UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The preparation of intermediates for the production of etoposide by the use of bio-transformations with… Jarvis, Terence C. 1991

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

Item Metadata

Download

Media
831-UBC_1991_A1 J37.pdf [ 9.15MB ]
Metadata
JSON: 831-1.0059760.json
JSON-LD: 831-1.0059760-ld.json
RDF/XML (Pretty): 831-1.0059760-rdf.xml
RDF/JSON: 831-1.0059760-rdf.json
Turtle: 831-1.0059760-turtle.txt
N-Triples: 831-1.0059760-rdf-ntriples.txt
Original Record: 831-1.0059760-source.json
Full Text
831-1.0059760-fulltext.txt
Citation
831-1.0059760.ris

Full Text

THE PREPARATION OF INTERMEDIATES FOR THE PRODUCTION OF ETOPOSIDE B Y T H E USE OF BIO-TRANSFORMATIONS WTH PLANT CELL CULTURES AND PLANT CELL CULTURE EXTRACTS  By Terence C. Jarvis B. Sc., The University of Manchester, 1981 M.Sc, The University of East Anglia, 1982 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY  in T H E F A C U L T Y OF G R A D U A T E STUDIES (DEPARTMENT OF CHEMISTRY)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA May 1991 © Terence C. Jarvis 1991  In  presenting  degree freely  this  at the  thesis  in  partial  fulfilment  University  of  British  Columbia,  available for  copying  of  department  this or  publication  of  reference  thesis by  this  for  his  and study. scholarly  or  thesis  her  for  of  C  H  DE-6 (2/88)  / 6j  I dj  <r (  I further  9 H  L ^  requirements  purposes  that the  agree  may  be  It  is  financial gain shall not  The University of British Columbia Vancouver, Canada  Date  the  I agree  representatives.  permission.  Department  of  that  Library  an  advanced  shall make it  permission for  granted  by  understood be  for  allowed  the that  without  extensive  head  of  my  copying  or  my  written  ii  A B S T R A C T  This thesis investigates the use of plant cell cultures in combination with synthetic chemistry to provide a new and inexpensive route to etoposide, an anti-cancer drug (1). Two facets of plant cell culture technology were investigated The first of these involved the use of cell cultures as a reaction medium in which synthetic precursors were added to the cultures and the bio-transformation products analyzed. The second facet investigated was the use of cell cultures of Podophyllum peltatwn to directly produce podophyllotoxin (3) and 4'-demethylpodophyllotoxin (5) in the culture medium. These compounds could then be utilized as intermediates in the production of etoposide (1). The synthetic studies involved the use of the readily available aromatic aldehydes, for example 3,4-dihydroxybenzaldehyde (89), as starting materials. In an efficient route employing a tandem conjugate addition of 86 and 88 to the butenolide 87, the precursor 85 was prepared. By a similar route an efficient synthesis of the precursors 63 and 64 were performed. Biotransformation of the precursors 62,63 and 64 with the whole cell culture of Catharanthus roseus resulted in the production of 71 as the only identified product. Compounds 62, 63 and 64 were treated with horseradish peroxidase (HRP) to compare plant cell enzymatic processes, with a commercially available peroxidase enzyme. The compound 63 was found to undergo bio-transformation to give the cyclized product 69 in 15-19% yield. Bio-transformation of the precursor 63 using a CFE (cell free extract) prepared from the cell culture of Catharanthus roseus.also produced the compound 69. Optimization of the conditions for this cyclization afforded 69 in 70% yield. The bio-transformation of (62) with the CFE prepared from the cell culture of Catharanthus roseus afforded 79.  The precursor 85 was treated with the C F E prepared from the cell culture of Catharanthus roseus and found to cyclize to yield the product 115. The same cyclization was achieved by treatment of the precursor 85 with the C F E prepared from the cell culture of Podophyllum peltatum. The precursor 116 was bio-transformed using both Podophyllum peltatum whole cells and the C F E of Catharanthus roseus to produce the compound 118. The compounds 69, 115 and 118 all have the necessary fused ring structure present in the podophyllotoxins, however they do not have the correct stereochemistry at position 1. Further work is in progress in our laboratory to address this problem. The second part of this thesis is concerned with the hydroxylation of the butanolide 69 and deoxypodophyllotoxin (6) to produce an intermediate suitable for conversion into etoposide (1). Reactions were carried out with the C F E prepared from the cell culture of Catharanthus roseus using various co-factors. Procedures were employed to stabilize and characterize the enzyme preparations. Hydroxylation experiments were carried out with the cell culture of Tripterygium wilfordii using the precursors 6 and 69. Hydroxylation, using the C F E prepared from the cell cultures of Catharanthus roseus, of the precursors 133 and 134 obtained fron 6, afforded the hydroxylated product 139. Thefinalpart of the thesis involves the isolation of the lignans podophyllotoxin (3), deoxypodophyllotoxin (6), podophyllotoxone (60) and 4'-demethylpodophyllotoxin (5), from the cell suspension culture of Podophyllum peltatum. This is the firsttimethat these lignans have been isolated from the cell cultures of Podophyllum peltatum.  85  M  CH 0^  OCH3  "OCH3  3  OH  OH  62 R , R ' = - C H 63 R > C H R = H 64 R , R ' = H 2  3 f  68  R,R' = -CH -  69  R = C H , R" = H  70  R,R' = H  2  3  OCH3  115  6  VI  118  116  vii Table of Contents  Page Abstract  ii  Table of Contents  vii  List of Tables  xiii  List of Figures  xvi  List of Schemes  xviii  List of Abbreviations  xxii  Acknowledgements  xxv  Chapter 1:  Introduction  1.1  Plant Cell Cultures in the Production of Organic Compounds.  1  1.2  The Development of Etoposide (1) from Podophyllotoxin (3).  8  1.3  Synthetic Approaches to the Podophyllotoxins.  20  1.4  The Biosynthesis of the Podophyllotoxins.  30  1.5  Bio-transformations Involving Radical Type Ring Closure Reactions Catalysis by Peroxidases  35  Results and Discussion  1.6  39  Attempted Bio-Transformation of Dibenzylbutanolides (62, 63, 64) using the Whole Cell Suspension Culture of Catharanthus roseus.  1.7  Bio-transformations of the Precursors 62, 63 and 64 using Horseradish Peroxidase (HRP).  1.8  39  The Bio-transformation of rra/i5-2-(3",5"-Dimethoxy-4"-  45  viii hyaroxybenzyO-S-^'-hyckoxy-^-methoxybenzy^butanolide (63) 49  with the CFE of Catharanthus roseus. 1.9  The Bio-tt^sformation of rra/w-2-(3",5"-Dimethoxy-4"hyckoxybenzyl)-3-(3\4'-memylenedioxybenzyl)butanoli(ie (62) 54  with the CFE bf Catharanthus roseus. 1.10  Synthesis of rm«5-2-(3",5"-Dimethoxy-4"-hydroxybenzyl)-359  (3'-hydroxy-4'-fa0-propoxybenzyl)butanolide (85). 1.11  The Synthesis of rran5-2-(3",5"-Dimethoxy-4"-hydroxybenzyl) -3-(3\4'-dlhydroxybenzyl)butanolide (64) and Trans-2-(3",5"Dimethoxy-4"-hydroxybenzyl)-3-(3,4'-methylenedioxybenzyl) l  68  butanolide (62). 1.12  The Bio-transformation of Trans-2-(3" ,5"-Dimethoxy-4"hydroxybenzyl-3-(3'-hydroxy-4-i4'o-propoxybenzyl)butanolide ,  f  73  (85) with the CFE of Catharanthus roseus. 1.13  The Bio-transformation of Trans-2-(3" ,5 "-Dimethoxy-4"hydroxybenzyl)-3-(3'-hydroxy-4-wo-propoxybenzyl)butanolide l  77  (85) with the CFE of Podophyllum peltatum. 1.14  Other Work in our Laboratory. Synthesis of Trans-2(3",5"-Dimethoxy-4"-hydroxybenzyl)-3-(3-hydroxy-4-wopropoxybenzyl-a-hydroxy)butanolide (116). ,  1.15  Future Prospects  Chapter 2:  2.1  ,  79 80  Introduction  Hydroxylations using Enzymes. Results and Discussion  82 87  ix  2.2  The Attempted Hydroxylation of the Dibenzylbutanolide 69 and Deoxypodophyllotoxin (6) with Plant Cell Cultures and CFEs. 2.2.1  Utilizing the Cyclization Conditions (Section 1.8).  2.2.2  Hydroxylation Attempts Using Co-factors and Enzyme Stabilizers.  2.2.3  87 87 89  The Characterization of the Microsomal Fraction from Catharanthus roseus and Further Attempts at the Hydroxylation of 6.  2.2.4  Hydroxylation Attempts with the Cell Culture of Tripterygium wilfordii.  Chapter 3:  3.1  103  Introduction  The Isolation of the Podophyllotoxins from Plant Cell Cultures. Results and Discussion  3.2  95  114 117  The Isolation of Lignans from the Cell Culture of Podophyllum peltatum.  Chapter 4:  117  Experimental  4.0  General Introduction.  129  4.1  Synthesis of Precursors.  130  4.1.1  3-Hydroxy-4-wo-propoxybenzaldehyde (93).  130  4.1.2  4-Hydroxy-3-/i'o-propoxybenzaldehyde (94).  131  4.1.3  3,4-Di-wo-propoxybenzaldehyde (95)  132  4.1.4  3-Benzyloxy-4-i\s0-propoxybenzaldehyde (96)  132  4.1.5  3-Benzyloxy- l-bis(phenylthio)methyl-4-/i'opropoxybenzene (86).  133  4.1.6  3-((3'-Benzyloxy-4-i^o-propoxy)-a,a-bis(phenylthio)  4.1.7  benzyl)butanolide (97). 7ra/ij-2-(4"-Benzyloxy-3",5"-dimethoxybenzyl)-3-((3'benzyloxy-4'-w<9-propoxy)-a,a-bis(phenylthio)benzyl)  ,  butanolide (98). 4.1.8  ^ra/I^2-(4"-Benzyloxy-3^5"-dimethoxybenzyl)-3-((3,  ,  benzyloxy-4-j,so-propoxy)-a,a-bis(phenylthio)benzyl) l  butanolide (98). 4.1.9  rran5-2-(3",5"-Dimethoxybenzyl-4"-hydroxybenzyl) -3-(3'-hydroxy-4-j4o-propoxybenzyl)butanolide (85). ,  4.1.10  ,  rra/w-2-(3",5"-Dimethoxybenzyl-4"-hydroxybenzyl) -3-(3',4'-methylenedioxybenzyl)butanolide (62)  4.1.11 rra/w-2-(3",5 -Dimethoxy-4 -hydroxybenzyl)-3,,  ,,  (S'^'-dihydroxybenzylJbutanolide (64) 4.1.12 Demethylenedeoxypodophyllotoxin (133) The Propagation of the Cell Cultures and the Preparation of Cell Free Extracts. 4.2.1 Propagation of the Catharanthus roseus PlantCell Culture 4.2.2 4.2.3 4.2.4 4.2.5  Preparation of the Cell Free Extract from Plant Cell Cultures of Catharanthus roseus Propagation of the Podophyllum peltatum PlantCell Culture Preparation of the Cell Free Extract from Plant Cell Cultures of Podophyllum peltatum Propagation of the Tripterygium wilfordii Plant Cell Culture  Attempted Bio-transformation of Dibenzylbutanolides (62, 63,64) using the Whole Cell Suspension Culture of Catharanthus roseus Bio-transformations using Horseradish Peroxidase (HRP) 4.4.1  Initial Small Scale Experiment  4.4.2  Bio-transformation of 63 with HRP. Large Scale, Experiment 1  4.4.3  Bio-transformation of 63 with HRP. Large Scale, Experiment 2  The Evaluation of pH Dependency on the Ring Closure of 63  xi by the CFE of Catharanthus roseus  151  4.6  Attempted Ring Closure of 62 Using the CFE of Catharanthus roseus  154  4.7  Attempted Bio-transformation of 63 with CFE from Podophyllum peltatum  4.8  156  Bio-transformation of rra«5-2-(3",5"-Dimethoxy-4"hydroxybenzylJ-S-CS'-hydroxy^'-iw-propoxybenzy^butanolide (85) with the CFE of Catharanthus roseus.  4.9  157  4.8.1  Experiment 1 Small Scale.  158  4.8.2  Experiment 2 Large Scale.  159  4.8J  Experiment 3 Large Scale.  160  The Ring Closure of Trans-2-(3",5"-Dimethoxy-4"hydroxybenzyl-3-(3'-hydroxy-4'-wo-propoxybenzyl)butanolide (85) with the Cell Free Extract of Podophyllum peltatum Cell Culture.  161  4.10 Attempted Hydroxylation of 69 with the CFE of Catharanthus roseus. 4.10.1 Initial Experiment. 4.10.2 Co-factors Experiment 1. 4.10.3 Co-factors Experiment 2. 4.10.4 Co-factors Experiment 3. 4.11  Attempted Bio-Transformations of Deoxypodophyllotoxin (6) with the CFE of Catharanthus roseus.  4.12  The Use of Enzyme Stabilizers.  4.13  Characterization of the Cell Free Extract from Catharanthus roseus.  4.14  162 162 163 164 165  168 169  171  Attempted Bio-transformation of Deoxypodophyllotoxin (6) the S10 Fraction of the Catharanthus roseus Cell Extract.  173  xii  4.15  Evaluation of Hydroxylation of Deoxypodophyllotoxin (6) and 69 with the S10 Fraction of Catharanthus roseus with Various Co-factors. Small Scale.  4.16  4.17  4.18  174  Attempted Hydroxylation of Deoxypodophyllotoxin (6) with Whole Cells of the Tripterygium wilfordii Cell Suspension Culture.  175  Attempted Hydroxylation of 69 and 6 with the CFE Prepared from the Tripterygium wilfordii Cell Suspension Culture.  176  Attempted Hydroxylation of the Precursor 133 with the Whole Cells of the Tripterygium wilfordii Cell Culture.  4.19  4.20  Attempted Hydroxylations of 4'-Demethyl-6,7-demethylenedeoxypodophyllotoxin (134) and 6,7-demethylenedeoxypodophyllotoxin (133) using the CFE of Catharanthus roseus with Various Co-factors.  179  Attempted Hydroxylation of 4'-Demethyl-6,7-demethylenedeoxypodophyllotoxin (134) with the CFE of Catharanthus roseus.  4.21  177  The Isolation of Metabolites from Podophyllum peltatum Cell Culture. 4.21.1 Initial Experiments. 4.21.2 Investigation of the Major Constituents of the Podophyllum peltatum Cell Culture. Experiment 1 Shake Flask Culture and Microferm Culture. 4.21.3 Experiment 2. Large Scale Extraction of 5.5 L of Podophyllum peltatum Microferm Culture.  182  184 184  185 187  5.0  Appendix  190  6.0  References  203  Xlll  List of Tables  Table 1:  Toxicity, Cytostatic Potency and Anti-tumor Activity of the Podophyllum Lignans in Comparison with their Glycosides  Table 2:  Pharmacological Activities of some of the Benzaldehyde Condensation Products of the Podophyllum Glycosides  Table 3:  IS  Large scale bic>-transformations of the dibenzylbutanolide 63 with HRP  Table 4:  13  48  Optimum Conditions for the Cyclization of the Precursor 63 Using the CFE of Catharanthus roseus  Table 5:  51  The Bio-transformation bf 62 Using the CFE of  Catharanthus roseus Table 6:  The Bio-transformation of the Dibenzylbutanolide 85  Table 7:  The Conditions for the Attempted Hydroxylation  55 75  of 69 using the CFE of Catharanthus roseus Table 8:  The Attempted Hydroxylation of 69 Using Various Co-factors  Table 9:  96  The Specific Activities and Protein Concentration of the Centrifugation Fractions  Table 12:  91  The Major Factors Required for the Hydroxylation of Geraniol (126) and Nerol (128) as Measured by Coscia  Table 11:  89  Attempted Hydroxylation of 69 Using Various Co-factors and Stabilizing Agents  Table 10:  89  99  Small Scale Reactions Using the S10 Fraction and NADPH and /or FAD as Co-factors to hydroxylate 6 and /or 69  102  xiv  Table 13:  Attempted Hydroxylation of 69 and 6 with the CFE of Tripterygium wilfordii  Table 14:  Conditions for the Bio-transformation of 63 Using HRP  Table 15:  150  Bio-ttansformations of 63 Using Catharanthus roseus at Different pH Values  Table 17:  148  Recovery From the Bio-transformation of 63 with HRP  Table 16:  106  152  Peroxidase Activity of the CFE of Catharanthus roseus Prepared at Different pH Values  153  Table 18:  The Bio-transformation of 62  154  Table 19:  Extraction from the Bio-transformation of 62  155  Table 20:  Co-factor Experiment 2  165,  Table 21:  Extraction From Co-factor Experiment 3  167  Table 22:  Bio-transformation of 6  169  Table 23:  Buffer A  170  Table 24:  Buffer B  170  Table 25:  The Specific Activities and Protein Concentration of the Centrifugation Fractions  172  Table 26:  The S10 Fraction Using Co-factors  174  Table 27:  Attempted Hydroxylation of 69 and 6 with the CFE of Tripterygium wilfordii  Table 28:  Tripterygium wilfordii Whole Cell attempted Biotransformation  Table 29:  111  178  Small Scale Co-factor Experiment with Precursor 133  181  X V  Table 30:  Extraction of the initial isolation experiment from Podophyllum peltatum  Table 31:  184  Murashige and Skoog (M.S.) Medium (1962) for the Propagation of the Podophyllum peltatum Cell Culture  Table 32:  1-B5 Medium used for the Propagation of Catharanthus roseus  Table 33:  PRD2C0100 Medium for the propagation of the Tripterygium wilfordii Cell Culture  200 201  202  xvi  List of Figures  Figure 1:  The Structures of Etoposide ( 1 ) and Teniposide (2)  Figure 2:  Podophyllotoxin Lignans  10  Figure 3:  Lignan Glycosides  12  Figure 4:  Podophyllotoxin Benzylidene Glycosides  14  Figure 5:  4'-Demethylepipodophyllotoxin Benzylidene Glycoside ( 1 7 )  17  Figure 6:  Dibenzylbutanolide Precursors Initially Prepared for Bio-transformations  Figure 7:  35  The Product Isolated from the Biotransformation of  Figure 8:  7  6 4  45  A Possible Biosynthetic Intermediate in the Biosynthesis of Podophyllotoxin  54  Figure 9:  The Restricted Rotation on the wo-Propyl Group in 1 1 5  77  Figure 10:  The C O Absorption Spectrum of the P100 Fraction  100  Figure 11:  The C O Absorption Spectrum of the P100 Fraction as Measured by Coscia  100  Figure 12:  The Structure of 5-Methoxypodophyllotoxin ( 1 4 0 )  115  Figure 13:  The EI M S Fragmentation Pattern of 6 3 Showing the Important Fragments  Figure 14:  190  The EI M S Fragmentation Pattern of 6 9 Showing the Important Fragments  191  Figure 15:  Single crystal X-ray structure of 6 9  192  Figure 16:  !H-NMR and N O E difference spectrum of 9 3  (400 M H z , CDCI3) Figure 17:  EI M S spectrum of 8 5  Figure 18:  *H-NMR spectrum of 8  193 194  5 (400 M H z , CDCI3)  195  Figure 19:  ^-NMR spectrum of 115 (400 MHz,  Figure 20:  EI MS spectrum of 115  Figure 21:  The EI MS Fragmentation Pattern of 133 Showing the  CDCI3)  Important Fragments Figure 22:  iH-NMR spectrum of 139 (400 MHz, CDCI3)  xviii  TABLE OF SCHEMES  Scheme 1:  The Production of Plant Cell Suspension Cultures  3  Scheme 2:  The Preparation of a Cell Free Extract (CFE)  5  Scheme 3:  The Mechanism of Action of a Spindle Poisons  9  Scheme 4:  The Synthesis of the Benzylidene Glycoside 17 of Epipodophyllotoxin (18)  Scheme 5:  The Isomerization of Picropodophyllin (4) and Podophyllotoxin (3)  Scheme 6:  22  The Use of the Diels-Alder Reaction in the Synthesis of Podophyllotoxin (3)  Scheme 8:  21  Gensler's Synthesis of Podophyllotoxin (3) and Picropodophyllin (4)  Scheme 7:  18  25  Rodrigo's Synthesis of Podophyllotoxin (3) and Epipodophyllotoxin (37)  26  Scheme 9:  The Synthesis of 4'-Demethylepipodophyllotoxin (18)  29  Scheme 10:  Some of the Early Intermediates in the Biosynthesis of Lignans  Scheme 11:  The Proposed Intermediates in the Latter Stages of the Biosynthesis of Podophyllotoxin (3)  Scheme 12:  32  The Biosynthetic Interconversion of Some of the Podophyllotoxins  Scheme 13:  The Mechanism of Action of a Typical Peroxidase  Scheme 14:  The Coupling Reaction of Catharanthine (65) and Vindoline (66)  Scheme 15:  31  34 36  38  The Proposed Bio-transformation of the Precursors (62,63 and 64)  39  xix  Scheme 16:  Cyclizarion of 63 to 69 using HRP  Scheme 17:  Cyclizarion of 63 to 69 using the CFE of  46  Catharanthus roseus Scheme 18:  49  The Proposed Mechanism for the Oxidative Cyclizarion of 63 to 69  52  Scheme 19:  The Chemical Cyclizarion of 76 to Give the Product 77  53  Scheme 20:  The Bio-transfomiation of 62 with Catharanthus roseus  Scheme 21:  The Proposed Mechanism for the Oxidative Demethylarion  55  of 62  58  Scheme 22:  The Retrosynthetic Analysis of 85  59  Scheme 23:  The Alkylation of 3,4-Dftydroxybenzaldhyde (89) Using the Method of Kessar  60  Scheme 24:  The Synthesis of 93,96 and 86  62  Scheme 25:  The Synthesis of 97 and 98  65  Scheme 26:  The One-step Tandem Addition Route to 98  66  Scheme 27:  The Major Fragments Produced in the EI MS of Compound 85  67  Scheme 28:  The Initial Synthetic Route to Precursors 62,63 and 64  69  Scheme 29:  The Synthesis of 62 and 64 via a Thioketal  72  Scheme 30:  The Proposed Bio-transformation of 85 to 115 and Conversion to an Etoposide Intermediate  Scheme 31:  The Cyclizarion of 85 Using the CFE of Catharanthus roseus to Produce 115  Scheme 32:  74  75  The Cyclizarion of 85 to 115 Using the CFE of  Podophyllum peltatum Scheme 33:  79  The Synthesis of Precursor 116 and Subsequent Cyclizarion to Product 118  81  XX  Scheme 34:  Two Examples of Hydroxylation Reactions Using P-450s  83  Scheme 35:  The Preparation of the Microsomal Fraction  84  Scheme 36:  The Electron Transfer System From NADPH to P-450  85  Scheme 37:  The Mode of Action at the Active Site of P-450  86  Scheme 38:  The Proposed Hydroxylation of Precursor 69 Using the CFE of Catharanthus roseus  88  Scheme 39:  The Oxidative Denaturing of Cysteine  90  Scheme 40:  The Oxidation of Dithiothreitol (124)  91  Scheme 41:  The Proposed Hydroxylation of Deoxypodophyllotoxin (6) Using the CFE of Catharanthus roseus  Scheme 42:  The Enzymatic Hydroxylation of Geraniol (126) and Nerol (128)  Scheme 43:  The Procedure used to Prepare the Microsomal Pellets  Scheme 44:  The Bio-d^sformation of Tobacco Cembraniods with the Cell Culture of Tripterygium wilfordii  Scheme 45:  98  104  105  The Preparation of the Precursors 133 and 134 From Deoxypodophyllotoxin (6)  Scheme 47:  96  Proposed Bio-transformations using Tripterygium wilfordii  Scheme 46:  94  108  The Proposed Degradation of 1,2-Dihydroxy Aromatic Compounds  109  Scheme 48:  The Proposed Hydroxylation of the Precursors 133 and 134  111  Scheme 49:  The Synthesis of 138 and 139 From Podophyllotoxin (3)  112  Scheme 50:  The Microbial Hydroxylation of Deoxypodophyllotoxin (6) to Podophyllotoxin (3)  Scheme 51:  The Extraction Procedure used for the Initial Experiments for the Isolation of Lignans from the Cell Culture of  113  XXI  Podophyllum peltatum Scheme 52:  118  The Extraction Procedure for the Isolation of the Podophyllotoxins from the Podophyllum peltatum Cell Culture  Scheme 53:  120  Chromatographic Separation of the Combined Shake Flask and Mecroferm Bioreactor Extracts  122  Scheme 54:  The Modified Large Scale Extraction Procedure  124  Scheme 55:  Chromatographic Separation of the Extracts From the 5.5 L Microferm Bioreactor  125  List of Abbreviations Ac  acetyl  aq Ar  aqueous aryl  br f-Bu c C  broad tert-butyl concentration (g/100 ml) Celsius  CFE  Cell Free Extract  cm* 5  wavenumber chemical shift  d DCC  doublet dicyclohexylcaiftodiimide  dd ddd DMF DMSO DNA DTT e  doublet of doublets doublet of doublet of doublets N, N-dimemylformamide dimrthyl sulfoxide Deoxyribonucleic acid (UtWothreitol extinction coefficient  EDTA EI MS  ethylene diamine tetraacetic acid Electron Impact Mass Spectroscopy  Et FAD FDA FMN g G GC LCMS  ethyl flavoadenine dinucleotide Food and Drug Adrriinistration flavoadenine mononucleotide gram gravitational constant gas-liquid chromatography Uqmd chromatography-mass spectrometry  hr hv  hour wavelength  HMDS HMPA  hexamethyldisilylazide hexamemylphosphorattiamide  1  xxiii HPLC  high performance liquid chromatography  HRMS  high resolution mass spectroscopy  HRP  horseradish peroxidase  Hz  hertz  TJD50  inhibitory dose SO per cent  IR  infrared  J X  coupling constant wavelength  L  litre  LDA  lithium dnsopropylamide  LD50  Lethal Dose 50 per cent  m  multiplet  mM  nrillimolar  M  molar  M  +  molecular ion  mg  milligrams  MHz  megahertz  min  minute  uX,  microlitre  mL  millilitre  mmol  millimole  mol  mole  MS  mass spectrometry  mu  molecular unit  m/z  mass to charge ratio  Mp  melting point  m/z  mass to charge ratio  v  frequency  NADPH  nicotinamide adenone dinucleotide phosphate (reduced form)  NHI  non-iron heme protein  nm  nanometre  NMR  nuclear magnetic resonance  NOE  nuclear Overhauser effect  PCC  pyridinium cMorochromate  Ph  phenyl  ppm  parts per million  xxiv q  quartet  R  alkyl substituent  Rl  refractive  index  rpm  revolutions per minute  s  singlet  t  triplet  THF  tetrahydrofuran  TLC  thin layer chromatography  TMS  trimethylsilyl (or tetramethylsilane)  p-Ts  para-tosyl  UV  ultraviolet  w/v  weight per volume  ACKNOWLEDGEMENTS  I wish to express my gratitude to Professor James P. Kutney for his guidance and encouragement throughout the course of my research, and in the preparation of this thesis. I am also very grateful to the following list of people who have provided important support and services during the preparation of this thesis : Dr. Raymond Anderson, Dr.Charles Stone and Dr. Darlene Anderson, George Weetman, Philip Gunning, Joanne Weetman, Jan Palaty and Mark Aston. I would also like to thank Mr Gary Hewitt and Faye Hutton for many helpful discussions and for their assistance in the preparation of plant cell cultures. I am also indebted to Marshall Lapawa, Peter Borda and the staff of the UBC NMR facility for their technical assistance. Finally, I am deeply indebted to my parents for their support and encouragement  1 CHAPTER 1 INTRODUCTION 1.1  PLANT CELL CULTURES IN THE PRODUCTION OF ORGANIC COMPOUNDS  The production of useful compounds by plant cell culture biotechnology is an expanding area of science. Plants and their extracts have been used by mankind for centuries as medicines. As organic chemistry developed, many of the individual components of these preparations were isolated and their pharmacological modes of action were investigated. Thousands of organic molecules have now been directly isolated from plants and the synthesis of these natural products has provided a challenge to organic chemists. Research in plant biotechnology is directed toward manipulating plants so they are able to produce useful compounds in high yields, and to enable the plants, or their extracts to be exploited as biochemical "reagents" for the economical production of useful chemicals.  1  Several approaches may be taken toward the commercial production of chemicals from plants. The isolation of compounds from the native plant is the most direct approach, however, several problems may be encountered. The physical inaccessibility of the area in which the plant grows, or political factors, may limit the availability of the plant. Seasonal fluctuations will affect plant growth, and may also affect the build-up of specific metabolites in the plant. Many plants take years to reach maturity or accumulate metabolites and this can severely limit productivity. Also, production may be limited by the extraction procedure, which could require large masses of undifferentiated plant material, making procedures long and expensive. The low yields obtained from such isolation procedures result in processes which are often inappropriate for the commercial production of organic compounds. The use of plant cell culture techniques can eliminate many of these problems  and thus allow the utilization of plants as a potential warehouse of chemicals and reagents for the synthetic organic chemist. In 1898, Haberlandt was the first to attempt to culture single cells of various 2  plants. Despite his initial insight into the potential of cell cultures, he did not observe any cell division. In the next 30 years, many others tried to propagate cell cultures, and 2  although cultured cells were kept alive for extended periods of time, no cell division was observed. In 1934, White, successfully propagated a cell culture from the roots of the 2  tomato. In later work, White made use of growth factors such as the B vitamins and the 2  auxin, indol-3-yl-acetic acid for growth of cultured roots. Cell cultures have now been established for many plants. Scheme 1 gives an outline of the processes involved in establishing a typical plant cell culture. The initial 3  stage involves the removal of a section of plant tissue under aseptic conditions. This material, known as the explant is then placed on nutrient agar. The nutrient agar typically consists of a dissolved carbohydrate such as sucrose, inorganic salts and small amounts of plant growth auxins. The agar base consists of gelatin. The explant may be a piece of root, leaf or another part of the plant After incubation a callus culture is produced which is a mass of disorganized cells growing on the solid nutrient. A sample of this callus may then be removed and placed on fresh agar medium to produce a second callus. This process is called sub-culturing. If a sample of the callus is placed in a liquid nutrient medium a cell suspension culture is obtained. The liquid nutrient medium is prepared using similar constituents to those used to produce the nutrient agar, the base, however, is not a gelatin support as in the agar, but is aqueous. Cell suspension cultures can be sub-cultured in the same way as the callus culture. The ease with which a culture may be initiated will be dependent on the the type of plant chosen and thetissue'sorigin. The growth of the cells can be monitored by measurement of the cell mass at different culture ages. This monitoring enables determination of the time period between sub-cultures. Sub-culturing usually occurs just after the end of the growth phase.  3  Whole plant  •  •  Explant  •  •  Callus culture  •  •  Suspension culture Scheme 1. The Production of Plant Cell Suspension Cultures Knowledge of the growth characteristics of the plant cell culture is important in determining the culture age which will possess the necessary enzymes, to enable the biotransformation to be successfully accomplished. Once a cell culture has been established, growth may be rapid, and environmental factors such as light, temperature and gaseous exchange can be carefully controlled to promote growth. This eliminates the restrictions placed upon the plant by seasonal fluctuations. The use of sterile conditions removes the need for pesticides or herbicides, and production control enables yields and product quality to be closely monitored. Also, the yield of the compound of interest may become higher than that achieved from the intact  plant. In some cases compounds have been isolated from cell culture which were not 4  found in the original plant.5 The list of compounds which have been isolated from plant cell cultures is extensive and includes, alkaloids, vitamins and compounds used in 6  7  perfumery.8 The use of plant cells as a type of biochemical 'reagent' is a very exciting prospect. Bio-transformations involve addition of the substrate to the plant cell culture under sterile conditions. Next, there is an incubation period prior to the harvest of the cells and the recovery of the product. If the product is present in the culture broth, the cells can be filtered and resuspended in fresh medium. The product can then be extracted from the broth. However if the product is trapped in the cells, the cell material must be homogenized to break up the plant cells so the product may be extracted.9.10 There are many advantages to whole cell bio-transformations; they are simple to carry out, isolation may be easy if the product is contained in the cell broth and the co-factors necessary for the enzyme's function are generated in situ. Also, the cells are not destroyed by the process so enzyme integrity is preserved during the bio-transformation. Before bio-transformations are attempted there is usually an initial investigation of the plant to identify the metabolites which may be involved in the biosynthesis of the required compound. Based upon these results a suitable precursor can then be chosen and a bio-transformation attempted. This approach presupposes that the compound of interest is produced naturally by a plant and that a cell culture has been established for the species. If this is not the case, then bio-transformations can be attempted using plant cell cultures which have been found to have, or are presumed to have, the necessary enzyme systems for the desired bio-transformation. However, some problems can arise in whole cell bio-transformations. The substrate must have some degree of solubility in the cell culture growth medium, and must be able to diffuse or transport through the cell wall and membrane. The difficulties of product isolation are increased if the product does not diffuse out of the plant cells.  5 Another fundamental problem of whole cell culture bio-transformations is the complexity of the plant cell biochemistry. To produce the desired product there is a need for specificity 11  in the enzymatic transformations of the substrate. Often, many competing reactions can occur involving different bio-transformations and subsequent degradations to produce undesirable products. If the cells are disrupted and the contents extracted and purified, then any problems connected with cell membrane impermeability are eliminated and also the number of possible enzymatic reactions may be reduced. Typically the cell culture is filtered and the cells are disrupted by homogenization in the presence of a buffer at a low temperature (Scheme 2).  Cell Suspension Culture Filtration Cells Buffer added  Homogenization  Cool in ice bath  Homogenate  Cell Debris  Supernatant Cell Free Extract (CFE)  Scheme 2. The Preparation of a Cell Free Extract (CFE)  The disrupted cells can be visualized by the use of cell staining techniques, in which the degree of cell disruption, can be monitored by the examination of the suspension with a microscope. At this stage the suspension produced is called a homogenate. This homogenate is then centrifuged to remove cell debris.  The supernatant, after  centrifugation, is called the cell-free extract (CFE) and contains many of the enzymes originally present in the culture but they are now in a cell-free form. Various preparative techniques can be used to purify the CFE to enable isolation of classes of enzymes or individual enzymes. All of these preparations; crude CFE, partially purified enzymes and purified enzymes can be used to carry out bio-transformations. Partially purified enzymes and purified enzymes have also been immobilized onto solid supports and used to carry out bio-transformations.  12  Bio-transformations performed by CFE occur in an homogenous solution and so reaction conditions can be precisely controlled. Also, because less cell material is present during the bio-transformation isolation procedures are not as tedious as isolations from whole cell bio-transformations. There are problems associated with CFE preparations. A significant problem, on the production scale, is the extra number of manipulative steps involved in the "bio-reagent" preparation. Caution must be employed during the CFE preparation not to destroy the enzyme of interest, as many enzymes are very labile and subject to damage caused by osmotic shock or changes in pH or temperature. An additional problem is caused by some of the enzymes which are released during cell disruption, such as proteases. These enzymes will immediately begin using released proteins, and hence other enzymes as substrates for degradation. This can be prevented to some degree by cooling the cells and adding buffers and enzyme stabilizers and protease inhibitors before homogenization. The studies in this thesis are concerned with the exploitation of plant cell cultures as a potential tool for the production of etoposide (1) (Figure l ) .  13  Etoposide (1) is a  clinically used anti-tumor agent, and is produced from a class of lignans called the  podophyllotoxins, which occur naturally in several plant species. Teniposide (2) closely related compound which also pocesses anti-tumor activity.  Etoposide (1)  Teniposide (2) Figure 1. The Structures of Etoposide and Teniposide  8  1.2  THE DEVELOPMENT OF ETOPOSIDE (1) FROM PODOPHYLLOTOXIN (3)  The medicinal properties of the podophyllotoxins have been recognized for centuries. Natives on the American and Asian continents used the dried roots of Podophyllum emodi (Himalayan) and Podophyllum peltatum (American) as cathartics and anthelmintics. In 1820, podophyllin, obtained as the alcoholic extract of the rhizomes of these plants, was included in the United States and other pharmacopoeia.  14  Podophyllotoxin (3) (Figure 2) was isolated as the major constituent of podophyllin in 1880 by Podwyssotzki. In 1942, Kaplan described the effect of podophyllotoxin (3) on benign tumors, and four years later a mechanism of action for this anti-tumor activity was suggested.^ The active constituent of podophyllin was found to stop cell mitosis at the metaphase stage of cell division. Metaphase is a phase of cell division when the nuclear spindle is formed, the nuclear spindle is composed of protein microtubules. Protein microtubules were discovered around the sametimeas the podophyllin mode of action was discovered. They are cylindrical in shape and composed of a globular protein called tubulin. Tubulin exists in two forms that are in equilibrium, free tubulin 16  and tubulin in microtubules (Scheme 3). Some compounds such as vincristine, vinblastine and colchicine were found to bind to free tubulin causing it to precipitate, thus shifting the equilibrium away from microtubule formation and so stopping cell division. In 1951 the structure of podophyllotoxin (3) was determined and several other similar lignans were isolated (Figure 2). ? 1  Many common structural features were present in these new lignans, and this suggested the existence of some common biosynthetic precursor (Section 1.4). Some of the lignans were used clinically, in particular podophyllotoxin (3), which is still used topically, in the treatment of some venereal diseases. However, except for some superficial  skin cancers, the lignans were inferior to many of the anti-cancer drugs already available. Thus, in the 1950's the use of these compounds for the treatment of malignant cancers ceased.  Tubulin in Microtubules  Free Tubulin Podophyllotoxin Precipitation N o microtubules N o Nuclear Spindle N o Metaphase  N o C e l l Division  Scheme 3. T h e Mechanism of Action of a Spindle Poison Researchers at the Sandoz Ltd had successfully isolated some cardiac glycosides, <• such as digitalis, from plants, and they began an investigation of podophyllin, using enzyme inhibitors to isolate the glycosides of the podophyllotoxins. Four pure glycosides were initially isolated (Figure 3).  