeds. 13 Chapter II TAXONOMIC DISTRIBUTION OF SGT IN BRASSICA CEAE 14 INTRODUCTION A survey of sinapine content in the seeds of seventy-seven species of various taxonomic groups of Angiospermae families found that sinapine occurred in forty-four members of the Brassicaceae (Regenbrecht and Strack, 1985). A more detailed investigation of the distribution of total phenolic choline esters and sinapine in Brassica and ten allied genera showed that the sinapine content ranged from 0.8 ixmol/g seed (Eruca sativa) to 46 pmol/g seed (Hesperis matronalis) (Bouchereau et al, 1991). In a study of sinapine variability in the seeds of twenty-five genotypes of Brassica and S. alba, a positive correlation between sinapine content per seed and seed weight was observed (Wang, 1992). Among these investigated genotypes, seed weights ranged from 1 to 6 mg per seed, while sinapine contents correspondingly varied from 17 to 113 pg per seed. In addition to genetic variability, the sinapine content in brassicaceous seeds can also be influenced by environmental conditions (Kvartshava et al, 1980; Bouchereau and Larher, 1991; Wang, 1992). However, the diversity of sinapine distribution in the plant kingdom is primarily a result of long-term natural selection during evolution, and is only modestly affected by the geographic location where the plants grow (Clausen et al., 1985; Kraling etal., 1991). The earlier studies of sinapine biosynthesis established that sinapic acid cannot be directly used as a substrate to form sinapine. Instead, its energy-rich 1-O-glucose ester, 1- 0-sinapoyl-(3-D-glucose, serves as the immediate substrate for the formation of sinapine 15 in a reaction catalyzed by SCT. Accumulation of sinapine in seeds of brassicaceous species has been shown to be correlated positively with the occurrence of SCT (Regenbrecht and Strack, 1985). Sinapoylglucose, on the other hand, was found to be only transiently accumulated in young seedlings and developing seeds of Raphanus sativus (Linscheid et al, 1980) and S. alba (Kuhnl and Wellmann, 1979; Bopp and Ludicke, 1980). The enzyme (SGT) catalyzing the formation of l-0-sinapoyl-(3-D- glucose was later detected in young seedlings of R. sativus (Strack, 1980; Nurmann and Strack, 1981), but the distribution of SGT activity within Brassica and allied species has never been examined. If SGT activity is essential for sinapine formation, a close correlation between occurrence of this enzyme and accumulation of sinapine would be predicted. OBJECTIVE The objective of the present study was to investigate the degree of correlation between sinapine content and SGT activity in Brassicaceae. MATERIALS 1. Plant materials Seed samples used in this experiment were obtained either from Agriculture and Agri- Food Canada Saskatoon Research Center, Saskatoon, Saskatchewan or Department of Plant Science, University of British Columbia, Vancouver, British Columbia, Canada (Table 1). 16 Table 1. The names and sources of seeds of 38 genotypes from 13 species in Brassicaceae. Species Genotype Code Source Brassica carinata PGRC/E 200423 1 1990 Saskatoon Dodolla 2 1990 Saskatoon S-67 3 1990 Saskatoon PGRC/E 212825 4 1990 Saskatoon B. chinensis Mustard cabbage 5 UBC B. juncea R2121 6 1989 Saskatoon Commercial Brown 7 1988 Saskatoon Cutlass 8 1993 Scott Lethbridge 22A 9 1988 Saskatoon B. napus Bronowski 10 1989 UBC Westar 11 Cert. 89-7022321-41 Profit 12 Cert. 89-7000030-50 Argentine 13 1986 Scott Midas 14 1991 Scott B. nigra Type 4 15 1991 Scott Type 3 16 1989 Scott Type 1 17 1992 Scott Type 2 18 1989 Scott B. rapa R-500 19 1990 Scott Tobin 20 Cert. 91-7017242-41 AC Parkland 21 Cert. 90-7006098-41 Echo 22 1984 Scott Torch 23 1986 Scott 17 Table 1. continued. Species Genotype Code Source B. oleracea Native German 24 1993 GH increase Hybrid cabbage 25 UBC B. tournefortii Native Australia 26 Australia Eruca sativa R1841 27 India RL-Sask 28 1989 Saskatoon Raphanus sativus Native Poland 29 Poland Cherry belle 30 Ostal seed company Sinapis alba Tilney 31 1993 Saskatoon Thorney 32 1993 Saskatoon Ochre 33 1990 Saskatoon Gisilba 34 1993 Saskatoon S. arvensis Googale 35 1989 Goodale Arabidopsis thaliana M2-No. 72 36 UBC Columbia WT-2 37 UBC LandsbergWT 38 UBC 18 2. Chemicals All chemicals were reagent grade except chemicals used in high-performance liquid chromatography (HPLC), which were HPLC/UV grade. Most were purchased from Fisher Scientific Company or Sigma Chemical Company with the following exceptions. Sinapic acid (3,5-dimethoxy-4-hydroxyl-cinnamic acid, 98%) was purchased from Aldrich Chemical Company. Sinapine was isolated from S. alba cv. Ochre seed (Wang, 1992) and l-0-sinapoyl-(3-Z)-glucose was isolated from 3-day-old radish (R. sativus) seedlings according to the method of Vogt et al. (1993). 3. Equipment HPLC was conducted using a Gilson gradient system equipped with a model 715 UV detector, autosampling injector and a Nucleosil C18 reverse phase column (4.6 x 250 mm, 5 [i matrix, Alltech Chromatography Inc.). METHODS 1. Determination of sinapine by HPLC 1.1. Sample preparation Seeds (about 50 mg) were ground in a mortar with 0.75 ml 100% methanol, then transferred to a 2 ml screw-cap microcentrifuge tube. The mortar was rinsed with another 0.75 ml methanol which was combined with the homogenate in the tube. The tube was capped and the homogenate was allowed to stand for 1 hour at room temperature. The sample was then centrifuged at 14,000 x g for 20 min. The supernatant was transferred to 19 a HPLC sample vial and 10 ^il of the supernatant was injected onto a Nucleosil C18 reverse-phase column for the separation of sinapine and other phenolics. Three extracts were prepared for each seed sample. 1.2. HPLC analysis The HPLC analysis was carried out using a gradient elution system at room temperature. A linear gradient elution started from 0% to 50% solvent B (1% ort/io-phosphoric acid in acetonitrile) in solvent A (1% ortho-phosphoric acid in water) in 20 min, continuing from 50% to 80% B in A in 0.1 min and holding at 80% B for 1.9 min, then decreasing from 80% to 0% B in A in 0.5 min, and finally holding at 0% B in A for 2.5 min. The solvent flow rate was 1 ml/min and the eluant was monitored at 330 nm with a sensitivity of 0.1 absorbing unit full scale (AUFS). 1.3. Calculation of sinapine content An external standard method was used to calculate sinapine content. The standard curve was generated by injecting a series of known concentrations of sinapine (0, 15, 30, 50 and 100 pM). Sinapine content from a unknown sample was then calculated using the following equation: 6 CSin = [(RFx A) x (Ve/Vi) xDF/10 ] /W (Equation 1) where: CSin (pxnol/g seed) = sinapine content RF (pmol/mV*) = response factor (the regression coefficient of standard curve) A (mV2) = sinapine peak area 20 Ve (ml) = total volume of the extract V; (ml) = injection volume DF = dilution factor 106 (pmol/(a.mol) = conversion factor W (g) = weight of seeds used for extraction The sinapine content per seed was calculated as the following: 3 Sinapine (nmol/seed) = CSin x 10 x Ws (Equation 2) where: CSin (p:mol/g seed) = sinapine concentration from Equation 1 103 (nmol/jxmol) = conversion factor Ws (g/seed) = individual seed weight 2. SGT assay for dry seeds and seedlings 2.1. Seedling growth Seed samples (~1 g) from each genotype were placed in 9 cm Petri dishes with two layers of pre-wetted Whatman No. 1 filter paper and allowed to germinate at room temperature (~ 23° C) for 60 h in darkness. 2.2. Protein sample preparation Samples of dry seeds (~ 0.1 g), or fresh seedlings (~ 1 g), were ground to a fine powder in a pre-chilled mortar with liquid nitrogen. One ml extraction buffer [100 mM Tris/HCl, pH 6.5 with 4% (w/v) of insoluble polyvinylpolypyrrolidone (PVPP), 10 mM |3- 21 mercaptoethanol (2-ME) and 5% (v/v) glycerol] was added immediately and mixed. The homogenate was transferred to a 2 ml screw-cap microcentrifuge tube and extracted by rolling at 4° C for 1 h. The supernatant obtained by centrifugation at 14,000 x g and 4° C for 20 min was transferred to a 1.5 ml microcentrifuge tube and centrifuged again at 14,000 x g and 4° C for 10 min. The protein extract was then desalted by using spin- column chromatography (Penefsky, 1977). The spin-column consisted of a 1 ml disposable polypropylene syringe packed with Sephadex G-25 (Pharmacia Biotech Inc.). The column was equilibrated with 20 mM Tris/HCl buffer including 10 mM 2-ME and 5% glycerol, pH 6.5. and then spun in a clinical benchtop centrifuge with swinging bucket rotor (3000 x g for 2 min). Protein extract (200 ixl) was then transferred to the top of the gel matrix. Once the protein extract had soaked into the Sephadex bed, the column was spun at 3000 x g for 2 min. The filtrate (desalted protein extract) was collected in a 1.5 ml screw-cap microcentrifuge tube and stored at -70° C until the SGT activity assay was carried out. 2.3. Determination of protein concentration Protein concentrations of the seed and seedling extracts were measured using the Bradford dye-binding method (Bradford, 1976), with bovine serum albumin as the standard. 22 2.4. Determination of SGT activity SGT activity was normally assayed by HPLC, measuring the amount of sinapoylglucose formed in an enzymatic reaction. The reaction was carried out in a 1.5 ml HPLC sample vial containing 50 [il protein extract. The reaction was started by adding 25 [il substrate mixture (1.5 mM sinapic acid, 1.5 mM UDP-glucose, 30 mM 2-ME, 30% glycerol and 240 mM MES [2-[7V-morpholino]ethane sulfonic acid], pH 6.0), and the tube was incubated at 37° C for 30 min. The reaction was stopped by adding 250 (0,1 acetonitrile, leaving it at room temperature for 10 min, and then adding 675 u.1 distilled water. For HPLC analysis, 100 pi of this solution was injected onto a Hypersil C8 reverse phase column (4.6 x 250 mm, MOS 5 \i, Alltech Chromatography Inc.). The HPLC analysis was carried out using a gradient elution system at room temperature. A linear gradient elution started from 30% to 50% solvent B (2% glacial acetic acid in acetonitrile) in solvent A (2% glacial acetic acid in distilled water) in 2 min, holding at 50% B for 0.5 min, then continuing from 50% to 80% B in A in 0.5 min and holding at 80% B for 0.5 min, then decreasing from 80% to 30% B in A in 0.5 min, and finally holding at 30% B in A for 2.5 min. The solvent flow rate was 1.2 ml/min and the eluant was monitored at 330 nm with a sensitivity 0.1 AUFS. 2.5. Calculation of SGT activity An external standard method was used to calculate sinapoylglucose concentration. The standard curve was generated by injecting a series of known concentrations of 23 sinapoylglucose (0, 11, 22, 55 and 110 \xM). Sinapoylglucose concentrations of the samples were calculated using the following equation: CSG = ( RF x A ) x ( Vh / V;) (Equation 3) where: CSG (pmol) = amount of sinapoylglucose RF (pmol/m Vz) = response factor (the regression coefficient of standard curve) A (m V ) = the sinapoylglucose peak area Vh (ml) = final sample volume in the HPLC vial V; (ml) = injection volume Total SGT activity in the seed/seedling extract was calculated as follows: SGT activity (pkat) = (CSG / Tr) x (Ve / Vr) (Equation 4) where: CSG (pmol) = amount of sinapoylglucose in the reaction from Equation 3 Tr (sec) = reaction time in second Ve (ml) = total volume of protein extract Vr (ml) = volume of protein extract for the enzymatic reaction Thus, SGT specific activity (pkat/mg protein) = total SGT activity (pkat) / total protein (mg) (Equation 5) 24 RESULTS 1. Sinapine accumulation Sinapine content in the seeds was calculated based on both individual and bulk seeds except in the case of Arabidopsis, which was only calculated as bulk seeds. Based on bulk seed weights, the sinapine content ranged from 5 to 52 p,mol/g seed. The highest level was found in B. oleracea, while the lowest was found in is. sativa (R1841, Table 2). On an individual seed base, B. napus cv. Bronowski had the lowest sinapine content (9 nmol/seed) and B. oleracea line hybrid cabbage had the highest sinapine content (383 nmol/seed). In general, S. alba had the highest average level of sinapine among the investigated species (31 [xmol/g seed or 210 nmol/seed). 2. SGT activity SGT activity was measured for both mature seeds and 60-h-old seedlings. SGT activity was detected in dry seeds of most species, but not in line R2121 from B. juncea, or in any of the four tested genotypes of S. alba (Table 2). The highest SGT activity in dry seed was found inE. sativa genotype 28 (RL-Sask), which had an activity at 663 pkat/g seed or 23 pkat/mg protein. The lowest detectable SGT activity was also found in an E. sativa line, R1841 (4.4 pkat/g seed or 0.2 pkat/mg protein). Following germination, the seedlings were analyzed for SGT at a common time point (60 h). SGT activity in 5. alba still could not be detected 60 h after germination, but was found in all genotypes of other species (Table 2). The highest SGT activity was found in 25 CO CO CD o 1— r— "3" 00 CO a 00 CO CM oo Is- 00 CM LO oo LO o (0 1 d CM ,— d 1— d CM d CO CM 1 CM d in g 4-* c C\J CO oo CM CT) o oo Is- r— \> c LO CO o CD CM oo o LO CO LO o r— CD '53 (C0O 00 00 CM CD CM oo co 00 CO U +-* seed l d "i— CM ^ d CM d < o O a »*— s D) CM LO CD co o I - LO a> CM CO 00 "a5 a co CM o CM o CO CM o oo o . E CO a. CO o O d d d d d d d d d d d d to a •a H a o O) 00 00 CM o LO O co co IS- CO oo c s oo LO cn l-» o o I - CD co LO CO CO (5 Se e CQ CO a> Cxi c\i CM CO oo 5 d d i— d d CM 00 LO co CM CD LO CO C\J CD a CM o CD CM CO CM CM CO 00 o CNJ ,_ s CO CO l_ 00 I-' D) LO CM CO d co " d d d d c LO o oo CM 00 O) CD LO r— as CM CJ) c CO CD •sr LO eedl i ce oo co co co CM CO CO CM oo 00 00 Cvi co a) 00 d CO d co CM d CD r-~ d d H-» 3 CM 00 LO CD CNJ OT > W O < O) oo r-- CD 00 oo o LO cn 00 CD cn LO 00 LO o CM oo co 00 cn CD H a a CO CO LO co d d CM d LO d d CM co a. CM c CM r- CM CO o cn LO Is- LO o CO CO See d O o o o oo LO CM o o CO cn A rs — CU CM CM ai d I-" d CM LO 00 CM co d d d CM CD r— d CD CNJ oo CO CM oo CO LO i— CM a> 00 CM o r— LO CM CD CD r-- O) CO 00 a o CO CM CT) o r-_ cn CM o •a s a> CO CO CO CM I-" ,_ r—' CO d CM d d d d a> CM OT c o CO h- co CD CD LO o CD 00 cn CM O) 00 00 cu E c LO 00 CD LO r-- r- CD r- CM c c cc CO d co CM oo d CM 00 CM 00 CM co o cu CM co LO CO LO r— CO co oo r- 00 o £ LO co c s 03 o CO CD I - CM LO LO CO '5. XI a CO 00 oq 05 LO cn 00 - CO LO CM co cn CO a> CO T CM CM CM CM cu o CO CO T CO CO CM CM c to d " ~ d wl O) « 00 o CD Is- o CO Is- LO LO CO 1< 2 I- 26 co CO CO CM T— LO CO o o CM O co T— CM 00 CD LO CM CM CD 00 CM CM 1_ oq r-- cn co co O d ,~ LO co d d CM d d d c _ ^ 1^ c o CO CD h~ CO CM cn cn c CO co 00 LO 00 CM CM 00 LO CM Q) oo cn -cr "5! co CO LO CM CD CO CM CM d LO d CO W 1— CM 1— ,— o CM I— a ' o> co LO IS- LO co CM CO o cx> h- LO E S t> Q o o O CD o I - CD O o T-| CM n CO O d 1_ d d d d d d d d d d a T3 a> c o CO CO o LO oo CO co 00 CD ca LD LO co cn 00 CM co co CM CO CO oo CO CD O CO CO CM LO ,—• CO C\i T—" d co d d CM co cn o cn CM o oo CM IS- CM LO LO a cn CD LO CM h~ o CM CD CM CD CM T—- D) CD d co CO CD 1— 00 CM CO CM CO , CM ,_ c ~ 1=5 CD CO co o co CD •G CD CD CM cn v S CD ca CO CO LO I - o oo o LO oo 00 CD LO CO CO o od d cd d dS d d co r—' d a> I to CM 00 LO LO 1 CO I - CO 1 3 oo co .2 -3- LO CM cn CO CO CM o LO LO ^— cx> o> a oo o CM CD CD -4- CJ) CM CM CD CM fl] co cd d oo LO cd CM d LO CO CM CO cn CO 1— CM CM CO a. co o c CD CX> CO CM LO CM CM o LO CM o o to ca 00 CD co CD r-- o LO CD O) o CD co d 00 d d LO oo CM T—" o LO O CO CM LO CO CO i— CO CM CD oo r~- LO CD co CO r-~ LO CO O O CD o LO LO CD CD LO T3 CO oo co a> o CD d CO CD C\i d d CM CO a> ,~ CM w o oo LO CO co r- CM cn oo CM cn c CM CD CO LO CO CM a> LO LO CD cn CM E ca o 00 CO \i CO c CD c\i CD co CO d d C d d d LO o LO CD CO CD 00 00 T— i— CO LO LO co 1 r-- o co CM cn o LO LO cn 00 CO CD CM CO i— LO T-| CM CD a cn CO CO co d CO d \i CM d d CM d d ,_ ,_ CM ,_ CO C D) c CNJ CD CM CD 00 cn CO co CD cn CD "5 ca CO oo 00 5 00 oq cx> CD CO CD o ai oS d CO LO LO CM LO d E CM CM CM CM CM CM CM CM LO ^— 3. co eg | c? CD CD 5 Q. 3 O CD (5 o O CO ^ CM < a. .a CO 1 c CD CD CD O •— CD CC CD c^ CD Q. CL Q. LO -Q CL O 5 «> .5 ^ O u o o u > 0! T- j io r hhhCh < UJ I- Z I Z OC OC Z U 27 0 O 0 0 co LO 00 00 s Q 0 O 0 0 CM I - LO CO 0 d d d d T~ d d TJ c 0 0 0 0 CO CM 00 c CO (0 0 0 0 0 LO 00 CO s CO CO CM I-' "53 CO 0 d d d c\i o 1 Q. O) 0 0 0 0 CM CO LO E Q 0 0 0 0 O LO Is- T— is CO d d d d d d d d co TJ CO 0 0 0 0 CM CO Is- • 0 c CO r- CO 0 0 0 0 00 (0 T~— CO d d d d d CM T~ 0 0 0 0 CD a 0 0 0 0 Is-; CT> CD CO d d d d 1— d d d c 1=5 0 0 0 0 CD 0 Is- c 0 0 0 0 Is- CO CD , ® 10 CO d d d d d d d CO 1— LO CO CD 1 CO CO O) 0 0 0 0 CO a 0 0 0 0 0 CO q 55 CO d d d d d d ,—' CM Q. TJ CO • 0 0 0 0 0 0 CD Is- W c 0 0 0 0 0 0 (0 d d d d d d d d CO co T— 00 CM 1 CO CD CO •<* •3- 00 CD LD Is- LO TJ a CO CO d d CO CM T— w o CO 0 CO E c a> CO LO Is- CO < < < c co d d CM CM d 0 Is- CO 00 CM CM 1 00 O CD O CO CM co Is- TJ Is-; CO a C\J O CD 00 CM O C\l CO CO r d d d ' ~ d d d CO D) O c CO LO •sf a> co 1— CO CD LO CM s E CO CNJ d d d I-' d d CO CO CO CM 1— CM CO ^ E? a. CO o ^ CO CO CO c a> i CO o 141 CO o (5 —I t— CD CD 28 A. thaliana Landsberg WT (164 pkat/g fresh tissues), while the lowest was found in B. nigra Type 4 (1.8 pkat/g fresh tissues). On a specific activity basis, B. rapa line R-500 showed the highest SGT specific activity (26.5 pkat/mg protein), while E. sativa line R1841 showed the lowest (0.5 pkat/mg protein). SGT activities in the seedlings (pkat/g fresh tissues) were lower than those in the dry seeds (pkat/g seed) (Fig. 3A). However, the protein concentrations were also much lower in seedlings than in seeds (Table 2), so that an overall increase in SGT specific activities was observed (Fig. 3B). The relationship between SGT total activity and specific activity showed a positive correlation (r = 0.9554, in seeds; r = 0.7650, in 60-h-old seedlings). The correlation between SGT activities and sinapine levels was analyzed for the thirty- eight brassicaceous genotypes examined, but no significant correlation was observed between SGT activities and sinapine levels, based on either bulk seeds or individual seed. SGT activity also did not correlate with seed size. DISCUSSION Plants synthesize large amounts of secondary metabolites via the phenylpropanoid pathway. These secondary metabolites occur in plants mainly in the form of conjugates (Barz and Koester, 1981), and the esters of phenolic acids form one of most abundant conjugate classes (Barz et al, 1985). It has been demonstrated that CoA activation of the cinnamic acid side-chain reaction plays an important role in the formation of phenolic 29 (anssji usaji 6/ie>td) AIJAIJOB J_OS (inaicud 6uj/iB>|d) AIJAJPB ojipads j_os 30 acid esters (Zenk, 1979 and Gross, 1981). However, the biosynthesis of conjugates can also proceed via 1-O-acyl glucose derivatives of phenolic acids. Recent reports indicate that this acyl-glucose mediated carboxyl-group activation is a widespread alternate pathway (Michalczuk and Bandurski, 1980; Tkotz and Strack, 1980; Gross, 1982 and 1983; Dahlbender and Strack, 1986). The acyl-glucose formation of phenolic acids is catalyzed by UDP-glucose-dependent glucosyltransferases. In the sinapic acid ester pathway, the UDP-glucose-dependent glucosyltransferase is SGT, which plays a pivotal role in providing substrate for sinapine (Stack et al, 1983) and sinapoylmalate (Tkotz and Strack, 1980) biosynthesis. However, SGT activity has only been demonstrated to occur in B. oleracea (Corner and Swain, 1965) and R. sativus (Strack, 1980; Nurmann and Strack, 1981; Mock and Strack, 1993). In this investigation, SGT activity was detected in 34 genotypes of 12 species containing sinapine. The universal occurrence of SGT suggests that this enzyme, as expected, is specifically involved in sinapine biosynthesis and metabolism. An unexpected exception observed in this study is S. alba, which failed to yield extractable SGT activities from either dry seeds or 60-h-old seedlings (Table 2) of all 4 examined genotypes. This result suggests that sinapine metabolism in S. alba may differ from that in the other investigated brassicaceous species. The patterns of sinapine synthesis and metabolism in S. alba have therefore been examined in more detail (Chapter III). 31 SGT activity could not be detected in the dry seeds of R2121 (B. juncea), although it was detected in the 60-h-old seedlings. SGT activity detected in dry seeds may represent the enzyme synthesized during seed development. The SGT protein may lose its activity due to seed aging or seed storage conditions, which could partially explain lower enzyme levels in seeds of same genotypes. SGT activity detected in the seedlings may represent a combination of de novo synthesized enzyme and activated pre-existing SGT. The trend of increasing SGT activity in the seedlings is very obvious. SGT specific activities in most of genotypes increased from 1 to 23-fold after germination for 60 h, but this pattern could be influenced by the variation in seed sources, in ripening conditions, in moisture, protein and oil contents, and other post-harvest storage conditions. Single time-point reporting of SGT activity in the seedlings also makes interspecific comparisons difficult because of the differences in timing of seed germination and seedling development. No correlation between sinapine content and the level of SGT activity was observed in the present study. This may reflect a number of factors: (1) SGT is not an enzyme which directly catalyses the formation of sinapine, so a correlation is not necessarily predicted; (2) The SGT activities measured in this investigation were from the dry seed and young seedlings, and thus did not necessarily represent the SGT levels at the stage of seed development in which sinapine is synthesized and accumulated; (3) SGT activity may be required for other, non-sinapine-related biosynthetic processes, and its overall level may therefore reflect a combination of needs, including its function in sinapine biosynthesis and metabolism. 32 Chapter III THE DIVERSITY OF SINAPIC ACID ESTERS, ENZYMOLOGY AND DEVELOPMENTAL EXPRESSION OF SGT IN B. NAPUS AND S. ALBA 33 INTRODUCTION The general characteristics of plant secondary metabolism have been intensively reviewed (Wiermann, 1981; Davin and Lewis, 1992; Herrmann et al., 1995). The occurrence of secondary metabolites can vary with the plant tissue, as well as with the species. The presence and the level of individual compounds may also change with the developmental stage of the plant. Sinapine is a classic example of a plant secondary product. In the Brassicaceae, sinapine has only been found to accumulate in the seed tissue (Schultz and Gmelin, 1952 and 1953). The physiological function of this compound is still uncertain, but it has been reported that sinapine is rapidly degraded in seedling tissue during germination (Tzagoloff, 1963a; Strack, 1977; Strack et al, 1978; Clausen et al, 1982 ). Since no free choline accumulated, and no choline oxidase activity was detected in seedling tissue, it has been suggested that the choline released from sinapine during its catabolism is further metabolized (Tzagoloff, 1963a). A [14C]-choline feeding experiment using R. sativus seedlings indicated that liberated choline was partially incorporated into phosphatidylcholine, an intermediate in phospholipid metabolism (Strack, 1981). These observations inspired the suggestion that sinapine may serve as reserve supply of choline for the biosynthesis of phosphatidylcholine during the time of seedling development. Other studies on the biological activities of sinapine in vitro have shown that sinapine and its metabolic derivatives, e.g. sinapic acid and sinapoylglucose, are able to inhibit auxin oxidase activity. Thus, sinapoyl derivatives may function to protect IAA, and could thereby enhance the effect of auxin on seedling growth (Kefeli et al, 1977). However, an Arabidopsis mutant deficient both in sinapoylmalate 34 accumulation in leaf tissue, as well as sinapine accumulation in seed tissue, proved to germinate and develop as rapidly as the wild-type (Chappie et al, 1992). This indicates that neither of the proposed roles, if they operate at all, is crucial to plant growth and development. The biological function of sinapine in vivo thus remains to be established. Besides sinapine, several other conjugates of sinapic acid have been identified in brassicaceous species (Table 3). These sinapoyl-conjugates are all believed to be either precursors or metabolites of sinapine, and most of them do not accumulate to a significant level. The major exception is sinapoylmalate, which is one of the main methanol- extractable compounds in seedlings of R. sativus and B. napus. The long-term accumulation of sinapoylmalate has not been examined, so it is unclear whether this compound will be further metabolized. One possible physiological function for the sinapoyl-conjugates in the vegetative tissues may be increased protection from UV-B damage, since these compounds all have high extinction coefficients in this wavelength range. A mutant of Arabidopsis deficient in sinapoylmalate synthesis exhibits red chlorophyll fluorescence under UV (365 nm) light, whereas the wild-type shows a blue-green fluorescence (Chappie et al, 1992). This result indicates that sinapoylmalate may serve as a light filter to absorb potentially damaging UV radiation before it reaches the photosynthetic apparatus. 35 co CJ) o CNJ oo CJ> CJ) CJ) o o o CNJ 0o0 T— CNJ 00 CJ) CJ> „ CJ) COoo COoo CO CO CD CJ) • CO "3- -sj- 00 o •~ 00 00 i— CO O CO CO - - - CD o T3 co co TO 1— u T3 TO c cz 3 CC _l CD _ t— CD CD CD co CO CO 03 co co 0 T3 T3 ea i T3 T3 T3 CD CD CD 0 CD . i a n c: pi e CD pi e hei i he r SZ .c SZ CD CC o o CL CL o o CL o o o o o CD k_ CL CO co CO co co to w co CO sz cz sz •4—' •+—' c: c c CO O o CO Str a CQ O _l CO O CO CO '_J l_l l_l 1 X .to "co CO .c CD CO co .o "5 CD Co c co , 2 to co to € 3 .co § o CD •a Q. I Q. 1 .CO co S J2 "co CO CO CO | to >^ CO to C CO TO to O Ifl It ±: to c S to OC CQ OC CO 0c C/j < OC CQ ^ to CO to CO cz CD 0 T5 to > > 0 T3 CO CO 0 CD _0 JD to CD cz CD to o3 o3 3 CO CD to to 5 (0 to CO (0 CO CO o CO CO CO CO CZ cz c c 2 3 to != to CO to T3 T3 CD T5 T3 CO CD 0 0 > 0 0 0 0 CO 0 0 0 0 CfllO 0 0 CO CO — T3 to to to to E CD 0 •g 'to 0 c: co 0 0 >» 0 c to CL to 'c o O o 0 I O o sz 0 o J3 J2 CL to CO _to CD D) CO o 1 CD to 0 o >. O CTJ >. o o Q « O 0 o Q. i CO jg CQ. O) CL CO 3_ s> E CO _co CZ I 'c >» >. o CO o o o cz o CL O CL CO O CL CL CO i CO 1 CO cz CO CZ 0 cz cz CNJ CO co cz CO CO co" 'a. co co CO c CO 36 The enzymology of sinapine biosynthesis and metabolism has been intensively studied in R. sativus. The proposed reaction sequence based on this work is summarized in Fig. 2 (Chapter I). Four enzymes, SGT, SCT, SE and SMT, are proposed to be directly involved in this process, but SGT is unique in being required at two points. SGT activity was first detected in vitro in extracts from leaf tissue of B. oleracea (Corner and Swain, 1965), and was later characterized more extensively from seedlings of R. sativus (Strack, 1980; Nurmann and Strack, 1981; Mock and Strack, 1993). The highest level of SGT activity in R. sativus was found in the cotyledons of 2-day-old seedlings, although 32% of this maximal activity could also be detected in the dry seed (Strack, 1980). SCT activity was first demonstrated in vitro using R. sativus extracts (Strack et al, 1983), and was later characterized and purified from R. sativus, S. alba and B. napus (Grawe and Strack, 1986; Vogt etal, 1993). The activity of SCT is positively correlated with in vivo accumulation of sinapine during seed development, with the highest activity occurring at the dark-green seed stage in R. sativus (Strack et al, 1983) and B. napus (Vogt et al, 1993). SCT purified from B. napus was shown to be a homo-dimer consisting of two 28 kDa subunits. The general properties of SCT preparations from S. alba, R. sativus and B. napus are very similar (Strack etal, 1983; Grawe and Strack, 1986; Vogt etal, 1993). 37 SE activity has been investigated in seedlings of S. alba (Tzagoloff, 1963a and 1963b; Clausen et al., 1985) and also in R. sativus seedlings (Nurmann and Strack, 1979; Strack et al, 1980). Maximal SE activity was found two to four days after germination, depending on the species. SE partially purified from both S. alba and R. sativus was reported to differ from the general class of plant choline esterases, although the enzymes from these two sources also differed from each other in many respects, including enzyme compartmentation, substrate specificity and pH optimum. SMT is the most thoroughly studied enzyme of all those involved in sinapine pathway. It was first detected in vitro in extracts from cotyledon and leaf tissues of R. sativus (Tkotz and Strack, 1980; Strack, 1982), and later extensively characterized in those tissues (Strack et al, 1985; Sharma and Strack, 1985; Strack and Sharma, 1985; Strack et al, 1986; Grawe et al., 1992), as well as in seedlings of B. napus (Strack et al, 1990). Maximal SMT activity was obtained from seedlings of R. sativus two weeks after germination. Analysis of tissue and subcellular localization showed that SMT was located in the vacuoles of cells from the epidermal layer tissue in R. sativus (Strack et al, 1985; Sharma and Strack, 1985; Strack and Sharma, 1985). SMT activity in R. sativus was found to be light-induced, and also influenced by nitrate level (Strack et al, 1986). Purification of R. sativus SMT to homogeneity showed that this enzyme existed as two isoforms with molecular weights of 51 and 52 kDa (Grawe etal, 1992). 38 Despite the information available concerning the characteristics of individual enzymes involved in sinapine metabolism, a systematic analysis of the relationship between the levels of known sinapoyl metabolites and the activities of the relevant enzymes during the various developmental stages of a brassicaceous plant life cycle, has yet to be conducted. OBJECTIVES The objectives of this study were: (1) to elucidate the relationship between the pattern of accumulation of sinapoyl-conjugates and the dynamics of enzyme activities associated with sinapine metabolism and biosynthesis, throughout the life cycle of a brassicaceous plant; (2) to investigate the possibility that an alternative pathway for sinapine biosynthesis may exist, by comparing the patterns of sinapoyl ester accumulation in B. napus and S. alba; and (3) to establish whether the pattern of SGT expression is consistent with its proposed role in sinapine biosynthesis and metabolism mB. napus. MATERIALS 1. Plant materials Seeds of B. napus cv. Westar and S. alba cv. Ochre were obtained from Agriculture and Agri-Food Canada Saskatoon Research Center, Saskatoon, Saskatchewan, Canada. 