18  OH  OH  OR  Podophyllotoxin (3): R = C H  OCH  3  3  Picropodophyllin (4)  4-Demethylpodophyllotoxin (5): R = H OH  a-Peltatin (8): R = H Figure 2. Podophyllotoxin Lignans The pharmacology and toxicity of these compounds has been investigated (Table 1). The LD50 test, conducted with mice, showed the glycosides to be less toxic than the aglycones. The toxicity of the aglycones had been a serious clinical problem. The LD50 test measures the dose of compound required to kill 50% of the mice in the tested population. Thus, the more toxic compounds can be administered in the lower doses in  order to kill 50% of the mice, when compared to the doses required for less toxic compounds. Two other assays were carried out but their results were less favorable. The first of these was the ID50 P-815 test which is an in vitro measure of the cytostatic potency, or the capability to inhibit cell proliferation in mouse tumor cells. The lower these values the greater the activity of the compounds tested. As the figures indicate (Table 1) the glycosides were less effective than the aglycones in this test. Anti-tumor activity was also measured using the L-1210 test involving the inoculation of mice with a specific form of leukemia. These mice are then treated with the compound of interest, and the average increase in the mouse life-spans are then measured and compared to the life-spans of a control group of mice. The glycosides were again found to have a reduced activity in comparison with the aglycones. Many modifications have been made to these glycosides to increase cytostatic potency. One aim was to decrease the ease of glycosidic cleavage occurring in the intestinal tract of the animals. Modifications were made to the sugar hydroxy groups, including selective blocking of the hydroxy groups of the sugar, by condensation with benzaldehyde to produce the condensation product shown in Figure 4.  OR  Podophyllotoxin-P-D-glucopyranoside (9) R = C H  3  4'-DememylpodophyUotoxin-P-D-glucopyranoside (10) R  p-Peltatin-P-D-glucopyranoside (11) R = C H a-Peltatin-P-D-glycopyranoside (12) R = H Figure 3. Lignan Glycosides  3  TABLE 1  TOXICITY, CYTOSTATIC POTENCY AND ANTI-TUMOR ACTIVITY OF T H E P O D O P H Y L L U M LIGNANS IN C O M P A R I S O N W I T H T H E GLYCOSIDES™  Compound  Toxicity LD50  BDP-815  L-1210  mouse mg/Kg  Lig/rnL  % increase in life span  Podophyllotoxin (3)  35  0.005  35  4'-Demethylpodophyllotoxin (5)  >120  0.007  10  PicropodophylUn (4)  0.2  Deoxypodophyllotoxin (6)  52  0.002  27  P-Peltatin (7)  27  0.001  23  a-Peltatin (8)  13  0.001  2  (9)  297  6  7  2  0  (10) (11)  >200  5  (12)  >200  10  14 H  OR  Podophyllotoxin series. (13) R = C H (14) R = H  3  H  OR  Peltatin series. (15) R = C H (16) R = H  3  Figure 4. Podophyllotoxin Benzylidene Glycosides  15  TABLE 2  P H A R M A C O L O G I C A L ACTIVITIES  OF SOME OF THE  BENZALDEHYDE CONDENSATION PODOPHYLLUM  Compound  R  number 13  CH  14  H  15  CH  16  H  3  3  Proesid™  PRODUCTS OF T H E  GLYCOSIDES  LD50 mouse  ID50P8I5  L-1210 % increase  mg/Kg  Lig/mL  in life span.  240  3  5  160  0.8  27  360  3  6  350  1  2  214 iv  0.5  65  The products were found to be five to ten times less toxic than the podophyllotoxins (compare with Table 1), whilst having a higher cytostatic potency than the unprotected glycosides (Table 2).  However, they were described as having "borderline or.  msignificant" anti-tumor activity. The relatively poor anti-tumor activity but high inhibition of mitosis for these compounds, suggested that they had a mode of action other than as a spindle poison. A spindle poison is a compound which has a mode of action which stops the formation of microtubules prior to cell division (see page 9). The benzylidene compound 13 (called SPG-400) has been investigated further and found to have good chemical stability and good absorption in the gastrointestinal tract. In 1959 it was used clinically and found to have a positive effect in some patients suffering from cancer.  The twist in this story now began to develop. Throughout this period, because of the low cost and ease of preparation of the podophyllin extract, it was used as a mixture to produce the benzylidene glycoside 13. The mixture produced consisted of:- 80% podophyllotoxin benzylidene glycoside (13) -10% 4'-demethylpodophyllotoxin benzylidene glycoside (14) - 10% others including free glycosides, aglycones and unidentified products. This preparation was known commercially as Proesid™, and in 1962 a thousand patients suffering from cancer were treated with this mixture and 12% of them showed improvement. Comparison of the activities of this mixture (Table 2) with the pure components showed that the crude rnixture had a higher pharmacological activity toward tumor cells. This mixture was then investigated extensively to find the "active ingredient". Eventually, in 1965, by a process of systematic chromatography a new compound 17 was isolated (Figure 5). The "active ingredient", 4'-demethylepipodophyllotoxin benzylidene glycoside (17) has an ID50 P-815 of 0.007 u,g/mL, which means that in the in vitro assay 0.007 |Xg/mL of the compound is required to inhibit cell proliferation of the mouse tumor cells used in the assay. The glycoside 17 has an L-1210 value of 97%, which means that the life spans of the mice used in this in vivo.test were increased by 97% when treated with the glycoside 17.in comparison to the life spans of the control group. The test involves initial inoculation of all of the mice with L-1210 leukemia. The problem facing investigators at the time was the limited availability of this compound which was present in minimal amounts in the plant. Research emphasis was placed on development of a synthetic procedure and in 1968 the first stereoselective synthesis of podophyllotoxin (3) was accomplished (Section 1.3).  17 H Ph  o  \<o CH o  OCH3  3  OH  Figure 5. 4'-Demethylepipodophyllotoxin Benzylidene Glycoside (17)  The "active ingredient" 17 could now be synthesized. Podophyllotoxin (3) was converted into 4'-demethylepipodophyllotoxin (18) which was then acylated and the derivative 19 converted in four steps to 4'-demethylepipodophyllotoxin benzylidene glycoside (17) (Scheme 4).  19  The initial epimerization of position 1 of podophyllotoxin  (3) is not necessary in this route because glycosylation of the oc-isomer of 19 also produces the ^-glycoside 20. Using this route, over sixty condensation products of the glycoside were prepared and assayed for anti-tumor activity 20 at Sandoz. Teniposide (2) was first clinically tested in 1967 and etoposide (1) was tested in 1971. The results were initially encouraging, but by the mid-1970's cancer research was no longer a priority at Sandoz. The compounds were licensed to the Bristol Meyers company in 1978 where they continued development and, in 1983 etoposide (1) was licensed in the United States and approved by the FDA for the treatment of testicular cancer. It is now also used in the treatment of small cell lung cancer, hepatoma and other tumor diseases.2l The podophyllotoxins have also been found to possess anti-viral, insecticidal and phytotoxic activities.22  18 OH  1) HBr 2) BaC0  3  CH O 3  18  Podophyllotoxin (1) ClC0 CH Ph 2  2  CH OAc 2  2,3,4,6-tetra-Oacetyl-p-D-gluco pyranose. BF .Et 0 3  2  CH3O  CH3O  OC0 CH Ph  OC0 CH Ph  20  19  2  2  2  Zn(OAc)  2  21 Scheme 4. The Synthesis of the Benzylidene Glycoside 17 of 4'-Demethylepipodophyllotoxin (18)  2  19  CH OH  CH OH  2  2  21  22  PhCHO  17  Scheme 4 (continued)  The anti-cancer properties of etoposide are believed to arise from the compounds ability to interfere with DNA topoisomerase II. Topoisomerase U is an enzyme which creates reversible double-stranded breaks in DNA. These breaks prevent any tangling occurring during DNA replication.23 In summary, this is one of the growing examples of "folk remedies" finding a place in modem medicine. It is interesting to note that initial clinical tests with plant extracts did not look positive. It was the almost accidental isolation of a minor component of the mixture, which led to not only a potent compound, but one with a different mechanism of action. Further development of this compound led to the discovery of etoposide (1) and teniposide (2).  1.3  SYNTHETIC APPROACHES TO THE PODOPHYLLOTOXINS  The podophyllotoxins are, as previously mentioned, members of the group of compounds called lignans and thus they have a common biosynthetic origin (Section 1.4). The structures are very rigid because of the fusion of the C ring to the aromatic B ring and the attachment of the trans-fased lactone D ring.  Podophyllotoxin (3)  Podophyllotoxin (3) is the parent compound. The molecule is a very rigid molecule, this rigidity is caused by fused nature of the adjacent five and six membered rings (rings A, B, C and D) in the molecule. The Eringis attached in a cis orientation with the carbonyl group which adds to the relative thermodynamic instability of the structure in comparison with other structural isomers. Compounds in the picropodophyllin (4) series have a cw-fused lactone ring. Under basic conditions podophyllotoxin (3) will readily undergo isomerization to form predorninantly picropodophyllin (4) with an equilibrium constant of 37.0 at 31°C (Scheme S).  24  This has led to speculation that picropodophyllin  (4) is not a natural product but an artifact produced during the extraction and isolation of the lignans.25  OCH  OCH  3  Podophyllotoxin (3)  3  Picropodophyllin (4)  Scheme 5. The Isomerization of Picropodophyllin (4) and Podophyllotoxin (3)  These structural features presented a challenge for the synthetic organic chemist. Gensler, in 1966, published the first total synthesis of podophyllotoxin (3) (Scheme 6). 26  The carbon backbone of the podophyllotoxin structure was constructed by a condensation reaction between the ketone 24 and diethyl succinate to produce the unsaturated ester 25. This ester was further elaborated to produce the anhydride 26, which was cyclized using tin (TV) chloride giving the intermediate 27. The lactone ring was constructed by the reaction of the intermediate 27, with sodium hydride and ethyl formate, to give the enol 28. The enol 28 was then further elaborated to give the diol 30 which was then cyclized under acidic conditions to produce the lactone 31.  The key intermediate 31 was  hydrolyzed to the corresponding acid 32 which was resolved with quinine and then lactonized to give the (-) enantiomer of 31.  The tetrahydropyranyl ether of  picropodophyllin 33 was then treated with trimethylphenyl sodium to produce the kinetic enolate which was quenched to give, after hydrolysis, a mixture of podophyllotoxin (3) and picropodophyllin (4) in a ratio of 9:11.  22 SnCLt  ArCOCl  Diethyl succinate f-BuOK  1) NaOH, separation  P  ^  \  : O  2) H , Pd-C 3) CH3COCI 2  C0 Et 2  ,C0 Et  J !  o  2  ^  25  1) SnClj 2) H+ EtOH  O  NaH C0 Et  HC0 Et  o  <  C0 Et  2  2  2  Ar  Ar  27  28  H , Pt0 2  29 Ar = 3,4,5-trimethoxyphenyl Scheme 6. Gensler's Synthesis  2  p-TsOH Tetrahydropyran 33 Scheme 6 (continued)  24  OH  Ar  0  Picropodophyllin (4)  Scheme 6 (continued)  The Diels-Alder reaction has been employed by a number of authors for the synthesis of podophyllotoxins.27 Scheme 7 shows the strategy of one such route. The six step synthesis from the rra/ts-2-arylbenzocyclobuten-l-ol (34) gave (±)-podophyllotoxin (3) in 11% overall yield. The correct stereochemistry at four centers is achieved by the intramolecular cycloaddition reaction of intermediate 35 to produce the compound 36. A route by Rodrigo28 produced epipodophyllotoxin (37) in 11 steps (6%) and podophyllotoxin (3) in 12 steps (9.4% yield: Scheme 8). The Diels-Alder strategy was used once again to build the carbon skeleton and a different approach used to solve the  stereochemical problems. The initial step from the starting material 38 involves a cycloaddition to give the cyclic ether 39. H  Ar  Ar  0  Podophyllotoxin (3)  36  Ar = 3,4,5-trimethoxyphenyl Scheme 7. The Use of the Diels-Alder Reaction in the Synthesis of Podophyllotoxin (3)  This ether 39 is then further elaborated to produce the diol 42. The acid 44, produced from the diol 42 by protection of the diol as an acetonide 43 and then saponification, proved to be the key intermediate in the synthesis. When the acid was treated with dilute aqueous acid for 24 hours, the acetonide was cleaved and subsequent treatment with DCC (dicyclohexyldicarbodiimide) yielded epipodophyllotoxin (37). However, if the acid 44 was treated with aqueous acid for 48 hours, then neopodophyllotoxin (45) was produced. On the treatment of 45 with NaOH followed by treatment with DCC podophyllotoxin (3) was produced.  26 0 ^ ^ ^ s ^  C  H  ( °  C  H  3 ) 2  CH 0 CC=CC02CH3  P  AcOH, trace  \>  3  2  C0 CH 2  3  C0 CH 2  3  Ar  39  38  1) H , Pd 2  2) MeONa  CH OH 2  C0 CH 2  LiEt BH  3  3  C0 CH  C0 CH 2  2  3  3  Ar  Ar  40  41  Raney-nickel  CH OH  T T +  2  rT, acetone  o ^ C0 CH  CO CH 2  2  3  Ar  Ar  42  Ar = 3,4,5-trimethoxyphenyl  43  aq. NaOH  44 Scheme 8. Rodrigo's Synthesis of Podophyllotoxin (3) and Epipodophyllotoxin (37)  3  27  Epipodophyllotoxin (37)  Podophyllotoxin (3)  Ar = 3,4,5-trimethoxyphenyl Scheme 8 (continued)  This procedure provided a route to the correct stereochemistry for podophyllotoxin (3). The synthesis of podophyllotoxin (3) via a picropodophyllin type intermediate has now been used in several syntheses.  27  Other syntheses of (±) podophyllotoxin (3) have been published " together with 3  some asymmetric syntheses. ! Deoxypodophyllotoxin (6), and epipodophyllotoxin 3  32  (37)33 have also been synthesized. Other routes into the lignan system have also been investigated and are discussed in a review by Whiting.34 The etoposide precursor 4'-demethylepipodophyllotoxin (5) was synthesized by Kende35 in 1981 with an overall yield of 2.4% for 13 steps. Scheme 9 shows the strategy used to build the carbon skeleton; the stereochemical problem of the lactone was solved by Gensler's methodology. The conversion of podophyllotoxin (3), epipodophyllotoxin (37) or 4'-demethylpodophyllotoxin (5) to etoposide (1) has already been outlined (Scheme 4), in which the glycosylated intermediate 22 is condensed with acetaldehyde.  36  The syntheses described (Schemes 6,7, 8, and 9) are not suitable for scale up to produce podophyllotoxin (3), and hence etoposide (1) on a kilogram scale. Other more practical routes to produce large quantities of podophyllotoxin (3) are under investigation, including the use of plant cell cultures (Section 3.1). The development of a plant cell culture of a Podophyllum species which could produce podophyllotoxin (3) on a large scale would be the most direct approach to the problem. A second approach is the investigation of bio-transformations by plant cell cultures using suitable precursors. Such processes may be more amenable to scale up than the corresponding synthetic processes. Any investigation of the use of bio-transformations in the production of the podophyllotoxins must consider their biosynthesis. It is this information which will direct research toward the most potentially rewarding precursors.  18 Ar = 3,4,5-trimethoxyphenyl Scheme 9. The Synthesis of 4'-Demethylepipodophyllotoxin (18)  1.4  THE BIOSYNTHESIS OF THE PODOPHYLLOTOXINS  The common structural features of the podophyllotoxins and other lignans suggests that they are derived, biosynthetically, from a common intermediate or pathway. Some of the intermediates involved in this pathway are shown in Scheme 10.37 The aromatic rings are formed from shikimic acid (51) which is enzymatically transformed into (L)phenylalanine (52) and (L)-tyrosine (53). These intermediates are then converted by further enzymatic processes into 4-hydroxycinnamic acid (54) and then into ferulic acid (55). Two of these nine carbon units are then coupled together in a head to head fashion to produce the lignan intermediate 56 (Scheme 11). Experiments involving the use of precursors which have been prepared with carbon* labels in the molecules, have been use to trace the uptake of various metabolites 4  by the Podophyllum species. These labeling experiments have shown that two units of (L)-phenylalanine (52) are incorporated into the podophyllotoxin skeleton during the biosynthesis of the lignan molecules. (L)-Tyrosine (53) has not been incorporated in any experiments but it is a possible precursor. Also ferulic acid (55) has been found to be incorporated38 into podophyllotoxin (3) and 4'-demethylpodophyllotoxin (5). Other experiments involving labeled compounds have shown that matairesinol (57) and yatein (58) are incorporated into podophyllotoxin (3). The incorporation of yatein (58) into podophyllotoxin (3), involves a stereoselective cyclization in which a cis relationship is produced between the protons at the 1 and 2 positions. These results are summarized in Scheme 11 which presents the proposed intermediates in the biosynthesis of podophyllotoxin (3).  Scheme 10. Some of the Early Intermediates in the Biosynthesis of Lignans.  32  Scheme 11. The Proposed Intermediates in the Latter Stages of the Biosynthesis of Podophyllotoxin (3)  Further experiments also with labelled compounds have shown that 39  hydroxylations of deoxypodophyllotoxin (6) and 4'-demethyldeoxypodophyllotoxin (59) occur to give podophyllotoxin (3) or 4'-demethylpodophyllotoxin (5), respectively (Scheme 12). Also podophyllotoxone (60) and 4'-demethylpodophyllotoxin (61) have been shown to be produced from podophyllotoxin (3) and 4'-demethylpodophyllotoxin (5), respectively. No reactions have been detected which convert compounds in the podophyllotoxin series to compounds in the 4'-demethylpodophyllotoxin series or vice versa. Matairesinol (57), a compound which may be an intermediate earlier in the biosynthetic pathway, has been suggested as the possible common intermediate, but this has not been confirmed by radio-labelling experiments. The results of these experiments gives an indication of the precursors involved in the biosynthesis of the podophyllotoxins. These precursors provide valuable structural information for the design of suitable compounds for bio-transformation reactions. However, in considering which precursors are suitable for a potential biotransformation a number of factors must be taken into account. Firstly, the precursor must be easily synthesized on a large scale. Secondly, the precursor must also be sufficiently advanced in the biosynthetic pathway that it is transformed toward the desired product and not along other enzymatic branches of the biochemical tree. Only in cases where the precursor is very inexpensive can low incorporation into the product be tolerated. Precursors structurally similar to yatein (58) were chosen as the initial compounds for our investigation. A stereoselective ring closure is then required to produce compounds with the podophyllotoxin structure. Compounds (62,63,64) (Figure 6) were chosen for the initial study because of their ease of preparation. The necessary oxidative coupling reaction can be carried out by a class of enzymes called peroxidases.  Scheme 12. The Biosynthetic Interconversion of Some of the Podophyllotoxins  35 HO.  RO  OCH  OCH3  3  OH  OH  63 R = H 64 R = C H  62  3  Figure 6. Dibenzylbutanolide Precursors Initially Prepared for Bio-transformations  1.5  BIO-TRANSFORMATIONS INVOLVING R A D I C A L T Y P E RING C L O S U R E R E A C T I O N S : C A T A L Y S I S B Y PEROXIDASES  Peroxidases are a common class of enzymes found in the plant and animal kingdom. Molecular weights of the enzymes are typically about 50,000 mu and their prosthetic group is usually the ferriprotoporphyrin LX group. " Peroxidases use peroxides 4  as co-factors and behave as one-electron oxidizing agents for an extensive variety of substrates. The main role of peroxidases in plants is believed to be the removal of hydrogen peroxide from the plant cell. * Hydrogen peroxide is harmful to the cells. Some 4  evidence also suggests that peroxidases may be involved in the removal of the harmful species, superoxide, from the plant cells.  42  As a candidate for exploitation in plant cell  cultures they have many desirable features; the enzymes are often very stable, their mode of action is well established and they require inexpensive co-factors such as hydrogen peroxide.  36  The mechanism of action of a typical peroxidase is shown in Scheme 13.  43  Mechanism A is true for reactions at low concentrations of hydrogen peroxide (10-6-1O  4  M). The native enzyme is oxidized to Compound 1 by hydrogen peroxide, or another peroxide, and the compound is then able to oxidize various organic substrates (AH) to provide Compound 2. In the oxidation of a further molecule of substrate (AH) Compound 2 is transformed back to the native enzyme. When the concentration of hydrogen peroxide increases the mechanism becomes more complex (mechanism A and mechanism B). Compound 2 is oxidized further to give Compound 3, which can, by loss of an oxygen molecule and reduction of hydrogen peroxide cycle back to give Compound 2.  44  Compound 3 is unreactive toward most electron donors, and can be converted into a 45  denatured form of the enzyme called Compound 4. There is some evidence to suggest that Compound 3 can be converted back to the native enzyme.  46  H 0 2  HoO  2  Native enzyme  Compound 1  B Compound 3  Ferrous enzyme 0  2  Denatured enzyme Scheme 13. The Mechanism of Action of a Typical Peroxidase  In our laboratory a plant cell culture line of Catharanthus roseus has been established and found to contain peroxidase iso-enzymes. These enzymes were identified by examination of the coupling reaction of catharanthine (65) and vindoline (66) to produce 3\4-anhydrovinblastine (67) in 25% yield (Scheme 14). When immobilization l  47  of the enzyme was carried out this yield was increased to 38%. Thus, it was envisioned 48  that the Catharanthus roseus cell culture would contain the enzymes necessary to cyclize yatein type compounds to produce podophyllotoxin like structures.  38  Scheme 14. The Coupling Reaction of Camaranthine (65) and Vindoline (66)  RESULTS AND  1.6  DISCUSSION  ATTEMPTED BIO-TRANSFORMATION OF DIBENZYLBUTANOLIDES (62, 63, 64) USING THE WHOLE C E L L SUSPENSION CULTURE OF CATHARANTHUS  ROSEUS  It is known that the cell suspension culture of Catharanthus roseus contains peroxidase enzyme systems. Thus, it was envisioned that the cell culture would bio47  transform the precursors (62,63 and 64) into the corresponding cyclized compounds (68, 69 and 70) (Scheme 15). It is important to note that the ring open and corresponding ringclosed compounds have, conventionally, a different numbering system employed in their respective structural diagrams.  RO.  R'O  RO.  Catharanthus roseus 5'  7 •i  O  OCH  CH 0 3  Whole cells  R'O  OCH3  3  OH  OH  62 R,R' = - C H 63 R ' = C H , R = H 64 R,R' = H 2  3  68 R,R' = - C H 69 R'=CH3,R "=H 70 R,R' = H 2  ,  Scheme 15. The Proposed Bio-txansformation of the Precursors (62, 63 and 64)  These precursors were initially chosen as models, because they possess many of the structural feature present in yatein (58) and matatairesinol (57) and they were easily  synthesized. Yatein (58) and matatairesinol (57) have been established as intermediates in the biosynthesis of the podophyllotoxins. The initial synthesis of the three precursors (62, 63,64) is discussed in Section 49  1.11 of this thesis. The preparation of the dihydroxydibenzylbutanolide 63 was optimized by J Palaty.50 Later during the work of this thesis a new route to two of these compounds (63,64) was developed and this is also discussed in Section 1.11. Bio-transformations using whole cell suspension cultures have some advantages over other types of bio-transformations, such as transformations utilizing CFE (cell free extracts). The procedures are easy to carry out and the products can often be extracted from the culture broth. Also, as the cells are intact they provide in vivo the necessary cofactors for their enzyme systems. The major disadvantage of using whole cell suspensions, for bio-transformations, is often the difficulty of extracting the products from the cell material. The peroxidase activity of the Catharanthus roseus cell line has been correlated with culture age5l as shown in Graph 1. In general peroxidase activity is measured by performing the oxidation of pyrogallol to purpurogallin, the latter being measured spectroscopically (UV absorption at 420 nm). The activity of the peroxidase enzymes is expressed in terms of units. One unit is defined as the amount of enzyme required to produce 1 mg of purpurogallin in 20 seconds, in a phosphate buffer (0.1 M) at pH 6.3 kept at 20°C. Also shown on Graph 1 is the variation of fresh cell weight with cell culture age. The cells reach a stationary phase after about 12 days. The stationary phase is a phase of culture growth in which the mass of cells stops increasing. After this phase the cells will begin to die unless they are transferred to fresh nutrient medium. Seven day old culture was chosen for the bio-transformation as it would provide cells approaching a maximum of peroxidase activity. During this stage of growth it is believed enzymatic processes in the plant cells are directed toward the production of secondary metabolites.52  Graph 1  The variation of fresh weight and peroxidase activity of the Catharanthus roseus cell culture with age.  A pure sample of each of the precursors 62,63 and 64 was dissolved in ethanol and added to the 12 day old Catharanthus roseus cell suspension culture under sterile conditions. The flasks containing the cell culture were then incubated on shaker beds. Samples were removed from the cell culture (1.0 mL) and were used for analysis by HPLC. For the HPLC analysis a sample was removed after 24, 48 and 72 hours of incubation and mixed well with methanol (1.0 mL). The samples were injected onto a reverse phase C-18 column (solvent: methanol: water 45 : 55, each with 0.1% acetic acid, flow rate 1.0 mL/minute). The HPLC data was used to follow the disappearance of precursor, if any occurred, and to monitor the formation of products. Some precipitation was noted when the sample of the precursor 62 was added to the cell culture. Although some precipitation occurred, some of the precursor may have remained in solution and so would be able to pass through the cell membrane and undergo  bio-transformations. Although some loss of the precursor 62 was indicated by HPLC analysis, much of the material remained unreacted. No new products could be observed by HPLC analysis. The analysis did indicate that the two precursors 63 and 64 had been metabolized, but again, no products could be observed. Although the HPLC analysis of the reaction mixtures did not show any new peaks, it is possible that product peaks may be hidden beneath the many peaks which are produced from the material extracted from the plant cells. These peaks could be observed by the HPLC analysis of the samples from the control experiments. Chromatographic separation was carried out on the extracts (extraction solvents : dichloromethane followed by ethyl acetate) from the cultures and these samples were analyzed by !H-NMR and EI MS. The data for the dihydroxybutanolide 63 will be discussed in some detail, so as to establish the spectral details which were considered important in the search for a ring closed product. The EI MS data for the precursor 63 (see Appendix, Figure 13) shows a peak at 388 which corresponds to the signal produced by the molecular ion peak. The major fragment with a m/z value of 167 corresponds to a fragment consisting of the aromatic ring D plus the benzylic methylene group. This fragment then further fragments to produce a fragment with a m/z value of 153 which is consistent with the loss of the methylene group. Another major fragment is found with a m/z value of 137 and corresponds to the aromatic ring A together with the methylene group. As cyclization to produce compound 69 involves effectively the loss of two hydrogen atoms (2 mass units) we would expect the molecular ion peak of the cyclized compound 69 to have a m/z value of 386. The 1H-NMR spectrum shows a peak at 8 6.49 ppm which appears as a doublet of doublets with coupling constants of 8 Hz and 1 Hz. The 8 Hz coupling is a typical coupling constant for the coupling between two ortho protons on an aromatic ring. The coupling constant J = 1 Hz, is typical of a coupling constant for the meta coupling between two protons on an aromatic ring. Thus the signal at 8 6.49 ppm corresponds to the proton  at position 6' of compound 63. At 8 6.73 ppm a one proton doublet occurs with a coupling constant of 8 Hz. This peak is caused by the proton at position 5' and it is coupled to the proton at position 6'. The other aromatic proton on the ring A at position 2' appears as a doublet at 5 6.62 ppm with a coupling constant of 1 Hz. This proton shows typical meta coupling to the proton at position 6'. The remaining aromatic protons on the other aromatic ring produce a two proton singlet at 8 6.37 ppm. The aliphatic protons at position 7" are very diagnostic in the analysis of any possibleringclosures. In compound 63, for example, they occur at 8 2.45-2.65 ppm as a multiplet together with the protons at positions 2 and 3. The aliphatic protons at positions 7' gave a one proton doublet of doublets at 8 2.88 ppm with coupling constants of 14.6 and 5.7 Hz, and a one proton peak at 8 2.93 ppm with coupling constants of 14.6 and 6.4 Hz. The 14.6 Hz coupling constant is consistent with geminal coupling between the two protons at position 7', whilst the other coupling constant is due to coupling with the proton at position 3. The protons adjacent to the oxygen of the lactoneringat position 4 give a two sets of signals which are found downfield of the other aliphatic proton signals. A doublet of doublet with coupling constants of 7.5 and 3.0 Hz is found at 8 4.15 ppm and represents one of the protons at position 4 which is coupled to the other proton at position 4 and the proton at position 3. The other proton at position 4 gives a signal which is overlapped by the signal at 8 3.873.89 ppm which includes the singlet peak caused by the methoxy protons at positions 4', 3" and 5". At 8 5.24 and 5.58 ppm two broad singlets occur which correspond to the two hydroxyl protons at positions 4" and 3'. These signals disappear when the sample is shaken with deuterium oxide and the spectrum remeasured. The samples from the chromatography of the attempted bio-transformation of the precursor 63 did nm show any of the characteristic 1H-NMR peaks for the precursor 63 and no peaks were detected which could be assigned to a ring closed product, such as 69. Also the EI MS did not show a peak at m/z 386. It appeared that extensive decomposition of 63 had occurred to a complex mixture of products, and thus this study was abandoned.  The samples from the attempted ring closure of the precursor 64 were also examined by comparison of the !H-NMR and EI MS data with the data obtained for the starting precursor. A ring closed product was not identified and starting material was not recovered. The only product identified, in low yield (5%) was compound 71 (Figure 7) which was readily identified from its spectroscopic data, when compared to the data of compounds 63 and 64. The EI MS for 71 gave a peak caused by the molecular ion at m/z 388 and fragments at 167 and 137. The molecular ion of the starting precursor is 374. This corresponds to a compound similar to the precursor 63. The iH-NMR data was also very similar to that of the precursor 64 with the only differences in the spectra involving the peaks for the protons at positions 2' and 5'. A peak at 5 6.53 ppm occurred as a doublet of doublet and corresponds with the proton at position 6'. The 2' proton gave a signal at 8 6.42 ppm which is upfield in comparison to the peak corresponding to 2' in the precursor 63. This peak was a doublet with a coupling constant of 3 Hz which is consistent with meta aromatic coupling. A peak at 8 6.82 ppm was assigned to the proton at position 5* and it occurred as a doublet. This pattern is consistent with a compound similar to the precursor 63 with the methyl group now positioned on the 3' hydroxyl group. This compound 71 is presumably formed by reaction of the precursor 64 with methionine enzymes known to exist in plants. The low recovery of material may be explained by the oxidation of 64 to very polar compounds which were not extracted in the extraction process. Although theringclosure of the precursors (62, 63,64) had not been achieved it was a encouraging to find that they were taken in by the cells of Catharanthus roseus and, in turn, metabolized. It was therefore decided to carry out bio-transformations using initially the same precursors, and utilizing a enzyme preparation known to contain peroxidase type enzyme systems. These enzyme systems were believed to be important in the proposed oxidative coupling reaction.  45  2'  CH 0, 3  7 O  HO  OCH  3  OH  71 Figure 7. The Product Isolated from the Bio-transformation of 64  1.7  BIO-TRANSFORMATIONS USING  HORSERADISH  O F T H E P R E C U R S O R S 6 2 , 6 3 , A N D 64  P E R O X I D A S E (HRP)  Horseradish peroxidase (HRP) is a commercially available enzyme preparation. 53  Peroxidases, as previously mentioned (Section 1.5), are usually stable enzyme systems. The commercial peroxidase is only slightly affected by boiling, and can be refrigerated to preserve the enzymatic activity for long periods oftime.In the presence of peroxides, the peroxidases can catalyze the oxidations of many compounds. Hydrogen peroxide was chosen as the co-factor for the bio-transformations. Preliminary experiments previously carried out in our laboratory involved the 54  small scale (18 mg of precursor) bio-transformation of the three precursors (62, 63, 64) with HRP and hydrogen peroxide as the co-factor. After extraction, the products were analyzed by LCMS. As previously discussed, the peak corresponding to the molecular ion of the precursor 6 3 has a m/z value of 388. Some of the samples from the HRP  46 experiment carried out with the precursor 63 showed a peak in the MS at m/z 386 which corresponds to the molecular ion peak of the ring closed product 69. These preUrmnary results from the small scale HRP experiments, together with evidence of substantial metabolism of the precursor 63 with Catharanthus roseus whole cells, suggested that the dihydroxydibenzylbutanolide 63 was the most suitable compound for further evaluation as a precursor. Thus this precursor 63 was used in a series of biotransformations using HRP. The dihydroxydibenzylbutanolide 63 was dissolved in a suitable ethanol/water (15 mL/135 mL) mixture and then treated with a solution of the HRP preparation in buffer (0.1 M phosphate, pH 6.3) in the presence of hydrogen peroxide (1.1 equivalents) (Scheme 16). The HRP solution provided 415 units of peroxidase per mmol of substrate. Two control experiments were also carried out, one without the enzyme, that is, in the presence of hydrogen peroxide, and the second in the presence of HRP only. Only the starting precursor was identified, (!H-NMR and EI MS), in the extracts recovered from the control experiments.  HO  HO  CP CH 0  "\< O  3  HRP H 02  5 | A  4 ^ li B  CH O 3  2  OCH3  CH3O  CO  0CH3  OH  OH  63  69  Scheme 16. Cyclization of 63 to 69 using HRP  After extraction and purification of the reaction mixture the major product was identified as  l-(3\5-dimethoxy-4'-hydroxylphenyl)-6-hydroxy-3-hydroxymethyl-7,  methoxy-l,2,3,4-tetrahydro-2-naphthoic acid-y-lactone (69). The structure of compound 69 was assigned, in part by consideration of the spectral data for the precursor compound 63 which has been discussed in Section 1.6. The product showed an IR absorption at 1775 cm- which is consistent with the presence of the lactone carbonyl group. The EI MS 1  gave a molecular ion peak with a m/z value of 386 which is two mass units less than the molecular ion peak of the precursor 63. The EI MS also showed a low degree of fragmentation in comparison to the the fragmentation pattern of the precursor 63 (see Appendix Figure 14) and the peaks at 167, 153 and 137 are absent or greatly reduced in relative intensity. The fragmentation pattern for compound 69 shows many low intensity fragments between m/z 232 and the peak at m/z 386 (see Appendix Figure 15), these fragments are not present for the ring open compound 63. The fragment at m/z 232 corresponds to a fragment containing the rings A, B and C. The 1H-NMR spectrumof the product 69 was characteristically different from that of the precursor 63 which has been previously discussed in Section 1.6. The aromatic protons of compound 69 produce three singlets in the 1H-NMR spectrum as after ring closure no ortho or meta coupling is possible between the aromatic protons. The protons a position 2' give a singlet for two protons at 8 6.47 ppm which compares to a value of 8 6.37 ppm. The protons at positions 5 and 8 occur as one proton singlets at 8 6.34 and 8 6.73 ppm. The trans stereochemistry of the lactone which was present in the precursor, was retained in the product. This was shown by analysis of the signals produced by the aliphatic protons. The proton at position 2 appears as a doublet of doublets at 8 2.50 ppm with coupling constant of 14 Hz and 11 Hz. This suggests a trans relationship across the lactone and a trans relationship between the proton at position 2 and the proton at position 1. The proton at position 1 appears as a doublet at 8 4.10 ppm with a coupling constant of 11 Hz. This indicates that the cyclization has proceeded according to Scheme 16 with a  trans-trans relationship between carbon atoms 1,2 and 2,3. Further experiments were carried out on a larger scale (Table 3). As shown, the yield of the cyclized product 69 was low and was not increased by using a greater amount of hydrogen peroxide. Although the cyclizarion was successful the stereochemistry at the newly formed ring junction was trans, and this is not consistent with the structures of the desired podophyllotoxins.  