2. Chemicals Acetonitrile and glacial acetic acid were HPLC grade and purchased from Fisher Scientific Company. UDP-glucose, dithiothreitol (DTT), CoASH, ATP, Tris, malic acid, 39 choline chloride and PVPP were purchased from the Sigma Chemical Company. PD-10 desalting columns were obtained from Pharmacia. Sinapic acid (3,5-dimethoxy-4- hydroxyl-cinnamic acid, 98%) was purchased from the Aldrich Chemical Company, sinapine was isolated from seeds of S. alba cv. Ochre as described previously (Wang, 1992) and l-0-sinapoyl-(3-Z)-glucose was isolated from 3-day-old radish seedlings according to the method of Vogt et al. (1993). 0-Sinapoyl-Z,-malate isolated from Arabidopsis leaves was a gift from Dr. Clint Chappie, Purdue University, West Lafayette, Indiana, USA. METHODS 1. Plant growth conditions and sampling To produce 1 to 5-day-old seedlings, seeds were germinated on two layers of wetted filter papers in Magenta jars (Sigma Chemical Company) under cool white fluorescent lights (30 W, Sylvania) at 23° C. After germination for 1, 2, 3, 4 or 5 days, the seedlings were harvested, and frozen in liquid nitrogen, then stored at -70° C until analyzed. To obtain tissues of other ages, individual plants were grown in the greenhouse in pots (14 x 20 cm) with a soil mix containing 7% (v/v) peat moss and 10 g 14-14-14 (ratio of N: P: K in the fertilizer) control release fertilizer (Osmocote Sierra Chemical). A SON-T high-pressure sodium lamp (400 W, Philips, Netherlands) was used from 6:00 am to 8:00 pm to provide additional illumination. The average irradiance level at the plant growth surface was > 400 pEinsteins/m /sec. The plants were watered daily until mature. Plant 40 tissues were harvested at different growth stages, separated according to morphology, frozen immediately in liquid nitrogen and stored at -70° C for further use. 2. HPLC analysis of methanol-soluble phenolics 2.1. Sample preparation The sample preparation was the same as described in Method 1.1 (Chapter II), except that 50-100 mg immature seeds or 400-800 mg vegetative tissues was used for the methanol extraction. 2.2. HPLC analysis An acidified gradient solvent system was used for the HPLC separation of phenolic compounds. The initial solvent composition consisted of 80% solvent A (2% glacial acetic acid in distilled water) and 20% solvent B (2% glacial acetic acid in acetonitrile). After injection of 50 ul methanol extract onto a Nucleosil C18 reverse phase column (4.6 x 250 mm, 5 p,, Alltech Chromatography Inc.), the solvent gradient began with a linear increase to 40% solvent B in 10 min and held at 40% B for 5 min, and a further increase to 90% B in 0.1 min . After 1.9 min at 90% B, the gradient decreased to 20% B over 2 min. The flow rate was 0.7 ml/min. Detector wavelength was set at 330 nm with a sensitivity 0.1 AUFS. The separations were carried out at room temperature. 41 2.3. Qualitative and quantitative determination of sinapic acid derivatives The identification of peaks observed in the methanol extracts was based on comparison of their retention times to those of the authentic standards. The external standard method was used for quantitative determination of sinapic acid derivatives. The calculation of concentrations was based on Equation 1 (Method 1.3 in Chapter II) using individual response factors (RF) generated from standard curves for each compound. 3. Enzymatic assays of SE, SGT, SCT, SMT and 4CL 3.1. Sample preparation for protein extraction Plant tissues (about 4 g from different organs at different growth stages) were ground to a fine powder in liquid nitrogen, and then extracted with 6 to 8 ml ice-chilled extraction buffer consisting 4% insoluble PVPP (w/v), 5% glycerol (v/v) and 100 mM K2HP04/KH2P04 (KPi buffer), pH 7. After removing tissue debris by centrifugation, the extracts were desalted on PD-10 columns, and then diluted with appropriate buffers according to the enzymes being assayed. The details of preparation of protein extracts are outlined in Appendix A. Protein concentrations were determined by the Bradford dye- binding assay (Bradford, 1976). 3.2. Determination of SE activity by HPLC The enzymatic reaction was started by adding 20 pi protein extract (or 20 pi water for control) to a 1.5 ml HPLC sample vial containing 60 pi 200 mM Tris/HCl, pH 8.5 and 20 pi 6 mM sinapine bisulphate in water. After incubation at 37° C for 60 to 90 min, the 42 reaction was stopped by addition of 250 ul acidified acetonitrile (3 pi 20% glacial acetic acid in water mixed with 247 pi acetonitrile) to neutralize the reaction mix to pH 6.8. After standing at room temperature for 10 min, 650 pi distilled water was added and mixed well. For HPLC analysis, 50 pi of this reaction mix was injected onto the Nucleosil C18 reverse phase column. The enzymatic reaction product, sinapic acid, was separated by HPLC using the conditions of Method 1.2 (Chapter II). The SE activities were calculated using the integration area (A), response factor (RF) of sinapic acid and the Equations 3, 4 and 5 of Method 2.5 (Chapter II). 3.3. Determination of SGT activity by HPLC SGT activity in extracts from all tissues was determined as described in Method 2.4 and 2.5 (Chapter II). 3.4. Determination of SMT activity by HPLC The enzymatic reaction was started by adding 20 pi protein extract to a 1.5 ml HPLC sample vial containing 45 pi 100 mM KPi buffer, pH 6.3, 25 pi 4 mM sinapoylglucose and 10 pi 1 M Z-malate in the same buffer (pH 6.3, adjusted with 5 M NaOH). Z-malate was replaced by distilled water in the control incubation. The reaction was carried out at 37° C for 3 to 5 hours and stopped by addition of 250 pi acetonitrile. After standing at room temperature for 10 min, 650 pi distilled water was added and mixed well. For HPLC analysis, 100 pi of this reaction solution was fractionated by HPLC using the conditions of Method 2.4 (Chapter II). Based on the area of the sinapoylmalate peak, the 43 SMT activity was calculated using Equation 3, 4 and 5 in Method 2.5 (Chapter II) in which the integration area (A) and response factor (RF) of sinapoylmalate were employed. 3.5. Determination of SCT activity by HPLC The enzymatic reaction was started by addition of 20 ul protein extract to a 1.5 ml HPLC sample vial containing 20 u.1 100 mM KPi buffer, pH 7, 50 ul 2 mM sinapoylglucose and 10 ul 100 mM choline chloride in the same buffer. Choline chloride was replaced by the same volume of 100 mM KPi in the control incubation. The reaction was carried out at 37° C for 2.5 to 4 hours and stopped by addition of 250 u.1 acetonitrile. After standing at room temperature for 10 min, 650 u.1 distilled water was added and mixed well. For HPLC analysis, 50 ul of this reaction solution was injected. The enzymatic reaction product, sinapine, was separated by HPLC using the conditions of Method 1.2 (Chapter II). SCT activity was calculated using Equation 3, 4 and 5 of Method 2.5 (Chapter II) in .which the integration area (A) and response factor (RF) of sinapine were employed. 3.6. Determination of 4CL activity by HPLC The substrate and buffer system for the 4CL reaction were based on the method of Knobloch and Hahlbrock (1975). A substrate mix (2 ml), prepared just before use, contained 55 mg ATP and 3 mg DTT dissolved in 750 u.1 500 mM Tris/HCl, pH 7.8, 200 ul 500 mM MgCl2, 1 ml 10 mM /?-coumaric acid and 50 ul 4 M KOH. The enzymatic reaction was started by adding 20 ul protein extract to a 250 u.1 HPLC sample vial 44 containing 20 ul substrate mix, 10 ul 6.6 mM CoASH and 150 ul 500 mM Tris/HCl, pH 7.8. CoASH was replaced by 10 pi distilled water in the control incubation. The reaction was carried out at room temperature for 0.5 to 3 hours and stopped by addition of 50 pi acetonitrile. After standing at room temperature for 10 min, 50 pi reaction solution was injected onto the HPLC column. The enzymatic reaction product,/?-coumaroyl-CoA, was separated by using the conditions as described in Method 1.2 (Chapter II). The 4CL activity was calculated using Equation 3, 4 and 5 of Method 2.5 (Chapter II) in which the integration area (A) and response factor (RF) of /?-coumaroyl-CoA were employed. RESULTS 1. Profiles of methanol-soluble phenolics in various tissues at different growth stages 1.1. Pattern of phenolic compound accumulation in young seedlings During seed germination, five major UV-absorbing (330 nm) compounds were detected in extracts of B. napus while six were detected in S. alba seedling extracts (Fig. 4). Four of the compounds in B. napus could be tentatively identified as sinapoylglucose (peak 5, retention time [RT] 6.68 min), sinapine (peak 7, RT 7.78 min), sinapoylmalate (peak 12, RT 9.98 min) and sinapic acid (peak 14, RT 10.76 min), based on co-chromatography with authentic standards. The changes in levels of these compounds with seedling age indicate that they are metabolically active. The levels of the fifth compound (peak 24, RT 13. 26 min) detected in J3. napus extracts did not change as markedly as the others. 45 A. O-d-old seedlina G. O-d-old seedling 1 7 B. 1-d-old seedling H. 1-d-old seedling 7 7 C. 2-d-old seedlina I. 2-d-old seedling 5 7 24 4 -AAJ JL D. 3-d-old seedlina J. 3-d-old seedlina o co co l f CD o 24 c 5 CO 24 i o 1 14 w _j\JlAJ A < E. 4-d-old seedling K. 4-d-old seedlina 12 7 24 4 24 J\XJ i F. 5-d-old seedlina L. 5-d-old seedling I 2 4 5 24 1 1 i—i—i—i—i—i—i—'—r i—1—i—1—i— i r 0 5 10 15 20 0 5 10 15 20 B. napus cv. Westar S. alba cv. Ochre Time (min) 46 Peak 24 was also found in S. alba seedling extracts, but its levels in this species changed much more actively during tissue development than it did in B. napus. The increase in levels of other two unidentified compounds (peak 3 and peak 4) in S. alba appeared to be correlated with a decline in sinapine (peak 7) levels. In 5-day-old S. alba seedlings, two more unidentified compounds (peak 15 and peak 21) appeared. These two compounds were also found later in the leaf tissue. The quantities of identified sinapoyl-conjugates accumulating in seedlings of each species are summarized in Fig. 5. As expected, sinapine levels decreased dramatically in young seedlings of both B. napus and S. alba. However, the rate of sinapine degradation in S. alba was slower than that in B. napus. By 5 days, there was no detectable sinapine in B. napus seedlings, whereas in 5-day-old S. alba seedlings, the sinapine concentration still remained at 1100 nmol/g tissue. Sinapoylglucose was detected in dry seeds and young seedlings of B. napus, but it was not detectable in S. alba. The sinapoylglucose levels reached a maximum in B. napus 2 days after germination, and then declined (Fig. 5A). Sinapoylmalate could also be detected in the dry seeds of B. napus, but again, this conjugate could not be detected at all in S. alba extracts. The level of sinapoylmalate increased gradually in B. napus seedlings and it remained as a predominant UV- absorbing compound in the vegetative tissues. Only a trace of free sinapic acid could be detected in young seedlings of B. napus. 47 CD 50000 3 in 40000 4 Z; 30000 4 20000 4 100004 60000 IL S. alba cv. Ochre • SinA 50000 4 • SinG • SinM 40000 • Sin 30000 20000 4 10000 0 Jl 0 1 2 3 4 5 5-day-old Hypocotyl Cotyledon Days after germination ( Day ) Fig. 5. Changes in levels of sinapoyl-conjugates in young seedlings of B. napus cv. Westar and 5. alba cv. Ochre during germination. 48 1.2. Pattern of phenolic compound accumulation in various tissues of adult plants 1.2.1. B. napus cv. Westar The profiles of UV-absorbing compounds detected in HPLC chromatograms of extracts from various tissues of adult plants of B. napus are shown in Fig. 6. Sinapoylmalate (peak 12) could be detected in stem (Fig. 6B), leaf (Fig. 6C), flower axis (Fig. 6D) and silique wall (Fig. 6G&H) tissues. Sinapoylglucose (peak 5) was only detectable in the stems. An early eluting unidentified compound, peak 2, was prominent in almost all the investigated tissues except the roots. The profiles of UV-absorbing compounds in buds (Fig. 6E) and flowers (Fig. 6F) were complex and very similar to each other. More than ten compounds could be detected at 330 nm, three of which (peak 2, 8, and 12) appeared in the stem and leaf extracts as well. However, the UV-absorbing compound patterns of bud and flower differed greatly from those of other tissues. Most notably, sinapoylmalate (peak 12) was not a dominant compound in bud and flower tissues. Although more than 12 UV-absorbing compounds were found in the chromatograms of vegetative tissue extracts of B. napus, only two (peaks 5 and 12) could be identified as sinapoylglucose and sinapoylmalate. The others remained to be unidentified. Tracing the profiles of UV-absorbing compounds in leaves at different stages of development (upper to lower leaf position) in 28-day-old B. napus plants revealed a pattern of early complexity, shifting to a simpler profile (Fig. 7). Nine major UV- absorbing compound were found in juvenile leaves (Fig. 7A and B), including 49 Time (min) Fig. 6. HPLC chromatograms of methanol extracts from various tissues of B. napus cv. Westar. Peak 5 is tentatively identified as sinapoylglucose and peak 12 as sinapoylmalate. The other peaks are unidentified. 50 A. 6/7th leaf mix F. 9/10th leaf mix 2 1 12 23 2 17 18 25 15 B. 5th leaf G. 8th leaf 15 21 12 2 C. 3rd leaf H. 6th leaf 15 21 D. 1st leaf 12 E. Cotyledon J. Cotyledon 12 1 1 I ' I ' 1 ~i—1—r r n— —i— r ~T~ 20 0 5 10 15 20 0 5 10 15 B. napus cv. Westar S. alba cv. Ochre Time ( min) Fig. 7. HPLC chromatograms of methanol extracts from leaves of 28-day-old B. napus cv. Westar and S. alba cv. Ochre plants. Peak 12 is tentatively identified as sinapoylmalate. 51 sinapoylglucose (peak 5), sinapoylmalate (peak 12) and peak 2. Four of the unidentified compounds found in juvenile leaves quickly disappeared and were not detectable in older leaves (Fig. 7C and D). Peak 2 levels also declined, but more slowly. Sinapoylmalate was also a prominent compound in juvenile leaves, but it remained constant and became predominant in fully expanded leaves. The major UV-absorbing compound in 28-day-old cotyledons was sinapoylmalate. Overall, this pattern indicates that sinapoylmalate is the likely end-product of sinapine metabolism in B. napus. The quantitative analysis of identified sinapoyl-conjugates in various tissues of adult plants of B. napus is summarized in Fig. 8. No free sinapic acid or sinapine was found in these tissues, and sinapoylglucose was only found in juvenile tissues, such as the stem apex and the primary leaves. 1.2.2. S. alba cv. Ochre The HPLC profiles of phenolic compounds in various tissues of S. alba plants are shown in Fig. 9. The profiles of UV-absorbing (330 nm) compounds in the vegetative tissues of S. alba are very different from those of B. napus. None of peaks detected in vegetative tissues of S. alba matched the retention times of the available sinapoyl esters, and no further attempt was made at identification. There were two major UV-absorbing compounds (peak 15 and peak 21) in the leaf tissues of S. alba (Fig. 7F-I). Peak 21 had a higher level in juvenile tissues (Fig. 7F and G) while 52 lUB|d-ABp-/6 LU9JS luB|d-Aep-^6 looy 3 Xep-09 HEM anbnis 13 ABp-rjfr ||BM anbins lUB|d-ABp-9g S8ABa-| s • —• jaMoy png -M o ox 9XB J9M0y •4-1 3 m c lUE|d-ABp-82 .w ta iBai mm *~ •S jUB|d-ABp-82 saAE9| paxjiftj aj •w luB|d-Aep-82 CS OJD JB3| js u 3 liiB|d-AEp-82 O uopa|X}oo u }UB|d-ABp-8S wajs o §• )UB|d-ABp-82 jooy 0 )UB|d-ABp-g i saAEan }UB|d-ABp-g i AS a uopa|Ajoo }UB|d-ABp-g i, tuajs '5 • c u 1 § (enssii useji 6 / |owu ) uojiBJiueouoo ^ I cc s oi°5 IX o 53 A. 28-day-Root E. Bud 20 I22 28 27 A 1 11 B. 28-day-Stem 21 15 C. 28-day-Leaves G. 5-day-Silique wall 15 21 21 15 D. Flower axe H. 40-dav-siliaue wall 21 J2 I—'—I—1—I—1—I—'—I—1 I 0 4 8 12 16 20 0 4 8 12 16 20 Time (min) Fig. 9. HPLC chromatograms of methanol extracts from various tissues of S. alba cv. Ochre. 54 peak 15 had a higher level in older tissues (Fig. 7H and I). The pattern of change in these two compounds suggests that they could be metabolically related. In contrast to B. napus, the fate of sinapine metabolized in the cotyledons of S. alba was uncertain since no dominant phenolic metabolites could be recognized in 28-day-old cotyledons (Fig. 7J). 1.3. Profiles of phenolic compounds accumulated in developing seeds The patterns of UV-absorbing compounds detected in extracts prepared at different stages of seed development are summarized in Fig. 10. In B. napus, sinapic acid (peak 14) was detected as a dominant compound five days after flowering in the ovule tissues (Fig. 10A). Sinapoylglucose (peak 5) became prominent in developing seed ten days after flowering (Fig. 10B). The level of sinapic acid was continually increasing at this stage. Twenty days after flowering, sinapine (peak 7) was found as the major sinapoyl- conjugate while the sinapoylglucose appeared as a minor compound. Sinapic acid remained as a major component at this stage, but gradually dropped in concentration to become a minor compound thirty days after flowering (Fig. 10D and E). In S. alba, four major UV-absorbing compounds could be detected in the ovule tissues five days after flowering (Fig. 10F). Although peak 2 and peak 21 appear to be the compounds in bud, flower and leaf tissue extracts as well, peak 13 (RT 10.15 min) and peak 26 (RT 14.19 min) are apparently unique to young developing seeds. These four compounds were greatly diminished twenty days after flowering, by which time sinapoylglucose (peak 5), sinapic acid (peak 14) and sinapine (peak 7) could be detected 55 A. 5 DAF F. 5 DAF B. 10 DAF 14 C. 20 DAF H. 20 DAF D. 30 DAF I. 30 DAF 1 7 14 E. 50 DAF J. 50 DAF 7 i—1—i—1—i—1—r -r i—1—i—1—i—' r 20 0 5 10 15 20 0 5 10 15 B. napus cv. Westar S. alba cv. Ochre Time ( min) Fig. 10. HPLC chromatograms of methanol extracts from developing seeds of B. napus cv. Westar and S. alba cv. Ochre. Developing seeds were harvested based on the days after flowering (DAF). Peak 5 is tentatively identified as sinapoylglucose, peak 7 as sinapine, and peak 14 as sinapic acid. The other peaks were not identified. 56 (Fig. 10H). However, both sinapoylglucose and sinapic acid occurred, and remained at very low levels, whereas sinapine was the dominant compound accumulated after this stage (Fig. 101 and J). The patterns of sinapoyl-conjugate accumulation during seed development are summarized in Fig. 11. The patterns found in the developing seeds of B. napus and S. alba were very similar, although several unidentified compounds (i.e. peak 2, 13, 21 and 26) accompanied the known sinapoyl-conjugates in S. alba seeds at the early developmental stages. Whether these unidentified compounds are involved in sinapine biosynthesis in S. alba is unknown. 2. Developmental pattern of expression of SE, SGT, SMT, SCT and 4CL 2.1. SE, SGT, SMT, SCT and 4CL activities in young seedlings The changes in specific activities of SE, SGT, SMT, SCT and 4CL during growth of young seedlings of both species are presented in Fig. 12. In dry seeds of B. napus, all these enzyme activities were detected except SMT (Fig. 12A), while in dry seeds of S. alba, only SE activity was detectable (Fig. 12B). During the seedling development, SE in r both B. napus and S. alba reached its peak three days after germination. SGT reached its peak two days after germination in B. napus seedlings, but it could not be detected in any of investigated seedling stages in S. alba. In B. napus, SCT activity was only detectable during the first 24 hours post-germination, while it was not detectable at all in any seedling stages of S. alba. SMT activity was only detectable three days after germination 57 50000 A. B. napus cv. Westar 45000 40000 • SinA I 35000 • SinG • SinM 30000 • Sin 25000 20000 15000 10000- 5000- 0 Days after flowering ( Day) Fig. 11. Changes in levels of sinapoyl-conjugates in developing seeds of B. napus cv. Westar and S. alba cv. Ochre. 58 0 1 2 3 4 5 Days post-germination Fig. 12. Comparison of SE, SGT, SCT, SMT and 4CL specific activities in young seedlings of B. napus cv. Westar and S. alba cv. Ochre. 59 in B. napus seedlings. The specific activity of this enzyme (about 1 pkat/mg protein) was about 10-fold lower than those of SE, SGT, SCT and 4CL in young seedlings. No SMT activity was ever detected in S. alba seedlings. High activity (30-120 pkat/mg protein) of 4CL was observed in B. napus seedlings, but it was about 10-fold lower in S. alba. The patterns of enzyme activities observed in B. napus and S. alba seedlings are consistent with the profiles of the sinapoyl-conjugates found in both species. It indicates that the linkage between sinapine metabolism and sinapoylmalate biosynthesis in B. napus is similar to that reported for R. sativus. However, sinapine metabolism in & alba is strikingly different from that in B. napus. 2.2. SE, SGT, SMT, SCT and 4CL activities in vegetative tissues of adult plant SE activity was only detectable in bud and flower tissues of the adult plants of B. napus (Fig. 13), while in S. alba, it was also found in the leaf tissues (Fig. 14). SGT activity was found in all juvenile tissues of B. napus (Fig. 13), but only traces were detected in primary leaf tissue of S. alba (Fig. 14). Neither B. napus nor S. alba vegetative tissues yielded any detectable SCT activity. SMT activity was found in all vegetative tissues of B. napus, but was not detected in S. alba. In both B. napus and S. alba, 4CL was not detectable in mature leaf, flower axis, bud and flower tissues, but it was very active in the extracts from root and stem. 60 p|o-Aep-06 jooy J9M0IJ > U M s png I e =Q o 9XB J9M0IJ 3 ga 3 p|O-Aep-09 S9AB8-| paxi|/M *C cu p|0-Aep-8Z I |> 'S o p|o-/tep-8Z CS U S9AB9"] P9XI|AJ -a c CS pio-ABp-82 1B9"1 IS I. p|0-ABp-gi. ^ a. LU91S hi LU O O 5 O PIO-ABP-91. co co co co "3" • • • • w 5 looy a 1 o o o o o LD o LO o LO CM • T .SP s ( uiaiojd 6iu / iB>)d ) AIJAJIOB ojipgds fa TJ 61 p|O-Aep-06 jooy u jsMoy u O u> pna Q | 53 <>3 9XB J8M0U 3 pio-Aep-09 3 O XjUJ jB8-| CO "C (0 pio-Aep-82 6/8 lean p|0-Aep-82 u xjoi jee-| u i p|o-Aep-8c3 H «J c H g pio-Aep-gi. o S. UJ91S in jf pio-Aep-gi. looa O « owa s& Qm *- s ( ujeiojd 6UJ / jexd ) AijAipe ojjpeds 62 In this context, adult plants of S. alba essentially differ from those of B. napus in lacking SMT and SGT activities, an absence that is correlated with the failure of S. alba to accumulate sinapoylmalate in its vegetative tissues. The identities of the predominant UV-absorbing compounds found in the vegetative tissues of S. alba are unknown, but it would appear that their synthesis does not involve SGT or SMT activities. 2.3. SE, SGT, SMT, SCT and 4CL activities in fruit The tissues of the fruit were analyzed as two separate parts, the silique wall and seed. Neither SE nor SCT activities were detected in extracts of developing silique wall tissues of either B. napus (Fig. 15A) or S. alba (Fig. 15B). In .6. napus, SGT activity in silique walls was only detectable at an early stage of fruit development (e.g. 5-day-old silique), whereas this enzyme activity was not detected in any silique wall tissues of S. alba. SMT activity was found in developing silique wall tissues of B. napus, but not in S. alba. Only at the stage ten days after flowering in S. alba, or twenty days after flowering in B. napus, 4CL activity appear in the silique wall tissues, but it then could not be detected in the tissues at forty days after flowering in either species. In seed tissues of B. napus, SE activity was detectable even at early seed developmental stages (Fig. 16A), but it was not found until later seed developmental stages in S. alba (Fig. 16B). SGT was detectable at all seed developmental stages in B. napus, but was only found twenty days after flowering in S. alba. SCT activity appeared in the seed tissues twenty days after flowering in both B. napus and S. alba. The SCT levels 63 20 A. B. napus cv. Westar £ 15 10 li 20 B. S. alba cv. Ochre • SE • SGT o 15 - ] • SCT • SMT B4CL 10 1 1 i i l 10 20 30 40 50 60 Days after flowering ( Day) Fig. 15. Comparison of SE, SGT, SCT, SMT and 4CL activities in silique walls of B. napus cv. Westar and S. alba cv. Ochre during seed development. 64 10 20 30 40 50 60 Days after flowering Fig. 16. Comparison of SE, SGT, SCT, SMT and 4CL activities in developing seeds of B. napus cv. Westar and S. alba cv. Ochre. 65 remained high for more than twenty days in B. napus during seed development, but in S. alba it was only moderately active for about two weeks before quickly declining as the seeds matured. Developing seed tissues of both B. napus and S. alba lacked any SMT activity. Only traces of 4CL activity were detectable in the extracts of B. napus seeds at the early developmental stage, and it was not found in S. alba at all. The patterns of these enzyme activities suggest that the sinapine accumulated in the seeds of both B. napus and S. alba is synthesized de novo in the seed tissue and not transported from other tissues, since the full array of enzymes involved in sinapine biosynthesis is found only in developing seeds. 3. Developmental expression of SGT in B. napus cv. Westar SGT activity reached its maximum in young seedlings of B. napus about two days after germination. Further investigation showed that the majority (79%) of this activity was located in cotyledon tissues (Fig. 17). No SGT activity was detectable in seed coat. The distribution of SGT activity in all available leaves of 50-day-old adult plants is shown in Fig. 18. SGT was very active in the juvenile tissues, and maintained a lower but relatively constant level in fully expanded leaves. It is interesting to note that petiole tissue did not show any detectable SGT activity. Both reproductive tissues (bud and flower) showed modest SGT activity. Further analysis of SGT distribution in different parts of the flower indicated that only the stamen and pedicel tissues possessed detectable SGT activity (Fig. 19). However, the SGT activity in these tissues was very low (2-6 pkat /g tissue or 0.3- 66 Whole seedling Seed coat Hypocotyl Cotyledon Type of tissue Fig. 17. Distribution of SGT in 68-h-old seedlings of B. napus cv. Westar. Numbers above the base represent the percentage of total SGT activity in seedlings from which the individual tissues were derived. 67 Fig. 18. Distribution of SGT in leaves from different positions on 50-day-old plants of B. napus cv. Westar. SGT activity was determined for all individual leaves at this growth stage. The order of leaf position was designated according to the order of their appearance during plant development, with 12-14th representing the youngest leaves. 68 Whole Flower Petal Ovary Stamen Sepal Pedicel Stigma Style Type of tissue Fig. 19. Distribution of SGT in different tissues of flowers of B. napus cv. Westar. 69 0.6 pkat /mg protein). So far, the possibility that it may represent a non-specific reaction, which is catalyzed by glucosyltransferases other than SGT in these tissues, can not be excluded. During seed development, SGT activity in the developing seeds showed a dynamic pattern of change (Fig. 20). The total SGT activity increased as the seed matured, whereas the SGT specific activity increased in the first four weeks after flowering, and then gradually declined, probably because of the massive increase in seed storage proteins. The developmental pattern of expression of SGT activity throughout the life cycle of B. napus plants is presented in Fig. 21. It is clear that the juvenile tissues, such as young seedlings or primary leaves, provide a good source of SGT for enzyme purification. DISCUSSION This is the first report to compare the dynamics of sinapoyl-conjugate accumulation and associated enzymology throughout the life cycle of brassicaceous plants. Earlier studies of sinapine biosynthesis and metabolism have either focused on a specific growth stage, or a particular enzyme, or a specific phenolic metabolite (Tzagoloff, 1963a; Strack, 1977; Bopp and Ludicke, 1980; Strack etal, 1985; Bouchereau etal, 1992). The results of this comparison show how diverse the metabolism of phenolic conjugate biosynthesis can be, even between two relatively closely-related species. At least 28 compounds having UV absorption at 330 nm can be detected in various tissues of either B. napus or S. alba during the plant development. Only four of them, i.e. sinapic acid, sinapine, 70 Days after flowering Fig. 20. Changes in SGT activity and protein content in developing seeds of B. napus cv. Westar. 71 Fig. 21. Developmental expression of SGT throughout the life cycle of a B. napus cv. Westar plant. SGT specific activity was calculated based on the protein content in assayed tissue at each specific growth stage. 