TABLE 3  Large Scale Bio-transformations of the Dibenzylbutanolide 63 with H R P  Experiment  Hydrogen peroxide  HRP  Yield of 69  1  1.9 equivalents  690 units/mmol  19%  2  3.9 equivalents  720 units/mmol  15%  The low yield may result from over oxidation of the precursor 63 by the HRP enzyme. This would be consistent with the total loss of precursor which occurred during the reactions. Optimization of the reaction parameters may result in a higher yield of the product 69. Rather than continue with the above studies, experiments involving the biotransformation of 63 were carried out using the CFE prepared from the Catharanthus roseus cell suspension culture as previously established in our laboratory.55 The CFE was known to contain peroxidase enzymes, and since control of the culture in terms of growth, age etc was possible, a control of enzyme production was also possible. In turn, it was hoped that a higher yield of cyclization could be achieved under such controlled conditions, and that the product would have the desired stereochemistry of the podophyllotoxin system. This study was initiated by another co-worker (J. Palaty, M. Sc. thesis, UBC,  49 1990) while the above experiments with HRP were underway. Various aspects of these studies were continued by me and the results are documented in the next section.  1.8  T H E B I O - T R A N S F O R M A T I O N O F TRANS-2-{3",5"DIMETHOXY-4"-HYDROXYBENZYL)-3-(3 -HYDROXY-4',  M E T H O X Y B E N Z Y L ) - B U T A N O L H ) E (63) W I T H T H E C F E O F  CATHARANTHUS  ROSEUS  The precursor 63 was treated with the CFE prepared from 11 day old Catharanthus roseus culture in the presence of hydrogen peroxide (11 equivalents) (Scheme 17). Enough CFE was added to provide 140 units of peroxidase per mmol of substrate and the reactiontimewas 120 minutes. The reaction parameters were chosen based on previous work involving the coupling of 3\4-anhydrovinblastine (67) (see Section 1.5). l  Scheme 17. Cyclizarion of 63 to 69 using the CFE  of Catharanthus roseus  The cyclized product 69 was isolated in 58% yield and was found by 1H-NMR analysis and EI MS to be identical with the product 69 from the HRP cyclization experiment. The assignments for the spectral data were discussed in Section 1.7. With more material now in hand this compound was further characterized. The product 69 was found to be racemic and a single crystal X-ray diffraction was obtained (see Appendix 56  Figure 16). The stereochemical assignments made by 1H-NMR coupling assignments (Section 1.7) were thus confirmed as the X-ray diffraction results clearly showed a trans relationship between the substituents at the 1 and 2 positions of the product 69. A proposed mechanism for this cyclization is shown in Scheme 18. The mechanism is based on the known oxidative properties of peroxidase enzymes. It is 57  suggested that an initial single electron oxidation of the substrate occurs (Scheme 13) to give the phenoxy radical intermediate 72. Evidence suggests that phenoxy radicals may undergo disproportionation to produce the quinone 73.  58  The quinone 73 may then be  cyclized by intramolecular nucleophilic addition and subsequent proton exchange (Route A), or undergo further oxidation by peroxidase to produce the phenoxy radical 74 which can then cyclize to give the radical 75. The cyclized radical 75 (Route B) then picks up a hydrogen radical to give the cyclized product 69. Initially a series of reactions were carried out to narrow down the pH range for the optimized reaction. These experiments involved the preparation of CFE's in phosphate buffers (0.1 M phosphate) at five different pH values, 5.95, 6.49, 7.15, 7.78 and 8.15. HPLC analysis (reverse phase C-18 column, solvent: methanol: water 45 : 55, each with 0.1% acetic acid, flow rate 1.0 rnlVminute) was used and the ratio of the precursor 63 to the product 69 at the end of the reaction time (160 minutes) was measured. The results showed that at pH values above 7.15 substantial amounts of the precursor 63 remained at the end of the reaction. Disappearance of the precursor 63 was rapid (less than five rninutes) at pH's 5.95 and 6.49 but a second unknown product was also formed. Thus, the best yield of product 69 was found when the pH was between 5.95 and 7.1. The  optimum pH was further narrowed down by other experiments performed in our laboratory in which CFE's were prepared using phosphate buffers over a smaller pH 59  range. The optimum values for the number of hydrogen peroxide equivalents and the units of peroxidase per mmol of precursor required, in order to give a optimum yield of the product 69 have also been established by a series of experiments. These values were obtained from experiments involving the monitoring of samples by HPLC and the use of the calibration curves previously mentioned. The results are summarized in Table 4. Using these optimized conditions yields of 70-80% of the product 69 were obtained.  TABLE 4  Optimum Conditions for the Cyclization of the Precursor 63  Using the C F E of Catharanthus roseus pH  6.3  Hydrogen peroxide equivalents  2.0  CFE units of peroxidase per mmol of 63  250  Reactiontimein minutes  180  Thus the peroxidase enzymes of the Catharanthus roseus cells in the form of a CFE were found to cyclize the precursor 63 to 69 in a much higher yield than HRP, but once again, as with the HRP enzyme system, the cyclization did not result in the desired stereochemistry. The product 69 has a trans-trans relationship with respect to carbon atoms 1, 2 and 3. This is probably the most thermodynamically stable structure of the possible products of ring closure. Indeed this was the outcome when a similar compound 76 was cyclized chemically (Scheme 19) although a direct comparison between a 60  chemical and enzymatic process is to be done with caution. Furthermore, different precursors were employed."  52  Scheme 18. The Proposed Mechanism for the Oxidative Cyclization of 63 to 69  The precursor 63 was chosen for the reasons previously mentioned and because of the requirements suggested by the proposed mechanism of cyclization (Scheme 18). Thus, it would seem desirable to use a precursor with an hydroxy group in this position.  Scheme 19. T h e Chemical Cyclization of 76 to G i v e the Product 77 However, an intermediate 78 (Figure 8) has been suggested as a possible biosynthetic precursor to podophyllotoxin (3). This intermediate 78 already contains the methylene bridge which is required by the podophyllotoxin structure. Thus, the second candidate chosen for bio-transformation with the CFE was the 3',4'-methylenedioxydibenzylbutanolide  62 which had already shown a degree of  metabolism by the whole cells of the Catharanthus roseus cell culture. It was hoped that this precursor would undergo the necessary ring closure and give a product 68 (see Section 1.6).  The precursor 62, with only one hydroxyl group present, may be less  activated towards attack by the peroxidase enzyme system and hence, may react more selectively, thus, yielding a compound with the correct stereochemical requirements for an etoposide intermediate. The stereochemical requirements being the presence of a trans-  fused lactone ring together with a cis relationship between the substituent at the 1 and 2 positions of the cyclized product.  Figure 8. A Possible Biosynthetic Intermediate in the Biosynthesis of Podophyllotoxin  1.9  THE BIO-TRANSFORMATION OF TRANS-2-(3",S"DIMETHOXY-4"-HYDROXYBENZYL)-3-(3^4 ,  METHYLENEDIOXYBENZYL)BUTANOLIDE (62) WITH THE CFE OF CATHARANTHUS  ROSEUS  ^ra™-2-(3^5"-Dimethoxy-4"-hydroxybenzyl)-3-(3 ,4 ,  ,  methylenedioxybenzyl)butanolide (62) (preparation given in Section 1.10) was dissolved in a rnixture of ethanol (40 mL) and water (80 mL) and treated with the C F E prepared from Catharanthus roseus (11 days old in 0.1 M phosphate at p H 7.35) in the presence of hydrogen peroxide (3.7 equivalents) (Scheme 20). On addition of the C F E some precipitation, presumably of the precursor, did occur.  Scheme 20. The Bio-transformation of 62 with Catharanthus  roseus Two experiments at different pH, 6.3°1 and 7.35 were conducted. The experiments were performed using the amounts shown in Table 5. The experiment at pH 6.3 was quenched after 120 minutes, whilst the experiment at pH 7.35 was quenched after 60 minutes.  TABLE  5  Bio-transformation of  62 Using the CFE of Catharanthus  roseus Exrjeriment  Hydrogen peroxide  CFE  Yield of 79  Yield of 62  pH6.30  6.5 equivalents  460 units/mmol  25%  25%  pH7.35  3.7 equivalents  396 units/mmol  9%  61%  The product obtained in both reactions was identified as the conjugated lactone 79 (Scheme 20). The EI MS of compound 62 shows a peak caused by the molecular ion at m/z 386. The major fragments, namely the "benzylic" systems give rise to peaks at m/z 167 and 135. The !H-NMR spectrum of 62 is similar to the spectrum of the previously discussed precursor 63. The aromatic protons produce a pattern of signals which is  distinctive of a 1, 3,4 substituted aromatic ring. The proton at position 2* gives a signal at 8 6.60 ppm which is a doublet with a coupling constant of 2.0 Hz, whilst the proton at position 6' appears as a doublet of doublets at 8 6.45 ppm with coupling constants of 8.5 and 2.0 Hz. The other aromatic proton in the 1,3,4 substituted aromatic ring, at position 5', givesriseto a signal at 8 6.73 ppm which is a doublet with a coupling constant of 8.0 Hz. The majority of the aliphatic protons at positions 2, 3 and 7", give signals which occur as a multiplet centered at 8 2.88 ppm. The protons at position 4 occur as a pair of doublets of doublets at 8 3.88 and 4.12 ppm. A distinctive feature of the 1H-NMR of compound 62 is the set of signals at 8 5.94 ppm which occurs as a multiplet and is indicative of the presence of the methylene-dioxy protons. Examination of the EI MS for the product 79 revealed a peak at m/z 384 which corresponds to a molecule with two protons less than the precursor 62. A fragment at m/z 135 was visible and is characteristic of the ring A component. However unlike the precursor 62 no fragment was found which corresponds to the ring D component indicating that fragmentation at position 7" was not occurring. The 1H-NMR spectrum of the product 79 further indicated thatringclosure had not occurred. This was suggested by the observation of a signal at 8 6.7 ppm which appeared as a doublet (J = 7 Hz) and was assigned to a proton at position 5'. This is indicative of an aromatic proton ortho coupled to a second proton. This ortho coupling would not be observed if cyclization has occurred. The aromatic protons at positions 2' and 6' occurred as a multiplet at 8 6.56-6.65 ppm. A two proton signal was visible at 8 5.85-5.97 ppm which was indicative of the methylene-dioxy protons. Also at 8 7.48 ppm a signal was observed corresponding to the olefinic proton at position 7". Based on the Dreiding models of the compounds and the result of the oxidation of a similar yatein derivative the 62  stereochemistry of the more stable olefin is probably E. This stereochemical assignment was not confirmed. The experiment at pH 7.35 was carried out as a result of the observation that experiments at higher pH sometimes increased the recovery of material  from the bio-transformations. However, at the higher pH the yield of the product was decreased to 9%. It is suggested that the conjugated lactone 79 is produced via the quinone 73 (Scheme 18). Although the poor recovery in these bio-transformations did not allow complete analysis to be achieved, this precursor 62 was dismissed as a possible substrate for cyclization with the CFE of Catharanthus roseus cell cultures as no cyclized product 68 was detected. The poor recovery of material from this bio-transformation may be explained by the occurrence of another reaction caused by peroxidase enzymes. It is known from electron spin resonance spectroscopy that both hydroxy radicals and superoxide radicals are 63  produced by the lactoperoxidase enzyme when in the presence of high hydrogen peroxide concentrations.  Scheme 21 shows the proposed mechanism for the oxidative  demethylation of the precursor 62 to produce a quinone structure 80, these reactions are a known peroxidase catalyzed process. If a radical similar in structure to 80 is generated 64  from either the precursor 62 or product 79, it does not undergo cyclization but is attacked by a superoxide radical to form a peroxy anion 82. This species could be protonated to give 83 and subsequently attacked by water to give the orr/io-quinone 84. A reactive quinone of this type may then be further degraded by other oxidase enzymes present in the CFE and this would explain the poor recovery of material from this bio-transformation. With these results, emphasis was placed upon the synthesis of a new precursor which would hopefully cyclize to give a product suitable for conversion to etoposide (1). The new precursor was required to have an hydroxy group in the 3* and 4" positions as this would assist radical formation and presumably assist ring closure, and also after ring closure be easily transformed into etoposide (1).  Scheme 21. The Proposed Mechanism for the Oxidative Demethylation of 62  1.10  SYNTHESIS OF rflA/VS-2-(3",5"-DIMETHOXY-4 M  HYDROXYBENZYL)-3-(3'-HYDROXY-4'-is.9-PROPOXYBENZYL)BUTANOLIDE (85)  The butanolide 85 was prepared utilizing a convergent synthesis from a readily available starting material. The tandem conjugate addition is based on the procedure of Ziegler and Schwartz65 and is shown retrosynthetically in Scheme 22. A one-pot reaction was proposed in which the three units shown could be connected to produce a trans-fused lactone system.  85 88  Scheme 22. The Retrosynthetic Analysis of 85  3,4-Dmydxoxybenzaldehyde (89) was chosen as the starting material as it is readily available. Kessar has reported the selective alkylation of 3,4-dihydroxybenzaldehyde 66  (89) using sodium hydride as a base. Using the conditions shown (Scheme 23) they were able to obtain both of the alkylated isomers separately 90 and 91, and could separate each of these compounds from the di alkylated material 92. CHO  DMSO 1 hour NaH 1.1 equivalents  89  CHO  CHJ CH3O  CH3I  NaH DMSO 2.2 equivalents 1 hour CH 0. 3  CHO  CHO  CH3O  CH3O  92  Scheme 23.The Alkylation of 3,4-Dihydroxybenzaldehyde (89) Using the Method of Kessar  The 3,4-dihydroxybenzaldehyde (89) was treated under the same conditions as those employed by Kessar using iso-propyl iodide as the alkylating agent and 1.1 equivalents of sodium hydride as the base. The reaction did not proceed cleanly and a mixture of products was obtained which included both monoalkylated isomers (90,91). Potassium carbonate, which is milder and easier to handle was tried as the base (1.1 equivalents), and the reaction was carried out at room temperature. A mixture of both monoalkylated isomers 93 (77%) and 94 (3.5%) and some dialkylated material 95 (4.9%)  was obtained (Scheme 24). However, the mixture could be readily separated using column chromatography with a mixture of benzene : ethyl acetate : acetic acid, 90 : 5 : 5 as the eluent. The 4-iso-propoxy-3-hydroxybenzaldehyde (93), which was isolated as an oil, was the expected major product and this was confirmed by !H-NMR spectroscopy. The 1H-NMR spectrum of the product 93 showed a characteristic six proton doublet, with a coupling constant of 6.0 Hz, at 8 1.40 ppm which is produced by the two methyl groups of the iso-propyl group. The heptet signal with the same coupling constant (J = 6.0 Hz) was produced by the methine proton of the wo-propyl group and it was found at 8 4.75 ppm. The aromatic protons gave a one proton doublet at 8 6.95 ppm corresponding to the proton at position 5, and a two proton multiplet at 8 7.42 ppm corresponding to the protons at positions 2 and 6. The irradiation of the signal in the 1HNMR spectrum corresponding to the methine proton of the iso-propoxy group at 8 4.75 ppm gave a NOE enhancement of the signal at 8 6.95 ppm. This signal is a doublet with a coupling constant of 8 Hz and, as mentioned, is caused by the presence of the aromatic proton at position 5. Thus the H-5 proton is in a position adjacent to the wo-propoxy group and hence the wo-propoxy group is in the 4 position. Further NOE irradiations confirmed this substitution pattern (see Appendix Figure 16). The EI MS for compound 93 showed a peak at m/z 180 for the molecular ion and a large peak at m/z 137 which indicates a mass loss of 43 from the molecular ion, this corresponds to the loss of the isopropyl group. The !H-NMR spectrum of compound 94 is almost identical to the 1H-NMR spectrum of compound 93. The proton at position 5 is a doublet at 8 7.05 ppm, slightly to lower field when compared with the corresponding signal in 93. The EI MS for compound 94 showed a peak at m/z 180 for the molecular ion and a large peak at m/z 137 which indicates a mass loss of 43 from the molecular ion, again corresponding to the loss of the wo-propyl group.  The dialkylated product 95 was characterized by the EI MS and 1H-NMR spectra. The EI MS showed a peak produced by the molecular ion at m/z 222 and a large fragment at m/z 180. The 1H-NMR spectrum showed the occurrence of two wo-propyl groups in very similar environments by the presence of a twelve proton doublet at 8 1.42 ppm. Two separate heptet signals, corresponding to the methine protons of the iso-propyl groups, were visible at 8 4.55 and 4.65 ppm.  CHO  CHO  „  2  H O ^ ^ A ] ^CHO  89  II  6 93 PhCH Cl 2  K C0 2  PhCH 0 2  CH(SPh)  2  PhCH 0 2  BF .Et 0 3  2  PhSH  Scheme 24. The Synthesis of 93,96 and 86  3  CHO  Since the substituted benzaldehyde 93 was required on a large scale, and the reaction with the milder base was more convenient the alkylation was routinely carried out using potassium carbonate as a base. After purification, the 4-w«5-propoxy-3-hydroxybenzaldehyde (93) was treated with benzyl chloride and potassium carbonate to give the substituted benzyl ether 96 as an oil. The 1H-NMR spectrum of the benzyl ether 96 showed a six proton doublet at 8 1.40 ppm and a one proton heptet at 8 4.70 ppm which correspond to the protons of the wo-propyl group. A characteristic singlet at 8 5.2 ppm was visible corresponding to the two benzylic protons of the benzyl ether group. A doublet at 8 7.0 ppm was produced by the aromatic proton at position 5 whilst all of the other aromatic protons gave a multiplet signal at 8 7.25-7.50 ppm. The EI MS of 96 showed a peak caused by the molecular ion at m/z 270. The dithioacetal 86 was obtained by treatment of the substituted benzylic ether 96 with thiophenol in the presence of boron trifluoride.etherate.  This compound was  identified by spectroscopic methods. The 1H-NMR showed a characteristic one proton singlet at 8 5.30 ppm which suggests the presence of the methine proton adjacent to the sulphur groups. The expected signals previously described for the wo-propyl group and the benzyl group were present in the spectrum. The aromatic proton at position 5 gave a one proton doublet at 8 6.90 ppm with a coupling constant of 8 Hz. The other aromatic protons gave rise to a fifteen proton multiplet at 8 7.10-7.42 ppm. The EI MS fragmentation pattern showed a peak for M+-SPh with a m/z value of 363 which is expected because of the weakness of the S-C bond. The compound was carefully purified by chromatography, to remove traces of thiophenol, prior to use in the alkylation reaction. The conjugate addition of the dithioketal 86 to the but-2-en-4-olide (87)°7 was carried out using n-butyllithium as a base (Scheme 25). The n-butyllithium reacts with the dithioacetal 86 to generate a 2-lithio-l,3-dithane68 which then undergoes a conjugate addition to the but-2-en-4-olide (87). The product 97 was obtained in low yield (15%). The substituted butanolide 97 was identified by an IR absorption at 1770 cm-1 which is  consistent with the presence of a y-lactone.  The 1H-NMR of the product showed  characteristic signals for a 1,3.4 substituted aromatic system as previously described. A doublet signal at 8 6.82 ppm identifies the proton at 5' with a coupling constant of 8 Hz. At 8 7.05 ppm a doublet of doublets with coupling constants of 8 Hz and 2 Hz corresponds to the aromatic proton at position 6'. A doublet at 8 1.38 ppm and a heptet at 8 4.57 ppm indicated the presence of the wo-propyl group, whilst the two protons at position 2 occurred as separate doublets of doublets, with coupling constants of 12 and 8 Hz at 8 2.53 and 2.73 ppm. The 3-((3-benzyloxy-4'-propoxy)-a,a-bis(phenylthio)benzyl)butanolide (97) was ,  then alkylated with the bromide 88.69 HMDSfliexamethyldisilazane)and «-butyllithium were used to generate LiHMDS which is the base for the reaction, HMPA was used as a co-solvent. The reaction proceeded to give the desired butanolide 98 in 62% yield. The product 98 was characterized by spectral data. The 1H-NMR of the product showed a six proton doublet at 8 1.38 ppm together with a one proton heptet at 8 4.59 ppm which indicates that the wo-propyl is present in the product The proton at position 2 gives a multiplet at 8 2.90-2.96 ppm and the adjacent proton at position 3 also produces a multiplet at 8 3.20-3.35 ppm. The two non-equivalent protons at position 4 give rise to two sets of signals, both doublet of doublets at 8 3.39 ppm and 8 4.27 ppm. The signal at 8 4.27 ppm shows coupling constants of 11 and 3 Hz. The existence of a coupling constant as small as 3 Hz suggests that this signal is caused by the proton which is cis to the proton at position 3. The signal at 8 3.39 ppm exhibits coupling constants of 11 and 8 Hz which indicates the existence of geminal and trans coupling to this proton. The aromatic protons gave rise to signals consistent with the presence of a 1, 2, 4 substituted ring as was seen in the 1HNMR for compound 97. The EI MS of the compound showed a peak at m/z 595 which corresponds to the fragment produced on the loss of two thiophenol groups from the product 98.  65 The one step tandem addition (Scheme 26) gave a more efficient conversion (65%) to the dibenzylbutanolide 98 than the two step process (9.3%), if water was rigorously excluded from the reaction vessel and very pure dithioacetal 86 was used as the starting precursor. The procedure is based upon literature examples of similar tandem additions.  70  PhS  SPh  OCH Ph 2  88 PhS  Scheme 25. The Synthesis of 97 and 98  SPh  Reaction of the  /ro^-2-(4"-benzyloxy-3",5"-dlmemoxybenzyl)-3-((3'-benzyloxy-  4'-wo-propoxy)-a,a-bis(phenylthio)benzyl)butanolide  (98) with Raney-nickel caused  cleavage of the benzyl ether groups and removed the dithioketal function in one step to give the target dibenzylbutanolide 85 in 90% yield. The dibenzylbutanolide 85 was characterized by consideration of the spectral data found for the compound and by comparison with the previously identified dibenzylbutanolide 69.  Scheme 26. The One-step Tandem Addition Route to 98  Like compound 69, the product 85 showed characteristic 1H-NMR signals for the aromatic protons. The proton at 2* produced a doublet, with a coupling constant of 2 Hz at 8 6.60 ppm and is coupled to the proton at position 6', which appears as a doublet of doublets with coupling constants of 8 and 2 Hz, at 8 6.45 ppm. This proton is in turn coupled to the proton at position 5' which gave a doublet with a coupling constant of 8 Hz at 8 6.73 ppm. Also visible in the spectrum were the signals for the protons of the isopropyl group which appeared as a doublet at 8 1.35 ppm and a heptet at 8 4.51 ppm. Only one of the protons at position 4 in compound 85 was visible as a doublet of doublets at 8 4.12 ppm with coupling constants of 8 and 6 Hz. The other proton at position 4 was  overlapped by the signal at 8 3.78-3.90 ppm caused by the methoxy protons. The signals at 8 5.40 and 5.65 ppm were found to disappear when the sample was shaken with some deuterium oxide and the spectrum retaken. These peaks were assigned to the hydroxyl protons. The EI MS showed characteristic fragments of m/z equal to 416, 374 and 167 corresponding to the fragments shown in Scheme 27 (see Appendix Figure 18).  Scheme 27. The Major Fragments Produced in the EI MS of Compound 85  In summary the dihydroxybenzaldehyde  dibenzylbutanolide  85  was prepared from 3,4-  (89) in 5 steps with an overall yield of 10%.  This  dibenzylbutanolide 85, thus prepared, was then used for bio-transformation reactions.  1.11  T H E SYNTHESIS OF r*AiVS-2-(3 \5"-DIMETHOXY-4",  HYDROXYBENZYL)-3-(3\4'DIHYDROXYBENZYL)BUTANOLIDE (64) AND  TRANS-2-(3",S"-  DIMETHOXY-4"-HYDROXYBENZYL)-3-(3\4'METHYLENEDIOXYBENZYL)BUTANOLIDE (62)  The three precursors 63 64 and 65 were initially prepared in our laboratory by the synthetic route shown in Scheme 28. The route involves the Stobbe condensation of dimethyl succinate with the appropriate aldehyde to produce the corresponding a, (3unsaturated ester, which can then be reduced by magnesium in methanol to give the intermediates 104,105 or 106. In the remaining steps the lactone ring is formed , the bottom aromatic ring is attached in an alkylation reaction, and the benzyl protecting groups are removed if necessary to give the precursors 62,63 and 64. More recent work in our laboratory by J Palaty (M. Sc. thesis UBC 1990) has optimized the preparation of the dihydroxydibenzylbutanolide 63. The development in our laboratory (Section 1.10) of a new synthetic pathway to dibenzylbutanolide compounds led to a new route to two of the previously prepared precursors, 62 and 64 (Scheme 29).  CHO  C0 CH 2  Dimethylsuccinate NaOCH  R'O  46 RR' = - C H 90 R = H, R' = C H 89 R = H, R' = H  C0 H  R'O  3  2  101 RR' = - C H 102 R = H, R' = C H 103 R = H,R' = H  2  3  2  3  3  Mg/CH OH 3  C0 CH  l)KOH  2  2) Ca(BH ) 3) HCl 4  107 RR' = - C H 108 R = H, R' = C H 109 R = H, R' = H  2  3  C0 H  R'O  2  104 RR' = - C H 105 R = H,R' = C H 106 R = H, R' = H 2  2  3  3  CH Br 2  OCH,  R'O  10% Pd-C, H  2  CH O 3  OR"  110 RR' = - C H 111 R = H,R' = C H 112 R = H, R' = H 2  62 RR' = - C H 63 R = H,R' = C H 64 R = H, R' = H 2  3  3  Scheme 28. The Initial Synthetic Route to Precursors 62, 63 and 64  Piperonal (46) has been converted to the dithioketal 11371 in 98% yield and this ketal was then used in a tandem conjugate addition reaction with the bromide 88 and butenolide 87 to produce the intermediate 114 in a yield of 55%. This intermediate was then treated with an excess of Raney-nickel to give the monohydroxydibenzylbutanolide 62 in 70% yield. The Raney-nickel cleaved the dithioketal group and the benzyl ether in a single step. The monohydroxydibenzylbutanolide 62, was characterized by comparison of the spectral data with the data from authentic material, previously prepared by the method outlined in Scheme 28. The 1H-NMR spectrum showed a two proton multiplet at 8 5.94 ppm which corresponds to the methylene-dioxy protons. The aromatic protons gaveriseto the familiar pattern of a doublet of doublets for the proton at position 6', a doublet with a coupling constant of 8.0 Hz for the proton at position 5' and a doublet exhibiting only meta coupling (J = 2.0 Hz) for the proton at position 2*. The two proton singlet at 8 6.39 ppm corresponds to the aromatic protons at positions 2" and 6". The EI MS of the product 62 gave a spectrum with much fragmentation. The peak due to the molecular ion was visible at m/z 386. The monohydroxydibenzylbutanolide 62 was then converted in 73% yield to the trihydroxydibenzylbutanolide 64 by treatment with boron trichloride at -78"C for 2.5 hours. The trihydroxydibenzylbutanolide 64 was characterized by comparison with authentic material previously prepared by the method outlined in Scheme 28, and showed characteristic signals in the 1H-NMR spectrum. A pattern, previously discussed, for a 1,3,4 trisubstituted aromatic ring was present but no signal was present at 8 5.94 ppm, indicating the loss of the methylene-dioxy protons. Two broad peaks which corresponded to three protons were present at 8 7.05 and 7.64 ppm, and were found to have disappeared if the 1H-NMR spectrum was measured after the addition of deuterium oxide to the sample. This indicated the presence of three hydroxy protons at positions 3', 4' and 4".  This route to the two dibenzylbutanolides 62 and 64 is more efficient than the original route. The original route involved 5 or 6 steps respectively and gave the 72  dibenzylbutanolides 62 and 64 in yields of 9.5% and 18%, respectively. The new route involved 3-4 steps and gave the dibenzylbutanolides 62 and 64 in yields of 38% and 26%, respectively.  72  64 Scheme 29. T h e Synthesis of 62 and 64 via a Thioketal  1.12 THE BIO-TRANSFORMATION OF TRANS-2-(3",5"DIMETHOXY-4"-HYDROXYBENZYL)-3-(3 -HYDROXY-4'-wo,  PROPOXYBENZYDBUTANOLIDE (85) WITH THE CFE OF CATHARANTHUS  ROSEUS  The cyclization of dihydroxydibenzylbutanolide 63 in high yield, together with the lack of cyclization for the monohydroxydibenzylbutanolide 62 suggested the requirement for hydroxy groups in the 4 positions of the aromatic rings. However, as mentioned previously (Section 1.8) the precursor 63 cyclized to give a product with the incorrect stereochemistry for the production of podophyllotoxin (3) or etoposide (1). Furthermore, even if the stereochemistry of the ring closure of this precursor could be changed, the product would not be easily converted into an etoposide intermediate. Conversion would require the selective removal of one methoxy group at position 7 whilst leaving the methoxy groups at positions 3' and 5' unaffected. The trihydroxydibenzylbutanolide 64 was not considered a suitable candidate for cyclization as any bio-transformations attempted with this precursor led to a loss of precursor with no identifiable ring closed products (Section 1.6). The likely cause of this is degradation of the aromaticringsinto very polar products (see Section 2.2.4). It was envisioned (Scheme 30) that  rran.s-2-(3",5"-dimethoxy-4"-  hydroxybenzyl)-3-(3 -hydroxy-4 -wo-propoxybenzyl)butanolide (85) could be cyclized l  ,  using the CFE of Catharanthus roseus. The cyclized product 115 could then be treated with boron trichloride to remove the wo-propyl group. Existing chemistry could then be 73  used to convert the compound 70 to the cyclized compound 68 with a methylene bridge in place.  74  This intermediate 68 could then be converted to 4'-demethylpodophyllotoxin (5)  which could be readily converted, with existing chemistry, to etoposide (1). The stereochemistry of the cyclization may proceed conveniently to the 1-2 cis, 2-3 trans system  under the correct bio-transformation conditions. If this did not occur consideration could be given to changing the stereochemistry by synthetic methods.  Scheme 30. The Proposed Bio-tranformation of 85 to 115 and Conversion to an Etoposide Intermediate  A series of bio-transformations (Scheme 31) were carried out using the conditions which had previously been optimized for the dihydroxydibenzylbutanolide precursor 63  (Table 6). The initial experiment with the precursor 85 was conducted on a small scale and the cyclized product 115 was isolated in 74% yield, based on recovered starting material. Some starting material 85 (27%) was recovered.  85  115  Scheme 31. The Cyclization of 85 Using the CFE of[Catharanthus roseus to Produce 115  TABLE 6  The Bio-transformation of the Dibenzylbutanolide 85 Small scale  Large scale 1 Large scale 2  Precursor weight mgs  26  148  190  Hydrogen peroxide equivalents  2.0  2.0  2.0  Peroxidase units per mmol of 85  256  250  250  Reactiontimein minutes  30  30  30  Yield of 115  74%  52%  75%  The 1H-NMR of the product 115 (See Appendix Figure 19) showed three singlets at 8 6.34,6.45 and 6.72 ppm which indicates that no adjacent ortho hydrogens are present on the aromaticrings,this is consistent with the data previously described for the cyclized compound 69. The signal for-the wo-propoxy methyl protons occurs as two doublets at 8  1.15 ppm and 8 1.24 ppm and each doublet corresponds to a three proton signal. This is proposed to be caused by the restricted rotation of the wo-propoxy group. The rotation is restricted by the bulky aromatic B ring which is free to rotate above the plane of the aromatic Aring,thus interfering with the wopropoxy group (Figure 9). This restricted rotation is absent in theringopen precursor 85. The 1H-NMR signal which appears at 8 4.06 ppm is a doublet with a coupling constant of 11.5 Hz, is assigned to the proton at position 1. Due to the large coupling constant this proton is proposed to be in a trans relationship to the proton at position 2. The proton at position 2 gives a signal at 8 2.47 ppm which is a doublet of doublets with coupling constants of 11.5 and 13.5 Hz. This suggests that the proton at position 2 is in a trans orientation to both the proton at position 1 and the proton at position 3 and hence the stereochemistry of the product is as described in Scheme 30. The proton at position 3 shows a very complex pattern and appears as a multiplet at 8 2.53-2.70 ppm. Two doublets of doublets at 8 2.88 and 2.98 ppm are due to the protons at position 4. Both sets of signals show the same coupling constant of 15 Hz which is the geminal coupling constant between the two protons at position 4. The second coupling for the two sets of signals is very different (11 and 4.5 Hz) which indicates that, as expected, one of the protons is cistothe proton at position 3 whilst the other is in a trans orientation. The other signals observed were in agreement with the suggested structure. The EI MS of the product 115 shows a distinct peak caused by the molecular ion at m/z 414 which was peak matched by HRMS to a molecular formula of C23H26O7. The fragmentation pattern of the product is consistent with that of the cyclized product 115 and shows characteristic differences when compared to the fragments visible in the EI MS of the precursor 85 (see Appendix Figure 20).  77  HO.  5  4  3  o  OCH  3  OH  115 Figure 9. The Restricted Rotation of the wo-Propoxy Group in 115  The yields shown in Table 6 are comparable to the yields for the cyclization of dihydroxydibenzylbutanolide 63. The mechanism for cyclization is proposed to be identical for the two butanolides, (Scheme 18) as the wo-propyl group is not involved in the mechanism of the reaction.  1.13 T H E B I O - T R A N S F O R M A T I O N O F TRANS-2-(3",S"DIMETHOXYBENZYL-4 -HYDROXY-)-3-(3 -HYDROXY)-4'-isoM  ,  P R O P O X Y B E N Z Y D B U T A N O L I D E (85) W I T H T H E C F E O F  PODOPHYLLUM PELTATUM  Using the methods described in Section 1.1 a cell culture of Podophyllum peltatum was propagated using MS medium (50% strength) (see appendix Table 31) with 1.5% 55  sucrose added. The innoculum volume used was 15 mL of the harvested aggregates and embryos for every 250 mL of the medium. The period between sub-cultures was 17 days. This cell culture is very different in nature to the Catharanthus roseus cell culture. The Catharanthus roseus cell culture is a true suspension culture, as the cells are  undifferentiated and homogeneously dispersed throughout the liquid medium. However, the cells of the Podophyllum peltatum culture are found in differentiated clumps and it is best described as an embryonic cell culture. For several reasons the embryonic cell culture is more difficult to manipulate. The culture is slow growing, only increasing the cell mass by 50% in 17 days and so requires more material for sub-culturing to maintain the cell culture. Also, the hard nature of the cell masses causes it to be more difficult to produce a reproducible CFE from the culture than from Catharanthus roseus. However, despite these disadvantages it was still of great value to carry out biotransformations with the Podophyllum peltatum cell culture using the precursor 85. It was envisioned that as the cells were producing podophyllotoxin (3)(see Section 3.2), that the enzymes would be present which would not only give a cyclized product but would also give a product with the desired stereochemistry. A small scale bio-transformation was carried out with the precursor 85 (Scheme 31). The precursor (6 mg) 85 was dissolved in a mixture of ethanol and water (8:17, 1.1 mL) and treated with the CFE (257 units of peroxidase activity per mmol of substrate) in the presence of hydrogen peroxide (2.3 equivalents). Cyclization once again occurred and the product, indicated by TLC and HPLC, was the butanolide 115. The yield as measured by consideration of the areas of the peaks from the HPLC analysis of the reaction mixture was 20%. This bio-transformation was later repeated on a large scale and the cyclized product 115 was isolated in 19.4% yield. The 76  trans stereochemistry at position 1 and position 2 was again indicated by the !H-NMR coupling constants for the proton at position 1 and position 2. The failure of this cyclization reaction to give a product with the stereochemistry found in podophyllotoxin (3) suggests that either our precursor is not reaching the correct enzyme, or the precursor is not a suitable substrate for the enzyme system.  