72 sinapoylglucose and sinapoylmalate, have been quantitatively analyzed because of the availability of standards and structural information. Some of the unidentified compounds may also be sinapoyl-conjugates or other related hydroxycinnamoyl-conjugates, since an array of these has been reported from the Brassicaceae (Linscheid et al, 1980; Clausen et al, 1982; Larsen et al, 1983; Clausen et al, 1983; Strack et al, 1984; Bouchereau et al, 1991). In the absence of authentic standards, however, the observed changes in their levels are interesting, but uninterpretable at this point. The sinapoyl-conjugate pattern at the early germination stage in B. napus is similar to that reported in R. sativus (Linscheid et al, 1980; Strack, 1982) except that the rate of sinapoylmalate accumulation appears to rise more quickly in B. napus. This result is in agreement with earlier studies (Bouchereau et al, 1992; Vogt et al, 1993). Whether sinapoylmalate simply accumulates in brassicaceous tissues, or is further metabolized, had not been established in previous studies. Since the entire life cycle of the plant has been examined in this study, it is possible to conclude that sinapoylmalate is likely to represent the end-product of sinapine metabolism in B. napus since it is the predominant compound accumulated in the mature leaves and cotyledons until they die. Sinapoylmalate may serve as a reserve of both sinapate and malate in B. napus, in support of lignification and organic acid metabolism. Examination of the tissue distribution of sinapoylmalate in R. sativus cotyledons found that it was localized in the epidermal layer of this tissue (Strack et al, 1985). This suggests that sinapoyl-conjugates could also serve as a light filter to protect the plant from UV damage. The present study would support 73 this putative physiological function, since sinapoylmalate is the predominant UV- absorbing compound in the leaf tissues over the entire life cycle of B. napus plants. Both SGT and SMT activities were found in the tissues accumulating sinapoylmalate, especially in the juvenile tissues, a pattern consistent with the proposed pathway of sinapoylmalate biosynthesis iaR. sativus (Tkotz and Strack, 1980) and B. napus (Strack et al, 1990). It is obvious, however, that sinapine metabolism in S. alba differs from that in B. napus. There are no other reports concerning the fate of sinapine metabolized during seed germination in S. alba. The unidentified compounds, peak 3, 4 and 24 (Fig. 4J-L), detected in young seedlings of S. alba could be the initial sinapine metabolites, since their levels increased in concert with the rapid degradation of sinapine in young seedlings of S. alba. They appear to be further metabolized completely during the aging of the cotyledons (Fig. 7J). Unlike in B. napus, the major UV-absorbing compounds in the vegetative tissues of S. alba are peak 15 and 21 (Fig. 7F-I and Fig. 9). Whether these compounds are sinapoyl-conjugates requires further investigation. Since they are so prominent in S. alba tissues, they may play a role in S. alba similar to that played by sinapoylmalate in B. napus. How does general phenylpropanoid metabolism affect the fluctuation of sinapoyl- conjugates during seed germination? A study using the PAL-specific inhibitor, a- aminooxy-(3-phenylpropionic acid, during the germination of R. sativus showed that inhibition of PAL did not affect the formation of sinapoylglucose, although the 74 biosynthesis of anthocyanins, flavonols and a ferulic acid derivative were severely depressed (Strack et al, 1978). This result indicates that esterified sinapic acid present in the tissues during seedling development is not synthesized de novo through the phenylpropanoid pathway. A quantitative analysis of sinapoyl-conjugate levels revealed that the total pool of sinapoyl-conjugates remained constant, on a molar basis, throughout the early seedling stage (up to seven days after germination) in B. napus (Bouchereau et al, 1992), suggesting again that de novo synthesis of sinapic acid through phenylpropanoid metabolism does not significantly contribute to the fluctuation in levels of the different sinapoyl-esters immediately following seed germination. Those fluctuations are mainly the result of interconversions between existing sinapoyl- conjugates. Following the discovery that sinapine is rapidly degraded during brassicaceous seed germination, studies of SE have focused primarily on the tissues from the early seedling stage (Tzagoloff, 1963a and 1963b; Nurmann and Strack, 1979; Strack et al, 1980; Hadacova et al, 1981; Clausen et al, 1985). Hydrolysis of sinapine by SE yields both choline and sinapate, both of which may be important to the young seedlings. In relation to the regulation of plant growth, an analysis of the biological activity of sinapic acid derivatives showed that a combination of sinapine and IAA applied exogenously to wheat coleoptile pieces did not stimulate the tissue growth, whereas sinapic acid or sinapoylglucose plus IAA could stimulate the growth by 30% over the control containing only IAA (Kefeli et al, 1977). This stimulatory effect was, however, only observed at 75 specific concentrations of sinapic acid derivatives. Inhibition of the growth was observed when the concentrations of sinapic acid derivatives were over certain levels (e.g. 1.5 mM, for sinapic acid). If this response also occurs in vivo in the Brassicaceae, the rapid degradation of sinapine in germinating seeds may be relevant in a scenario such as this: SE expression may be induced at the beginning of germination by an increase in one or more phytohormones, such as auxin, gibberellin or cytokinin. The free sinapic acid, resulting from the hydrolysis of sinapine, could enhance auxin activity in the young seedling tissue. This model may also provide an explanation for the higher concentrations of sinapoyl-conjugates, and the higher activity of SGT in juvenile tissues, such as primary leaf and stem apex, since a detoxification process driven by SGT and SMT activities could re-esterify the free sinapic acid and thus reduce the stimulatory effect or avoid inhibition. Substantial SE activity was also observed in the bud and flower tissues of both species, even though no sinapine was detected. The nature and the physiological function of this esterase activity is not known at this stage. SE activity was also found in the developing seeds, especially a fairly high SE activity in the seeds at later developmental stage of S. alba. Since sinapine is accumulating at the same time, it must be assumed that either sinapine turns over rapidly during the accumulation phase, or that SE and sinapine are located in separate compartments at the subcellular or tissue level. Although sinapine is being rapidly degraded during the early germination stage in B. napus, SCT activity is still detectable at a level similar to that found in the mature seeds (Fig 16A). This observation is in agreement with the report in earlier study (Bouchereau 76 et al, 1992). The SCT activity found at the early germination stage may represent residual enzyme produced during seed development. In contrast to S. alba, no SCT activity was detectable in either the seedlings, or the mature seeds. In a much earlier study for S. alba, Tzagoloff reported the increased de novo synthesis of sinapine in vegetative tissues two weeks after germination (Tzagoloff, 1963a), but this was not substantiated in the present study. This conflict may be due to the use of lower resolution analysis techniques in the previous study that were unable to distinguish sinapine from other sinapoyl-conjugates or phenolics, or it could reflect differences between S. alba genotypes. Sinapine has been consistently shown to accumulate in the developing seeds between ten to twenty days after flowering in B. napus and S. alba (Fig 15, this study; Kuhnl and Wellmann, 1979; Bopp and Ludicke, 1980; Vogt et al, 1993), as well as in R. sativus (Strack et al, 1983). Since sinapine is a major product of phenylpropanoid metabolism in brassicaceous seed tissue, it could be predicted that PAL activity would be correlated with sinapine accumulation. Such a pattern has been reported (Kuhnl and Wellmann, 1979), and [14C]-phenylalanine has been shown to be incorporated into sinapine during seed development in S. alba (Bopp and Ludicke, 1980). The sinapoyl metabolite and enzyme activity profiles observed in the present study are entirely consistent with the pathway of sinapine biosynthesis established earlier in R. sativus (Strack et al, 1983). Both sinapic acid and sinapoylglucose transiently accumulate before sinapine appears at 77 an early stage of seed development in B. napus (Fig. 10A and B), and the enzymes required for sinapine biosynthesis from sinapic acid are also present at this stage. These results do not, however, preclude the possibility that an alternative pathway may also lead to the formation of sinapine in some instances. This is an important point in the context of manipulation of sinapine biosynthesis through plant genetic engineering. Kuhnl and Wellmann, in their earlier investigation of sinapine biosynthesis in developing seeds of S. alba, proposed that /7-coumaroyl-shikimate might be a key precursor for the formation of sinapine (Kuhnl and Wellmann, 1979). An enzyme activity capable of catalyzing the formation of /j-coumaroyl-shikimate from /7-coumaroyl-CoA and shikimic acid was detected in the developing seeds of S. alba, and they therefore proposed thatp- coumaroyl-shikimate might be converted to a /.-coumaroyl choline ester by an acyl transferase. Further 3 and 5-hydroxylation and methylation would then take place to form sinapine. In the present study, sinapic acid and sinapoylglucose did not appear in the developing seeds of S. alba until twenty days after flowering; instead, four other unidentified aromatic compounds (peaks 2, 13, 21 and 26) were found during the early stage of seed development (Fig. 10F). The size of peaks 2 and 13 remained constant, while the size of peak 21 decreased with advancing seed development. The most interesting compound is perhaps peak 26 whose level in the seeds increased up to at least ten days after flowering, and then declined. The chemical structure of this compound is unknown, so it is not possible to conclude whether it might serve as a precursor for the sinapine biosynthesis. If /.-coumaroyl-shikimate is involved in sinapine biosynthesis in S. 78 alba as proposed by Kuhnl and Wellmann, 4CL activity would presumably be required as part of this enzymatic reaction sequence, and should be detectable in developing seeds. However, 4CL activity was not detected at all in this tissue in the present study, whereas both SGT and SCT, the key enzymes for sinapine biosynthesis, were detected at the stage during which sinapine rapidly accumulated. Sinapic acid and sinapoylglucose also appeared in the extracts from the S. alba seeds of this stage, at a low but constant level. This pattern is consistent with the previous report of Bopp and Ludicke (Bopp and Ludicke, 1980). The apparent lack of marked accumulation of sinapic acid and sinapoylglucose in this species may simply reflect a higher turnover rate for these metabolites in this particular species. Based on these results, it seems likely that the mechanism for sinapine synthesis via sinapoylglucose established in B. napus and R. sativus also operates in S. alba. SGT has been a particular focus in this study because of the central role that it plays both in sinapine biosynthesis in the seed and in sinapoylmalate biosynthesis in the vegetative tissues. The products of the glucosyltransferase-catalyzed reaction, 1-O-acyl-glucose esters of phenolic acids, are a class of secondary metabolites found in many species of plants (Harborne and Corner, 1961). The transient accumulation of these hydroxycinnamoyl-glucose esters cannot be considered merely as products of detoxification reactions (Kojima and Uritani, 1973; Schlepphorst and Barz, 1979; Berlin and Witte, 1982; Strack et al, 1984; Barz et al, 1985). On the contrary, these 1-0-acyl- glucose esters are essential metabolically-active intermediates in plant secondary 79 metabolism (Kojima and Uritani, 1972; Strack, 1977; Koester et al, 1978; Strack et al, 1984). The enzymatic mechanism underlying the biosynthesis of 1-O-glucose esters by UDP-glucose-dependent glucosyltransferases has been established as a frequently used alternative to ester formation via CoA thioesters (Molderez et al, 1978; Michalczuk and Bandurski, 1980; Tkotz and Strack, 1980; Gross, 1983; Shimizu and Kojima, 1984; Villegas and Kojima, 1986; Bokern and Strack, 1988). In the present study, SGT activity was found to change in parallel with the changes in sinapoylglucose accumulation in B. napus as well as in S. alba. SGT was particularly active at the young seedling development stage as sinapine was hydrolyzed, and in juvenile tissues where sinapoylmalate was formed during vegetative growth. It was also active in developing seeds during sinapine synthesis in B. napus. Since S. alba does not accumulate sinapoylmalate, SGT activity could not be detected in any vegetative tissues in this species, but nevertheless, it appeared in the developing seeds where sinapoylglucose was required for the biosynthesis of sinapine in S. alba. 80 Chapter IV ENZYMOLOGY OF UDP-GLUCOSE:SINAPIC ACID GLUCOSYLTRANSFERASE FROM B. NAPUS 81 INTRODUCTION Among the enzymes required in sinapine metabolism and biosynthesis, SGT catalyses the formation of l-0-sinapoyl-(3-D-glucose from sinapic acid and UDP-glucose (Strack, 1980). Since sinapoylglucose is an essential substrate for both sinapine (Strack et al, 1983) and sinapoylmalate (Tkotz and Strack, 1980) biosynthesis, SGT plays an important role in linking the two patterns of metabolism of sinapic acid derivatives in Brassicaceae (Fig. 2). As a participant in two different pathways that operate at different stages of plant development, and in different tissues, the regulation of SGT activity is also of considerable interest. Although sinapine has been known as a remarkable chemotaxonomic character in many members of the Brassicaceae family, especially in Brassica and its closely-related species (Hegnauer, 1964; Voskerusa and Kolovrat, 1989; Bouchereau et al, 1991), SGT activity has only been demonstrated in leaf tissue of B. oleracea (Corner and Swain, 1965), and in seeds and young seedling tissues of R. sativus (Strack, 1980). Outside the Brassicaceae family, SGT activity has also been detected in extracts of in vitro cultured carrot cells (Halaweish and Dougall, 1990). This SGT activity led to the formation of 1-0-sinapoylglucose, which in this case is believed to serve as an intermediate in the synthesis of sinapoyl anthocyanin accumulated in the carrot cells. SGT has previously been partially purified and characterized from seedlings of R. sativus (Strack, 1980; Nurmann and Strack, 1981; Mock and Strack, 1993). As a class, glucosyltransferases such as SGT are not abundant proteins in plant tissues, and they are thus difficult to purify to homogeneity. A small number of UDP-glucose-dependent 82 glucosyltransferase proteins have been purified to varying degrees as summarized in Table 4. Although the structures of the aglycones, i.e. the sugar acceptors, differ dramatically, as might be expected given the diversity of plant secondary metabolites, the sugar donors are quite similar. All reported sugar donors are nucleotide-activated sugars, rather than glucose itself or low-energy derivatives, such as glucose-l-phosphate. In most cases, UDP-glucose is the only sugar donor that can be used, although thymidine-5'- diphospho-glucose (TDPG) is also a substrate in some systems, including mung bean (Barber, 1962) and radish (Strack, 1980). Adenosine-5'-diphospho-glucose (ADPG) has been reported to be preferred over UDP-glucose only in the wheat germ system, while cytidine-5'-diphospho-glucose (CDPG) and guanosine-5'-diphospho-glucose (GDPG) are seldom used (Hosel, 1981). The glucosyltransferases characterized to date also share some other general properties: (1) they are monomeric proteins with molecular weights about 40 ~ 55 kDa; (2) they often display a broad aglycone specificity; (3) they require a reducing reagent e.g. DTT or 2-ME, to maintain activity; (4) they require no other co-factors for activity; and (5) they are very sensitive to Cu++ and Zn++. OBJECTIVE The objective of this study was to purify and characterize SGT from seedlings of B. napus as a necessary step toward cloning the gene encoding this enzyme. 83 CM 00 CO 00 OO CD o cn LO 00 co CO oo cn o C\J cn oo oo oo X2 oo o cx> CD 00 CD ca N 05 co W 00 i- ' CD i_ E u -t—' cn . FT TO CD c o CO a> TO co CO oo - TO TO T3 oo CD oS c CD o3 m cn "(0 0) c: co c: CD -t—' 0) 1 cz oo" CD CO CD DC CD CD OT E N ' i CO CD OT w 3 co 'Ma N CD E o o o L_ CO CQ £ 0 O CO z •c "53 CL CO• CO CD CD w OT CD CD OT 3 CD "ro ' CO cz g w OT 3 co 1Q CD ^—' o O CO OT > E CD o CD E CO o E CD -4—o ' CD CO CD o "CD 0 CO CC CO O DC O O 3 c c to .2 o CO cCD (O 6 CO o CD I o .I3S 6 CO TO 3 u 5 C a CD 3 CO .C TO CO 8 -Q CO TO O . OT 0 0 -g -g 'w to o o 0 0 o to jg 3 0 o -g co 0 o to i_i to to jg 1 o O u 0 0 o o _D> 3 o >•. Z3 CO T> S to J2 ro O CO 0 _g> 0 O c: CZ co cz >^ o o 0 E 8 © o o '•g > 0 S c CL ro _3 o 'c co o co 'c > g o ro cz H— 0 ro O =5 >. < -a O o 1 CO OT Q O cn. CO O X o3 84 oo oo O CO CO CX> Is- CO CO s oo 00 I - CO 5) co co CO c\J CO CD T- o cn cn CO CO co cx> CO CO CO .. k_ cn 0> TO CO CO cn CZ cz cz •4—' o 03 CO . ^ CO c CO TO co ^i: —C.O CQ N o3 co 0> CD 08 TO CD ' co = o3 CO CO cz Q. co CD CD _cz CD cz CD CD o3 0) X . o cz CD CD CO cz CO 'E o o O O k_ CO CD o3 CO TO x: DC E ZJ JZ CD CL JD cz o j*. CD D) CD 'cz *CO c co JZ o 2£o CD CO 0 CD 'CD SI N CQ Q CO X > DC CQ CD >- O CO co co CO CD CO CO CO O) 3 CO cz CZ 0 W 3 jz Z5 cz cz 0 JD 0 o = = ^ CO C«O = = OT CD CZ TO TO TO TO TO o co 0 0 CO k CD 0 0 •+—• o 0 CD CD "CD 0 0 o 0 0 CO O CO CO DC QL CO CO CO O 5 1 CO CO co CO -Q -Q 3 0 CO .2 Si o CO JD ca o 2 .CO 3 c OT .CO 3 co co o co 0 Q. 0 3 I co .2 ca c "5 CO "5 CO O .Q> c 0 co 3 co CO 0 CO o -5 JO CO o a c 2, O co CO ,<0 o .8 .o JD CO CO CL & 0 55 co 0 CO CO -c co c CO co c co CJ E co 0 CO o .Q. .c 3 CD co .cj c CO 5 0 C,1Q JD CO G N 6 CQ CO 0 0 0 0 CO 0 •g JO CO o 0 co •g 'co o 0 •g sz CO 0 CO o CO 0 o cz •g 'co CL •g 0 0 JO CO 0 •g 0 'co •g o 'co o •g CO CO o o CJ 'co JO o 'co o sz jo o CO ~a o 3 o o o O -2 Q. co o co _3 JO o o 3 _3 CO JD _i_ 1 JO JO) 0 jg C CO jo 0 CO 6 CO +-> 0 0 O CO N co CO 1 1 c o o CZ 0 CZ 0 C\J 0 o o CO 0 co s o 3 c o o ' I - JZ\ cz lu c d e cz "3 •g T3 o 2 CO o 'N CO '3 * 2'cz CZ CO O CO 'o s •i: CD — o cr 0 10 - JO >* CO 'x X o co co CL 0 cz CD o o o0 CO CD O o o ZJ E CJ ZJ N zs CoO cz < o 0 > CO > L— •a Z3 CZ CO cz *1_ o CO "CO O 0 CO 0 cz CO CO 0 LL o3 X CD CO < -z. a CQ 85 CD CD CJ> "3- CO CO CD CD CO CD CO N CO CO LO CO TJ CD CO cu i _T o CO CO o c o3 ._ o c CO T5 co cu > CD CO o o CO LU co CD cu CO O CL - CL CL o3 o (5 03 -t—o• CD o l_ CD 3 i_ CO 3 CO ^ TJ co O) CU H—' o CO CD _c _g 3 CD 1 O CO T3 CD -I— TJ CD 0 TJ CO CD (J 05 CD ^ CD CD CD "CD CD CO 08 q CO CO o UJ c 0 •5 CO co s c c CO .CO co CD 1 "co CJ I -2 CD TO CD CO CO CO s CU CO CD *o 0 CO co I cu CD CO CD o Q. co 8 t .8 CD CO I c E -5 CO o CO 1 s I 5 o c -S QQ o CO CD 0). 0 _C0 o o CD g c g CO •g o 'co 'co o o o CO CD o o o T3 CD TJ _3 _3 o 'co _3 86 MATERIALS 1. Plant materials B. napus cv. Westar was obtained from Agriculture and Agri-Food Canada, Saskatoon Research Centre. 2. Chemicals UDP-[(3-£>-glucose (U- C)] was obtained from New England Nuclear. The protease inhibitor kit was from Boehringer Mannheim. Macro-Prep ceramic hydroxyapatite matrix was from Bio-Rad Laboratories. Hydrazide Affinity-Filter was purchased from Affinity Technology. Immobilon-PSQ polyvinylidene fluoride (PVDF) membrane was purchased from Millipore. Fast protein liquid chromatography system (FPLC), EAH-Sepharose 4B beads, Superdex 75 matrix, Polybuffer 74, Pre-packed PD-10, HiTrip Blue affinity, Prep- Mono Q HR 10/10, Mono P HR 5/20 and Superose 12 HR 10/30 columns were from Pharmacia Biotechnology Canada. Microsep centrifugal concentrators (10 kDa-cut-off) were purchased from Filtron Technology Corp. Flat-bottom 96-well plates were obtained from Becton, Dickinson & Company. Polyclonal anti-pea HSP70 serum was a gift from Dr. E. Vierling, Department of Molecular and Cellular Biology, University of Arizona, Tucson, AZ, USA. Anti-HSP70 monoclonal antibody SPA-810 was purchased from StressGen Biotechnologies Corp., Victoria, BC, Canada. Anti-thiohydroximate glucosyltransferase poly- and monoclonal antibodies were a gift from Drs. J.W.D. GrootWassink and D.W. Reed, Plant Biotechnology Institute, NRC, Saskatoon, SK, Canada. All other chemicals were from Sigma unless otherwise stated. 87 METHODS 1. Determination of SGT activity 1.1. Multi-well plate method A simple semi-quantitative assay method using a multi-well plate reader was designed for quick screening of SGT activity in the fractions obtained from column chromatography procedures. Each sample well of a flat-bottom 96-well plate was loaded with 25 - 50 pi enzyme solution (replaced with water for control) and 25 pi substrate mixture (1.5 mM sinapic acid, 1.5 mM UDP-glucose, 30 mM 2-ME, 30% glycerol and 240 mM MES, pH 6.0). After incubation for 30 - 60 min at 37° C, the reaction was stopped by addition of 200 pi CAPS (3-[cyclohexylamino]-l-propanesulfonic acid) buffer, pH 11.0. These basic conditions result in a differential bathochromic shift in the absorption spectra of sinapic acid and sinapoylglucose, which allows measurement of the concentration of sinapoylglucose (yellow color developed) at 405 nm in a multi-well plate reader. For quantitative analysis, a standard curve was generated with the OD405 nm and the corresponding concentration of sinapoylglucose standard (i.e. 0, 0.25, 0.50 and 1.25 pmol/well). 1.2. HPLC method The HPLC method described in Method 2.4 and 2.5 in Chapter II was used as a precise quantitative assay for SGT activities when it was necessary. 88 1.3. Radioisotope method In some cases, an ultra-sensitive assay was needed when SGT activity was low, or when aglycones other than sinapic acid were being tested. A radioactive substrate mix was prepared at a ratio of 1 pi UDP-[|3-Z)-glucose (U-14C)] (0.025 pCi, 125 pmoles UDP- glucose per pi) to 24 pi standard substrate mixture (1.5 mM sinapic acid, 1.5 mM UDP- glucose, 30% glycerol, 5 mM DTT or 30 mM 2-ME and 240 mM MES, pH 6.0). The radioactivity assay for SGT activity was started by incubating 50 pi enzyme solution (replaced with distilled water in the control reaction) together with 25 pi radioactive substrate mix in a 2 ml screw-cap microcentrifuge tube at 30° C for 30 min. The reaction was stopped by adding 10 pi 6 N HC1. Sinapoylglucose produced in the reaction was extracted with 1 ml ethyl acetate by vortexing, then centrifuged at 14,000 x g for 1 min. The upper layer (0.7 ml) was transferred to a liquid scintillation vial and mixed with 5 ml scintillation cocktail (EcoLite, ICN Biomedical, Inc.). The radioactivity of 14C-labeled sinapoylglucose was measured by a liquid scintillation counter (LSC). SGT activity was calculated using the following equations: Csg (pmol) = {[ (Rs-Rc)/E]/S} xDsgx(Va/Vm) (Equation 6) where: Csg (pmol) = sinapoylglucose produced in the reaction Rs (cpm) = per minute count reading for sample Rc (cpm) = per minute count reading for control E (cpm/dpm) = 1, counting efficiency of LSC S (dpm/pmol) = 440, specific activity of radioisotope used in reaction 89 14 Dsg = 289, dilution factor of C-labeled UDPG in total added UDPG Va (ml) = 1.0, volume of added ethyl acetate for the extraction Vm (ml) = 0.7, volume of ethyl acetate used for the counting Total SGT activity of a protein solution is calculated as: SGT activity (pkat) = (Csg / Tr) x (Ve / Vr) (Equation 7) where: Csg (pmol) = amount of sinapoylglucose in reaction from Equation 6 Tr (sec) = reaction time in seconds Ve (ml) = total volume of protein solution Vr (ml) = volume of protein solution assayed SGT specific activity (pkat/mg protein) can be calculated using Equation 5 in Chapter II. 1.4. Spectrophotometry method This method was used to study SGT kinetics and reversibility in vitro, when monitoring of real-time kinetics was required. Since the X max (332.6 nm) of sinapoylglucose, the product of enzymatic reaction, is very close to the X max (324.4 nm) of sinapic acid, it is difficult to monitor the production of sinapoylglucose at its X max in the presence of high concentrations of the substrate. However, the absorbance spectrum of sinapoylglucose extends to a longer wave length than that of sinapic acid. A X 355 nm was chosen to monitor the production of sinapoylglucose, since there is minimal interference from the absorbance of sinapic acid at this wave length. The reaction mix, including 240 pi 50 mM 90 2-ME (or DTT) in 50% glycerol, 15 pi 44 mM sinapic acid in 20% methanol, 15 pi 44 mM UDP-glucose, and 150 pi 150 mM MES (pH 6.0), was added to each of two 1.5 ml (1 cm path length) quartz glass cuvettes, and then mixed with 150 pi enzyme solution in the sample cuvette and 150 pi heat-denatured enzyme in the reference cuvette. The cuvettes were immediately placed in a dual-beam spectrophotometer (UV-160A UV-Vis spectrophotometer, Shimadzu), and the absorbance at 355 nm was recorded for 5-15 min. In order to obtain the molar extinction coefficient for sinapoylglucose at A. 355 nm, 2 pi 4 mM sinapoylglucose in 40% methanol was added sequentially into the sample cuvette with the same reaction system (except no enzyme added) for 5 times and the absorbance recorded after each addition of sinapoylglucose. A standard curve was generated by using the amount of sinapoylglucose added in the cuvette and the corresponding OD355 nm. SGT activity was calculated using the following equations: C (pmol) = (At-A0)/(exL) (Equation 8) where: Csg (pmol) = sinapoylglucose produced in the reaction At = absorbance (355 nm) after reaction time t An = absorbance at initial reaction time t•.o e (pmol" cm" ) = extinction coefficient of sinapoylglucose at 355 nm L (cm) = path length of cuvette Total SGT activity of a protein solution was calculated as follows: SGT activity (pkat) = ( Csg / TA) x (Ve / Vr) (Equation 9) 91 where: Csg (pmol) = amount of sinapoylglucose in reaction from Equation 8 TA (sec) = reaction time (t -10) in seconds Ve (ml) = total volume of protein solution Vr (ml) = volume of protein solution assayed 2. Induction of SGT 2.1. Induction of SGT by sinapic acid B. napus cv. Westar seeds (0.1 g) were germinated in a 5 cm Petri dish with two layers of Whatman No. 1 filter paper wetted with an induction buffer (pH 6.5) containing 0, 0.1, 1, 5 or 10 mM sinapic acid in MS macronutrient solution (1.65 g/1 NH4N04, 1.90 g/1 KN03, 4.40 g/1 CaCl22H20, 3.70 g/1 MgS047H20 and 1.70 g/1 KH2P04). All treatments were repeated three times. The filter paper was maintained moist daily with the induction buffer. After germination under cool white fluorescent light at 23° C, seedlings were harvested after 1, 2, 3, 4 or 5 days. The seedlings were rinsed with distilled water, and patted dry on filter paper. The fresh weight of the seedlings from each treatment was recorded. Seedlings were flash frozen in liquid nitrogen and stored at -70° C before assaying SGT activity. The preparation of protein extracts from the seedlings, the determination of protein concentrations, and the assay of SGT activities were as described in Methods 2.2, 2.3 and 2.4 in Chapter II, respectively. 92 2.2. Induction of SGT by light For 1 and 2-day-old seedlings, ~3 g of B. napus cv. Westar seeds was germinated at room temperature on moistened 3MM filter paper in Petri dishes (10 x 1.5 cm). For 3, 4 and 5- day-old seedlings, germination was carried out in Magenta boxes (6x6x10 cm, Sigma Chemical Company). Both Petri dishes and Magenta boxes were divided into three sets for treatment under three different light conditions. One set was grown under continuous white light (40 cm below 2 x 30 W Sylvania cool white fluorescent tube, Lethonia Lighting Canada). The second set was exposed to an UV lamp (20 cm from 2 x 15 W, 302 nm UV light in TM-36 Chromato-UVE Transilluminator, Ultra Violet Products Limited, San Gabriel, CA) for 15 min each day. When not exposed to the UV light, the seedlings were kept in darkness. The third set of containers was wrapped with aluminum foil and the seedlings were grown in complete darkness. The seedlings were harvested 1, 2, 3, 4 or 5 days after germination and the fresh weight of the seedlings from each treatment was measured. All treatments were replicated three times. Treated seedlings were flash-frozen in liquid nitrogen and stored at -70° C before assaying SGT activity. The preparation of protein extracts from the seedlings, the determination of protein concentrations, and the assay of SGT activities were as described in Method 2.1. 93 2.3. Induction of SGT by heat shock B. napus cv. Westar seeds (~3 g) were germinated on moistened 3MM filter paper in Petri dishes (10 x 1.5 cm) at 23° C in darkness. After 44 h of germination, the seedlings were divided into four sets and treated with four different temperature regimes. The first set was grown at 30° C for 2 h, the second set at 37° C, and the third set at 45° C. The fourth set was maintained at 23° C. The seedlings from each treatment were grown at 23° C again after the heat shock. When the seedlings were 60 h old, the heat shock treatments were repeated again, but this time for 4 h. The seedlings were then grown at 23° C until harvested at 70 h. The harvested seedlings were divided into two parts for protein extraction. Preparation for protein extract was as described in Method 2.2 of Chapter II, except that the extraction buffer for one group of seedlings contained a set of known protease inhibitors: 74 pM antipain dihydrochloride, 0.1 pM aprotinin, 130 pM bestatin, 50 pM chymostatin, 14 pM E-64, 1 mM EDTA-Na2, 1 pM leupeptin, 1 mM Pefabloc SC, 1 pM pepstatin and 0.1 mM phosphoramidon. The extraction buffer for the other group of seedlings did not contain any protease inhibitors. SGT activity in each extract was determined by the HPLC method as described in Method 2.4 (Chapter II). 3. Subcellular localization of SGT A non-ionic detergent (Triton X-100) was used in the protein extraction buffer to investigate whether SGT was associated with the membranes. The proteins in 60-h-old seedlings were extracted as described in Method 2.2 (Chapter II), except that the 94 extraction buffer included 0.2% Triton X-100. After extraction and centrifugation, both the pellet and supernatant were assayed for SGT activity by the HPLC method. To investigate which subcellular compartment might contain SGT, ultracentrifugation was used to resolve the different classes of subcellular organelles in extracts from 60-h- old seedlings. The seedlings were homogenized by a Waring Blender in an extraction buffer (5 mM DTT, 5% glycerol, 4% PVPP and 50 mM Bis-Tris/HCl, pH 7.0) at a ratio of 1 part seedlings to 2 parts buffer (w/v). The homogenate was mixed in a tube by rolling it from end-to-end at 4° C for 30 min, and then passing the resulting suspension through four layers of Miracloth. The filtrate was centrifuged at 800 x g for 5 min and the supernatant was passed through two layers of Miracloth again to remove lipid residues. Fractionation was carried out using 75% Percoll (Pharmacia) system with two different sample loading methods. Three 0.5 x 2" Beckman Ultra-Clear Quick-Seal ultracentrifuge tubes were filled with 4.5 ml 75% Percoll, including 5% glycerol, 2 mM DTT, 0.25 M sucrose and 50 mM Bis-Tris, pH 6.5. Ten pi density marker bead suspension was added to one of three tubes, followed by 360 pi extraction buffer. After the density marker beads were mixed with the Percoll medium, the tubes were sealed, as described in the Beckman instructions. To the second tube, 370 pi seedling extract was carefully layered on top of the Percoll solution. To the third tube, 370 pi seedling extract was added and mixed completely with the Percoll solution by shaking. These tubes were sealed and ultracentrifuged in a Beckman NVP 90 rotor at 30,000 rpm (60,000 x g) and 4° C for 30 95 min with both slow acceleration and deceleration. The positions of visible density marker bands were recorded after centrifugation. Fractions (1 ml) were collected from the bottom of each tube containing extract, and the protein concentration and SGT activity of each fraction were determined as described in Methods 2.3 and 2.4 (Chapter II). 