Scheme 32. The Cyclization of 85 to 115 Using the CFE of Podophyllum peltatum  1.14  OTHER WORK IN OUR LABORATORY. SYNTHESIS OF TRANS2-(3 ^5 -DIMETHOXY-4"-HYDROXYBENZYL)-3-(3 ,  M  ,  HYDROXY-4'-is0-PROPOXYBENZYL-a-HYDROXY)BUTANOLIDE (116)  In our laboratory rran5-2-(3",5"-dimethoxy-4"-hydroxybenzyl)-3-(4'-j5opropoxy-S'-hydroxybenzyl-a-hydroxyJbutanolide (116) has been conveniently prepared?  6  from the dithioketal 98 intermediate (Scheme 33) by removal of the dithioketal functionality with boron trifluoride.etherate and mercuric oxide. The ketone 117 produced was identified by 1H-NMR spectroscopy and by the molecular ion peak by EI MS at 610 mass units. Reduction with tri-r-butoxylithium aluminium hydride gave the epimeric alcohols 118 in a ratio of 10 : 1 with the 4-fJ isomer predominating. The benzyl groups were then removed by palladium catalyzed hydrogenolysis to yield the /rfln5-2-(3",5"-dimethoxy4"-hydroxybenzyl)-3-(4-wo-propoxy-3'-hydroxybenzyl-a-hydroxy)butanolide ,  quantitative yield.  (116) in  Bio-transformation of this precursor 116 (Scheme 33) has been recently carried out by others, and under optimized conditions, ring closure occurred to give 118 in 84% 76  yield using the CFE prepared from the Catharanthus  roseus cell culture. The bio-  transformation was men also carried out using the whole cells of Podophyllum peltatum and the same cyclized product 118 was obtained in a 51% yield although optimization of conditions has not yet been completed. Thus, the precursors investigated so far 62, 85 and 116 have undergone an oxidative cyclization to produce the podophyllotoxin skeleton. The cyclizations are achieved in good yields with a short reaction time. The cyclized products, particularly 118, will be evaluated further for their utilization in developing an efficient route to etoposide (1).  1.15 FUTURE PROSPECTS  The above studies have provided an excellent basis for the directions of future research. With the Podophyllum peltatum cell line now developed and shown to produce podophyllotoxin (3), 4'-demethylpodophyllotoxin (5), podophyllotoxone (60) and deoxypodophylloxin (6) (see Section 3.2), it is clear that the relevant enzymes for the above ring closure and in the correct stereochemistry are present. The studies with 116 and related precursors are currently underway by other coworkers in our laboratory.  PhS S P h  118  116  Scheme 33. The Synthesis of Precursor 116 and Subsequent Cyclization to Product 118  CHAPTER 2 INTRODUCTION  2.1  HYDROXYLATIONS  USING  ENZYMES  The successful oxidative ring closure of compounds with the yatein (57) skeleton would lead to products similar to deoxypodophyllotoxin (6). The compound would then require benzylic oxidation before conversion to etoposide (1) could be achieved. Enzymatic oxidations are common bio-transformations and so the possibility of the use of plant cell cultures to cany out the hydroxylation was investigated. The cytochrome P-450s are well known hydroxylating enzymes common to many plants and animals. They are hemoproteins with a characteristic absorbance at 450 nm in the U V spectra of their CO adducts. This absorption was first reported by Williams in 1955, and then independently by Klingerberg and Garfinkel in 1958.78 In 1962 Omura 77  and Sato confirmed the heme nature of the protein and gave the class of compounds their name "Pigment with an absorption at 450 nm". Research has been extensive since those initial reports and many P-450s now have known amino acid sequences. In 1982 P450cam. was analyzed by X-ray crystallography.79 The prosthetic group of the P-450 is  the common ferroprotoporphyrin LX group but the specificity of the reactions of the P-450 is controlled by the arnino acid groups surrounding the active site. The P-450 proteins are ubiquitous in nature and many of their hydroxylation reactions are well characterized.80 In animals, cytochrome P-450 are responsible for steroid metabolism, drug detoxification, fatty acid hydroxylation and the oxidation of polycyclic aromatic hydrocarbons into potent carcinogens.81 In plants P-450 are responsible for many transformations such as the conversion of kaurene (119) to kaurenol (120) (Scheme 34)  82  and the 4-hydroxylation of cinnamic acid (121) to 4-hydroxycinnamic acid (54) , which 83  is an intermediate in the lignin biosynthetic pathway (Scheme 16). Also, many cytochrome  P-450 have been shown to be involved in the oxidation of alkenes, to form epoxides and allylic alcohols.  84  P-450  Kaurene (119)  Kaurene-19-ol (120) C0 H  CO H 2  2  P-450  HO  4-Hydroxycinnamic acid (54)  Cinnamic acid (121)  Scheme 34. Two Examples of Hydroxylation Reactions Using P-450s  As mentioned many P-450s are found throughout nature and they are generally isolated by fractional centrifugation (Scheme 34). Although some P-450s are found in the mitochondrial fractions, most are found in the microsomal fraction. The microsomal fraction is isolated as a supernatant after centifugation at 100,000 G for 30 minutes, as depicted in Scheme 35; but the exact conditions may vary for different systems. During the bio-conversions discussed cytochrome P-450 is behaving as a monooxygenase. The mono-oxygenation occurs according to Equation 1 and involves the incorporation of one oxygen atom into the substrate.  85  DH  2  +  0  2  +  XH  2  Equation 1  DH(OH)  +  X  +  H 0 2  The enzyme system has an absolute requirement for a reducing co-substrate (XH2).  Cell homogenate C = centrifugation G.x 10 / minutes C3/10  C 0.8 /10  Nuclear fraction  C 25 /10 Heavy microsomal fraction  C 100 / 30  Light microsomal fraction Microsomal-*fraction  Light mitochondrial fraction C 100/ 180  Mitrochondria Lysosomes Peroxisomes Scheme 35. The Preparation of the Microsomal Fraction  Nicotinamide adenine dinucleotide phosphate (NADPH) is the usual reducing cofactor. However, the pyridine nucleotides are two electron donors whilst cytochrome P450 can only accept one electron at atime,thus other transfer agents are involved in the  85 electron shunt. A typical system for mitochondrial or bacterial P-450s is shown in Scheme 36.  FpH  2  NADPH  NADP  Fp = flavoprotein Scheme 36. The Electron Transfer System From NADPH to P-450  The NADPH undergoes a two electron transfer to a flavoprotein, usually FMN (flavoadenine mononucleotide) or FAD (flavoacknine dinucleotide) which then interacts with a NHI (non-iron heme protein), usually adrenodoxin or rubredoxin (in plants the noniron protein shunt is absent). These proteins can then undergo a single electron transfer to the P-450 system which is then able to oxidize the substrate. A more detailed look at the function of the P-450 system is shown in Scheme 37. The important factors are as follows:1. the substrate binds to the ferric complex, 2. a one electron reduction then occurs to give an Fe2+ species. The electron is provided by the flavoprotein system, 3. an oxygen molecule then binds to this complex, 4. again a one electron reduction occurs, 5. the product is now formed and the ferric state regenerated to be recycled. 1  5  H 0+ 2  3 -3+  Fe=0 S  2H " 4  o  2  one electron 4  5 SO (product) Scheme 37. The Mode of Action at the Active Site of P-450  There are problems associated with the use of P-450 systems in plant cell biotransformations particularly i f a C F E or other enzyme extract is required. The most significant problem is the labile nature of the enzyme systems. Also, on a large scale the cost of enzyme isolation may be prohibitory and the co-factors required are expensive. However, success has been achieved in immobilizing some P-450 enzymes and it may become possible to recycle the co-factors.86 Work is also being carried out to develop model compounds which will mimic the action of cytochrome P-450S.  87  R E S U L T S AND DISCUSSION  2.2  T H E ATTEMPTED HYDROXYLATION OF T H E D I B E N Z Y L B U T A N O L I D E (69) AND D E O X Y P O D O P H Y L L O T O X I N (6) WITH P L A N T C E L L C U L T U R E S AND CFEs  2.2.1 UTILIZING T H E C Y C L I Z A T I O N CONDITIONS (SECTION 1.8)  It was decided to use the cyclized precursor 69 as a model compound for the next step in the proposed route to a suitable etoposide intermediate.  It was hoped that the  molecule would be hydroxylated at the benzylic position 4, by a suitable enzyme system. The stereochemistry of position 1 could then perhaps be changed by chemical methods to the necessary stereochemistry for the podophyllotoxins. Initially the precursor 69 was treated with the CFE of Catharanthus roseus (Scheme 38). It was anticipated that radical formation could occur to produce an intermediate 122 which could react further to produce the hydroxylated product 123. It was hoped that the Catharanthus roseus system could carry out the bio-transformation as peroxidase enzymes are present. In the previous bio-transformations the enzyme system was presumably involved in the cyclization reaction of the dibenzylbutanolides 62 and 85 and this reaction occurred over a smalltimeperiod, thus, for hydroxylation to occur the reaction mixture was stirred for several hours. The parameters chosen for the reaction are shown in Table 7. The conditions are very similar to those used in the optimized cyclization of the dibenzylbutanolide precursors 62 and 85.  When the CFE (prepared using the method previously outlined: cell at age 10 days, 200 units of peroxidase per mmol of precursor) was added to the precursor 69 precipitation (presumably of the precursor) occurred but it was assumed that some of the precursor  would remain in solution. The reaction mixture was left stirring for 20 hours, which is considerably longer than the time required for theringclosure reaction to occur. However HPLC analysis (samples were injected onto a reverse phase C-18 column : solvent : methanol: water 45 :55, each with 0.1% acetic acid, flow rate 1.0 mL/minute) showed no consumption of precursor 69 and no formation of any new product.  123 Scheme 38. The Proposed Hydroxylation of Precursor 69 Using the CFE of Catharanthus roseus  TABLE 7  The Conditions for the Attempted Hydroxylation of 69 Using  the C F E of Catharanthus roseus Reaction flask Hydrogen peroxide equivalents  1.9  Peroxidase units per mmol of 69  200  Reaction time in hours  20  2.2.2  H Y D R O X Y L A T I O N A T T E M P T S USING C O - F A C T O R S AND ENZYME  STABILIZERS  It was decided to repeat the hydroxylation attempt but with two significant changes; other co-factors beside hydrogen peroxide would be used, and more polar solvents, such as DMSO, employed to dissolve the precursor 69. The other co-factors were chosen as they are identified as possible co-substrates in hydroxylation reactions, primarily by oxygenase enzymes.88 Table 8 shows a summary of Experiment 1. The flasks were allowed to incubate for 28 hours.  TABLE 8  The Attempted Hydroxylation of 69 Using Various Co-factors Flask number  Experiment 1  1  2  3  4  Peroxidase units per mmol of 69  245  245  245  245  Co-factor  FMN  Ascorbic acid NADPH  H2O2  No hydroxylation product was observed by HPLC analysis, in the manner previously described, in any of the reaction flasks, only the starting precursor 69 was identified. As previously discussed, the cytochrome P-450 mono-oxygenase enzymes are able to hydroxylate a wide range of substrates. A feature of the P-450 system is the necessity for reducing co-factors in order for the catalytic cycle to be complete, and the need for molecular oxygen which would be provided by the hydrogen peroxide present. However, a major problem with the use of mono-oxygenase enzymes is the loss of enzymatic activity due to modifications at the active site. The most common of these denaturing effects is the oxidation of the thiol groups of the amino acid cysteine (Scheme 39),89 which is found at the active site of the enzyme.90  -S-S- Disulfide Protein-SH  -SOH -S0 H 2  Scheme 39. The Oxidative Denaturing of Cysteine These oxidations are promoted by divalent ions, such as those of some heavy metals. These can be successfully removed by the use of complexing agents such as EDTA which is an excellent ligand. Another protective method involves the addition of controlled amounts of compounds which have thiol groups present. These thiol compounds will then react with any activated oxygen present before reaction with the thiol group of the cysteine amino acids can occur. Dithiothreitol (DTT) (124) is one such reagent and is able to provide protection of proteins, in concentrations of 0.5-1.0 mM, for up to 24 hours (Scheme 40).  91  91  SH  +  H 0 2  SH  124  125  Scheme 40. The Oxidation of Dithiothreitol (124)  An experiment was then carried out (Experiment 2) using the reducing co-factors as in Experiment 1, but also employing two compounds to aid in the stabilization of the enzyme system (EDTA and DTT). The CFE from the cells of Catharanthus roseus (age 11 days) was prepared using the method previously mentioned. The various co-factors and stabilizing agents were added as shown in Table 9.  TABLE 9  Attempted Hydroxylation of 69 Using Various Co-factors and Stabilizing Agents  Flask number 1  2  3  Precursor 69 20 mg  20 mg  20 mg  CFE  20 mL  20 mL  20 mL  Hydrogen peroxide  5 mL  5 mL  Ascorbic acid 50 mg NADPH  80 mg  80 mg  EDTA  20 mg  20 mg  DTT 124  10 mg  10 mg  The percursor 69 (60 mg) was dissolved in acetone (5 mL) and this solution was used for three separate experiments each utilizing 450 peroxidase units (from the CFE) per mmol of precursor. The analysis using HPLC (method previously mentioned) showed clearly that the precursor 69 was not metabolized and no new products were identified. Three larger scale experiments were carried out (Experiment 3) utilizing the CFE prepared from Catharanthus roseus (cell age 12 days), in the hope of detecting small amounts of an hydroxylated product. Thefirstof these experiments involved the use of NADPH (40 mg) as a co-factor, and an acetone (13 mL) and water (87 mL) mixture to dissolve the precursor 69 (200 mg). CFE (300 mL, peroxidase activity 1.2 units) was added to initiate the reaction. The reaction mixture was left to stir at room temperature for 25 hours and samples were removed for HPLC analysis (as previously described). The HPLC results did not indicate that a reaction had taken place. After 25 hours the mixture was treated with celite and then filtered and extracted with dichloromethane. The remaining aqueous solution was then extracted with ethyl acetate. Although the material was not purified a large portion (196.8 mg) of the starting precursor 69 was isolated in the dichloromethane extract. By TLC, the ethyl acetate fraction contained many different compounds and by comparison with the TLC from the control experiment these compounds likely included compounds present in the plant cell material. The high resolution EI MS for this fraction showed the presence of a peak of m/z 402 which corresponds with the molecular ion for product 123. The *HNMR did not show any protons in the region 5 6.0 to 8.0 ppm and if aromatic lignan compounds were present we would expect to found some peaks in this region. However, this material was not purified and no further evidence of the product 123 was found. The second experiment (CFE 300 mL, peroxidase activity 1.2 units), involved the use of a large quantity of ascorbic acid (600 mg) as a co-factor. The precursor 69 (104  mg) was dissolved in a acetone (13 mL) and water (87 mL) mixture before the addition of the co-factor and the CFE. The results were comparable with those for the first experiment (NADPH as co-factor) but on this occasion EI MS data did not provide a peak to indicate the presence of the product 123. The final experiment utilized the precursor 69 (100 mg) again in acetone (9 mL) and water (60 mL) but ferrous sulphate (100 mg) and hydrogen peroxide (10 mL) were used as the co-factors. The reaction mixture was treated in the same way as discussed previously and extracted using the procedure outlined. Once again the starting material was isolated in the dichloromethane fraction. The 1H-NMR of the ethyl acetate fraction did not show any peaks in the aromatic region of the spectrum. Although high resolution EI MS had indicated the possible presence of the product 123 it was clear from these experiments that the precursor 69 was not being metabolized to any significant degree and so that line of approach was abandoned. Although it was feasible that mono-oxygenase enzyme was being denatured in some way, it was decided to examine first the possibility of a bio-transformation of a different precursor. The new precursor was deoxypodophyllotoxin (6), and a sample was supplied to us by Japanese workers where it was available in reasonable quantities by extraction from the seeds of Hernandia ovigera .92 If precursor 6 was successfully hydroxylated in the 4 position it would lead directly to podophyllotoxin (3) or epipodophyllotoxin (37) (Scheme 41) both of which could be converted into etoposide (1). A small sample of deoxypodophyllotoxin (6) (33 mg) was dissolved in acetone (5 mL) and treated with the CFE prepared from the Catharanthus roseus cell culture (cell age 14 days: 1760 units of peroxidase activity per mmol of the precursor) in the presence of hydrogen peroxide (9 equivalents). When the CFE was added to the reaction mixture some precipitation occurred. This was found to be a consistent problem with the use of deoxypodophyllotoxin (6) as a precursor. The reaction mixture was stirred at room temperature for 18 hours during whichtimesamples were removed for HPLC analysis (the  details have been previously discussed).  The mixture was then extracted with  dichloromethane and the remaining aqueous solution extracted with ethyl acetate. After chromatography, the presence of starting material (27 mg, 82%) was confirmed by !HNMR analysis in which a comparison was made with a sample of deoxypodophyllotoxin (6). A very similar experiment was set up using deoxypodophyllotoxin (6) (30 mg) as the precursor and utilizing the same CFE as prepared for the above experiment but, using hydroquinone (150 mg) as a co-factor. The reaction was followed and worked up in the same way as the above reaction and yielded the recovery of starting material (22 mg, 73%), with no new products being isolated.  