4. Purification of SGT A series of column chromatography techniques was used for the SGT purification. All buffers contained 5% glycerol and a reducing reagent (2-ME or DTT) at the concentration indicated. The following procedures were used for a large scale purification and took 40 days to complete. All procedures were performed at 4° C whenever possible. Step 1. Protein extraction: Freshly harvested 65-h-old seedlings (800 g) were flash- frozen in liquid nitrogen and ground to a fine powder with a coffee mill containing a small amount of sea sand. The powder was extracted with 1500 ml of ice-chilled 100 mM potassium phosphate buffer (KPi, pH 7.0), including 5% glycerol, 0.5 mM phenylmethanesulfonyl fluoride (PMSF), 15 mM 2-ME and 4% insoluble PVPP for 30 min. The homogenate was filtered through two layers of Miracloth and centrifuged at 15,000 x g for 30 min. The supernatant was filtered again though two layers of Miracloth to remove lipid residues. Step 2. Protamine precipitate adsorption: Protamine sulphate (2% in 100 mM KPi buffer, pH 7.0) was added to the supernatant to give a final concentration of 0.1%. The 96 mixture was stirred for 30 min, then centrifuged at 20,000 x g for 20 min. The supernatant was filtered through two layers of Miracloth and dialyzed in an 8 kDa-cut-off dialysis tube against 2000 ml 10 mM KPi buffer pH 7.0, including 15 mM 2-ME. After dialysis for 6 h, the buffer was replaced with fresh buffer and dialysis was continued overnight. The dialyzed protein extract was centrifuged at 20,000 x g for 15 min. Step 3. Differential surface binding: The supernatant from Step 2 was divided into 10 portions (150 ml each) and each portion was loaded individually on a Macro-prep hydroxy apatite column (HTP, 16 x 200 mm, flow rate 4 ml/min) equilibrated with 10 mM KPi buffer by FPLC (Pharmacia Chemical Company). The unbound proteins were removed by washing with 250 ml loading buffer, and SGT was eluted with a linear gradient of KPi buffer (15 mM 2-ME and 400 mM KPi, pH 7.0, flow rate 8 ml/min) from 10 to 200 mM in 320 ml. The fractions (12 ml) were collected and assayed for SGT activity by the multi-well plate method. The SGT-active fractions were pooled and dialyzed against Mono Q loading buffer (20 mM 2-ME and 20 mM Bis-Tris, pH 6.5) overnight. The buffer was changed with the fresh Mono Q loading buffer and dialyzed for another day. The dialyzed HTP fractions were concentrated in the dialysis tubes by water absorption with the dry polyethylene glycol powder (MW 8,000). The dialyzed and concentrated HTP fractions were centrifuged at 20,000 x g. Step 4. Anion exchange: The supernatant from Step 3 was divided into 8 portions (25 ml each). One portion at a time was loaded onto an anion exchange column (Prep-Q HR, 10 97 x 100 mm, flow rate 2 ml/min) which was equilibrated with the Mono Q loading buffer. The unbound protein was removed by washing with 100 ml loading buffer, and SGT was eluted with a linear gradient of NaCl (2 M NaCl in loading buffer, flow rate 2 ml/min) from 0 to 200 mM in a 160 ml volume. The fractions (5 ml) were collected and assayed for SGT activity by the multi-well plate method. The SGT-active fractions were pooled and precipitated by ammonium sulfate at 80% of saturation for 3 h. After centrifugation at 20,000 x g for 30 min, the supernatant was discarded. The pellet was stored at -20° C until used in the next step. Step 5. The first size exclusion: The protein pellet obtained in Step 4 was dissolved in Superdex 75 loading buffer (2 mM DTT, 150 mM NaCl and 20 mM Tris/HCl, pH 6.5), and divided into seven portions (2 ml each). One portion of protein solution at a time was loaded onto a Superdex 75 gel filtration column (10 x 1020 mm, flow rate 0.5 ml/min) equilibrated with the Superdex 75 loading buffer. Using a ConSep 1000 LC (Millipore Corp.), SGT was eluted with 100 ml of the same buffer at 0.5 ml/min. The fractions (5 ml) were collected and assayed for SGT activity by the multi-well plate method. The SGT-active fractions were pooled and used directly in the next step. Step 6. Affinity binding: The pooled gel filtration fractions were divided into two portions (50 ml each). One portion at a time was loaded onto a blue dye affinity matrix column (HiTrip Blue, 16 x 25 mm, flow rate 0.5 ml/min) equilibrated with a HiTrip Blue . loading buffer (2 mM DTT and 20 mM Bis-Tris/HCl, pH 6.5). Using the FPLC system, 98 the unbound proteins were removed by washing the column with 60 ml loading buffer, and SGT was eluted with a linear gradient of NaCl (2 M NaCl in the loading buffer, flow rate 5 ml/min) from 0 to 2 M in a 50 ml volume. The fractions (4 ml) were collected and assayed for SGT activity by the HPLC method. The SGT-active fractions were pooled and precipitated by ammonium sulfate at 70% of saturation for 3 h. After centrifugation at 20,000 x g for 30 min, the supernatant was discarded, and the pellet was stored at -20° C until used in the next step. Step 7. The second size exclusion: The protein pellet from Step 6 was dissolved in the Superdex 75 loading buffer and divided into 3 portions (2 ml each). The second size exclusion was carried out on the Superdex 75 gel filtration column using the same conditions as described in Step 5 except that the flow rate was 0.17 ml/min. The fractions (1 ml) were collected and assayed for SGT activity by the HPLC method. The SGT- active fractions were pooled and precipitated by ammonium sulfate at 80% of saturation for 3 h. After centrifugation at 15,000 x g for 1.5 h, the supernatant was discarded. The pellet was dissolved in a Mono P loading buffer (2 mM DTT and 25 mM 2-methyl- piperazine, pH 6.3), and then desalted on a PD-10 column equilibrated with the Mono P loading buffer. Step 8. Chromatofocusing: The desalted protein solution from Step 7 was divided into 3 portions (2 ml each). One portion at a time was loaded onto a chromatofocusing column (Mono P HR, 5 x 200 mm, flow rate 0.5 ml/min) equilibrated with the Mono P loading 99 buffer. The column was eluted by FPLC with 42 ml Polybuffer 74 (2 mM DTT and 10% Polybuffer 74, pH 4.5), and SGT was separated by a linear gradient of pH from 4.7 to 5.7, which was generated on the column by Polybuffer 74. In order to neutralize the acidity of the elution buffer, the fractions (1.5 ml) were collected in 2 ml microcentrifuge tubes containing 0.2 ml 250 mM Tris/HCl, pH 6.5 and assayed for SGT activity by the HPLC method. The SGT-active fractions were pooled and concentrated to a 2 ml volume using a 10 kDa-cut-off Microsep centrifugal concentrator. Step 9. The third size exclusion: The concentrated protein solution from Step 8 was divided into 3 portions (0.7 ml each) and subjected to a third size exclusion on the Superdex 75 gel filtration column. The gel filtration conditions were the same as described in Step 7. The SGT activity in each fraction (1 ml) was detected by the radioassay method. The fractions were then stored at - 20° C. 5. Physical and chemical characterization of SGT A partially purified SGT preparation from purification procedure Step 7 was used for the characterization of the enzymatic properties, unless otherwise indicated. 5.1. Determination of molecular weight The molecular weight of native SGT (from Step 8) was determined by FPLC on a Superose 12 HR gel filtration column (10 x 30 mm), which was calibrated by a set of native reference proteins (Boehringer Mannheim Biochemical), including bovine serum 100 albumin (67.0 kDa), egg white albumin (45.0 kDa), chymotrypsinogen A (25.0 kDa) and ribonuclease A (13.7 kDa). The buffer system was the same as described in Step 7, and the flow rate was 0.2 ml/min. The SGT activity in each fraction (0.2 ml) was determined by the HPLC method. The molecular weight of denatured SGT was determined by 10% discontinuous sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Laemmli, 1970). The Bio-Rad Mini-Protean II electrophoresis system was used for gel casting and electrophoresis. The gel was stained by Coomassie blue R-250 dye, and the standard molecular weight markers for SDS-PAGE were a set of low molecular weight protein mix (Bio-Rad), including phosphorylase B (97.4 kDa), bovine serum albumin (66.2 kDa), ovalbumin (45.0 kDa), carbonic anhydrase (31.0 kDa), soybean trypsin inhibitor (21.5 kDa) and lysozyme (14.4 kDa). 5.2. Determination of SGT pH stability and pH optimum In order to investigate SGT stability at different pH values during the storage, the buffer in a partially purified SGT preparation was changed to 20 mM Tris/HCl (pH 7.0), 25 mM 2-methyl-piperazine (pH 6.3) or 10% Polybuffer 74 (pH 5.1) using PD-10 columns. All buffers contained 5% glycerol and 2 mM DTT. The protein solutions from each treatment were stored at - 20° C and sampled 1, 3, 6 and 12 days later. The treatment with 10% Polybuffer 74 was also sampled 0.5, 1, 2, 3 and 4 h after incubation on ice. The SGT activity of each sample was assayed by the HPLC method. 101 In order to investigate the optimum pH for SGT activity, 400 mM Tris/HCl buffers at pH 6.0, 7.0, 7.5, 8.0, 8.5, 9.0 and 9.5, 400 mM MES/NaOH buffers at pH 5.4, 6.0 and 6.5, and 400 mM NaAc/HAc buffers at pH 3.5, 4.0, 4.5, 5.0 and 5.5 were employed. The partially purified SGT preparation (40 pi) was incubated with 25 pi 50 mM 2-ME in 50% glycerol and 25 pi appropriate pH testing buffer in a 1.5 ml HPLC vial at 4° C for 2 h. The mixture was then incubated with 5 pi 10 mM sinapic acid in methanol and 5 pi 10 mM UDP-glucose in distilled water at 37° C for 30 min. The reaction was stopped by addition of 250 pi acetonitrile, and then made up to a final volume of 1 ml with distilled water. The SGT activity of each treatment was determined by the HPLC method. 5.3. Determination of SGT thermal stability and temperature optimum In order to investigate SGT stability at different temperatures during the storage, the partially purified SGT preparation was stored at 23°, 4° and - 20° C, and sampled 2, 4, 8, 24 h, and 2, 4, 7 and 30 days later. The SGT activity of each sample was determined by the HPLC method. To establish the optimum temperature for SGT activity, the partially purified SGT preparation (50 pi) was first pre-incubated at 4°, 23°, 32°, 37°, 42°, 50° or 65° C in a 1.5 ml HPLC sample vials for 15 min, and then mixed with 25 pi substrate mix. The reaction was incubated at the same temperature as pre-incubation for another 30 min, stopped by addition of 250 pi acetonitrile, and then made up to a final volume of 1 ml with distilled water. The SGT activity of each treatment was determined by the HPLC method. 102 5.4. Requirement for metal ions Various metal ions, including NaCl, KC1, CaCl2, CoCl2, CuCl2, FeS04, HgCl2, MgCl2, MnCl2 and ZnS04, were tested for their effects on SGT activity. EDTA was added to one set of reactions after the pre-incubation of the metal ion with the SGT preparation in order to test the reversibility of the metal ion effects. The partially purified SGT preparation (50 pi) was pre-incubated at 4° C for 1 h with 25 pi 50 mM 2-ME in 50% glycerol and 25 pi 5 mM metal ion solution made up in 250 mM MES buffer (pH 6.0). Where reversibility was to be tested, the mixture was mixed with 25 pi 40 mM EDTA in 250 mM MES (pH 6.0). Otherwise it was mixed with 25 pi 250 mM MES buffer (pH 6.0). After incubation at 4° C for 1 h, the SGT substrates (5 pi 10 mM sinapic acid in methanol and 5 pi 10 mM UDP-glucose in distilled water) were added and incubated at 37° C for 1.5 h. The reaction was stopped by addition of 250 pi acetonitrile, and then made up to a final volume of 1 ml with distilled water. The SGT activity of each treatment was determined by the HPLC method. 5.5. Requirement for reducing reagents Reducing reagents (ascorbic acid, 2-ME or DTT) were tested at concentrations ranging from 0.1 to 20 mM when they were used individually, or at 5 mM each when they were applied as a combination of two. The SGT source was the crude extract from 65-h-old seedlings, which was prepared as described in Method 2.2 (Chapter II), except that no reducing reagent was included in the extraction buffer. The desalted seedling extract (50 pi) was first pre-incubated in a HPLC sample vial at 4° C for 1 h with 25 pi 50% glycerol 103 plus 25 (0.1 reducing reagent solution prepared in 0.4 M MES buffer (pH 6.0). The mixture was then incubated at 37° C for 30 min with 12.5 ul 5 mM sinapic acid in 10% methanol and 12.5 JJ.1 5 mM UDP-glucose in distilled water. The reaction was stopped by addition of 250 ul acetonitrile, then made up to a final volume of 1 ml with distilled water. The SGT activity of each treatment was determined by the HPLC method. 5.6. Determination of SGT substrate specificity ADP-glucose, CDP-glucose, GDP-glucose, TDP-glucose, UDP-glucose, UDP-galactose, UDP-mannose, UDP-xylose and (3-D-glucose were used to test the SGT specificity for sugar donors. Cinnamic acid, jO-coumaric acid, caffeic acid, 5-hydroxyl-ferulic acid, ferulic acid, sinapic acid and syringic acid were used to test the SGT specificity for sugar acceptors. In order to investigate specificity for the sugar donor, 50 ul partially purified SGT preparation was incubated at 37° C for 15 min with 25 u.1 50 mM 2-ME in 50% glycerol, 5 ul 0.4 M MES buffer (pH 6.0), 25 ul 2.6 mM sinapic acid in 10% methanol and 25 ul 2.6 mM testing sugar donor solution made up in 0.4 M MES (pH 6.0). The reaction was stopped by addition of 250 ul acetonitrile, then made up to a final volume of 1 ml with distilled water. The SGT activity was determined by the HPLC method. To test SGT specificity for the sugar acceptor, 50 ul partially purified SGT preparation was incubated in a screw-cap microcentrifuge tube at 30° C for 60 min with 12 ul 2.6 104 mM testing sugar acceptor solution made up in 20% methanol in 0.4 M MES (pH 6.0), 12 pi 3.125 mM UDP-glucose solution (31.25 mM DTT, 50 mM 2-ME, 50% glycerol and 0.4 M MES, pH 6.0) and 2 pi UDP-[(3-£>-glucose (U-14C)] (0.025 pCi, 125 pmoles UDP- glucose per pi). The reaction was stopped by addition of 10 pi 6 N HC1 and the enzymatic product was extracted with 1 ml ethyl acetate. The mixture was vortexed and centrifuged at 15,000 x g for 1 min. The upper layer (0.7 ml) was transferred to a scintillation counting vial and mixed with 5 ml scintillation cocktail (EcoLite, ICN Biomedical, Inc.). The radioactivity of 14C-labeled glucose ester was measured by LSC. 5.7. Determination of the inhibitory effect of UDP-glucose analogues UDP-galactose, UDP-mannose, UDP-xylose, UDP, TDP and (3-Z)-glucose were tested for their inhibitory effect on SGT activity. The partially purified SGT preparation (50 pi) was pre-incubated in a HPLC sample vial at 4° C for 1 h with 25 pi 50 mM 2-ME in 50% glycerol and 25 pi 10 mM UDP-glucose analogue in 0.4 M MES (pH 6.0). The mixture was then incubated with 12.5 pi 5 mM sinapic acid in 10% methanol and 12.5 pi 5 mM UDP-glucose at 37° C for 30 min. The reaction was stopped by addition of 250 pi acetonitrile, and then made up to a final volume of 1 ml with distilled water. The SGT activity was determined by the HPLC method. 5.8. Determination of effects of other inhibitors A^-ethylmaleimide, iodoacetic acid (Na salt), j9-hydroxyl-mercuribenzoic acid, phenylmethanesulfonyl fluoride, 3,5-dihydroxybenzoic acid and 4-hydroxyl-3,5- 105 dimethoxy-benzaldehyde azine, were tested for their ability to inhibit SGT activity. The partially purified SGT preparation (50 u.1) was pre-incubated at 4° C for 2 h with 25 u.1 50 mM 2-ME in 50% glycerol, 25 u.1 5 mM inhibitor solution made up in 5% methanol in 250 mM MES buffer (pH 6.0) and 25 ul 0.4 M MES buffer (pH 6.0). The mixture was then incubated at 37° C for 15 min with 12.5 ul 5 mM sinapic acid in 20% methanol and 12.5 u.1 5 mM UDP-glucose. The reaction was stopped by addition of 250 ul acetonitrile, and made up to a final volume of 1 ml with distilled water. The SGT activity was determined by the HPLC method. 5.9. Determination of SGT reversibility 5.9.1. From the forward reaction ( SinA + UDPG SinG + UDP ) The enzymatic reaction was carried out in a 1.5-ml cuvette (A, B or C) containing 240 ul 50 mM 2-ME in 50% glycerol, 15 ul 44 mM sinapic acid in 20% methanol, 15 ul 44 mM UDP-glucose and 780 ul 150 mM MES (pH 6.0). The reaction was started by adding 150 u.1 partially purified SGT preparation to cuvettes A and B (the sample cuvettes), and 150 ul heat-denatured SGT preparation was added to cuvette C (the reference cuvette). The absorbance (355 nm) of the reaction solution was recorded at five minute intervals. After 40 min, 15 ul 100 mM TDP was added to cuvette A, 15 ul 100 mM UDP was added to cuvette B and 15 u.1 distilled water was added to cuvette C. These additions were designed to reverse the reaction direction in cuvettes A and B, as follows: SinA + TDPG ^- SinG + TDP, or SinA + UDPG <- SinG + UDP. The decrease in absorbance at 355 nm was monitored for another 40 min at five minute intervals. 106 5.9.2. From the reverse reaction ( SinA + UDPG To each of four 1.5-ml cuvettes (A, B, C and D), 240 pi 50 mM 2-ME in 50% glycerol, 2 pi 44 mM sinapoylglucose, 6 pi 100 mM UDP (except to the cuvette A) and 802 pi 150 mM MES (pH 6.0), were added. The reaction was started by adding 150 pi partially purified SGT preparation to cuvettes A, B and C, and 150 pi heat-denatured SGT preparation to cuvette D. The absorbance (355 nm) of the reaction was recorded at five minute intervals. After 40 min, 15 pi 100 mM UDP was added to cuvette A, 15 pi 100 mM TDPG was added to cuvette B, 15 pi 100 mM UDPG was added into cuvette C and 15 pi distilled water was added into the cuvette D. The reverse reaction in the cuvette A was started (SinA + UDPG <— SinG + UDP). The direction of the reaction in cuvette B and C changed to an opposite way: SinA + TDPG SinG + TDP, or SinA + UDPG -» SinG + UDP. The absorbance at 355 nm was monitored for another 40 min at five minute intervals. 5.10. Investigation of affinity matrices binding for SGT Three types of immobilized substrate and its analogues were used for testing their affinity binding for SGT. The structures of immobilized ligand are shown in Fig. 22. 5.10.1. UDP-glucose-hydrazide membrane UDP-glucose was coupled to a cellulose membrane through a vicinal hydroxyl-hydrazide coupling method. The UDP-glucose was dissolved in 5 ml 100 mM sodium acetate (pH 5.0) to a final concentration at 10 mM, and then mixed with 0.25 ml 200 mM sodium 107 A. Immobilized UDP-glucose on hydrazide membranes B. Immobilized UDP-glucuronic acid on agarose beads o C. Immobilized sinapic acid on Sepharose beads CH=CH- Sepharose bead H3CO Fig. 22. Structures of three affinity matrices for binding SGT. 108 periodate (NaI04). The mixture was incubated in darkness at 4° C for 1 h. After the incubation, 90 u.1 4 M ethylene glycol was added, and the mixture was incubated again in darkness at 4° C for 25 min. A hydrazide cellulose membrane cartridge (8x10 mm) was washed with 25 ml ice-chilled distilled water and 25 ml ice-chilled 100 mM sodium acetate (pH 5.0), then incubated with the ligand by circulating 5.3 ml oxidized UDP- glucose through it at a flow rate 2 ml/min and 4° C for 8 h. The cartridge was washed with 100 ml 1 M NaCl, then flushed with 1 M NaCl at 2 ml/min and 4° C for 1 h. The immobilized UDP-glucose membrane cartridge was stored at 4° C until used for the SGT binding test. A dialyzed seedling extract of B. napus cv. Westar (2 ml; prepared as described in Method 2.2 Chapter II) was loaded by FPLC at a flow rate 0.1 ml/min onto the UDP- glucose coupled cartridge, which had been pre-equilibrated with a loading buffer (2 mM DTT, 5% glycerol and 20 mM Tris/HCl, pH 7.0). The unbound proteins were removed by washing with 10 ml loading buffer. The cartridge was washed with 10 ml 0.3 M NaCl in the loading buffer at a flow rate 0.5 ml/min, then eluted with 20 ml 20 mM UDP-glucose in the loading buffer. All filtrates and eluant were assayed for SGT activity by the HPLC method. 5.10.2. UDP-glucuronic acid-agarose UDP-glucuronic acid-agarose beads (100 mg, UDPGA-Agarose, Sigma) were re- hydrated with 5 ml distilled water in a 10 ml syringe column. The swollen matrix (~ 2 109 ml) was washed with 100 ml loading buffer (2 mM DTT, 5% glycerol and 20 mM Tris/HCl, pH 7.0), and excess liquid was removed by vacuum filtration. The dialyzed seedling extract (2 ml) was added to the matrix and incubated on ice for 30 min. The mixture was then shaken gently for 30 min. The protein solution was removed from the gel matrix by centrifugation at 1,000 x g for 2 min. The gel matrix in the column was washed with 15 ml loading buffer, then eluted with 40 ml 5 mM UDP-glucose in the loading buffer. All filtrates and eluant were assayed for SGT activity by the HPLC method. 5.10.3. Sinapic acid-Sepharose Sinapic acid was coupled to a EAH-Sepharose 4B gel matrix (Pharmacia) by use of a carbodiimide derivative method. The EAH-Sepharose matrix (-12 ml) was first washed with 250 ml 0.5 M NaCl, and then 1 L distilled water. The matrix was dried by draining out the water and transferred into a 50 ml tube containing 10 ml 100 mM sinapic acid in dimethyl formamide (DMF). The pH of the mixture was adjusted to pH 4.5 by 1 M HC1, and 2 ml 1.25 M A^A^'-dicyclohexylcarbodiimide (DCC) in DMF was added drop by drop over a period of 15 min. The pH of the mixture was maintained at pH 4.5 - 5.0 by addition of 0.1 M NaOH as required for 1 h. The mixture was shaken from end-to-end on a roller shaker at room temperature overnight. The gel was washed sequentially with 100 ml DMF, 250 ml 100% ethanol, 250 ml 1 M NaCl, 1 L distilled water and 100 ml loading buffer (2 mM DTT, 5% glycerol and 20 mM Tris/HCl, pH 7.0) in order, then stored at 4° C until used in the SGT binding test. 110 The dialyzed seedling extract (2 ml) was added and mixed with 2 ml moist but filtered and sinapic acid coupled affinity matrix in a 10 ml syringe column. The mixture was incubated on ice for 30 min, then shaken gently at 4° C for 30 min. The protein solution was removed from the gel matrix by centrifugation at 1,000 x g for 2 min. The gel matrix in the column was washed with 15 ml loading buffer, then eluted with 5 ml 5 mM UDP- glucose in the loading buffer. The column was washed again with 20 ml loading buffer and eluted with 5 ml 10 mM sinapic acid in the loading buffer, or eluted with a stepwise gradient consisting of 5 ml each of 0.1, 0.2, 0.3 and 0.5 M NaCl in loading buffer. All filtrates and eluant were assayed for SGT activity by the HPLC method. 5.11. Determination of SGT substrate kinetics The apparent Km values of the enzyme for its substrate were determined by varying one substrate (sinapic acid or UDP-glucose) from 0.1 to 2.0 mM at a fixed concentration of the second substrate (UDP-glucose or sinapic acid). The partially purified SGT preparation (50 pi) was incubated with 25 pi 50 mM 2-ME in 50% glycerol, 25 pi 0.4 M MES buffer (pH 6.0), 5 pi sinapic acid in 100% methanol (2.2, 11, 22 or 44 mM) and 5 pi UDP-glucose (2.2, 11, 22 or 44 mM) in a 1.5 ml HPLC sample vial at 37° C for 15 min. The reaction was stopped by addition of 250 pi acetonitrile, then made up to a final volume of 1 ml with distilled water. The sinapoylglucose produced in each reaction was determined by the HPLC method. The kinetic parameters and reaction mechanism were analyzed using the EZ-Fit program (Dr. Frank W. Perrella, Glenolden Laboratory, E.I. DuPont de Nemours & Co.). ill 6. Recovery of SGT from native PAGE Laemmli's discontinuous SDS-PAGE system (Laemmli, 1970) was modified to a non- SDS (native) condition for recovering SGT activity in the PAGE gel. The preparation of PAGE gel was as described by Laemmli except that SDS was not included, and the separation gel contained 0.15% 2-ME. The gel was subjected to pre-electrophoresis as follows. To the upper buffer reservoir, 70 ul 10 mM glutathione was added and mixed with 140 ml running buffer (24.8 mM Tris and 191.8 mM glycine, pH 8.3). The gel was electrophoresed at constant current (3 mA/gel) for 2 h. After electrophoresis, the running buffer was decanted and the gel was stored at 4° C overnight. A protein preparation (from Step 5, Method 4) concentrated by a 10 kDa-cut-off Microsep centrifugal concentrator was mixed with a native sample loading buffer (10 mM DTT, 40% glycerol and 50 mM Tris/HCl, pH 6.8) at a ratio 1:1, and loaded on the pre-electrophoresed gel. To the upper buffer reservoir (140 ml running buffer), 140 ul 100 mM thioglycolate was added and electrophoresis was conducted at constant current (15 mA/gel) for 40 min at 4° C. After electrophoresis, the gel was removed from the glass plates, rinsed with distilled water and then soaked in a re-naturation buffer (2 mM DTT, 5% glycerol and 50 mM Tris/HCl, pH 6.0) at 4° C for 15 min. The gel was sliced to ~ 2.5 mm strips from the top of separation gel, and each of the gel strips was chopped to small pieces and incubated with 50 ul re- naturation buffer and 25 ul radioactive substrate mix (as described in Method 1.3) in a 2 ml screw-cap microcentrifuge tube at room temperature overnight. The procedures involving radioisotope assay for SGT activity were as described in Method 1.3. 112 7. Determination of amino acid composition and sequence A highly purified SGT preparation from Step 9 in Method 4 was concentrated in the 10 kDa-cut-off Microsep centrifugal concentrator, then fractionated by 10% SDS-PAGE. Protein peptides on the gel were then transferred to an Immobilon-PSQ PVDF membrane and individual bands were cut out for the analysis of amino acid composition and sequence. Pre-electrophoresis and SDS-PAGE electrophoresis were as described above in Method 6, except that the running buffer and the gel contained 0.1% SDS, and the loading buffer included 20% glycerol, 1% SDS, 5% 2-ME, 0.1% bromophenol blue and 50 mM Tris/HCl, pH 6.8. After electrophoresis, the gel was removed from the glass plates and soaked in blotting buffer (10% methanol in 10 mM CAPS, pH 11.0) for 5 min. The Immobilon-PSQ PVDF membrane was saturated in 100% methanol for a few seconds, then rinsed with the blotting buffer. The gel was sandwiched between the PVDF membrane and 2 layers of 3MM filter paper and electro-blotted with the blotting buffer at constant current (212 mA) at 4° C for 2.5 h. After the blotting, the PVDF membrane was rinsed with the distilled water and saturated with 100% methanol for a few seconds, then stained briefly in a Coomassie Brilliant Blue dye solution (0.1 % Coomassie Blue R-250 and 1% acetic acid in 40% methanol). The stained PVDF membrane was de-stained immediately in 50% methanol to remove the dye from the background, and then rinsed extensively with distilled water. The protein band of interest was sliced and stored in a microcentrifuge tube for the further analysis. 113 Protein on the PVDF strip was hydrolyzed using gas-phase techniques, and the amino acid composition was analyzed using an Applied Biosystems 420A derivatizer-analyser system at the Protein Microchemistry Centre, University of Victoria, Victoria, BC. Canada. For N-terminal amino acid sequence, the intact peptides blotted on the PVDF were analyzed on an Applied Biosystems 470A gas-phase sequencer equipped with on• line PTH-AA analyzer at the Protein Microchemistry Centre, University of Victoria. For the internal amino acid sequence, the protein on the PVDF strip was digested in situ with trypsin at the Harvard Microchem, The Biological Laboratories, Harvard University, Cambridge, MA. USA. The tryptic peptides were purified by HPLC on a reverse-phase column, and sequenced using an Applied Biosystems 477A gas-phase sequencer equipped with on-line PTH-AA analyzer 8. Protein sequence analysis The amino acid sequences obtained were compared with those in the protein sequence data bases, including PDB (Brookhaven Protein Data Bank), SWISSPROT (SWISS- PROT Protein Database), PIR (Protein Identification Resource Protein Sequence Database) and GENPEPT (CDS translations from GenBank). The database searches were carried out by using the BLAST search program from the National Center for Biotechnology Information, USA, through the Internet. The alignment analysis of protein sequences was conducted with PC/Gene nucleic acid and protein sequence analysis software (IntelliGenetics Inc., Geneva, Switzerland). 114 9. Production of polyclonal antibodies Both native and denatured SGT preparations were used as antigens in attempts to produce anti-SGT polyclonal antibodies in young New Zealand White rabbits housed at the Animal Care Center, University of British Columbia. All injection, bleeding and harvesting procedures were carried out by the Animal Care Center staff. 9.1. Production of polyclonal antibodies with native antigen An extensively purified SGT preparation from Step 9 in Method 4 was concentrated using a 10 kDa-cut-off Microsep centrifugal concentrator. The concentrated protein preparation (170 u.g protein in 200 u.1 phosphate buffered saline [PBS, 137 mM NaCl, 2.7 mM KC1, 4.3 mM Na2HP04 and 1.4 mM KH2P04, pH 7.0]) was mixed with 1.2 ml complete Freund's adjuvant and 300 ul PBS, then injected subcutaneously into the neck of a rabbit at four sites. Four weeks later, a boost was conducted by an intramuscular injection with a mixture of the concentrated protein and incomplete Freund's adjuvant on the back legs of the rabbit at two sites (80 u.g protein / 0.5 ml / site). A blood sample was taken twelve days after the boost injection for testing the immune response. The second boost was carried out fifteen days after the test bleeding, by repeating the intramuscular injection on the back legs of the rabbit at two sites (50 u.g protein / 0.5 ml / site). A second test bleed followed nine days after the second boost injection, and a whole body bleeding was conducted seven days after the second test bleeding. 115 9.2. Production of polyclonal antibodies with denatured antigen A concentrated SGT preparation from Step 9 in Method 4 was fractionated by 10% SDS- PAGE and visualized by Coomassie Blue dye staining. The desired protein band (~ 45 kDa) was sliced out of the gel, placed in a microcentrifuge tube, macerated with a plastic piston, and then mixed with an appropriate volume of PBS (pH 7.0). The primary immunization was conducted by subcutaneous injection with the mixture of macerated gel and complete Freund's adjuvant at 4 sites (35 u.g protein / 0.25 ml / site) on the neck of a rabbit. Three weeks later, a boost consisting of a mixture of the mashed gel and incomplete Freund's adjuvant was injected intramuscularly at two sites (25 \xg protein / 0.5 ml / site) on the back legs of the rabbit. A blood sample was taken ten days after the boost injection. The second boost was conducted fifteen days after the test bleed by repeating the intramuscular injection at two sites (25 u.g protein / 0.5 ml / site) on the back legs of the rabbit. A second test bleed followed ten days after the second boost injection, and the whole body bleeding was conducted seven days after the second test bleeding. 10. Immunodetection of SGT by Western blot Electrophoresis of the -partially purified SGT preparation was conducted using the discontinuous SDS-PAGE (10% gel) system (Laemmli, 1970). After electrophoresis, the gel was rinsed with blotting buffer (Laemmli's running buffer plus 20% methanol and 0.02% SDS), then blotted to a WestTran PVDF membrane (Schleicher & Schuell Inc.) at constant current (212 mA) and 4° C for 2.5 h. After electro-blotting, the PVDF membrane 116 was blocked with 5% instant skim milk in Tris buffered saline (TBS, 150 mM NaCl and 50 mM Tris/HCl, pH 7.6) at room temperature for 1 h. An antibody solution (anti-SGT polyclonal or other antibodies being tested for cross-reactivity) was added directly to the blocking solution to make a dilution of 1:500 - 1:2,000. The PVDF membrane was incubated at room temperature with gentle shaking overnight. The blocked PVDF membrane was first washed with 50 ml TBS at room temperature for 20 min, then washed with 50 ml TBS plus 0.05% NP-40 for 20 min, and finally washed with 50 ml TBS for 20 min. The PVDF membrane was then incubated with a secondary antibody solution (1:1,000 - 1:3,000 dilution of goat anti-rabbit or anti-mouse IgG alkaline phosphatase conjugates with 5% instant skim milk in TBS) at room temperature for 1 h. After the incubation, the PVDF membrane was washed with TBS again as described above, then stained with a color development mixture of 10 ml 100 mM Tris/HCl buffer (pH 9.5) containing 50 mM MgCl2, 100 mM NaCl, 44 pi nitroblue tetrazolium (NBT) (BRL) and 32 pi 5-bromo-4-chloro-3-indolyl phosphate (BCIP) (BRL). 11. Immunoprecipitation of SGT activity A partially purified SGT preparation (50 pi) was mixed with different volumes (5 - 50 pi) of antibody solution (Pre-immune rabbit serum was used as a control). This mixture was made up with distilled water to a total volume of 100 pi in a 0.5 ml microcentrifuge tube. The mixture was shaken by rolling at 4° C for 3 - 15 h, then transferred to a mini- immunobead column, consisting of a 20 pi filter-pipette tip (GT-9003 series, Gordon Technologies Inc.). The mini-immunobead column was prepared by placing 7 mg of dry 117 protein A-Sepharose beads (Pharmacia) in the filter-pipette tip, and allowing it to re- hydrate in 100 u.1 TBS (pH 6.5) for 2 h. After the beads were swollen, the TBS was drained out by centrifuging at 1,800 x g for 1 min. The mixture of antigen and antibody was then incubated with the drained-dried protein A-Sepharose beads at 4° C for 3 - 6 h. The column was placed in a 2 ml microcentrifuge tube and centrifuged at 1,800 x g for 1 min. The filtrate was collected in the tube and assayed for SGT (unbound or excess SGT) activity by the HPLC method. The immunobeads in the column were washed with 2 x 150 ul TBS (pH 6.5) plus 0.05% NP-40, then washed with 3 x 150 ul TBS (pH 6.5) to remove unbound proteins. The column was centrifuged at 1,800 x g for 1 min to remove excess liquid. A radioactive SGT substrate mix (24 ul standard substrate mix plus 1 ul 14C-UDP-glucose prepared as described in Method 1.3) was added to the column and incubated with the immunobeads at room temperature for 8 h. The column was centrifuged at 1,800 x g for 1 min, and the filtrate was collected and assayed for radioactivity as a measure of the activity of SGT bound on the immunobeads. RESULTS 1. Induction of SGT 1.1. Substrate - sinapic acid effect on SGT The growth of the B. napus seedlings was not significantly affected (a = 0.05) by the presence of sinapic acid in the growth medium at concentrations lower than 5 mM (Fig. 23). SGT activity reached its peak two days after germination in all treatments except in the case of 10 mM sinapic acid (Fig. 24). There was no obvious difference in the SGT 118 119 cu .* t- I- cu) CL) x I £ CD 8 & M £ .5 TJ CD > T3 C O CD CD CO 50 CO C(-D 73 a o 03 TJ O CD 6fl XI TJ CD 'a. H CD OS se T3 '35 o X +3 •S E 1) d o £ is u — fa 4) TJ s s d CD X .3 O o in X CO c '> © -a CD CD d CD H ©' k_ CD CD US CD CO T3 CO § GO >> 41 CO sl Q B 9 9 .s % d U S .2 ea -d 3 H * O O o GO CD •s i X T3 TJ &b C '3 CD oo S x 0> co bO CD CS TJ d CD •B "M «*- .s c .g £ S CD CD bO o CD SB CD cd XI fa c CD TJ vi_ ^ £ 2 ° .* $ S | (inaiojd 6LU / iB>)d ) AJJAJIOB ojiioeds ios fa x S 120 activity patterns between treatments except that the seedlings treated with 10 mM sinapic acid had a SGT activity peak that was delayed and increased in scale. The patterns of sinapoylglucose accumulation in seedlings grown in the different regimes were also affected. The time of greatest accumulation was shifted from the second day to the fourth day of germination (Fig. 25). 1.2. Light effect on SGT activity Different light conditions did not significantly affect (a = 0.05) either the expression pattern or absolute level of SGT activity (Fig. 26), although the treatments modified other traits, such as morphology and chlorophyll development. 1.3. Heat stress effect on SGT activity The growth of the seedlings was reduced after heat shock at 45° C. But, the SGT activity in the heat-shocked seedlings did not significantly differ (a = 0.05) from the SGT activity in the seedlings grown under normal conditions (Fig. 27). 2. Subcellular localization of SGT No difference in SGT activity was observed whether or not Triton X-100 was in the extraction buffer, but the specific activity was higher in the protein extract prepared without the detergent (Table 5). In both cases, the majority of SGT activity was found in the soluble portion of the extract. In the ultracentrifugation experiment, two different sample loading methods were used to detect an association of SGT with a specific 121 122 40 Days of germination Fig. 26. Effect of different light conditions on SGT activity during the seed germination. The seeds of B. napus cv. Westar were germinated under continuous white light, in complete darkness, and in darkness broken by exposure to UV (15W, 302 nm) light for 15 min each day. The seedlings were harvested at different ages, and the SGT activity in the seedlings was determined by the HPLC method. The data presented are the mean of three replicates. 123 70 i 60 4 25 30 37 45 Temperature (0 C ) Fig. 27. Effect of heat stress on SGT activity during B. napus seed germination. Seeds of B. napus cv. Westar were germinated at 23° C and exposed to short periods of heat stress at 30, 37 and 45° C (see Method section). The seedlings were harvested after 70 h germination, and their SGT activity in the seedlings was determined by the HPLC method. The data presented are the mean of three replicates. 124 Table 5. Recovery of SGT from soluble and insoluble portions of seedling extract prepared with or without 0.2% Triton X-100. The data presented are the means of three replicates. Sample Protein SGT activity Specific activity ( mg/ml) ( pkat/ml) ( pkat/mg protein ) Without Triton X-100 Supernatant 3.45 ± 0.06 68.7 ± 7.4 19.92 Pellet 0.3 With Triton X-100 Supernatant 4.30 ± 0.28 61.0 ±6.8 , 14.19 Pellet 0.2 125 subcellular compartment. When the SGT preparation was mixed with the Percoll medium, SGT activity was distributed evenly cross the different density zones after ultracentrifugation (Fig. 28A); whereas when the seedling extract was layered on the top of the Percoll medium, 90% of the SGT activity was detected in the upper zone where the filtrate density was less than 1.042 g/ml (Fig. 28B). 3. Purification of SGT After a 9-step purification procedure over 40 days, SGT was purified 135-fold with an overall yield of 0.007% of the original proteins and 1% of the total activity in the initial seedling extract (Table 6). It should be pointed out that the increased purity could actually be much higher than 135-fold, since the SGT protein will have lost much of its enzymatic activity during this time-consuming purification process. The SGT proteins were eventually separated from the tremendous amount of seedling proteins by a series of column chromatography techniques (Fig. 29). It appears that there were no isoforms of SGT in the seedlings, since only one peak of SGT activity was resolved in each of the chromatography processes. Size exclusion on the Superdex 75 column proved to be a very efficient step to separate SGT from the other proteins (Fig. 29C, E and Fig. 30A), while at the same time desalting the protein preparation. After the last chromatography step, the protein profiles from SGT-active fractions were analyzed on 10% SDS-PAGE (Fig. 30B). By comparing the protein profiles and the SGT activities in the corresponding fractions, a protein band that appeared to be slightly smaller than 45 kDa on SDS-PAGE, was identified as the likeliest candidate for SGT enzyme. In some fractions, e.g. fraction 126 ( |LU / 6 ) AJISU8Q CM co o co o CO -3- S c* c o o •5 S 2 * S. M f 35 o 9IAJQ o- g 00 *•> co i- LO co o o 2CD —' c«*- o. CD oo T3 ;> :H un CD "•4—* Xi o S PQ H cd o. —I O CO 2 o S 0 CO «S T3 O 15 o u CD CO X CD o CD >-> ts »-U o. a o o CM T3 CD <" CD O- ob 09 «* 6 x o LO CO CNJ o S TO CO ml o" 2 s O ( |LU / 6 ) A1JSU9Q O "H O CM CO O T- 05 1 rt So CD O CM is a O O i- co .5 CD oo C M T3 c°T 3 rt aiAia .3. CD CD o CD X P3 o CD CO i- o CD CD C GO CO 3 ^ o O S3 e O CD M c CO o CD _ CD Qo o © PQ CD o rt O CD 3 1— X2 3 cu •a o CD TO CM 00 CD 3 C 4-1 O .2 5 1 III •i— Q c LO CM 2 co CD o CD cs UOjJOBJJ 00 flj S • •1 rT 4> 127 Table 6. SGT purification schedule and protein profile of each step on SDS-PAGE. Purification step Total Activity Specific Activity Total protein Enriched (pkat) ( pkat/mg protein ) ( mg ) (fold ) Crude extract 250,427 37 6,760 1.0 Protamine absorption 239,733 37 6,445 1.0 Differential surface binding 151,749 45 3,368 1.2 (Macro-prep Hydroxyapatite) Anion exchange 102,368 115 894 3.1 ( Prep-Mono Q ) Gel filtration 62,450 220 284 5.9 (Superdex 75) Affinity binding 31,503 272 116 7.4 ( HiTrap blue dye ) Gel filtration 22,874 401 57 10.8 (Superdex 75) Chromatofocusing 12,515 544 23 14.7 (Mono P) Gel filtration 2,500 5000 0.5 135.1 ( Superdex 75) Recovery (%) 1 0.007 128 (uoipejj / ;e>|d) AjjAipe 19s W3J (wu 082)00 (LUU082)dO W rt s .--3 >—- t/j 'Z2 . - fJ rt 0) rt O- co (WW) DEN (uoipejj / )B>(d) AjiAipe 19s x 1 C rt O s 00 o © X .1 f a 00 X '•0 GO c >> rt CM • —I •g H o c 3 sg O CO cu rt o (wu 082)00 u (wu 082)00 rt •t in PH s 3 ^ O ( |/\|LU ) 9ieu.dsoud wnjssejod 03 X c (WW) DEN 1? 3 O s , 00 DO CU •» I Q c .g 0 co ^< c 3 e S3 <*- 1rt rt O II E £ rt a. ii rt — o OX) s-CU fa O DC DO rtS •*-> *7 <-> o Sc co -—. S- O .3 Ptm, * is u (wu 082)00 (wu 082)00 CN £ O M| PH 8 2 00 129 A 0.5 -i Protein r— 80 SGT activity E c § 0.3 h-40 CM • O 0.0 Fraction No. 47 49 51 53 55 57 T T T T T 40 80 120 160 200 240 280 Time (min) B Fraction No. M 43 45 49 51 53 55 57 M kDa 31.0 21.5 Fig. 30. Chromatogram, SGT activity pattern and protein profile on SDS- PAGE for the SGT preparation from the last step of purification. (A) The chromatogram of the 3rd gel filtration on the Superdex 75 column. The SGT activity of each fraction was measured by the radioactivity method. (B) The protein profiles of fractions from the 3rd gel filtration following electrophoresis through a 10% SDS-PAGE. Same volume (20 u.1) of each fraction was used for each lane on the gel. The gel was stained with Coomassie Blue and the low molecular weight marker mix (M) from Bio-Rad was used as the molecular weight standard. 130 No. 51, the putative SGT was apparently homogeneous (Fig. 30B). Attempts to recover SGT activity from a native PAGE only succeeded when the most sensitive radioisotope method was used to detect SGT activity in the gel slices. Only one SGT activity peak was detected (the gel slice No. 7 for 7% PAGE and the gel slice No. 5 for 10% PAGE), and the recovery of SGT activity was about 1%. 4. Physical and chemical characterization of SGT 4.1. Molecular weight Based on the calibration curve ( Y = - 9,651.68 X + 173,764, R2 = 0.9422 ) generated on the Superose 12 gel filtration column, the native SGT had a molecular mass of approximately 42,000 ± 1,000 Da (Fig. 31 A). SGT is apparently monomeric, if the polypeptide identified on the SDS-PAGE gels corresponds to the enzyme (fraction 51, Fig. 30B). The migration of the putative SGT band was slightly faster than that of ovalbumin (45,000 Da) on SDS-PAGE. Based on the calibration curve ( Ln Y = - 0.890548 Ln X + 13.2172, R2 = 0.9984 ), the denatured SGT protein had a molecular weight of approximately 42,000 ± 1,000 Da (Fig. 31B). 4.2. pi, pH stability and pH optimum A linear pH curve ranged from pH 4.7 to 5.7 was generated when two different pH buffers, 25 mM 2-methyl-piperazine (pH 6.3) and 10% Polybuffer 74 (pH 4.5) were used for chromatofocusing on a Mono P HR 5/20 column. The peak of SGT activity was 131 v« OO ^ 3 O PQ cS Xi Q c« %s P •B BC N s o • 2 H cu GO _> £ O CU w E cS >C E CS _, e E o M-i CU ' 00 CO k_ co > go o ^ a, x 'E c c » «' CU M O ChU X! CO f-H 'a § xa GO — o xl bcS o i_ QL 2 o g E _ X *j c CQ S-rS CU ^ O X ^ X c & o < e co & "2 a. 2 > ^ cS < "S • cu - CuU/ E S- W o 3 O o o o X O o o o « .S 1"c3S O o o o Q .S c O o o o o > r— LO CO •ra £ .S o r_ CS CU ( EQ ) lu JE|noe|0|Aj bo b oo- &aM .SPb u fa oo cu a" 132 detected in fractions at pH 5.1 ± 0.1, which can be taken as its apparent pi value. The stability of SGT at its pi was markedly reduced compared to pH 6.3 and 7.0 (Fig. 32). Sixty percent of the activity was lost within 1 h (Fig. 32B). Neutralization of the elution pH during the chromato focus ing was thus an important step to reduce the loss of SGT activity. Sodium acetate, MES and Tris buffers were used to determine the optimum pH for SGT activity. The optimum pH was found to be pH 6.0 when MES buffer was used (Fig. 33). No SGT activity could be detected when the assay pH was lower than pH 4.5 or higher than pH 8.5. 4.3. Thermal stability and optimum temperature SGT was not stable at room temperature and 92% of the original activity in the SGT preparation was lost after 24 h at room temperature (Fig. 34). It was fairly stable at 4° C, less than 10% of the original activity was lost after two days, but after 1 month at this temperature, less than 10% of the activity remained. SGT was reasonably stable at -20° C or lower. The optimum assay temperature was 32° C (Fig. 35). No activity could be detected when the reaction was taken at a temperature higher than 50° C. 4.4. Effect of metal ions No metal ion was absolutely required as a co-factor for SGT activity (Table 7). Mg++ and 133 Fig. 32. The stability of SGT during storage at different buffer pH. Partially purified SGT from seedlings of B. napus cv. Westar was stored in buffer at - 20° C (A), or kept on ice (B). After a pre-determined period of time, a sample of the protein solution was assayed for SGT activity. After incubation at 37° C for 30 min, the yield of sinapoylglucose was determined by the HPLC method. 134 4.0 6.0 8.0 PH Fig. 33. Determination of the optimum buffer pH for SGT activity. Partially purified SGT from seedlings of B. napus cv. Westar was first incubated with the test buffer at 4° C for 2 h and then mixed with the substrates. After incubation at 37° C for 30 min, the SGT activity of each treatment was determined by the HPLC method. The data presented are the means of three measurements. 135 L|1U0|A| L Asp f a> w CD CD E M 8 L| v L| Z B «3 L| 0 la T3 CU CU •M CU «U co • w (%) AjjAipB ios 9A|ie|ay cu 136 20 40 60 Temperature of enzymatic reaction (0 C ) Fig. 35. Determination of the optimum temperature for the SGT reaction. Partially purified SGT preparation from seedlings of B. napus cv. Westar was pre-incubated at either 4, 23, 32, 37, 42, 50 or 65° C for 15 min, and then mixed with the substrates. After incubation at the same temperature for 30 min, the SGT activity of each sample was determined by the HPLC method. 137 Table 7. Metal ion effects on SGT activity. Partially purified SGT from seedlings of B. napus cv. Westar was incubated with various metal ion solutions at 4° C for 2 h, then mixed with substrates. After reaction at 37° C for 1 h, the SGT activity of each treatment was determined by the HPLC method*. SGT activity ( pkat) Sample Relative activity Mean SD (n = 3) (%) Control 2.02 0.05 100.0 Control+EDTA 2.03 0.18 100.3 Sodium chloride 2.00 0.20 99.1 NaCI+EDTA 2.05 0.23 101.5 Potassium chloride 1.90 0.10 94.1 KCI+EDTA 1.92 0.17 94.9 Calcium chloride 1.95 0.00 96.4 CaCI2+EDTA 1.40 0.02 69.1 Cobaltous chloride 0.94 0.07 46.8 CoCI2+EDTA 1.62 0.11 80.3 Cupric chloride 0 0 0 CuCI2+EDTA 0 0 0 Ferrous sulphate 0.93 0.03 46.0 FeS04+EDTA 1.09 0.10 54.1 Mercuric chloride 0 0 0 HgCI2+EDTA 0 0 0 Magnesium chloride 2.22 0.36 110.1 MgCI2+EDTA 1.99 0.15 98.6 Manganese chloride 2.44 0.12 120.6 MnCI2+EDTA 1.84 0.08 90.9 Zinc sulphate 0 0 0 ZnS04+EDTA 1.10 0.13 54.5 * The pH of various metal ion solutions was adjusted to pH 6.0. The final concentration of metal ion was 0.93 mM and EDTA was 7.4 mM in the reaction solutions. 138 Mn++ (at 0.93 mM) showed a modest stimulatory effect on SGT activity, and this stimulation could be eliminated by addition of EDTA. SGT was inhibited by Zn++, Co++, Fe++, Cu++ and Hg++. The inhibitory effect by Zn++, Co++ and Fe++ was reversible by EDTA, but the inhibition by Cu++ and Hg++ was not. Zn++ was the most inhibitory divalent ion, and the inhibition would occur when the concentration of Zn++ in the SGT preparation was as low as 10 uM (Table 8). 4.5. Effect of reducing reagents Reducing reagent, such as 2-ME or DTT, was required to maintain SGT activity during the purification (Fig. 36). Glycerol ( >5%) was also found to be helpful. SGT activity was completely lost after three days storage at -20° C if there was no reducing reagent or glycerol in the buffer. The investigation for the effects of variety and concentration of reducing reagents showed that ascorbic acid had a slight inhibitory effect on SGT. However, this inhibition could be restored by addition of 2-ME or DTT. An increase in the concentration of DTT showed a stimulation effect. 4.6. Effect of other inhibitors Among various known -SH group inhibitors (A^-ethylmaleimide, iodoacetic acid and p- hydroxyl-mercuribenzoic acid), a serine reactive inhibitor (phenylmethane sulphonyl fluoride) and substrate analogues (3,5-dihydroxybenzoic acid and 4-hydroxyl-3,5- dimethoxy-benzaldehyde azine) tested, only /?-hydroxyl-mercuribenzoic acid (PHMB) showed a significant inhibitory effect (Table 9). 139 Table 8. Inhibitory effects of Cu++, Hg++ and Zn++ on SGT activity. Partially purified SGT from seedlings of B. napus cv. Westar was incubated with the metal ion at 4° C for 2 h and then mixed with substrates. After reaction at 37° C for 1 h, the SGT activity of each treatment was determined by the HPLC method. SGT activity ( pkat) Sample Relative activity Mean SD (n = 3) ( % ) Control 1.38 0.05 100.0 Cupric chloride 0.93 mM 0 0 0 0.093 mM 0 0 0 0.0093 mM 1.34 0.04 96.8 Mercuric chloride 0.93 mM 0 0 0 0.093 mM 0.68 0.62 49.0 0.0093 mM 0.87 0.48 63.3 Zinc sulphate 0.93 mM 0 0 0 0.093 mM 0 0 0 0.0093 mM 0.81 0.52 58.6 140 > QJ »5 TJ . O <+- co o cu CO t3 rJboJ *cor 6 .£ Xi u TJ co Q. 8 a a w -S cu S S r. 7> CXU 4) o • 1—1 co CO TJ £ -*—' cu CZ Ct> c CD . cu QJ CD "C XJ T3 CO o-x co CD k_ cn •2 o *2 c o =3 £\° I "d CD |~ • m « CU cu 03 cz CU o o H bo ~ ccj >, ^ CO a cu *j > a bO \s cz C O CD 9 o rt CO 3 H o TJ cz CU o 00 o «|§ .2 l M) ca .S X £ CJ O 3 TJ cu T. J U CU cj y co cu .3. S 'co w B « c . * .2 f> S x • CO 3 E ^ .£ ( % ) AijAjpe ios 0A!iB|9y 141 Table 9. Effects of other inhibitors on SGT activity. Partially purified SGT from seedlings of B. napus cv. Westar was incubated with each inhibitor at 4° C for 2 h, and then mixed with substrates. After reaction at 37° C for 15 min, the SGT activity of each treatment was determined by the HPLC method*. SGT activity ( pkat) Relative activity Mean SD (n = 3 ) ( % ) Control 2.05 0.06 100 N-ethylmaleimide 2.10 0.28 103 3,5-Dihydroxy-benzoic acid 2.04 0.14 99 4-OH-3,5-dimethoxy-benzaldehyde azine 1.98 0.19 96 lodoacetic acid, Na salt 1.95 0.20 95 Phenylmethanesulphonyl fluoride 1.95 0.28 95 /?-OH-mercuribenzoic acid 0.72 0.28 35 * The final concentration of inhibitors in the reaction solution was 0.8 mM. 142 4.7. Substrate specificity Different nucleotide activated glucose compounds (i.e. ADP-glucose, CDP-glucose, GDP-glucose, TDP-glucose and UDP-glucose) were tested for the specificity of SGT toward sugar donors. Only UDP-glucose and TDP-glucose could be used as a glucose donor (Table 10). On the other hand, SGT accepts a broader range of glucose acceptors, although within the range tested, sinapic acid showed to be the best acceptor. 4.8. Inhibitory effect of substrate analogues A strong inhibitory effect on SGT activity was found when the enzyme was incubated with UDP or TDP (Table 11). UDP-mannose, UDP-xylose and UDP-galactose showed less inhibitory effect, and fB-D-glucose was ineffective. 4.9. Reversibility of SGT SGT has the capability to catalyze both forward and reverse reactions, i.e. SinA + UDPG (or TDPG) <=> SinG + UDP (or TDP) in vitro (Fig. 37). Adding UDP or TDP to a reaction mixture in which formation of sinapoylglucose is being catalyzed by the SGT immediately drove the reaction backwards (Fig. 371). When the reaction solution did not include UDP (only sinapoylglucose and enzyme), the reverse reaction did not proceed until UDP was added (Fig. 3711). This result indicates that there were no other esterases contaminating the SGT preparation to hydrolyze sinapoylglucose, and that is only in the presence of UDP (or TDP), the glucose moiety on sinapoylglucose can be cleaved out, and then transferred to the nucleotide. 143 Table 10. The specificity of SGT for sugar donor and acceptor. Partially purified SGT from the seedling of B. napus cv. Westar was used. For the sugar donor test, sinapic acid was used as the sugar acceptor, and the formation of the product was determined by HPLC. For the sugar acceptor test, UDP-[(3-Z)-glucose (U-14C)] was used as the sugar donor, and the formation of the product was determined by the radioactivity assay. Sugar donor SGT relative activity Sugar acceptor SGT relative activity ( % ) ( % ) UDP-Glucose 100 Sinapic acid 100 TDP-Glucose 96 Ferulic acid 77 ADP-Glucose 0 5-OH-Ferulic acid 39 CDP-Glucose 0 Cinnamic acid 24 GDP-Glucose 0 /?-Coumaric acid 21 Caffeic acid 14 UDP-Galactose 0 . Syringic acid 10 UDP-Mannose 0 UDP-Xylose 0 Glucose 0 144 Table 11. Effects of UDP-glucose analogues on SGT activity. Partially purified SGT from seedlings of B. napus cv. Westar was incubated with each analogue at 4° C for 1 h and then mixed with substrates. After reaction at 37° C for 30 min, the SGT activity of each sample was determined by the HPLC method*. SGT activity ( pkat) Analogues Relative activity Mean SD (n = 3 ) (%) Control 7.77 0.56 100 (3-D-Glucose 8.02 0.35 103 UDP-Galactose 7.01 0.77 90 UDP-Xylose 6.43 0.23 83 UDP-Mannose 3.05 0.12 39 UDP 1.01 0.04 13 TDP 0.78 0.08 10 * The concentration of UDP-Glucose analogue used for inhibitory effect test was 2 mM. 145 cd CD o CO o CO 'E O a, ^ E CD o 1 o CNJ .3 "3. moo h o rt -r- CD o oq o o o ( wu ggg ) eoueqjosqv o 00 8 m CD o m ,-H —' CO ^ ^ E rt o o CM < OQ O -I- -r LO o o ( UJU ggg ) soueqjosqv a *0 146 4.10. Affinity matrices Immobilized UDP-glucose and UDP-glucuronic acid matrices could not bind SGT (Table 12), but an immobilized sinapic acid matrix did show an affinity for SGT. The bound SGT could be disassociated from the affinity matrix by either sinapic acid or NaCl buffer, but it could not be eluted by the UDP-glucose buffer. The SGT binding strength is very weak since SGT could be dissociated by NaCl at a concentration as low as 0.1 M. There was no significant difference between the protein profiles on the 10% SDS-PAGE of the SGT active fractions eluted by either sinapic acid buffer or by NaCl buffer. 4.11. Kinetic parameters The pattern of dependence of the reaction rate on both substrate concentrations showed that SGT displayed Michaelis-Menten kinetics (Fig. 38). Analysis of the kinetic parameters suggested that the SGT catalytic mechanism best fit a "random bi-bi" model. The kinetic parameters obtained by using EZ-Fit program showed that the K[UDP.glucose] was 0.24 mM, the K[sinapic acid] was 0.16 mM, and the Vmax was 10.6 pkat. 5. Amino acid composition and sequence The amino acid composition of the putative SGT band (42 kDa) detected on SDS-PAGE is summarized in Table 13. The protein profiles of the fractions eluting after the most purified SGT fraction showed a 33 kDa protein band (Fig. 30). The amino acid composition of this protein was also analyzed (Table 14). After automated Edman degradation of the N-termini of the intact 42 and 33 kDa polypeptides, both N-terminal 147 Table 12. Affinity binding of SGT by immobilized UDP-glucose membrane cartridge, UDP-glucuronic acid-agarose and sinapic acid-Sepharose. SGT preparation from seedling extract of B. napus cv. Westar was loaded on the affinity membrane cartridge or columns. The unbound proteins were removed by washing with the sample buffer, and the cartridge or columns were then eluted with UDP-glucose, or sinapic acid, followed by NaCl (refer to Method section for details). % Total SGT activity Samples UDPG-membrane UDPGA-Agarose SinA-Sepharose Load 100 100 100 100 Filtrate 97 92 0 0 Elution by UDPG 0 0.7 0 0 Elution by sinapic acid 79 Elution by NaCl 0 4 72 148 3 u Vi An s s O G T on d 00 oCD •73 CD CD • HH JS M-i CM 'C3 O &. CO un CD C rt O rt O "3 ti CoD- PH VH e n TS u 1-1 flu o PH C O CD CD CD s ap i VH CJ c H4—C> M °CO v-. o iff e O 'O CO u E rtC 3 CD «2 s CJ CD Cj C -3 CD rt o rt minati o >• rt CD "O a> *rt io n an d eac t © M CD T3 IZl CD CO 0 C O 3 t/3 CO «*. ~Sa rt UDP- : sta r w plot s o CD -< 42 t3_ E > PQ CD • 4- CO 3 > co Cj & CD • oHH e w na t 4-1 o -J 149 Table 13. Amino acid composition of the putative SGT protein (42 kDa) from seedlings of B. napus cv. Westar. Amino acid * Symbol Composition (%) Histidine H 1.27 Methionine M 2.37 Tyrosine Y 2.51 Arginine R 4.08 Proline P 4.11 Phenylalanine F 4.76 Lysine K 5.05 Serine S 5.11 Threonine T 5.62 Isoleucine 1 6.13 Leucine L 7.10 Glycine G 8.84 Valine V 9.22 Alanine A 10.17 ' Glutamic acid/glutamine E/Q 11.52 Aspartic acid/asparagine D/N 12.13 Total 100 * Cysteine and Tryptophan were destroyed during the sample treatment. 150 Table 14. Amino acid composition of a 33 kDa protein purified from seedlings of B. napus cv. Westar. Amino acid * Symbol Composition (%) Histidine H 0.98 Tyrosine Y 2.99 Arginine R 3.23 Methionine M 3.29 Threonine T 4.01 Phenylalanine F 4.16 Lysine K 4.21 Serine S 5.41 Isoleucine 1 5.42 Proline P 7.01 Leucine L 8.37 Alanine A 8.81 Glycine G 9.08 Valine V 9.29 Aspartic acid/asparagine D/N 9.48 Glutamic acid/glutamine E/Q 14.25 Total 100 * Cysteine and Tryptophan were digested during the sample treatment. 151 sequences were obtained (Table 15). In addition, the putative SGT protein blotted on the PVDF was also digested in situ by trypsin. The digested peptides were separated by HPLC and more than 100 peptides were resolved. Three of the digested peptides were sequenced. Four sequences containing 76 amino acid residues were obtained (Table 15), since one peptide peak yielded two discrete sequences. 6. Protein sequence homology and comparison Both N-terminal and internal sequences of the 42 kDa polypeptide revealed a high degree of homology to the 70 kDa heat shock protein family (HSP70), based on a data base search using the BLAST program. The sequence comparison of the putative SGT and Arabidopsis HSP70 cognate (HSC70, GenBank Accession No. S46302) are shown in Table 16. Only five out of a total 76 amino acid residues were different in the two sequences. The sequences from four internal peptides of the putative SGT covered the sequence range from the N-terminus to the 361st amino acid of the Arabidopsis HSC70. 7. Immunological responses Immunization of rabbits with either the native (for 1 rabbit), or denatured (for 2 rabbits) SGT or denatured 33 kDa protein (for 1 rabbit) failed to elicit the production of specific antiserum against SGT on Western blots. Immunotitration of SGT activity by these sera also failed. Cross reaction in Western blots with the anti-thiohydroximate glucosyltransferase polyclonal and monoclonal antibodies did not reveal any specific response (data not shown). I 152 Table 15. N-terminal and internal amino acid sequences of the putative SGT protein (42 kDa) and 33 kDa polypeptides purified from seedlings of B. napus cv. Westar. Intact peptides blotted on PVDF were used for the analysis of N-terminal sequences. Tryptic peptides were purified by RP-HPLC, and then analyzed for internal sequences. Putative SGT protein ( 42 kDa) N-terminal sequence K G E G P A I G I D L G T T Y S X V G V W Q H Internal sequence PT110-1 T L S S T A Q T T I E I D S L Y E G V D F PT73 DAG S I S G L N V L R PT84 FEE L N M D L F R PT110-2 I Q X L L Q D F X N 33 kDa protein N-terminal sequence I V K S D P V V S F R E T V L E R X V V T 153 Table 16. Alignment of amino acid sequences of the putative SGT (42 kDa) polypeptides purified from seedlings of B. napus cv. Westar and the heat shock cognate protein 70 isolated from A. thaliana. SGT KGEGPAI GIDLGTTYSX VGVWQH HSP 70 1 MSGKGEGPAI GIDLGTTYSC VGVWQHDRVE IIANDQGNRT SGT HSP 70 41 TPSYVAFTDS ERLIGDAAKN QVAMNPVNTV FDAKRLIGRR SGT HSP 70 81 FSDSSVQSDM KLWPFKIQAG PADKPMIYVE YKGEEKEFAA SGT HSP 70 121 EEISSMVLIK MREIAEAYLG VTIKNAWTV PAYFNDSQRQ SGT DAGSISG LNVLR HSP 70 161 ATKDAGVIAG LNVMRIINEP TAAAIAYGLD KKATTVGEKN •k * * SGT HSP 70 201 VLIFDLGGGT FDVSLLTIEE GIFEVKATAG DTHLGGEDFD SGT TL HSP 70 241 NRMVNHFVQE FKRKSKKDIT GNPRALRRLR TSCERAKRTL SGT SSTAQTTIEI DSLYEGVDF FEE LNMDLFR HSP 70 281 SSTAQTTIEI DSLYEGIDFY STITRARFEE LNMDLFRKCM SGT IQXLLQDFX HSP 70 321 EPVEKCLRDA KMDKSTVHDV VLVGGSTRIP KVQQLLQDFF * SGT N HSP 70 361 NGKELCKSIN PDEAVAYGAA VQGAILSGEG NEKVQDLLLL HSP 70 401 DVTPLSLGLE TAGGVMTTLI PRNTTIPTKK EQVFSTYSDN HSP 70 441 QPGVLIQVYE GERARTKDNN LLGKFELSGI PPAPRGVPQI HSP 70 481 TVCFDIDANG ILNVSAEDKT TGQKNKITIT NDKGRLSKDE HSP 70 521 IEKMVQEAEK YKSEDEEHKK KVEAKNALEN YAYNMRNTIQ HSP 70 561 DEKIGEKLPA ADKKKIEDSI EQAIQWLEGN QLAEADEFED HSP 70 601 KMKELESICN PIIAKMYQGA GGEAGGPGAS GMDDDAPPAS HSP 70 641 GGAGPKIEEV D The amino acid sequence of Arabidopsis HSC70 was from GenBank Accession No. S46302. The asterisk indicates non-identical residues between the two amino acid sequences. 154 Since the putative SGT showed a high degree of amino acid sequence homology to HSP70, an anti-pea HSP70 polyclonal serum was examined for cross-reactivity with the SGT protein. It was found that SGT activity could be titrated by the anti-pea HSP70 (Fig. 39). On Western blots, a weak signal from the crude seedling extract and a strong signal from partially purified SGT at 42 kDa position were specifically detected by the anti-pea HSP70 serum (Fig. 40A). As a positive control, the protein extract from heat shocked mouse muscle cell culture showed a signal at 70 kDa. No specific signals were detected by the pre-immune serum. A partially purified SGT preparation obtained earlier from immature green seeds by T. Vogt also showed a strong signal at the same band position on SDS-PAGE as the SGT purified from seedlings in the present study (Fig. 40B). These results are consistent with the possibility that the putative SGT may share some common epitopes with HSP70, since they share in at least the 360 amino acid sequence from the N-terminus. The protein extract from heat shocked seedlings gave a similar immunological response pattern as the control on Western blots (Fig. 41). The 42 kDa band was detected in all treatments whether protease inhibitors were included in the extraction buffer. An HSP70 signal could be detected in all heat-shocked seedling extracts when the anti-pea HSP70 serum was used. The HSP70 band was also detected in all seedling extracts by the monoclonal anti-HSP70 SPA-810. However, no signal was detected at 42 kDa in either partially purified SGT preparations or crude seedling extracts by this monoclonal antibody (Fig. 42). 155 8 H Pre-immune serum • Immune serum 5 10 15 20 Addition volume of serum (uJ) Fig. 39. Immunotitration of SGT activity by anti-pea HSP70 serum. Partially purified SGT preparation, 50 u.1, from seedlings of B. napus cv. Westar was first incubated with different volumes of anti-pea HSP70 polyclonal serum or pre-immune serum (as control) at 4° C for 2 h, and then incubated with 30 u.1 protein A Sepharose 4B beads at 4° C for 3 h. After removal of the protein A beads by centrifugation, the supernatant was assayed for SGT activity by the HPLC method. The data presented are the means of three replicates. 156 CC ^ CM o o LO r-i co LO CU 2 o3 CS cS CD CD CU 4—) o f s cu a. cS f_ cS ! ' 1 fc" o TJ o o 2 1 f r . cS S o +J cS £ o ^ CM .a CL oo X• a _ TJ c o a S bo C 1— H S 0_ DC TJ 0) o u m r CU CS •#-» i "a CU irj co LO CO CD co 3 £ £ co cu P s w > -p e H 3 5 es CX Cu ^ O TJ *3 c T3 3 >^^> O co ca C i-. TJ scS 2 a CX cu o o V o § cS 3 « o £ CS 3 C mi CO CM CO T3 * C cu CU © o cu IB W V— CO C3 Ofj ~ TJ C o S 2 fa CU , cS cS * 3 CZ! * O W 2 " 4) U CU TJ 5 C3 TJ M 4» fc. I CU ft DC O c E .•A5 fc" cu S 3 co £ cS •- g 3 s cS 3 £ ° 2 E e° S 00 lo 2| *2 TJ ft o 1 S 3 CS H s Ea E^ o TJ cCuS 5cS "«o 00 ^ CS TJ o TJ fc- ff cu S i c 3 TJ £ 3 all C cu 3 CX DC TJ cu C M O o 8 s ° -s g 'V o o .s CU U3 0 « CS !- co CO 1-. S 2 £ DO JS a, cs C cu cx e ? 8 o 3 £ t CU T Xi o o o 2 C 0 t ^ £ 5- "a3 ?5 - 5 1 cu O cS U i- q p cS X CO 12 cui QJ M TJ CwU 1 a S c .s ^ rt '53 ^ c * !s CU CQ £ o cu ^ ft 1 co —' "1 I ft o 5 cx fl fc" 22 CO cS TJ co co m ? 3 C CM O q LO rj TJ B 3= _3 CD LO * Cu cS TJ C\j 2 CD CD 00 00 TJ CU CU . CS *£ I CU CU 157 CO VH CD X T3 44 a o CD PH cd » CO |£ 00 CD 1^ M E © CO MH x o 2 cfa 3 9 ed 00 c £ .9 CD o Xi C S •i * CD Oh •HcdB C CD CD 3 -2 00 3 '3 co e O 1 sfa CD .3 CD no 00 g CD co S- CD -71 VH OH c CD cd T3' 1 a z, 43 — Cd E 3 I a i T3 o CO 6 CD ill S ed S -a PH O co c 2 ^ cd J2 CO CD Q cd c CD 8 & CO E e ° > c ca 35 cd CD • PH o CD CD H—> VH o £ g CO ' -a Cu ca CD CJ CO CD B 2 .MS 3- a3 CaDj 2CD CSO CcDo - cu X CD 53 * CO . -a „ 5S 3 5 -rt c O3 cd •HH cd CD CD co CO CD 3 CNJ 3 ill "© . .9 « .9 tS 2CD I CO W -a g Cd rH < C 4- g eq PH " H •a 3 I OX) 4-H oo o CN 'CD 00 00 co O 9 oo CNJ VH -y Q 9 -2 -a 4c-d X oo cdo 7,3g x Q LH CD -a H^'l fl 00 '5 u CJ 3 •a -a 9- 3 CD ^ o CD cd X cd 3 i3 c •3 CO M B CD 00 " •(-> X o E N « co co o« ca O CD >3 cd cd CD fl Ml O fl ; 8 CD' 9 a ca CO c« g cu 00 S > 'E -d « c co O > CD eo CD c Q, sC D M*H .9 U PH a**" co cd -a 2 2 « •39 g "S a LO — CD w « o- •» E cd *-> -a x cd CD -a cd M. CD U cu £ M CO o £ O. CD ed O E OH t3 X c * aS § PH X CD cd PL] T3 co 158 Western blot PVDF stained by with anti-human HSP70 Coomassie blue Heat treatment (0 C) (°C) L8 45 37 30 23 23N P M 23 37 kDa HSP70 97.4 - 66.2 - 45.0 - 31.0 Fig. 42. Comparison of Western blots generated using anti-human HSP70 (StressGen 810) monoclonal antibody and proteins from heat-treated seedling extracts. Seeds of B. napus cv. Westar were germinated at room temperature and exposed to heat stress for 2 h each day for 2 days. The seedlings were harvested after germination for 70 h and extracted with buffer containing a set of protease inhibitors. The extracted proteins were separated by SDS-PAGE and transferred to the PVDF membrane. Proteins were detected by Coomassie blue dye staining and immuno-detection was conducted using anti-human HSP70 monoclonal antibody. Sample L8 was protein extract from heat-shocked mouse muscle cell culture, P was partially purified SGT, M was standard molecular weight markers. All other lane labels represent the temperature of the heat treatment. Sample 23N was a control extraction in which no protease inhibitors were included in the extraction buffer. 159 DISCUSSION 1. Induction of SGT SGT activity in B. napus changes dramatically during plant growth and development, and is therefore under strict regulation. The mechanism of this regulation is, however, unknown. One of the environmental signals that might control the rate of sinapoyl ester formation is the perception of increased UV exposure, since these esters appear to provide at least part of the UV screen in foliar tissues. Another potential control factor might be the changes in the size of the intracellular pool of sinapate, arising either from de novo synthesis in the developing seed, or from sinapine hydrolysis in the germinating seed. To test these ideas, as well as the possible link between SGT and HSP70, young seedlings were exposed to three different regimes: (1) an exogenous supply of sinapic acid in the growth medium; (2) light and UV-B irradiation treatment during the growth of seedlings; and (3) heat shock stress on seedlings. The results showed that only the exogenous supply of a high concentration (e.g. 10 mM) of sinapic acid could induce an increase in SGT activity in the seedlings. The concentration of sinapine in the seeds of B. napus was - 40 pmol/g seed, and ~ 4 pmol/g (fresh tissue weight) of sinapine was detected in the young seedlings two days after germination. The level of free sinapic acid detected in 2-day-old seedlings was ~ 0.4 pmol/g. About 90% of sinapine was hydrolyzed and converted to other sinapoyl esters. Observations on the metabolites showed that normally expressed SGT activity (-500 pkat/g fresh tissue, 2-day-old seedling) in Brassica was sufficient to reesterify the endogenous free sinapic acid released from sinapine during the seed germination. The existing SGT activity was also sufficient to 160 detoxify exogenous sinapic acid taken up from the growth medium (containing up to 5 mM sinapic acid) by forming sinapoylglucose, a less toxic and more metabolically active form. As the consequence of feeding sinapic acid to the seedlings, the pattern of accumulation of sinapoylglucose changed (Fig. 25). The increase in the concentration of sinapic acid in the growth medium increased the level of transiently accumulated sinapoylglucose in the tissues. The peak of SGT activity was also shifted from the second day to the third or the fourth day of the germination when sinapic acid concentration was greater than 5 mM. This may be due to the high concentration of sinapic acid in the cell stimulating and retaining the SGT activity. Since the induction of SGT activity by 10 mM sinapic acid was less than 2-fold, the response is obviously different from the induction of UDP-glucose:salicylic acid glucosyltransferase in the roots of 5-day-old oat seedlings, in which the activity of the glucosyltransferase was induced 23-fold by incubating roots with 0.5 mM salicylic acid (Yalpani et al, 1992a and 1992b). The induction of the glucosyltransferases by exogenous supply of substrate has been demonstrated in several other species as well, such as the induction of UDP- glucose:cinnamic acid glycosyltransferase in a bean cell culture by cinnamic acid (1 mM) (Edwards et al, 1990); the induction of UDP-glucose:salicylic acid glucosyltransferase (6-fold) by salicylic acid (13.5 - 27 pg/g tissue) in tobacco leaves (Enyedi and Raskin, 1993) and in a cell culture of Mallotus japonicus (Tanaka et al, 1990), and the induction of UDP-glucose:isoflavone 7-0-glucosyltransferase (5-fold) by spraying a diphenyl ether herbicide (Acifluorfen) on soybean leaves (Cosio etal, 1985). 161 Environmental stresses are also known to induce glucosyltransferases in plant system. Examples of this type of induction have been reported in several species, including the induction of UDP-glucose:cinnamic acid glycosyltransferase by fungal elicitor in bean and alfalfa cell cultures (Edwards et al, 1990); the induction of UDP-glucose:salicylic acid glucosyltransferase (7-fold) by inoculation of tobacco mosaic virus on tobacco leaves (Enyedi and Raskin, 1993); the induction of UDP-glucose:quercetin glucosyltransferase (3-fold) and UDP-glucuronic acid:quercetin glucuronosyltransferase (240-fold) by irradiation of a dill cell culture with UV-B light (Mohle et al, 1985). A study on feruloylglucose metabolism in cell suspension cultures of Chenopodium rubrum found that the activity of UDP-glucose:hydroxycinnamic acid glucosyltransferase was increased by 4-fold when the cell culture was grown in darkness compared to culture grown in light (Bokern et al, 1991). However, in B. napus seedlings, the activity of SGT was not inducible by either UV irradiation or heat shock stress. 2. Subcellular localization of SGT The Percoll ultracentrifugation system provides an excellent self-generated density gradient in situ which can be used to study the subcellular localization of SGT protein. Using this system, SGT was found to behave like a soluble protein in the cytosol. This conclusion is based on the results from two different experiments. If SGT is associated with membranes, the proportion of the SGT activity in the membrane would be included in the extract with detergent. The SGT activity should be higher than that in the extract without the detergent after the separation of soluble and insoluble portions of the extract 162 by centrifugation. In the present study, the total SGT activity in the soluble portions of two extracts remained the same, while the specific activity in the soluble portion of the extract treated with Triton X-100 was lower. This suggests that some membrane-bound non-SGT proteins were released by Triton X-100 and the total protein content in the soluble portion of the extract was thereby increased, causing the decrease in the SGT specific activity. If SGT is membrane-bound or associated with a subcellular organelle, no matter how the protein preparation is loaded in (or on) the Percoll medium, after ultracentrifugation, SGT activity should be located in a discrete region where the medium density corresponds to the density of the target organelle. In the present investigation, the fact that SGT did not penetrate into a higher density zone when it was layered on the top of the Percoll medium indicates that SGT is soluble in the cytosol and its density is lower than any organelle's densities. Consistent with this, when the SGT preparation was mixed with the Percoll medium before centrifugation, the SGT activity was detected in all fractions of the resulting gradient. Most glucosyltransferases reported for the glucosylation of the secondary products appear to be freely soluble, and some of them have been confirmed to be located in the cytosol, such as UDP-glucose:anthocyanidin glucosyltransferase in Hippeastrum and Tulipa (Hrazdina et al, 1978), UDP-glucose:o- coumaric acid glucosyltransferase in Melilotus alba (Oba et al, 1981), UDP- glucosexoniferyl alcohol glucosyltransferase in Picea abies seedlings (Schmid et al, 1982), UDP-glucose:cyanohydrin glucosyltransferase in Triglochin maritima seedlings, (Hosel and Schiel, 1984), UDP-glucose:diosgenin glucosyltransferase in Solanum melongena (Paczkowski and Wojciechowski, 1994), UDP-glucose:esculetin 163 glucosyltransferase in Hordeum vulgare (Werner and Matile, 1985) and UDP-glucose: p- hydroxybenzoate glucosyltransferase in Lithospermum erythrorhizon (Yazaki et al, 1995). Membrane-bound or associated glucosy transferases are not unknown, but they are usually involved in the formation of glycolipids, polysaccharide and/or glycoproteins. These membrane-bound glucosyltransferases have been found to be associated with the endoplasmic reticulum, the Golgi membranes and the mitochondria (Durr et al, 1979), the plasma membranes (Hartmann-Bouillon and Benveniste, 1978 and 1987; Ullmann et al, 1987; Ury et al, 1989), and the microsomes (Drake et al, 1990 and 1991; Warnecke and Heinz, 1994). After the examination of the pathway for the enzymatic synthesis of polymethylated flavonol glucosides in vivo in Chrysosplenium americanum, Ibrahim et al. (1987) proposed the existence of a multi-enzyme system, including a set of aggregated and membrane-associated O-methyltransferases and O-glucosyltransferases. Although such a multi-enzyme system has not been isolated yet, there is good, albeit indirect, evidence to support this concept. In a study of sinapoylmalate biosynthesis in R. sativus, sinapoylglucose was proposed to be transported from cytoplasm into the vacuoles (Strack and Sharma, 1985; Sharma and Strack, 1985; Strack et al, 1986), a result which is consistent with the present data. 3. Physical and chemical properties of SGT Only a single SGT activity peak was detected in all fractions of the various chromatographic separations tested, including anion exchange, hydroxyapatite differential surface binding, hydrophobic interaction (not included in the purification 164 procedure), sinapic acid-affinity chromatography interaction (not included in the purification procedure), blue dye affinity binding, chromatofocusing and size exclusion. Recovery of SGT activity from a native PAGE gel also revealed a single peak. These results suggest that SGT has no isoforms. Detailed characterization of SGT purified from the 65-h-old seedlings showed that this enzyme shares many features with the SGT from R. sativus seedlings (Strack, 1980; Nurmann and Strack, 1981; Mock and Stack, 1993) and the wild carrot cell culture (Halaweish and Dougall, 1990), as well as with many other UDP-glucose-dependent glucosyltransferases from numbers of species, such as Ipomoea batatas (Shimizu and Kojima, 1984), Gardenia jasminoides (Mizukami et al, 1985), Cinchona succirubra (Khouri and Ibrahim, 1987), Euonymus alatus (Ishikura and Yang, 1994). These are all monomer species with a molecular mass of 40,000 - 50,000 Da, and require a sulfhydryl reducing reagent to maintain enzymatic activity. They have no absolute requirement for a divalent cation or other co-factors, and they are strongly inhibited by /7-hydroxyl-mercuric benzoic acid, by UDP and by some divalent ions, such as Zn++, Cu++ and Co++. In the present study, reducing reagents and glycerol were found to be essential to stabilize SGT during the protein purification. However, addition of ascorbic acid as an antioxidant to the enzyme preparation showed a negative effect. This result is in contrast to the behavior of UDP-glucose:cyanidin 3-0-glucosyltransferase from Daucus carota cell cultures (Petersen and Seitz, 1986). In Daucus carota cell culture system, ascorbic acid could stimulate the activity of the desalted glucosyltransferase preparation up to 14-fold, 165 and it was therefore considered to be a co-factor. This stimulatory effect was not observed when other reducing reagents, e.g. glutathione, 2-ME, DTT or cysteine, were included in the reaction. A similar effect was reported for the preparation of the hydrochinone- glucosytransferase from Impatiens balsamina (Miles and Hagen, 1968). It is worth mentioning that the negative effect of ascorbic acid on SGT in the present study could be restored by addition of 2-ME or DTT. The mechanism of this phenomenon is unclear at this point. A comparison of the effects of known -SH group inhibitors on SGT activity, showed that only PHMB was inhibitory. This may be an artificial result, since the effect of 2-ME (8.3 mM) in the reaction mixture was not taken into account. Previous studies have suggested that reducing reagents, such as 2-ME and DTT, could completely reverse the inhibition of glucosyltransferase activity produced by -SH group inhibitors (Strack, 1980; Guo and Poulton, 1994; Cheng et al, 1994; Ishikura and Yang, 1994), which could explain the apparent ineffectiveness of A^-ethylmaleimide and iodoacetic acid. The PHMB may react with the target thiol (s) on SGT to form a more stable product than those formed by N- ethylmaleimide and iodoacetic acid. Overall, however, the effects of reducing reagents and -SH group inhibitors indicate that one or more -SH groups in the SGT protein plays a pivotal role in its catalytic function. SGT catalyzes a freely reversible reaction. This reversibility of UDP-glucose-dependent glucosyltransferases has also been found in enzymes from Petroselinum (parsley) cell 166 suspension cultures (Sutter and Grisebach, 1975), R. sativus (Strack, 1980), Quercus robur (Gross, 1983), Zea mays (Leznicki and Bandurski, 1988), Phaseolus aureus (Drake etal, 1991) mdPicea abies (Schmid and Grisebach, 1982; Heilemann and Strack, 1991). The implication of such reversibility is that the glucose ester linkage is roughly as energy-rich as the glycosidic linkage between (3-Z)-glucose and UDP. This equivalence may be important for the preservation of energy during the metabolism of phenylpropanoid compounds. A similar reversibility has also been demonstrated for the reaction catalyzed by hydroxycinnamic acid-CoA-dependent transferases (Rhodes and Wooltorton, 1976; Ulbrich and Zenk, 1979 and 1980; Strack etal, 1986). The kinetic properties of SGT purified from B. napus seedlings suggest that the SGT catalytic mechanism best fits the "random bi-bi" model, whereas the enzyme from R. sativus was reported to use an "ordered bi-bi" mechanism (Mock and Strack, 1993), in analogy to the UDP-glucose:flavonoid glucosyltransferase in C. americanum (Ibrahim et al, 1987). The pattern of SGT binding to the substrate and substrate analogue ligands on affinity matrices supports the hypothesis that the catalytic mechanism used by SGT from B. napus may differ somewhat from that of the R. sativus enzyme The inhibitory effect of UDP-glucose analogues on SGT activity suggests that one binding site of SGT is specific for the nucleotide moiety of the sugar donor. However, UDPGA-agarose and immobilized UDP-glucose both failed to bind SGT under the conditions tested. On the other hand, SGT could be bound on an immobilized sinapic acid matrix, and could also be eluted by sinapic acid or NaCl, but not by UDP-glucose. This phenomenon suggests 167 that catalytic mechanism of SGT from B. napus seedlings does not fit "ordered bi-bi" model as proposed in R. sativus, in which SGT is first bound to UDP-glucose, then sinapic acid, and followed by a sequential release of sinapoylglucose and UDP. The nature of SGT catalytic mechanism may be more complicated than previously thought. 4. Amino acid sequence homology and immunological responses The partial amino acid sequences derived from purified SGT contained a total of 76 amino acid residues, including the N-terminus. Surprisingly, these sequences almost perfectly match the sequences of plant HSP70 proteins. Alignment analysis of the partial sequences obtained from the putative SGT peptides showed that only five of 76 residues differed from the corresponding regions of Arabidopsis HSC70. Although the N-terminal sequence also showed some homology with glucose regulated proteins (GenBank Accession: P22010, Q03684 and PI 1021) or ADP-ribosylation factors (GenBank Accession: P32889, P16587 and P35676), the degree of homology in these cases was much lower. There are several arguments that point to the existence of a HSP70-related SGT protein. First the purification procedure developed for SGT is a 9-step comprehensive process so that the possibility of random co-purification of SGT with a partially degraded HSP70 must be regarded as low. Second, the size exclusion data showed that the native form of SGT had a mass of - 42 kDa, and a major protein band of this size was also found on the SDS-PAGE of the most highly purified preparation. The appearance of this band was found to be closely correlated with SGT activity in the fractions of the gel filtration (Fig. 30). Third, the peptide sequencing of the 42 kDa 168 protein band did not show any strong evidence that the protein was contaminated with a second species. The only other protein candidate seems to be the 33 kDa band observed in fractions that displayed a lower SGT activity. However, the elution pattern of this 33 kDa band did not correlate well with the pattern of SGT activity in the gel filtration fractions. Its size was too small for it to qualify as a monomeric species, but it was too big to be a subunit of a homo-dimer because the molecular mass of the native SGT was about 42 kDa. The N-terminal sequence (21 residues) of the 33 kDa peptide had 70-91 % of identity with the internal sequences (amino acid 550 - 590) of known protein elongation factors (GenBank Accession No. P29691, PI3060, P09445, PI3639 and P05086). The 33 kDa protein is therefore not a degradation product from the 42 kDa protein. It thus appears that either (a) the 42 kDa band is a unique SGT gene product, or (b) this band represents a partially degraded HSP70 gene product that possesses SGT catalytic activity. If this 42 kDa protein is a unique SGT polypeptide, why does its sequence have such a high degree of homology with HSP70? If it is not a unique SGT polypeptide, which protein band is SGT ? The sequence of the 42 kDa protein band does not show homology with any published glucosyltransferase sequences (Tomaschewski et al, 1985; Furtek et al, 1988; Ralston et al, 1988; O'Reilly and Miller, 1989; Wise et al, 1990; Szerszen et al, 1994), but since other glucosyltransferases acting on this class of aglycones have yet to be sequenced, it is uncertain how much homology could be anticipated. 169 In total, four rabbits were immunized for the production of anti-SGT polyclonal antibodies, but without success. The immunogenicity of purified SGT to rabbits is apparently very weak. This phenomenon had also been observed in attempts to raise SCT antibodies (Vogt et al, 1993). Both poly- and monoclonal anti-thiohydroximate S- glucosyltransferase antibodies (GrootWassink et al, 1994) failed to cross-react with SGT, indicating that these two glucosyltransferases do not share common epitopes. Given the high degree of homology observed, it is not surprising that SGT strongly cross- reacted with a pea HSP70 polyclonal antiserum which had been raised against an amino- terminal fragment (amino acids 9-183) of pea HSP70 (DeRocher and Vierling, 1995). In order to investigate the possibility that the 42 kDa putative SGT might arise from proteolytic degradation of HSP70, a set of protease inhibitors was added to the extraction buffer. These failed to eliminate the appearance of the 42 kDa band in crude tissue extracts, strongly suggesting that the seedlings contained a HSP70-independent 42 kDa protein. On the other hand, the commercial monoclonal anti-HSP70 antibody (SPA-810) only recognized HSP70 (may be some HSC70s) bands in the seedling extracts. The partial epitope map developed for human HSP70 suggests that SPA-810 recognizes an epitope centered on residues 436-503 (StressGen, personal communication, 1995). Since the putative SGT is predicted to consist of about 386 amino acid residues, the failure of SPA-810 to recognize the SGT band is consistent with its target specificity. 170 On balance, the evidence appears to favor the hypothesis that the 42 kDa polypeptide purified from B. napus seedlings represents an independent protein. It is neither heat- inducible nor does it appear to be degraded from HSP70. Chromatographically, it fits the SGT activity elution pattern. It may be relevant that one of the attributes of the HSP70 protein is its ATPase activity (Schlossman et al, 1984; Rothman and Schmid, 1986). This ATPase activity has been shown to be located in the region between the N-terminus and amino acid residue 385 (Chappell et al, 1987). The biochemical and structural analysis of the mammalian cytosolic HSP70 has shown that an N-terminal 44 kDa proteolytic fragment retained ATP binding and ATPase activity (Flaherty et al, 1990; Flynn et al, 1991; Wilbanks et al, 1995). Whether this phenomenon is also correlated with a SGT catalytic activity, or whether SGT is evolved from HSP70 family is not yet known. The linkage between SGT and HSP70 is still an open question and will remain so until the completion of cloning and heterologous expression of a SGT cDNA. 171 Chapter V ATTEMPTS TO IDENTIFY A GENE ENCODING SGT 172 INTRODUCTION Even though sinapine and sinapoylmalate are the most abundant phenolics in seeds or vegetative tissues of most Brassica species, the physiological roles of these compounds are not clear yet. In considering a plant genetic engineering route to low sinapine germplasm, it is important to consider the consequences that may result from elimination of sinapine or sinapoylmalate by blocking its biosynthesis pathway. Fortuitously, there is a relevant model of such a situation available in the crucifer, Arabidopsis thaliana. An Arabidopsis mutant deficient in conversion of ferulic acid to 5-OH-ferulic acid in the general phenylpropanoid pathway has been shown to fail to accumulate sinapic acid esters in either the seeds or vegetative tissues (Chappie et al, 1992). Despite this deficiency, the mutant germinated and developed as rapidly as wild-type. This implies that it should be possible to genetically eliminate or reduce sinapine from Brassica seeds without a serious deleterious effect on plant growth. Identification and cloning of the SGT gene would not only facilitate the application of antisense technology to create a new genotype of B. napus with a reduced content of sinapine in seeds, but also provide the tools for more detailed analysis of the regulation of sinapine synthesis, metabolism and physiological function(s) in plants. To date, only a few genes encoding UDP-glucose-dependent glucosyltransferases have been described. They have been isolated from various organisms, including plants. The examples known to date are the genes encoding a T4 glucosyltransferase from bacteriophage (Tomaschewski et al, 1985), an ecdysteroid glucosyltransferase from baculovirus 173 Autographa californica (O'Reilly and Miller, 1989), a zeaxanthin glucosyltransferase from the non-photosynthetic bacterium Erwinia herbicola (Hundle et al, 1992), a flavonoid glucosyltransferase from maize (Zea mays) (Furtek et al, 1988; Ralston et al, 1988) and from barley (Hordeum vulgare) (Wise et al, 1990), an IAA glucosyltransferase from maize (Szerszen et al, 1994), the multiple secondary plant product glucosyltransferase from cassava (Manihot esculenta) (1994, EMBL Accession No. X77459), and a sterol glucosyltransferase from oat {Avena sativa) (Warnecke and Heinz, 1994). In mammals, ihe glucosylation of polar substrates is a detoxification process, and mainly catalyzed by UDP-glucuronosyltransferases which are associated with liver microsomes. Genes coding for glucuronosyltransferase have been described from several species, including humans (Jackson et al, 1987), rat (Mackenzie, 1987) and mouse (Kimura and Owens, 1987). Although glucosyltransferases share many physical and chemical properties, their aglycone specificity could have a substantial impact on the structure of the gene. A comparison of amino acid sequences of insect ecdysteroid glucosyltransferase, maize flavonoid glucosyltransferase and three mammalian glucuronosyltransferases indicated that glucuronosyltransferases showed strong similarities among themselves but not with glucosyltransferases (Hundle et • al, 1992). The genes encoding flavonoid glucosyltransferase isolated from maize and from barley also showed an overall 75% similarity (Wise et al, 1990). These results suggested that regions of conserved sequence might be useful for the isolation of the same gene from other species, but it is difficult to 174 use the DNA sequence information from one specific glucosyltransferase to clone the gene coding for another glucosyltransferase which uses a different aglycone as the substrate. OBJECTIVE N-terminal and internal sequences of the putative SGT purified from seedlings of B. napus showed a high degree of homology with the HSP70 protein family. The results of immunological analysis of SGT expression, using anti-HSP70 antibodies, suggest that an independent, HSP70-related, 42 kDa protein is produced in B. napus. The objective of this study was to investigate whether there is a separate gene coding for SGT, or whether SGT is a modified gene product derived from hsp70/hsc70. MATERIALS 1. Plant material Seeds of B. napus cv. Westar were from Agriculture and Agri-Food Canada, Saskatoon Research Centre. 2. Chemicals TRIzol RNA isolation reagent, Moloney murine leukemia virus reverse transcriptase (M- MLV-RT), First Strand cDNA Synthesis kit, Random Primers DNA Labeling kit, 1 kb DNA Ladder were purchased from BRL, Life Technologies, Inc. Avian myeloblastosis virus reverse transcriptase (AMV-RT), Tag DNA polymerase, polymerase chain reaction 175 (PCR) substrate mix, Xba I, EcoR I, and BamH I were purchased from Pharmacia Biotech. GeneClean DNA purification Kit was purchased from Bio 101. DNA oligo nucleotide synthesis kit was from Beckman. Agarose and Zeta-probe GT membrane were from Bio-Rad. HSP70 cognate of spinach chloroplast envelope protein cDNA clones (sce70N and 5ce70C) and Brassica immature seed HSC70 cDNA clone {bhsclQ) were a gift from Dr. K. Ko, Queen's University, Kingston, ON, Canada. The sce7(M contains the N-terminal half of scelQ, while the scelQC contains C-terminal half. All other chemicals were obtained from Sigma, unless otherwise indicated in the text. METHODS 1. Synthesis of BEK oligo-d(T)20 oligo adapter Poly (T) 40-mer oligo including BamH I, EcoR I and Kpn I (BEK) sites was synthesized on a Beckman Oligo 1000M instrument according to the manufacturer's instructions. The sequence of the BEK oligo adapter was: 5' CG GGATCC GAATTC GGTACC TTTTTTTTTTTTTTTTTTTT 3' BamH I EcoR I Kpn I 2. Synthesis of SGT specific N-terminal primers Degenerate primers, N-l and N-2, for hsp70/hsc70 and sgt PCR amplification were synthesized at the Nucleic Acid Protein Service Unit, University of British Columbia. The 25-mer oligo sequences were based on the N-terminal sequence of B. napus SGT (42 kD protein). The positions and the sequences of primers were as follows: 176 Position at N-terminus: 5' KGEGPAIGIDLGTTYSXVGVWOH 3 N-2 N-l N-l degenerate sequence: 5' GCTCTAGA GTI GGI GTI TGG CAA CA 3 Xba I site G N-2 degenerate sequence: 5' GCTCTAGA AAA GGI GAA GGI CCI GC 3' Xba I site G G The oligo was precipitated with ammonia-butanol after synthesis, and then dissolved in 500 pi water. 3. Isolation of total RNA from seedlings Seedlings grown at 23° C in the culture dishes were harvested at 52, 57 and 66 hours post-germination, respectively. RNA was isolated from the seedlings using TRIzol reagent according to the supplier's instructions as described below. Seedlings (~ 1 g) were ground with liquid nitrogen in a cold mortar, then transferred (~ 0.5 g tissue powder) into a 2 ml screw-cap microcentrifuge tube. TRIzol reagent (1.4 ml) was immediately added to the tube and mixed by shaking. The extract was left at room temperature for 10 min, then centrifuged at 12,000 x g and 4° C for 10 min. The supernatant was transferred into a fresh 2 ml screw-cap microcentrifuge tube. Chloroform (0.28 ml) was added and the tube was capped tightly, then shaken by hand for 15 sec and incubated at room temperature for 5 min. The extract was centrifuged at 12,000 x g and 177 4° C for 15 min. The aqueous (top layer) part (~ 60% of TRIzol volume) was transferred into a fresh 2 ml screw-cap microcentrifuge tube, isopropyl alcohol (0.7 ml) was added, mixed and incubated at room temperature for 10 min. After centrifugation at 12,000 x g and 4° C for 10 min, the RNA pellet was washed with 1.4 ml 75% ethanol by vortexing, then centrifuged at 7,500 x g and 4° C for 5 min. After removing the supernatant, the RNA pellet was air-dried by inverting the tube for 8 min. The RNA pellet was dissolved in 50 pi diethyl pyrocarbonate (DEPC) treated water and left for 10 min at room temperature with occasional mixing. The RNA was heated at 60° C for 10 min with occasional mixing. The dissolved RNA was stored at -70 ° C for further use. The purity and yield of isolated RNA were determined by using the GeneQuant RNA/DNA Calculator (Pharmacia Biotech). 4. Synthesis of single strand cDNA from total RNA by AMV-RT and M-MLV-RT RNA samples (4 pi, ~ 30 pg RNA) from seedlings were heated at 70° C for 5 min, then placed on ice. Substrates were added as follows: 5.0 pi 5 x first strand buffer (375 mM KC1, 15 mM MgCl2 and 250 mM Tris/HCl, pH 8.3), 2.0 pi 0.1 M DTT, 2.0 pi BEK oligo-d(T)20 (3.6 pg), 2.5 pi human placenta RNase inhibitor (HPRNI) (New England), 5.0 pi 2 mM dNTP, then 2.5 pi AMV-RT (19 U/pl, 2.5 pi = 48 U) or 3.0 pi M-MLV-RT (200 U/pl). The mixture was incubated for 90 min at 42° C for AMV-RT or at 37° C for M-MLV-RT. After the reaction was completed, 30 pi 0.1 M NaCl in 40 mM EDTA, 21 pi 2 M NaOH and 30 pi water were added and incubated at 46° C for 30 min. The reaction was mixed with 21 pi 1 M HC1, 21 pi 1 M Tris/HCl, pH 8, 153 pi 4 M NH4OAc 178 and 612 ul 95% ethanol, then left at -20° C overnight. The reaction was warmed to room temperature, then centrifuged at 16,000 x g for 20 min. The pellet (single strand cDNA, scDNA) was washed with a mixture of 50 ul 2 M NH4OAc and 100 ul 70% ethanol, then centrifuged at 16,000 x g for 20 min. The pellet was washed again with 200 ul ice-cold 70% ethanol, then centrifuged at 16,000 x g and 4° C for 20 min. The pellet was air-dried for 10 min, then resuspended in 50 u.1 water and stored at -20° C for further use. 5. Amplification of hsp70/hsc70 and sgt by PCR Standard PCR procedures were followed and performed on a Thermal Cycler (PHC-3, Techne). The PCR reaction included 0.003-0.15 \xg cDNA (the scDNA synthesized as above) and 12.3 ul PCR reaction mix, which included 5 ul 10 x PCR reaction buffer (500 mM KC1, 20 mM MgCl2 and 200 mM Tris/HCl, pH 8.4), 2 ul 2.5 mM dNTPs, 2 ul 5' primer (N-l 35 pmol/ul or N-2 27 pmol/ul), 3 ul 3' primer "(BEK oligo-d(T)20 15 pmol/ul), 0.3 u.1 Taq (5U/ul) and distilled water (made up to a total volume 50 ul) in a 0.5 ml PCR reaction tube. Mineral oil (50 ul) was layered on the top of the reaction mixture. Reaction tubes were preheated at 94° C for 4 min. The PCR cycling program consisted of two short cycles designed to allow initial binding of degenerate primers and synthesis of their corresponding strand, then followed by a long cycle for the PCR amplification of specific products. The short cycling program consisted of 3 cycles of denaturation at 94° C for 1 min, annealing at 50° C for N-2 primer (49° C for N-l primer) for 1 min and extension at 72° C for 1.5 min, then followed by 3 cycles of denaturation at 94° C for 1 min, annealing at 37° C (BEK oligo d(T)20 primer) for 1 min and extension at 72° C for 179 1.5 min. The long cycling program consisted of 30 cycles of denaturation at 94° C for 1 min, annealing at 65° C for 1 min and extension at 72° C for 1.5 min, followed by 1 cycle of extension at 72° C for 6 min. 6. Agarose gel electrophoresis PCR products were analyzed by 1-1.5% agarose mini gel electrophoresis using a TAE buffer (40 mM Tris/HCl, 20 mM sodium acetate and 1 mM EDTA, pH 7.4) system. The PCR product (10 ul) was mixed with 3 ul loading buffer (50% glycerol, 0.1 M EDTA, pH 8, 1% SDS and 0.1% bromophenol blue) and loaded on an agarose gel containing ethidium bromide at a concentration of 0.75 u,g/ml gel. The gel was electrophoresed at constant voltage (80 V) for 60-90 min. After electrophoresis, the DNA fragments were visualized on UV illuminator and photographed on the Gel Documentation System (IS- 500, Alpha Innotech Corp.). 7. Labeling hsp70/hsc70 probes with ct-32P-dATP by random primer labeling system 7.1. DNA preparation Spinach HSP70 cognate cDNA clones sce70N and sce70C and Brassica HSC70 cDNA clone bhsc70 were purified by the boiling method for mini-preparation of plasmid DNA (Engebrecht et ai, 1993). The sce70N and sce7QC were digested with EcoR I and the bhsc70 was digested with £coR I and Hind III using the Pharmacia One-Phor-All buffer system. Digested DNA was fractionated on a 1.5% agarose gel. Insert DNA fragments (1.1 kb from sce70N, 1.6 kb from sce70C and 2.1 kb from bhsc70) were sliced out of the 180 gel and purified using the Bio 101 GeneClean kit. The purified inserts were concentrated by SpeedVac and stored until further use as a probe. 7.2. Random primer labeling The probe DNA was labeled with 32P by using the BRL random primer labeling system. The probe DNA (50 ~ 150 ng 15 pi) was denatured by heating at 100° C for 5 min, then snap-cooled on ice. Labeling components were added as follows: 7.5 pi random primer buffer mix (0.67 M Af-[2-hydroxyethyl] piperazine-A^-2-ethanesulfonic acid [HEPES], 0.17 M Tris/HCl, 17 mM MgCI2, 22 mM 2-ME, 1.33 mg/ml bovine serum albumin (BSA) and 18 OD260 units/ml oligodeoxyribonucleotide primers [hexamer fraction], pH 6.8), 1 pi 0.5 mM dCTP, 1 pi 0.5 mM dGTP, 1 pi 0.5 mM dTTP, 3.5 pi water, 1 pi Klenow fragment (3 U/pl) and 5 pi [a-32P]-dATP (3000 Ci/mmol, 10 pCi/pl, Amersham Corp.). The reaction was incubated at room temperature for 2 h, then stopped by adding 5 pi 0.2 M Na2EDTA, pH 7.5, and mixed with 70 pi TE buffer (10 mM Tris/HCl and 1 mM EDTA, pH 8.0). The mixture was spun through a 1 ml syringe column packed with Sephadex G-50 to remove unincorporated nucleotides. The radioactivity of the labeled probe was measured by liquid scintillation counting. Labeled probes were then used in further Northern and Southern blots. 8. Northern blot analysis of seedling RNA Northern blot analysis was performed using the protocol of Davis et al. (1994). 181 8.1. Electrophoresis of RNA An agarose slab gel (150 x 100 x 5 mm) was prepared by mixing 1.25 g agarose with 106.3 ml DEPC treated distilled water and 12.5 ml 10 x MOPS (0.1 M Na acetate, 10 mM EDTA and 0.4 M 3-[A^-morpholino]-propanesulfonic acid, pH 7.0). The gel mix was boiled, then cooled to 50° C and mixed with 6.2 ml formaldehyde. The gel mix was poured into a gel cast in the fumehood and left until polymerized. The RNA sample (-10 pg in 5 pi DEPC treated water) was mixed with 6 pi loading dye (20 pi 10% xylene cyanol and bromphenol blue each in 1 ml deionized formamide), 2 pi formaldehyde and 0.6 pi 10 x MOPS, and heated at 60° C for 10 min, then placed on ice. After cooling, 1 pi ethidium bromide (1 mg/ml) was added to the sample and the mixture was loaded on the gel. The gel was electrophoresed with 1 x MOPS at constant voltage (70 V) for ~ 1.5 h in the fumehood. The RNA bands were visualized on the UV illuminator and photographed by the Gel Documentation System (Canberra-Packard Canada Ltd.). 8.2. Capillary transfer with sodium citrate buffer After electrophoresis, the gel was rinsed with water and incubated in 10 x sodium citrate buffer (SSC, 1.5 M NaCl and 150 mM tri-sodium citrate, pH 7.8) with gentle agitation at room temperature for 20 min. Zeta-Probe GT membrane was re-hydrated with distilled water, then soaked in 10 x SSC. The gel was transferred to a platform covered with three layers of 3MM filter paper, which was saturated with 10 x SSC. The SSC-soaked Zeta- Probe membrane was layered on the top of the gel, then covered with three layers of pre- • wetted 3MM filter paper (same size as the membrane, saturated with 10 x SSC), three 182 layers of dry 3MM filter paper and a stack of paper towels (7 cm high) topped with a weight (~ 300 g). The RNA transfer was allowed to proceed overnight. The membrane was then removed, rinsed with 2 x SSC, air-dried, and then placed on a gel dryer (HB1125B Slab Gel Dryer, Bio-Rad) at 80° C for 30 min. The membrane was wrapped in filter paper and stored in a sealed plastic bags at room temperature until used for hybridization. 8.3. Pre-hybridization The membrane blot (from Method 8.2) was wetted in 1% SDS, and incubated in a rotating hybridization tube (HB-10 Hybridiser, Techne) with 8 ml pre-hybridization buffer consisting of 4 ml formamide, 2 ml 20% SDS, 2 ml 2 M NaH2P04/ Na2HP04 in 4 mM EDTA, pH 7.2 and 160 pi BSA (50 mg/ml) at 42° C for 1 h. 8.4. Hybridization with probe The 32P-labeled probe was denatured by heating at 95° for 10 min, then added into the hybridization tube containing the membrane blot and the pre-hybridization buffer. Probe hybridization was conducted at 42° C overnight. After hybridization, the membrane blot was first washed with 2 x SSPE (2 mM EDTA, 0.36 M NaCl and 20 mM Na2HP04, pH 7.4) plus 0.3% SDS at room temperature for 20 min, then washed with 1 x SSPE plus 0.5% SDS for another 20 min. The blot was then transferred to a plastic bag and sealed. For autoradiography, the blot was sandwiched between two Kodak films (X-OMAT-AR, Kodak) and exposed at -70° C for 1-10 days. 183 9. Southern blot analysis of PCR products 9.1. Electrophoresis of PCR products The gel preparation and electrophoresis conditions are those described in Method 6 of this chapter. 9.2. Capillary transfer with 0.4 M NaOH After electrophoresis, the gel was rinsed with 0.4 M NaOH, and then transferred to the blot platform covered with three layers 3MM filter paper, which were saturated with 0.4 M NaOH. Zeta-Probe GT membrane was re-hydrated with 0.4 M NaOH, and then layered on the top of gel. The membrane was covered with three layers of pre-wetted 3MM filter paper (same size as the membrane, saturated with 0.4 M NaOH), three layers of dry 3MM filter paper and a stack of paper towels (7 cm high) topped with a weight (~ 180 g). The DNA was blotted by capillary transfer overnight. After blotting, the membrane blot was rinsed with 2 x SSC, air-dried, and then placed on a gel dryer at 80° C for 30 min. The dried membrane was wrapped by filter paper and stored in plastic bags at room temperature until used for hybridization. 9.3. Pre-hybridization The membrane blot was wetted in 1% SDS, then incubated in the hybridization tube with 8 ml pre-hybridization buffer consisting of 8 ml 0.25 M Na2HP04/ H3P04, pH 7.2, plus 7% SDS at 55° C for 5 min. 184 9.4. Hybridization with probe The 32P-labeled probe was denatured by heating at 95° for 10 min, then added into the hybridization tube containing the membrane blot and the pre-hybridization buffer for hybridization at 55° C overnight. After hybridization, the membrane blot was washed two times with 20 mM Na2HP04/ H3P04, pH 7.2, 5% SDS at 55° C for 30 min, followed by two washes with 20 mM Na2HP04/ H3P04, pH 7.2, 1% SDS at 65° C for 30 min. The blot was transferred to a plastic bag and sealed. For autoradiography, the blot was sandwiched between two Kodak films (X-OMAT-AR, Kodak) and exposed at room temperature for 2-4 h, or at -70° C overnight. RESULTS 1. RT-PCR products 1.1. Effect of template sources, concentrations and reverse transcriptases Two reverse transcriptases, AMV-RT and M-MLV-RT, were tested for synthesis of the first strand cDNA from total RNA isolated from the seedlings after germination for 52, 57 and 66 h, respectively. After PCR amplification using the SGT-specific degenerate oligo nucleotide N-2 plus the BEK oligo-d(T)20 as primers, a - 0.5 kb fragment was found in the 52 h sample synthesized by AMV-RT and a -0.9 kb fragment was found in the 52 and 57 h samples synthesized by M-MLV-RT (Fig. 43A). Since the expected gene size for the sgt was about 1.2 kb, 57 h scDNA appeared to be a suitable DNA template for PCR amplification with these primers. Dilution of the template DNA showed that the 185 o bO QJ 3. m o\ QJ !-Z ° I o < Z ffl o ct 10 CO cu bO t =! H .5 jS TJ 71 CO QJ bO -—- 'cn S QJ 3_ 3 w -a i I r>. o QJ cu o cu cu co CX c .. co c -a o w ^ cu ' CU o '§ •M Z X! | x © •*7" co C"O in r~- TJ TJ un o C ^ S § 03 ro LO O XI o o co o rt 3 CO 1 1 © TJ C\j -r T- o 186 fragment with a size at about -0.9 kb could still be specifically amplified when as little as 3 ng of the template was used (Fig. 43B). 1.2. Effect of Mg++ concentration and primers Increasing the Mg++ concentration from 1.5 mM to 4 mM did not significantly affect the specificity or yield of the amplified product (lane 2-4, Fig. 44A). When only the N-2 primer was included in the reaction, a - 1.3 kb fragment was produced (lane 6, Fig. 44A), while with only the BEK oligo-d(T)20 primer, no specific product was detected (lane 7, Fig. 44A). This indicates that the -0.9 kb fragment in lane 2-4 is a specific PCR product amplified by using the N-2/BEK oligo-d(T)20 primers and 57 h scDNA synthesized by M- MLV-RT. However, no specific PCR product was found when degenerate primer N-l was combined with the BEK oligo-d(T)20 primer and 57 h scDNA synthesized by M- MLV-RT in PCR reaction (lane 2-3, Fig. 44B) although some small non-specific fragments were obtained by single primer amplification (lane 4, Fig. 44B). Varying the PGR conditions did not improve these results. 1.3. Effect of annealing temperature In order to establish an optimum annealing temperature for PCR amplification of sgt- specific products, a range of temperatures from 50 to 70° C was investigated in combination with two concentrations of template. No specific PCR product was found when annealing temperature was lower than 65° C (lane 1-4, Fig. 45), whereas a -0.9 kb 187 9 > fl O O CO O LO 31 CO CNJ i— O ao :2 mi CL) co u CJ CD JS CO CO llll o H •PJ O O CO O LO ri C\i r- o m to CO <" w CcD dCU cd cd -^J J§S .5a 188 u o T3 o m rt c o c oo a Q o C cd oo u 5 O bO CU C3 cd w -a co a CD c c OT O a 03 g»u e I -a > W T3 cd ISCD O C c ? cd 3 < cd 0*0 —c w < T3 .9 -2 Q C < pq Q B Cd g g Z -a cd 1 CD -2 < 0 z c o-Z Q T3 E « CD S cd C fa 3 cd o o PM CO 189 fragment was amplified when annealing temperature was 65° C or higher (lane 5-10, Fig. 45). 2. Southern blot RT-PCR products amplified by the N-2/ BEK oligo-d(T)20 primers were separated on a 1% agarose gel, and analyzed by Southern blot using spinach HSP70 probes (sce70C, for the hsplO C-terminal, and ,sce70N, for the hsplO N-terminal) and the B. napus HSC70 probe, bhsc70. The -0.9 kb product of RT-PCR was detected by the bhsc70 probe, but not detected by the spinach ,sce70C and sce70N probes (lane 4, Fig. 46). This indicates that the -0.9 kb fragment amplified by PCR was related to Brassica hsclQ. Signals detected in lane 3 were the result of cross-hybridization with vector sequence. 3. Northern blot Northern blotting was used to establish whether transcripts with homology to hsclQ could be detected in seedling RNA. After hybridization with either the 32P-labeled spinach HSP70 probe sce70N or the B. napus HSC70 probe bhsc70, all RNA samples isolated from three different age seedlings showed a signal at the position of -2 kb fragment (Fig. 47C&D). This signal was not found when the blot was hybridized with the spinach HSP70 cognate C-terminal probe scelOC (Fig. 47B). This result indicates that hsc70 transcripts exist in Brassica seedlings, but that they may have more sequence divergence in their C-terminal region, compared with spinach sce70. 190 Pi u PM TJ I cn CU •< CO X XI Z cd o O 191 TJ OS CL, ^ TJ TJ bo eg QJ e 15 -Q O CL oo o TJ m ^ cd 8 CM < CM n «g Cd 3 • —i CM •a •Q TJ QJ CD 33 ? XS o 35 CD £1 co o s CD NJ z O cd o Cu 3 CD r- o <4J 192 DISCUSSION HSP70s are among the mostly highly conserved proteins known. These proteins are often heat-inducible, but certain HSP70 family members are also constitutively expressed, and are referred to as HSC70 proteins (Lindquist and Craig, 1988). HSP70/HSC70 proteins are believed to possess both an ATPase activity (Schlossman et al, 1984; Chappell et al., 1987) and a peptide-binding activity (Beckmann et al., 1990; Langer et al, 1992). The ATPase activity is found in the N-terminal 1-385 amino acid residues (Chappell et al, 1987; Wilbanks et al, 1995) and sequence analysis of this N-terminal region also reveals ATP binding domains (Flaherty etal, 1990; Ko etal, 1992; Wilbanks etal, 1995; Lee- Yoon etal, 1995). Peptide-binding activity, and the cognate binding domain, are located within the 388-554 amino acid residues in the C-terminal of the protein (Wilbanks et al, 1995; Morshauser et al, 1995). The consistent amplification of a -0.9 kb fragment by RT-PCR using SGT/HSP70- specific degenerate primer N-2 plus oligo-d(T)20 indicates that there is possibly a transcript in the target DNA population that is related to sgt or hsclQ. The Southern blot results show that this -0.9 kb RT-PCR fragment has some homology with Brassica HSP70 since it was recognized by bhsclO. On the other hand, the amplified band was not recognized by spinach HSP70 cognate sce70 probes. The size of this amplified fragment is also anomalous. The expected size of hsc70 is about 2 kb. Since the N-2 primer is designed to reflect the putative SGT (42 kDa protein) N-terminal amino acid sequence, which is identical to the N-terminal sequences of HSP70s according to the results of 193 protein database search, it should be able to prime the amplification of both hsp70/hsc7Q and sgt cDNA's if the corresponding genes are being expressed in these tissues. Based on these data, it seems that -0.9 kb fragment is a possible candidate for a sgt sequence, although its size is a little smaller than the predicted expected size (1.2 kb). If we assume that the 42 kDa protein corresponds to the SGT enzyme, then it follows that either two separate genes exist (sgt and fee 70), or a multi-functional fee 70 gene, which serves as a precursor for the formation of SGT, exists. The latter modification might happen at the mRNA level (transcription) or at the protein level (translation). Analysis of seedling total RNA on Northern blots revealed a strong signal at about the 2 kb position in all RNA samples isolated from three different age seedlings when probed with sce70N and bhsc70, whereas no signal was detected at any other mRNA sizes. This result suggests that at least one fec70 gene is being expressed in these tissues. If SGT does share a high degree of homology with HSP70, however, it seems there is no separate transcript population derived from sgt. This leads to the conclusion that SGT is likely a modified gene product of fec70. It is worth mentioning that sce70C (the C-terminal end of spinach hsp70 cognate) did not detect any sequences in either Southern or Northern blots. Since the RT-PCR amplified fragment was only -0.9 kb in size, it would not be expected to contain the C-terminal sequence of hsc70, and thus it could not be detected by sce70C. The negative Northern blot result implies that there may be considerable variation in the C-terminal region between Brassica fec70 and spinach sce70. The 194 diversity of C-terminal sequences between members of the HSP70 family has been noted in previous studies (Boorstein et al, 1994). In summary, the data from RT-PCR, including Southern blot of RT-PCR products and Northern blots, do not support each other. The RT-PCR data imply that there exists a transcript showing some degree of homology with fec70 and having a similar size as sgt in 57-h-old seedlings, but apparently no fee70. On the other hand, the Northern blot data demonstrate that fec70 transcripts exist in all seedlings, but reveal no sign of sgt-size transcripts. Since the -0.9 kb fragment was only significantly amplified when using 57-h- old seedling cDNA as the template (trace found when using 52-h-old seedling cDNA; not detectable at all when using 66-h-old seedling cDNA), whereas 2 kb signals were detected from all three RNA samples on the Northern blots, the conservative conclusion would be that the -0.9 kb fragment may not be part of a fec70 sequence. PCR is a technique that can often yield artificial results, and it is not possible to exclude the possibility that the PCR conditions might not be optimal for amplifying fec70. Similarly, the Northern blot conditions used might not be ideal for detecting sgt transcripts, because how much overall identity exists between the two genes is still unknown. Direct sequencing of this -0.9 kb fragment should cast more light on this question. 195 Chapter VI CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER RESEARCH 196 In Brassicaceae, the levels of SGT activity are closely correlated with active sinapine biosynthesis, although the absolute level of SGT activity is not correlated with the absolute sinapine content in the seed. SGT may or may not play a role in further metabolism of sinapine, depending on the species. SGT was found to be active in seedlings of all the investigated brassicaceous species except S. alba. A more detailed analysis of the metabolites accumulated during seedling development showed that the pattern of aromatic compounds observed in S. alba extracts was different from that presented inB. napus and/?, sativus. The major compound found in the vegetative tissues of S. alba is not sinapoylmalate, which is the predominant end-product of sinapine metabolism in B. napus. The almost universal occurrence of SGT in seedlings of these investigated species suggests that the pathway of sinapine metabolism among these species is similar to that in B. napus, and that the sinapic acid released from the sinapine is therefore likely to be re-esterfied by SGT to form energy-rich sinapoylglucose for further metabolic conversions, such as the formation of sinapoylmalate. SGT is expressed in all growth stages of B. napus plants, but is most active in the early germination and seed development stages, notably in the tissues of cotyledons, juvenile leaves and shoots. These tissues are the most active sites for the biosynthesis of either sinapine or sinapoylmalate. As a protein, SGT was found to be a monomeric polypeptide with an estimated molecular weight of 42 kDa and a pi at pH 5. SGT appears to be a cytosolic protein. This enzyme is 197 not inducible by environmental stresses such as heat shock or UV irradiation. Its general characteristics are similar to those of SGT from R. sativus as well as a number of other UDP-glucose-dependent glucosyltransferases. At its optimal pH (pH 6.0), SGT showed a Km for UDP-glucose of 0.24 mM and for sinapic acid of 0.16 mM. TDP-glucose can be used by this enzyme as a sugar donor as efficiently as UDP-glucose. Although a number of phenolics can be used as sugar acceptors, sinapic acid was shown to be the best substrate. SGT also catalyzes the reverse reaction using UDP, or TDP and sinapoylglucose to form UDP-glucose or TDP-glucose, respectively. No cofactors are required for SGT activity, but thiol reducing reagents and glycerol are required to stabilize the enzyme. SGT is strongly inhibited by known -SH group inhibitors, such as PHMB; by substrate analogues, e.g. UDP, TDP, UDP-mannose, and by divalent ions (Zn++, Cu++ , Hg++ , Fe++, Co++). The kinetic properties and substrate affinity chromatography data of SGT from B. napus suggest that its catalytic mechanism best fits a "random bi-bi" model, whereas the enzyme from R. sativus was found to use an "ordered bi-bi" mechanism (Mock and Strack, 1993). Since the partial amino acid sequences derived from purified SGT unexpectedly matched the sequences of HSP70 proteins, and the purified protein could also be detected using anti-pea HSP70/HSC70 antibodies, co-purification of SGT with HSP70/HSC70 cannot be totally excluded. Nevertheless, several lines of evidence suggest that 42 kDa SGT protein could be different from HSP70/HSC70. 198 Further study is recommended in the following areas: 1. Anti-SGT antibodies Since both native and denatured SGT have failed repeatedly to elicit antibodies in rabbits, it seems that the immunogenicity of purified B. napus SGT is very weak for this species. Using another species of animal, such as rat, guinea pig or chicken, or modifying purified SGT by conjugation with carrier proteins should be attempted for the production of anti- SGT antibodies. 2. Identification of a sgt gene Since a -0.9 kb fragment can be specifically and consistently amplified by degenerate primer N-2 and oligo-d(T)2o, and its sequence appears to be related to hsclO by Southern blot, cloning and sequencing of this -0.9 kb DNA fragment is required. This sequence information may lead to the cloning of the sgt gene or cDNA and could thus clarify the relationship between HSC70 and SGT. 3. Relationship between HSP70/HSC70 and SGT Since the putative SGT (42 kDa protein) showed such a high degree of homology with HSP70/HSC70, expression of the partially deleted bhsclQ (gene encoding Brassica HSC70) from the C-terminus, and assay of the expressed protein for SGT activity should be undertaken. This would help clarify whether SGT activity resides in an independent 199 gene (sgt) product, or has evolved as a second (catalytic) function for one or more members of the plant HSP70/HSC70 family. 200 BIBLIOGRAPHY Andersen, H. R. and Sorensen, H. 1985. Double low rapeseed meal in diets to young bulls. In: Advances in the Production and Utilization of Cruciferous Crops (ed. Sorensen, H.), Martinus Nijhoff Publ. Dordrecht, pp. 208-217. Andersen, P. E. 1985. Double low rapeseed meal in diets to dairy cows. In: Advances in the Production and Utilization of Cruciferous Crops (ed. Sorensen, H.), Martinus Nijhoff Publ. Dordrecht, pp. 218-221. Bajaj, K. L., Deluca, V., Khouri, H. and Ibrahim, R. K. 1983. Purification and properties of a flavonol-ring B glucosyltransferase from Chrysosplenium americanum. Plant Physiol. 72:891-896. Barber, G. A. 1962. Enzymic glycosylation of quercetin to rutin. Biochem. 1:463-468. Barz, W. and Koester, J. 1981. Turnover and degradation of secondary (natural) products. In: The Biochemistry of Plants: A Comprehensive Treatise. Vol. 7: Secondary Plant Product (ed. Conn, E. E.), Academic Press, New York, pp. 35-84. Barz, W., Koester, J., Weltring, K. M. and Strack, D. 1985. Recent advances in the metabolism and degradation of phenolic compounds in plants and animals. In: Annual Proceedings of Phytochemistry Society of Europe. Vol. 25: The Biochemistry of Plant Phenolics (ed. Van Sumere, C. F. and Lea, P. J.), Clarendon Press, Oxford, pp. 307-347. Baumker, P. A. Jutte, M. and Wiermann, R. 1987. The separation of two different enzymes catalyzing the formation of hydroxycinnamic acid glucosides and esters. Z. Naturforsch. 42c:1223-1230. Beckmann, R. P., Mizzen, L. A., Welch, W. J. 1990. Interaction of Hsp 70 with newly synthesized proteins: implications for protein folding and assembly. Science. 248:850- 854. Berlin, J. and Witte, L. 1982. Metabolism of phenylalanine and cinnamic acid in tobacco cell lines with high and low yields of cinnamoyl putrescines. J. Natural Products. 45:88- 93. Bille, N., Eggum, B. O., Jacobsen, I., Olsen, O. and Sorensen, H. 1983. The effects of processing on antinutritional rape constituents and the nutritive value of double low rapeseed meal. Z. Tierphysiol., Tierernahrung u. Futtermittelkd. 40: 148-163. Blair, R. and Reichert, R. D. 1984. Carbohydrate and phenolic constituents in a comprehensive range of rapeseed and canola fractions: nutritional significance for animals. J. Sci. Food Agric. 35: 29-35. 201 Bokern, M. and Strack, D. 1988. Synthesis of hydroxycinnamic acid esters of betacyanins via 1-O-acylglucosides of hydroxycinnamic acids by protein preparations from cell suspension cultures of Chenopodium rubrum and petals of Lampranthus sociorum. Planta. 174:101-105. Bokern, M., Wray, V. and Strack, D. 1991. Phenolic acid conjugates and betacyanin accumulation and activity changes of enzymes involved in feruloylglucose metabolism in cell suspension cultures of Chenopodium rubrum. Planta. 184:261-270. Bolwell, G. P. and Butt, V. S. 1983. Photoinduced changes in O-diphenol oxidase andp- coumarate hydroxylase activities in spinach beet seedlings and leaves. Phytochem. 22:37- 45. Boorstein, W., Ziegelhoffer, T. and Craig, E. A. 1994. Molecular evolution of the HSP70 multigene family. J. Mol. Evol. 38:1-17. Bopp, M. and Ludicke, W. 1980. Synthesis of sinapine during seed development of Sinapis alba. Z. Naturforsch. 35c:539-543. Bouchereau, A., Hamelin, J., Lamour, I., Renard, M. and Larher, F. 1991. Distribution of sinapine and related compounds in seeds of Brassica and allied genera. Phytochem. 30: 1873-1881. Bouchereau, A., Hamelin, J., Renard, M. and Larher, F. 1992. Structural changes in sinapic acid conjugates during seedling development of rape. Plant Physiol. Biochem. 30:467-475. Bouchereau, A. and Larher, F. 1991. A survey of phenolic choline esters in Brassica and allied genera seeds. In: Proceeding of 8th International Rapeseed Congress, Saskatoon, Saskatchewan, Canada pp. 705-710. Boudet, A. M., Lapierre, C. and Grima-Pettenati, J. 1995. Biochemistry and molecular biology of lignification. New Phytol. 129:203-236. Bradford, H. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254. Bugos, R. C, Chiang, V. L. C. and Campbell, W. H. 1991. cDNA cloning, sequence analysis and seasonal expression of lignin-bispecific caffeic acid/5-hydroxyferulic acid O-methyltransferase of aspen. Plant Mol. Biol. 17:1203-1215. Butler, E. J. and Fenwick, G. R. 1984. Trimethylamine and fishy taint in eggs. World's Poult. Sci. J. 40:38-51. 202 Chappie, C. C. S., Vogt, T., Ellis, B. E. and Somerville, C. R. 1992. An Arabidopsis mutant defective in the general phenylpropanoid pathway. The Plant Cell. 4:1413-1424. Chappell, T. G., Konforti, B. B., Schmid, S. L. and Rothman, J. E. 1987. The ATPase core of a clathrin uncoating protein. J. Biol. Chem. 262:746-751. Cheng, G. W., Malencik, D. A. and Breen, P. J. 1994. UDP-glucose:flavonoid O- glucosyltransferase from strawberry fruit. Phytochem. 35:1435-1439. Clandinin, D. R. 1961. Rapeseed oil meal studies. 4. Effect of sinapine, the bitter substance in rapeseed oil meal, on the growth of chickens. Poult. Sci. J. 40:484-487. Clausen, S., Larsen, L. M., Ploger, A. and Sorensen, H. 1985. Aromatic choline esters in rapeseed. In: Advances in the Production and Utilization of Cruciferous Crops. World Crops: Production, Utilization, Description (ed. Sorensen, H.), Martinus. Nijhoff Publ. Hague, pp. 61-72. Clausen, S., Olsen, O. and Sorensen, H. 1982. 4-hydroxybenzoylcholine: a natural product present in Sinapis alba. Phytochem. 21:917-922. Clausen, S., Olsen, O. and Sorensen, H. 1983. Separation of aromatic choline esters by high-performance liquid chromatography. J. Chromatogr. 260:193-199. Corner, J. J. and Swain, T. 1965. Enzymatic synthesis of sugar esters of hydroxy- aromatic acids. Nature. 207:634-635. Cosio, E. G., Weissenbock, G. and McClure, J. W. 1985. Acifluorfen-induced isoflavonoids and enzymes of their biosynthesis in mature soybean leaves. Plant Physiol. 78:14-19. Dabrowski, K. J., Rutkowski, A., Manimanna, S. G. and Rowicka, A. 1989. The reduction of glucosinolate and sinapine contents under industrial processing of rapeseed. Fat Sci. Technol. 91: 361-363. Dabrowski, K. J. and Siemieniak, B. 1987. Removal of sinapine from rapeseed using various solvents and extraction condition. In: Proceeding of 7th International Rapeseed Congress, Poznan, Poland, pp. 1476-1481. Dahlbender, B. and Strack, D. 1986. Purification and properties of 1- (hydroxycinnamoyl) -glucose: 1 - (hydroxycinnamoyl) -glucose hydroxycinnamoyltransferase from radish seedlings. Phytochem. 25:1043-1046. Davin, L. B. and Lewis, N. G. 1992. Phenylpropanoid metabolism: biosynthesis of monolignols, lignans and neolignans, lignins and suberins. In: Phenolic Metabolism in Plants (ed. Stafford, H.A. and Ibrahim, R. K.), Plenum Press, New York, pp. 325-375. 203 Davis, L., Luehl, M. and Battey, J. 1994. Preparation and analysis of RNA from eucaryotic cells. In: Basic Methods in Molecular Biology (ed. Norwalk, C. T.), Appelton and Lange, pp. 350-355. DeRocher, A. and Vierling, E. 1995. Cytoplasmic HSP70 homologues of pea: differential expression in vegetative and embryonic organs. Plant Mol. Biol. 27:441-456. Downey, R. K. and Robbelen, G. 1989. Brassica species. In: Oil Crops of the World (ed. Robbelen, G., Downey. R. K. and Ashri, A.), McGraw-Hill Publ. Co., New York, pp. 339-362. Drake, R. R., Kaushal, G. P., Pastuszak, I. and Elbein, A. D. 1991. Partial purification, photoaffinity labeling, and properties of mung bean UDP-glucose:dolicholphosphate glucosyltransferase. Plant Physiol. 97:396-401. Drake, R. R., Palamarczyk, G., Haley, B. E. and Lennarz, W. J. 1990. Evidence for the involvement of a 35-kDa membrane protein in the synthesis of glucosylphosphoryldolichol. Biosci. Rep. 10:61-68. Durr, M., Bailey D. S. and MacLachlan, G. 1979. Subcellular distribution of membrane- bound glycosyltransferases from pea stems. Eur. J. Biochem. 97:445-453. Edwards, R. and Dixon, R. A. 1991. Purification and characterization of 5-adenosyl-L- methionine: caffeic acid 3-O-methyltransferase from suspension cultures of alfalfa (Medicago sativa L.). Arch. Biochem. Biophys. 287:372-379. Edwards, R., Mavandad, M. and Dixon, R. A. 1990. Metabolic fate of cinnamic acid in elicitor treated cell suspension cultures of Phaseolus vulgaris. Phytochem. 29:1867-1874. Engebrecht, J. and Brent, R. 1993. Minipreps of plasmid DNA. In: Current Protocols in Molecular Biology (ed. Ausubel, F. M.), pp. 1.6.4-1.6.5. Enyedi, A. J. and Raskin, I. 1993. Induction of UDP-glucose:salicylic avid glucosyltransferase activity in tobacco mosaic virus-inoculated tobacco (Nicotiana tabacum) leaves. Plant Physiol. 101:1375-1380. Fenwick, G. R., Hobson-Frohock, A., Land, D. G. and Curtis, R. F. 1979. Rapeseed meal and egg taint: treatment of rapeseed meal to reduce tainting potential. Br. Poult. Sci. 20: 323-329. Flaherty, K. M., DeLuca-Flaherty, C. and McKay, D. B. 1990. Three dimensional structure of the ATPase fragment of a 70K heat-shock cognate protein. Nature. 346:623- 628. 204 Fleuriet, A., Macheix, J. J., Suen, R. and Ibrahim, R. K. 1980. Partial purification and some properties of a hydroxycinnamoyl glucosyltransferase from tomato fruits. Z. Naturforsch. 35c:967-972. Flynn, G., Pohl, J., Flocco, M. and Rothman, J. 1991. Peptide-binding specificity of the molecular chaperone BiP. Nature. 353:726-730. Furtek, D., Schiefelbein, J. W., Johnston, F. and Nelson, O. E. 1988. Sequence comparisons of three wild-type Bronze-1 alleles from Zea mays. Plant Mol. Biol. 11:473- 481. Gabriac, B., Werck-Reichhart, D., Teutsch, H. and Durst, F. 1991. Purification and immunocharacterization of a plant cytochrome P450: the cinnamic acid 4-hydroxylase. Arch. Biochem. Biophys. 288:302-309. Gadamer, J. 1896. Uber die Bestandteile des schwarzen und des weissen Senfsamens. J. Arch. der. Pharm. 235:44-83. Gadamer, J. 1897. Sinapine. J. Arch. der. Pharm. 235:92-102. Goh, Y. K., Shires, A., Robblee, A. R. and Clandinin, D. R. 1982. Effect of ammoniation of rapeseed meal on the sinapine content of the meal. Br. Poult. Sci. 23: 121-128. Goodwin, T. W. and E. I. Mercer. 1983. Introduction to Plant Biochemistry. Pergamon Press, New York. pp. 15-255. Grand, C. 1984. Ferulic acid-5-hydroxylse: a new cytochrome P-450 dependent enzyme from higher plant microsomes involved in lignin synthesis. FEBS Lett. 169:7-11. Grawe, W., Bachhuber, P., Mock, H. and Strack, D. 1992. Purification and characterization of sinapoylglucose:malate sinapoyltransferase from Raphanus sativus L. Planta. 187:236-241. Grawe, W. and Strack, D. 1986. Partial purification and some properties of 1- sinapoylglucosexholine sinapoyltransferase ("sinapine synthesis") from seeds of Raphanus sativus L. and Sinapis alba L. Z. Naturforsch. 41c:28-33. GrootWassink, J. W. D., Reed, D. W. and Kolenovsky, A. D. 1994. Immunopurification and immunocharacterization of the glucosinolate biosynthetic enzyme thiohydroximate S- glucosyltransferase. Plant Physiol. 105:425-433. Gross, G. G. 1981. Phenolic acids. In: The Biochemistry of Plants: A Comprehensive Treatise. Vol. 7: Secondary Plant Products (ed. Conn, E. C), Academic Press, London, pp. 301-316. 205 Gross, G. G. 1982. Synthesis of (3-glucogallin from UDP-glucose and gallic acid by an enzyme preparation from oak leaves. FEBS Letters. 148:67-70. Gross, G. G. 1983. Synthesis of mono-, di- and trigalloyl-(3-Z)-glucogallin-dependent galloytransferases from oak leaves. Z. Naturforsch. 38:519-523. Guo, L. and Poulton, J. E. 1994. Partial purification and characterization of Arabidopsis thaliana UDPG:thiohydroximate glucosyltransferase. Phytochem. 36:1133-1138. Hadacova, V., Hofman, J., Almeida, R. M., Vackova, K., Kutacek, M. and Klozova, E. 1981. Choline esterases and choline acetyltransferase in the seeds of Allium altaicum (Pall.) reyse. Biologia Plantarum. 23:220-227. Hahlbrock, K. and Scheel, D. 1989. Physiology and molecular biology of phenylpropanoid metabolism. Annu. Rev. Plant Physiol. & Plant Mol. Biol. 40:347-369. Halaweish, F. T. and Dougall, D. K. 1990. Sinapoyl glucose synthesis by extracts of wild carrot cell cultures. Plant Sci. 71:179-184. Harborne, J. B. 1980. Plant phenolics. In: Encyclopedia of Plant Physiology, Vol. 8: Secondary Plant Products (ed. Bell, E. A. and Charlwood, B. V.), Springer, Berlin Heidelberg, New York, pp. 329-402. Harborne, J. B. and Corner, J. J. 1961. Plant poly phenols. 4. Hydroxycinnamic acid- sugar derivatives. Biochem. J. 81:242-250. Hartmann-Bouillon, M. A. and Benveniste, P. 1978. Sterol biosynthetic capability of purified membrane fractions from maize coleoptiles. Phytochem. 17:1037-1042. Hartmann-Bouillon, M. A. and Benveniste, P. 1987. Plant membrane sterols: isolation, identification and biosynthesis. Methods Enzymol. 148:632-650. Haslam, E. 1993. Polyphenol complexion. In: Polyphenolic Phenomena (ed. Scalbert, A.), INRA, Paris, pp. 23-31. Hegnauer, R. 1964. Cruciferae. Chemotaxonomie der Pflanzen. 3:586-606. Heilemann, J. and Strack, D. 1991. Flavonol glucosyltransferase from Norway spruce needles. Phytochem. 30:1773-1776. Herrmann, K. M. 1995. The shikimate pathway as an entry to aromatic secondary metabolism. Plant Physiol. 107:7-12. 206 Hobson-Frohock, A., Fenwick, G. R., Heaney, R. K., Land, D. G. and Curtis, R. F. 1977. Rapeseed meal and egg taint: Association with sinapine. Br. Poult. Sci. 18: 539-541. Hobson-Frohock, A., Land, D. G., Griffiths, N. M. and Curtis, E. F. 1973. Egg taints: association with trimethylamine. Nature. 243: 304-305. Hosel, W. 1981. Glycosylation and glycosidases. In: The Biochemistry of Plants: A Comprehensive Treatise. Vol. 7: Secondary Plant Products (ed. Conn, E. E.), Academic Press, pp. 725-753. Hosel, W. and Schiel, O. 1984. Biosynthesis of cyanogenic glucosides: in vitro analysis of the glucosylation step. Arch. Biochem. Biophys. 229:177-186. Hrazdina, G., Wagner, G. J. and Siegelman, H. W. 1978. Subcellular localization of enzymes of anthocyanin biosynthesis in protoplasts. Phytochem. 17:53-56. Hundle, B. S., O'Brien, D. A., Alberti, M., Beyer, P. and Hearst, J. E. 1992. Functional expression of zeaxanthin glucosyltransferase from Erwinia herbicola and a proposed uridine diphosphate binding site. Proc. Natl. Acad. Sci. USA. 89:9321-9325. Ibrahim, R. K., Luca, V. D., Khouri, H., Latchinian, L., Brisson, L. and Charest, P. M. 1987. Enzymology and compartmentation of polymethylated flavonol glucosides in Chrysosplenium americanum. Phytochem. 26:1237-1245. Imperato, F. 1976. l-(3-glucosyloxy-4-hydroxycinnamoyl)glucose from Lantana hybrida. Phytochem. 15:1786. Ishikura, N. & Yang, Z., 1994. Multiple forms of flavonol 0-glucosyltransferases in young leaves of Euonymus alatus f. ciliato-dentatus. Phytochem. 36:1139-1145. Ismail, F., Vaisey-Genser, M. and Fyfe, B. 1981. Bitterness and astringency of sinapine and its components. J. Food Sci. 46: 1241-1244. Jackson, M. R., McCarthy, L. R., Harding, D., Wilson, S., Coughtrie, M. W. H. and Burchell, B. 1987. Cloning of a human liver microsomal UDP-glucuronosyltransferase cDNA. Biochem. J. 242:581-588. Jorgensen, R. 1994. The genetic origins of biosynthesis and light-responsive control of the chemical UV screen of land plants. In: Recent Advances in Phytochemistry, Vol. 28: Genetic Engineering of Plant Secondary Metabolism (ed. Ellis, B. E., Kuroki, G. W. and Stafford, H. A.), Penum Press, New York, pp. 179-192. Kefeli, V. I., Kutacek, M., Vatzkova, K., Makhachkova, I., Zmrgal, Z., Vlasov, P. V., Gus'kov, A. V. and Shapkin, V. I. 1977. Derivatives of sinapic acid in Kohlrabi 207 seedlings. Properties and biological activity. Soviet Plant Physiol (translated from Russian). 24:966-970. Kerber, V. E. and Buchloh, G. 1980. Sinapingehalte in speisesenf. Z. Lebensm. Unters. Forsch. 171: 355-356. Kerber, V. E. and Buchloh, G. 1981. Sinapinfreie und sinapinhaltige Brassicaceae. Beitr. Biol. Pflanzen. 55:377-384. Kimura, T. and Owens, I. S. 1987. Mouse UDP glucuronosyltransferase cDNA and complete amino acid sequence and regulation. Eur. J. Biochem. 168:515-521. Knobloch, K. H. and Hahlbrock, K. 1975. Isoenzymes of />-coumarate:CoA ligase from cell suspension culture of Glycine max. Eur. J. Biochem. 52:311-320. Ko, K., Bornemisza, O., Kourtz, L., Ko, W., Plaxton, W. C. and Cashmore, A. R. 1992. Isolation and characterization of a cDNA clone encoding a cognate 70 kDa heat shock protein of the chloroplast envelope. J. Biol. Chem. 267:2986-2993. Koester, J., Ohm, M. and Barz, W. 1978. Metabolism of anthranilic acid in plant cell suspension cultures. Z. Naturforsch. 33c:368-372. Kojima, M. and Uritani, I. 1972. Elucidation of the structure of a possible intermediate in chlorogenic acid biosynthesis in sweet potato root tissue. Plant Cell Physiol. 13:1075- 1084. Kojima, M. and Uritani, I. 1973. Studies on chlorogenic acid biosynthesis in sweet potato root tissue in special reference to the isolation of a chlorogenic acid intermediate. Plant Physiol. 51:768-771. Koukol, J. and Conn, E. E. 1961. The metabolism of aromatic compounds in higher plants. IV. Purification and properties of the phenylalanine deaminase of Hordeum vulgare. J. Biol. Chem. 236:2692-2698. Kraling, K., Robbelen, G. and Thies, W. 1991. Genetic variation of the content of sinapoyl esters in seeds of rape, B. napus. Plant Breeding 106: 254-257. Kuhnl, T. and Wellmann, E. 1979. Sinapin-biosynthese in mustard embryos. Ber. Deutsch. Bot. Ges. Bd. 92:741-748. Kuroda, H., Shimada, M. and Higuchi, T. 1981. Charaterization of a lignin-specific O- methyltransferase in aspen wood. Phytochem. 20:2635-2639. Kvartshava, L. S., Durmishidze, S. V. and Kefeli, V. I. 1980. Light action on sinapine accumulation and growth of kohlrabi cotyledons. Soviet Plant Physiol. 26: 704-706. 208 Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227:680-685. Langer, T., Lu, C, Echols, H., Flanagan, J., Hayer, M. K., and Hartl, F.-U. 1992. Successive action of DnaK, DnaJ and GroEL along the pathway of chaperone-mediated protein folding. Nature. 356:683-689. Larsen, L. M., Olsen, O., Ploger, A. and Sorensen, H. 1983. Phenolic choline esters in rapeseed: possible factors affecting nutritive value and quality of rapeseed meal. In: Proceeding of 6th International Rapeseed Congress, Paris, France, pp. 1577-1582. Lee-Yoon, D., Easton, D. Murawski, M., Burd, R. and Subjieck, J. R. 1995. Identification of a major subfamily of large hsp7-like proteins through the cloning of the mammalian 110-kDa heat shock protein. J. Biol. Chem. 270:15725-15733. Lewis, N. G. and Yamamoto, E. 1990. Lignin: occurrence, biogenesis and biodegradation. Annu. Rev. Plant Physiol. Plant Mol. Biol. 42:455-496. Leznicki, A. J. and Bandurski, R. S., 1988. Enzymic synthesis of indole-3-acetyl-l-0-(3- Z)-glucose. II. Metabolic characteristics of the enzyme. Plant Physiol. 88:1481-1485. Lindquist, S. and Craig, E. A. 1988. The heat shock proteins. Annu. Rev. Genet. 22:631- 677. Linscheid, M., Wendisch, D. and Strack, D. 1980. The structures of sinapic acid esters and their metabolism in cotyledons of Raphanus sativus. Z. Naturforsch. 35c:907-914. Lynn, D. G. and Chang, M. 1990. Phenolic signals in cohabitation: implications for plant development. Annu. Rev. Plant Physiol. Plant Mol. Biol. 41:497-526. Macheix, J. J. 1977. Biosynthese des esters glucoses des acides hydroxycinnamiques a partir d'uridine diphosphoglucose et des acides librez chez la pomme. CR. Acad. Sci., Ser. D 284:33-36. Mackenzie, P. 1987. Rat liver UDP- glucuronosyltransferase. J. Biol. Chem. 262:9744- 9749. Markham, K. R. 1982. Techniques of Flavonoid Identification. Academic Press, London, New York, pp. 15-35. Mazza, G. and Miniati, E. 1993. Anthocyanins in Fruits, Vegetables, and Grains. CRC Press, Boca Raton, pp. 6-10. 209 Michalczuk, L. and Bandurski, R. S. 1980. UDP-glucose:indoleacetic acid glucosyltransferase and indoleacetyl-glucose:myo-inositol indole-acetyl transferase. Biochem. Biophys. Res. Commun. 93:588-592. Miles, CD. and Hagen, C. W. 1968. The differentiation and pigmentation in flower parts. IV. flavonoid elaborating enzymes from petals of Impatiens balsamina. Plant Physiol. 43:1347-1354. Mock, H. and Strack, D. 1993. Energetics of the uridine 5'-diphosphoglucose:hydroxy- cinnamic acid acyl-glucosyltransferase reaction. Phytochem. 32:575-579. Mohle, B. Heller, W. and Wellmann, E. 1985. UV-induced biosynthesis of quercetin 3-0- (3-Z)-glucuronide in dill cell cultures. Phytochemistry. 24:465-467. Molderez, M., Nagels, L. and Parmentier, F. 1978. Time-course tracer studies on the metabolism of cinnamic acid in Cestrumpoeppigii. Phytochem. 17:1747-1750. Morshauser, R. C, Wang, H., Flynn, G. C. and Zuiderweg, E. R. P. 1995. The peptide- binding domain of the chaperone protein HSC70 has an unusual secondary structure topology. Biochem. 34:6261-6266. Nagels, L., Molderez, M. and Parmentier, F. 1981. UDPG: /?-coumarate glucosyltransferase activity in enzyme extracts from higher plants. Phytochem. 20:965- 967. Nicholson, R. L. and Hammerschmidt, R. 1992. Phenolic compounds and their role in disease resistance. Annu. Rev. Phytopath. 30:369-389. Nurmann, G. and Strack, D. 1979. Sinapine esterase. I. Characterization of sinapine esterase from cotyledons of Raphanus sativus. Z. Naturforsch. 34c:715-720. Nurmann, G. and Strack, D. 1981. Formation of 1-sinapoylglucose by UDP- glucose:sinapic acid glucosyltransferase from cotyledons of Raphanus sativus. Z. Pflanzenphysiol. 102:11-17. Oba, K., Conn, E. E., Canut, H. and Boudet, A. M. 1981. Subcellular localization of 2- ((3-D-glucosylhydroxy)-cinnamic acids and the related (3-glucosidase in leaves of Melilotus abla Desr. Plant Physiol. 68:1359-1363. Ohashi, H., Yamamoto, E., Lewis, N. G. and Towers, G. H. N. 1987. 5-hydroxyferulic acid in Zea mays and Hordeum vulgare cell walls. Phytochem. 26:1915-1916. O'Reilly, D. R. and Miller, L. K. 1989. A baculovirus blocks insect molting by producing ecdysteroid UDP-glucosyl transferase. Science. 245:1110-1112. 210 Paczkowski, C. and Wojciechowski, Z. A. 1994. Glucosylation and galactosylation of diosgenin and solasodine by soluble glycosyltransferase(s) from Solanum melongena leaves. Phytochem. 35:1429-1434. Penefsky, H. S. 1977. Reversible binding of Pi by beef heart mitochondrial adenosine triphosphatase. J. Biol. Chem. 252:2891-2899. Petersen, M. and Seitz, H. U. 1986. UDP-glucose:cyanidin 3-O-glucosyltransferase in anthocyanin-containing cell cultures from Daucus carota L. J. Plant Physiol. 125:383- 390. Pridham, J. B. and Saltmarsh, M. J. 1963. The biosynthesis of phenolic glucosides in plants. Biochem. J. 87:218-224. Ralston, E. J., English, J. J. and Dooner, H. K. 1988. Sequence of three bronze alleles of maize and correlation with the genetic fine structure. Genetics. 119:185-197. Regenbrecht, J. and Strack, D. 1985. Distribution of l-sinapoylglucose:choline sinapoyltransferase activity in the Brassicaceae. Phytochem. 24:407-410. Remsen, I. and Coale, R. D. 1884. Investigations on sinapine. I. sinapic acid. Am. Chem. J. 6:50-60. Rhodes, M. J. C. and Wooltorton, L. S. C. 1976. The enzymatic conversion of hydroxycinnamic acids to jo-coumaroylquinic and chlorogenic acids in tomato fruits. Phytochem. 15:947-951. Rothman, J. E. and Schmid, S. L. 1986. Enzymatic recycling of clathrin from coated vesicles. Cell. 46:5-9. Schlepphorst, R. and Barz, W. 1979. Metabolism of aromatic acids in plant cell suspension cultures. Planta Medica. 36:333-342. Schlossman, D. M., Schmid, S. L., Braell, W. A. and Rothman, J. E. 1984. An enzyme that removes clathrin coats: purification of an uncoating ATPase. J. Cell Biol. 99:723- 733. Schmid, D. and Grisebach, H. 1982. Enzymic synthesis of lignin precursors - purification and properties of UDP-glucose:coniferyl-alcohol glucosyltransferase from cambial sap of spruce (Picea abies L.). Eur. J. Biochem. 123:363-370. Schmid, D., Hammer, D. K., Ritterbusch, A. and Griserach, H. 1982. Appearance and immunohistochemical localization of glucosexoniferyl alcohol glucosyltransferase in spruce (Picea abies L. Karst.) seedlings. Planta 156:207-212. 211 Schultz, O.-E. and Gmelin, R. 1952. Papierchromatographie der senfolglucosidrogen. Z. Naturforsch. Teil. 7b: 500-508. Schultz, O.-E. and Gmelin, R. 1953. Papierchromatographie senfolglucosidhaltiger Pflanzen. Neue Ergebnisse (VI. Mitteilung uber senfolglucoside). Z. Naturforsch. 8b:151-156. Schwarze, P. 1949. The bitter substance of rapeseed. Naturwissenschaften. 36: 88-89. Sharma, V. and Strack, D. 1985. Vacuolar localization of l-sinapoylglucose:L-malate sinapoyltransferase in protoplasts from cotyledons of Raphanus sativus. Planta. 163:563- 568. Shimizu, T. and Kojima, M. 1984. Partial purification and characterization of UDPG:*1- cinnamate glucosyltransferase in the root of sweet potato, Ipomoea batatas Lam. J. Biochem. 95:205-212. Smith, C. E., Orr, J. D., Lynn, D. G. 1989. Chemical communication and the control of development. In: Natural product isolation: separation methods for anti-microbials (ed. Wagman, G. H. and Cooper, R.), Amsterdam, Elsevier, pp. 561-597. Sosulski, F. W. 1979. Organoleptic and nutritional effects of phenolic compounds on oilseed protein products: a review. J. Am. Oil Chem. Society. 56:711-715. Stapleton, A. Allen, P. V., Tao, H. P., Belknap, W. R. and Friedman, M. 1992. Partial amino acid sequence of potato solanidine UDP-glucose glucosyltransferase purified by new anion-exchange and size exclusion media. Protein Expr. Purif. 3:85-92. Strack, D. 1977. Sinapic acid ester fluctuations in cotyledons of Raphanus sativus. Z. Pflanzenphysiol. 84:139-145. Strack, D. 1980. Enzymatic synthesis of 1-sinapoylglucose from free sinapic acid and UDP-glucose by a cell-free system from Raphanus sativus seedlings. Z. Naturforsch. 35c:204-208. Strack, D. 1981. Sinapine as a supply of choline for the biosynthesis of phosphatidylcholine in Raphanus sativus seedlings. Z. Naturforsch. 36c:215-221. Strack, D. 1982. Development of l-0-sinapoyl-(3-glucose:Z-malate sinapoyltransferase activity in cotyledons of red radish (Raphanus sativus L. var. sativus). Planta. 155:31-36. Strack, D., Bokern, M., Berlin, J. and Sieg, S. 1984. Metabolic activity of hydroxycinnamic acid esters in cell suspension cultures of Chenopodium rubrum. Z. Naturforsch. 39c:902-907. 212 Strack, D., Ellis, B. E., Grawe, W. and Heilemann, J. 1990. Sinapoylglucose:malate sinapoyltransferase activity in seeds and seedlings of rape. Planta. 180:217-219. Strack, D., Hartfeld, F., Austenfeld, F. A., Grotjahn, L. and Wray, V. 1985. Coumaroyl-, caffeoyl- and feruloyltartronates and their accumulation in mung bean. Phytochem. 24:147-150. Strack, D., Knogge, W. and Dahlbender, B. 1983. Enzymatic synthesis of sinapine from l-0-sinapoyl-(3-Z)-glucose and choline by a cell-free system from developing seeds of red radish {Raphanus sativus L. var. sativus). Z. Naturforsch. 38c:21-27. Strack, D., Nurmann, G. and Sachs, G. 1980. Sinapine esterase. II. Specificity and change of sinapine esterase activity during germination of Raphanus sativus. Z. Naturforsch. 35c:963-966. Strack, D., Reinecke, J. and Takeuchi, S. 1986. Evidence for a relationship between malate metabolism and activity of l-sinapoylglucose:Z,-malate sinapoyltransferase in radish {Raphanus sativus L.) cotyledons. Planta. 167:212-217. Strack, D., Ruhoff, R. and Grawe, W.1986. Hydroxycinnamoyl-coenzyme-A:tartronate hydroxycinnamoyltransferase in protein preparations from mung bean. Phytochem. 25:833-837. Strack, D. and Sharma, V. 1985. Vacuolar localization of the enzymatic synthesis of hydroxycinnamic acid esters of malic acid in protoplasts from Raphanus sativus leaves. Physiol. Plant. 65:45-50. Strack, D., Tkotz, N. and Klug, M. 1978. Phenylpropanoid metabolism in cotyledons of Raphanus sativus and the effect of competitive in vivo inhibition of /--phenylalanine ammonia-lyase (PAL) by hydroxylamine derivatives. Z. Pflanzenphysiol. 89:343-353. Sun, Y. and Hrazdina, G. 1991. Isolation and characterization of a UDP-glucose:flavonol O 3-glucosyltransferase from illuminated red cabbage (Brassica oleracea cv Red Danish) seedlings. Plant Physiol. 95:570-576. Sutter, A. and Grisebach, H., 1975. Free reversibility of the UDP-glucose: flavonol 3-0- glucosyltransferase reaction. Arch. Biochem. Biophys. 167:444-447. Szerszen, J. B., Szczyglowski, K. and Bandurski, R. S. 1994. iaglu, A gene from Zea mays, involved in conjugation of the growth hormone, indole-3-acetic acid. Science. 265:1699-1701. 213 Tanaka, S. Hayakawa, K. Umetani, Y. and Tabata, M. 1990. Glucosylation of isomeric hydroxybenzoic acids by cell suspension cultures of Mallotus japonicus. Phytochemistry. 29:1555-1558. Tayaranian, H. and Henkel, H. 1991. Processing of rapeseed and meal to reduce the content of sinapine and glucosinolates. In: Proceeding of 8th International Rapeseed Congress, Saskatoon, Saskatchewan, Canada, pp. 1555-1560. Tkotz, N. and Strack, D. 1980. Enzymatic synthesis of sinapoyl-/_-malate from 1- sinapoylglucose and L-malate by a protein preparation from Raphanus sativus cotyledons. Z. Naturforsch. 35c:835-837. Tomaschewski, J., Gram, H., Crabb, J. W. and Rueger, W. 1985. T4-induced alpha- and beta-glucosyltransferase: cloning of the genes and a comparison of their products based on sequencing data. Nucleic acid Res. 13:7551-7568. Tzagoloff, A. 1963a. Metabolism of sinapine in mustard plant. I. Degradation of sinapine into sinapic acid and choline. Plant Physiol. 38: 202-206. Tzagoloff, A. 1963b. Metabolism of sinapine in mustard plant. II. Purification and some properties of sinapine esterase. Plant Physiol. 38:207-213. Ulbrich, B. and Zenk, M. H. 1979. Partial purification and properties of hydroxycinnamoyl-CoA:quinate hydroxycinnamoyl transferase from higher plants. Phytochem. 18:929-933. Ulbrich, B. and Zenk, M. H. 1980. Partial purification and properties of p- hydroxycinnamoyl-CoA:shikimate-/?-hydroxycinnamoyl transferase from higher plants. Phytochem. 19:1625-1629. Ullmann, P, Bouvier-Nave, P. and Benveniste, P. 1987. Regulation by phospholipids and kinetic studies of plant membrane-bound UDP-glucose sterol (3-£>-glucose transferase. Plant Physiol. 85:51-55. Uppstrom, B. and Johansson, M. 1985. Determination of sinapine in rapeseed. Sveriges Utsadesforenings Tidskrift. J. Swedish Seed Assoc. 95: 123-128. Ury, A. Benveniste, P. and Bouvier-Nave, P. 1989. Phospholipid-dependence of plant UDP-glucose sterol (3-D-glucosyl transferase. Plant Physiol. 91:567-573. Villegas, R. J. A. and Kojima, M. 1986. Purification and characterization of hydroxycinnamoyl D-glucose. J. Biol. Chem. 261:8729-8733. 214 Vogt, T., Aebershold, R. and Ellis, B. E. 1993. Purification and characterization of sinapine synthase from seeds of Brassica napus. Arch. Biochem. Biophys. 300:622-628. Voskerusa, J. and Kolovrat, O. 1989. Seed sinapine content of cruciferous oil plants (Obsah sinapinu v semeni brukvovitych oiejnin). Rostlinna Vyrob. 35: 1087-1093. Wallis, P. J. and Rhodes, M. J. C. 1977. Multiple forms of hydroxy-cinnamate: CoA ligase in etiolated pea seedlings. Phytochem. 16:1891-1894. Wang, X., 1992. Studies of Methodology of Sinapine Determination and Sinapine Variation in Brassica and Sinapis Species, A thesis of Department of Crop Science and Plant Ecology, University of Saskatchewan. Saskatoon, Saskatchewan, pp. 20-110. Warnecke, D. C. and Heinz, E. 1994. Purification of a membrane-bound UDP- glucose: sterol |3-£>-glucosyltransferase based on its solubility in diethyl ether. Plant Physiol. 105:1067-1073. Werner, C. and Matile, P. 1985. Accumulation of coumarylglucosides in vacuoles of barley mesophyll protoplasts. J. Plant Physiol. 118:237-249. Wiermann, R. 1981. Secondary plant products and cell and tissue differentiation. In: The Biochemistry of Plants: A Comprehensive Treatise. Vol. 7: Secondary Plant Product (ed. Conn, E. E.), Academic Press, New York, pp. 85-116. Wilbanks, S. M., Chen, L., Tsuruta, H., Hodgson, K. O. and McKay, D. B. 1995. Solution small-angle X-ray scattering study of the molecular chaperone HSC70 and its subfragments. Biochemistry 34:12095-12106. Wise, R. P., Rohde, W. and Salamini, F. 1990. Nucleotide sequence of the Bronze-l homologous gene from Hordeum vulgare. Plant Mol. Biol. 14:277-279. Wurtele, E. S., Thayer, S. S. and Conn, E. E. 1982. Subcellular localization of a UDP- glucose:aldehyde cyanohydrin (3-glucosyl transferase in epidermal plastids of Sorghum leaf blades. Plant Physiol. 70:1732-1737. Yalpani, N., Schulz, M, Davis, M. P. and Balke, N. E. 1992a. Partial purification and properties of an inducible uridine 5'-diphosphate-glucose:salicylic acid glucosyltransferase from oat roots. Plant Physiol. 100:457-463. Yalpani, N. Balke, N. E. and Schulz, M. 1992b. Induction of UDP-glucose:salicylic acid glucosyltransferase in oat roots. Plant Physiol. 100:1114-1119. 215 Yazaki, K., Inushima, K., Kataoka, M. and Tabata, M. 1995. Intracellular localization of UDPG:Q-hydroxybenzoate glucosyltransferase and its reaction product in Lithospermum cell cultures. Phytochemistry. 38:1127-1130. Zenk, M. H. 1979. Recent work on cinnamoyl CoA derivatives. In: Recent Advances in Phytochemistry. Vol. 12: Biochemistry of Plant Phenolics (ed. Swain, T., Harborne, J. B. and Van Sumere, C. F.), Plenum Press, New York, London, pp. 139-176. 216 u S cd H a t/2 H O CZ3 w 1/3 i- .© cj CS H e -t-> in CU co S } cu •oM u a cOC •M CO a cu 10 5 m M DT T s- 3 0 u l 1 M Fo r SGT : Tris/HCl , p H 6. 5 a u a cu s- in 3 T3 00 cu O © M °o —' Pw CWM o y s. -1. w < o O 'S ro H a CU a. 217 T3 " S1 O^ o 00 o. C 3 X T3 CS CU ^ cu cS •a o c 5 < cS c ts m E c U <; o ~ H ^ cu DC O '3 CM "cu : cu X E c 3 ^ cs O 83 OS Me (X« > w o- .5 x c co "C U -3 S ig o o o © W < o o _3 IT W cS CO S CS £ x LU s °a. o CQ to £ s E < PQ O O > CcS iB_ ^ Q O ©B X.2! PQ 00 c > XQ j 6 I- i—1 C- 4-1 TJ QJ TJ "2 * QJ .2 CO Z 2; M C <| X O Z TJ . i— —I Cfl Q CJ TJ « CU < x N C Z M QJ o 4j w Di 1 Di - in Uo £ -! 1 bO ZH' CM Q, oo 03 3 T QJ CJ •3 QJ Pi 0 _ 3 oI3 H J U fa x 2§ Di r= Q ^ co LO o o > co bO fa cu c i—1 s- 33 mS «ct« s >os H 03 O O w LO * 'CC T-TJl ^ rC- -PI o o co 2 o QJ 3. pq CD s w -e M QJ CO C\j T- ^ O to cd X o o I 3 sQJ *TJ CU co _N js • • o3 ^M. 0i>3 Qz 'co cx ^0 3 i?s ; cDoi QJ ^ x 6 —' QJ •is cu o c •2 QJ fa 03 vfa >> 7? co O -. co • .9 2 m O CD