Scheme 41. The Proposed Hydroxylation of Deoxypodophyllotoxin (6) Using the CFE of Catharanthus roseus  2.2.3 THE CHARACTERIZATION OF THE MICROSOMAL FRACTION FROM CATHARANTHUS  ROSEUS AND FURTHER ATTEMPTS  AT THE HYDROXYLATION OF (6)  Hydroxylation reactions have been achieved with enzymatic preparations from Catharanthus roseus by Coscia 93,94  m  1973 (Scheme 42). Hydroxylations have also been  carried out using the microsomal fraction prepared from the 5 day old seedlings of Catharanthus roseus. Geraniol (126) was hydroxylated to produce 10-hydroxygeraniol (127) and its cis isomer nerol (128) was hydroxylated to give 10-hydroxynerol (129). The mono-oxygenase characterized was found to be dependent upon NADPH and oxygen as co-factors. The method employed by Coscia involved the the grinding of the seedlings in a grinding medium consisting of sucrose, potassium chloride, magnesium chloride, EDTA, metabisulfite and DTT in a buffer at pH 7.6. After centrifugation a pellet was obtained from this mixture which was then homogenized in a buffer at pH 7.8 containing DTT, sucrose, EDTA and glycerol. The pellets thus prepared were used as the source of enzyme. Incubation was carried out in the presence of glucose-6-phosphate, glucosesphosphate dehydrogenase (commercially available) and NADP+ to regenerated the NADPH in situ.  The dependency on these co-factors is demonstrated by the values of percentage activity shown in Table 10. The activity was assayed by the measurement of the degree of hydroxylation of geraniol usingradio-labellingtechniques. The characterization of a monoterpene hydroxylase was also achieved from the cell suspension culture of Catharanthus roseus.  95  Once again NADPH and oxygen were  necessary co-factors in order to achieve hydroxylation.  96  R Catharanthus roseus  CFE  H2O2 126 R = CH OH, R' = H 128 R = H, R' = CH OH 2  2  127 R = CH OH, R' = H 129 R = H, R' = CH OH 2  2  Scheme 42. The Enzymatic Hydroxylation of Geraniol (126) and Nerol (128)  T A B L E 10  The Major Factors Required for the Hydroxylation of Geraniol (126) and Nerol (128) as Measured by Coscia Components present  10-Hydroxygeraniol formation pmoles/mg protein/minute  NADPH system plus oxygen  148  NADPH system  6.0  Oxygen  5.0  It was hoped that the hydroxylase activity of our enzyme preparation could be utilized in the benzylic hydroxylation of our precursors 6 and 69. Thus, we followed a procedure (Scheme 43) similar to that used in Coscia's characterization of hydroxylase from the cell culture. Many differences exist between this procedure and the previously discussed procedure used to prepare the CFE for the bio-transformation experiments involving the cyclization reactions. The cells from the Catharanthus roseus cell culture (3000 mL, age 11 days) were washed with the extraction buffer which contained sucrose (70 mM) and D-  mannitol (220 mM). The purpose of these sugars is to help to mimic the conditions present in the cell before homogenization. The pH of the buffer used in this procedure is 7.4, in contrast to the previous pH used (pH 6.3) which was optimized for the peroxidase catalyzed cyclization reaction (Section 1.8). The mixture wasfilteredand the wet cells (750 g) were then homogenized in the presence of the extraction buffer (600 mL). This homogenate was then centrifuged (Scheme 43) at 1,000 G for 10 minutes to produce the pellet (PI) and a supernatant (SI). The supernatant (SI) was further homogenized at 10,000 G for 10 minutes to produce a second pellet (P10) and a new supernatant (S10). The final centrifugation step involved the use of an ultra-centrifuge in order to centrifuge some of the supernatant (S10) at 100,000 G for 75 minutes. This provides us with the pellet (P100) and the final supernatant (S100). The pellets P10 and P100 were resuspended in a resuspension buffer (30 mL). The resuspension buffer was a phosphate buffer (pH 7.5) containing EDTA (1 mM), DTT (0.1 mM) and 10% w/v glycerol. The protein stabilizing agents EDTA and DTT have previously been discussed. The glycerol is believed to prevent enzyme denaturing by reducing the unfolding of proteins in an aqueous system.  96  Certain cytochromes when in their reduced form, are able to produce adducts on addition of CO which have a characteristic absorption at 450 nm. These enzymes, as mentioned previously, are called cytochrome P-450s. Thus, the measurement of such an absorption with a new enzyme preparation is an indication that a cytochrome P-450 system is present. Samples of the S10, P10, S100 and P100 fractions were used for the measurement of the CO difference absorption spectra (the samples were defined as in Scheme 43). The spectra for the P100 fraction are shown in Figure 10. A comparison of the absorption of the CO adduct obtained by Coscia can be made by examining Figure 11.  98 Cell culture Filter Homogenize  Centrifuge 1,000 G 10 minutes  Supernatant SI  Pellet PI  Centrifuge 10,000 G 10 minutes Supernatant S10  Pellet P10  Centrifuge 100,000 G 75 minutes  P100  Supernatant S100  Scheme 43. The Procedure used to Prepare the Microsomal Pellets  The two sets of curves in Figures 10 and 11 are almost identical. On addition of sodium dithionate, which reduces the enzyme, a shift of the Soret band occurs toward lower wavelength (a bathochromic shift). Then on addition of CO a small increase in absorption is observed. Although a small band is visible at 450 nm the absorption does not increase visibly on addition of CO. If such an increase in absorption was observed it would be a clear indication of the presence of a P-450 enzyme system. Therefore, we cannot identify or measure the P-450 content of our microsomal preparation. This result is consistent with the findings of Coscia, who noted that other preparations with similar absorption spectra, are known to contain cytochrome P-450 enzymes. The peroxidase specific activity was also measured for each fraction. The specific activity is defined as the peroxidase activity in units (measured as discussed previously) divided by the protein concentration of the sample in mg/mL. The results are shown in Table 11.  T A B L E 11  The Specific Activities and Protein Concentration of the Centrifugation Fractions  Fraction  [Protein] mg/mL  Specific activity  SI  1.48  0.99  S10  1.46  0.90  SI 00  0.86  1.65  P10  1.52  0.16  P100  1.72  0.00  100  2.6 2.4 -  +Dithionate +CO  2.2Initial  2.0 1.8 1.6 1.4 -  -—i—i—r—i—i—"i—1  390 410  430  450 470  X(pm)  4 9  0  5  1  0  5  3  0  Figure 10. The C O Absorption Spectrum of the P100 Fraction  0.6  +Dithionate  0.5 O.D.  .•CO •"f  0.4 Initial  0.3 V*  0.2  X* \> \ •» \ ^ \>  0.1  \  "N \  400  420  440  460 480 X  500  -.V  520  (nm)  Figure 11. The C O Absorption Spectrum of the P100 Fraction as Measured by Coscia  101 The results indicated that the peroxidase activity is concentrated in the soluble fractions, where the peroxidase activities values were consistent with the values measured previously from CFE preparations (typically 1.0-1.8 units). The specific activities were close to zero for the resuspended pellets. Bio-transformations were attempted using the supernatant from the 10,000 G centrifugation step. This fraction (S10) was chosen, as the volume of suspension available was considerably larger (800 mL) than the volume of the suspension available (30 mL) at the P100 stage. Also the CO absorption curves for the twofractionswere very similar as were the peroxidase activities. In an initial experiment, deoxypodophyllotoxin (6) (98.6 mg) was treated with the S10fraction,preparedfromthe Catharanthus roseus cell culture (cell age 10 days), in the presence of FAD (500 mg). FAD is, as mentioned previously, a co-factor involved in the P-450 catalytic cycle. The reaction mixture was stirred at room temperature for 24 hours after which it was quenched and extracted with dichloromethane. The remaining aqueous solution was then further extracted with ethyl acetate. Column chromatography of the fractions gave recovered deoxypodophyllotoxin (6) (95.6%), as identified by comparison of the spectral data with that of the authentic starting material. A series of small scale reactions were also conducted with the co-factors NADPH and FAD and utilizing the S10 fraction prepared as described previouslyfromthe cell culture (age 11 days) of Catharanthus roseus. It was hoped that under these conditions the enzyme system would carry out the hydroxylation of the ring closed precursor 69 and/or deoxypodophyllotoxin (6). Each of the two substrates were dissolved in acetone to provide a stock solution (1 mg per 0.2 mL). Co-factors and the S10fraction(24 mg of protein) were added to samples of the stock solution as shown in Table 12.  T A B L E 12 Small Scale Reactions Using the S 1 0 Fraction and NADPH and/or F A D as Co-factors to Hydroxylate 6 and/or 69  Flask  S10  6  69  A  21 mL  B  21mL  0.2 mL  C  21 mL  0.2 mL  D  21 mL  0.2 mL  E  21 mL  0.2 mL  F  21 mL  0.2 mL  G  21 mL  0.2 mL  H  21 mL  I  21 mL buffer  J  21 mL buffer  FAD  NADPH  16 mg 17 mg  16 mg  16 mg 17 mg  16 mg  17 mg  16 mg  0.2 mL  17 mg  16 mg  0.2 mL  17 mg  16 mg  The reactions were monitored by HPLC (as previously described) attimeszero minutes, 30 minutes, 100 minutes and 17 hours at which point the reaction was stopped, The samples from the reaction flasks of precursor 69 did not show any new peaks by HPLC analysis. The samples from the reaction flasks of deoxypodophyllotoxin (6) were "spiked" with authentic samples of podophyllotoxin (3) and no peaks which corresponded to podophyllotoxin, or any new peaks, were found. From "spiking " experiments on blank samples is was found that a 5-10% conversion to podophyllotoxin (3) could be identified. Thus, the metabolism of the precursors 6 and 69 by the Catharanthus roseus preparations was not observed. This lack of reaction could be caused by several factors; benzylic hydroxylases not being present, the enzyme system being too labile despite our precautions, or the precursors may be unsuitable. Lack of solubility, inability to pass  103 through the cell membrane or incompatibility with the enzyme active site could all be factors involved in precursor suitability.  2.2.4 HYDROXYLATION ATTEMPTS WITH THE CELL CULTURE OF TRIPTERYGIUM WILFORDII  Other work in our laboratory,97 concerned with the bio-transformation of tobacco cembranoids, was successful in achieving an hydroxylation reaction (Scheme 44). The cembranoid 130 was converted to a mixture of the two allylic alcohols 131, and the epoxide 132 in a ratio of 1:2.S, by incubation with the resuspended cell culture of Tripterygium wilfordii. No co-factors were added and the reaction was carried out at a pH of 7.5. The bio-transformation was also successful when the whole cells of Tripterygium wilfordii.v/QK employed for the reaction. No bio-transformation had been observed when the cembranoids were treated with the CFE of Catharanthus roseus. As this culture obviously has hydroxylation capabilities, we incubated deoxypodophyllotoxin (6) with the whole cell culture of Tripterygium wilfordii. (Scheme 44). Deoxypodophyllotoxin (6) (100 mg) was dissolved in acetone (20 mL) and added to the resuspended cells (1000 mL, resuspended in buffer at pH 7.5 containing 8% sucrose w/v) of Tripterygium wilfordii (age 16 days). Samples (15 mL) were removed from the reaction mixture after 24, 48 and 120 hours for HPLC analysis as described previously. Because one of the flasks used were found to be contaminated with bacteria, only 500 mL of the culture was harvested. After harvest the material from the flask was extracted with ethyl acetate to yield a crude extract (184 mg). Separation by chromatography, of a portion of this extract (70 mg) gave deoxypodophyllotoxin (6) (15 mg) as the only identified material. It appears that under the conditions used that deoxypodophyllotoxin (6) is not metabolized by the whole cells of Tripterygium wilfordii.  104  HO  OH  1 3 1 : 1 3 2 , 1 : 2.5  Scheme 44. The Bio-transformation of Tobacco Cembraniods with the Cell Culture of Tripterygium wilfordii  The hydroxylation of 69 and deoxypodophyllotoxin (6) was also attempted using the CFE prepared from the Tripterygium wilfordii cell culture (age 21 days) (Scheme 45). The CFE was prepared using the procedure previously discussed for the preparation of the CFE from Catharanthus roseus and is reviewed below. The cells (2 L) were filtered from the broth and washed with ice-cold distilled water. Phosphate buffer (0.1 M, pH 7.3, 270 mL) was then added and the suspension was homogenized. The homogenate was centrifuged (10,000 G for 30 rninutes) to provide the CFE. The peroxidase activity was measured together with the protein concentration to give a value for the specific activity of 7.3 units/mg protein/mL of solution, for the Tripterygium wilfordii. This value is considerably higher than the value normally found  105  for the CFE of Catharanthus roseus (typically 1.0-2.0). As shown in Table 13 two biotransformations were attempted using hydrogen peroxide as a co-factor and using the precursors 6 and 69.  |j  O Tripterygium wilfordii / cells  CH 0 3  OCH3  Tripterygium wilfordii OCH3 C F E * H0 2  2  "tt Tripterygium wilfordii  CH 0 3  H0 2  CH3O  OCHo  C  H O 3  2  CH3C  0CH3 OH  69  123  Scheme 45. Proposed Bio-transformations using Tripterygium wilfordii  Analysis of samples by HPLC in the manner previously described did not show any loss of precursor or the formation of any new products for either of the two precursors.  106  TABLE 13 Attempted Hydroxylation of 69 and 6 with the CFE of Tripterygium wilfordii Flask #  Precursor  CFE  Solvent 1  (69)(50mg)  Hydrogen peroxide  150 mL  10 mL  150 mL  10 mL  150 mL  10 mL  acetone (2.0 mL) (6)(50mg) acetone (1.0 mL) none acetone (1.5 mL)  One of the concerns of the research was that the solubility of the two precursors used in the hydroxylation attempts 6 and 69 was low. Precipitation usually occurred on addition of the CFE. Thus, two new precursors 133 and 134 were prepared which would retain the podophyllotoxin skeleton but be more soluble in aqueous media. The synthesis of these two compounds is shown in Scheme 46. Compound 134 was prepared by Arimoto98 in our laboratory.  Compound 133 was prepared by the addition of  deoxypodophyllotoxin (6), in dichloromethane, to a solution of boron trichloride also in dichloromethane. The reaction mixture was kept at -78 °C for 2 hours before being quenched. The complex mixture was then refluxed with acetone, water and calcium carbonate for 2 hours. After work-up and purification by column chromatography the product (133) was isolated in 67% yield. The compound was characterized by analysis of the spectral data. The 1H-NMR showed a pattern typical for the fused A and B rings as discussed previously. The proton at positions 5 and 8 appeared as one proton singlets, and the protons at positions 2' and 6' appeared as a two proton singlet at 5 6.42 ppm. The proton  at position 1 gave a doublet at 8 4.51 with a coupling constant of 4 Hz which is consistent with a coupling constant expected for a cis interaction between the proton at position 1 and position 2. The proton at position 2 appears as a one proton multiplet at 8 3.08 ppm. The protons at position 11 provide a distinctive pattern of two doublets of doublets which has been discussed previously. The two doublets of doublets occur at 8 3.96 and 4.42 ppm. The other aliphatic protons, at positions 3 and 4 are found as a multiplet at 8 2.77 ppm. The EI MS gave a peak with a m/z value of 386 which corresponds with the weight of the molecular ion. The fragmentation pattern is shown in (see Appendix, Figure 20). DemethylenepodophyllotDxin (133) (100 mg) was incubated with resuspended cell culture (age 21 days, 500 mL per 50 mg of precursor) from Tripterygium wilfordii. Half of the precursor was dissolved in DMSO while a second portion was dissolved in ethanol and water before addition to the cell culture, no precipitation was observed. The compound was incubated for 100 hours. The reaction mixture using DMSO as the solvent was not worked-up but showed no evidence of a new product using HPLC analysis (the HPLC analysis was used as discussed previously). The reaction mixture using ethanol and water as solvents was extracted and from the total mass of the extracts only a trace of the starting material could be identified. Many products and materials extracted from the cells were visible by TLC. This result suggests that decomposition of the precursor 133 to more polar products had occurred. This situation may reflect the high peroxidase activity of this cell culture (specific activity measurements are typically 5 to 10timesthose from the Catharanthus roseus cell line as mentioned previously), and the presence of the hydroxy groups at positions 6 and 7. This will make the precursor 133 very susceptible to oxidative degradation.  108  134 Scheme 46. The Preparation of the Precursors 133 and 134 From Deoxypodophyllotoxin (6)  Scheme 47 shows two possible routes of the degradation process for a typical 1,2dihydroxy aromatic compound 135. Route A ' 9 9  1 0 0  is the peroxidase catalyzed reaction to  form orr/io-quinones 136 and then diacids 137 which are very polar, may be further metabolized. Route B101 shows the direct conversion of 1,2 dihydroxy aromatic systems into diacids 137, this conversion is carried out by dioxygenases and involve the incorporation of two oxygen atoms into the substrate.  109  Route B  Scheme 47. The Proposed Degradation of 1,2-Dihydroxy Aromatic Compounds  The problem of over oxidation by the Tripterygium wilfordii cell culture directed the research to the Catharanthus roseus cell culture. A series of small scale experiments were conducted with the CFE of the Catharanthus roseus cell culture (age 10 days) utilizing the following co-factors, NADPH, FAD, FMN, MnCl2 and hydrogen peroxide. The two precursors 133 and 134 were investigated. In the experiments 2 mg of each precursor was used in the reaction flasks. For each precursor the following set-up was used: one flask contained only the precursor and CFE, one flask contained the precursor, CFE and one of the co-factors (NADPH, FAD, FMN, manganese chloride or hydrogen peroxide) making up the contents of five flasks and finally one flask contained the precursor, CFE and all of the co-factors. Samples were removed from the flasks over a period of 20 hours and were analyzed by HPLC to attempt to identify any formation of the hydroxylation products 138 and 139 (Scheme 48). The results showed a metabolism of the precursor 134, only in the presence of both CFE and hydrogen peroxide. This result was also found  for the precursor 133. In the experiment with compound 133 considerable loss of material occurred and a new HPLC peak at a similar retentiontimeto the expected product 138, was noted in the flask containing CFE and FAD. A sample of the expected products 138 and 139 has been prepared by Arimoto in our laboratory using the selective demethyleneation of the podophyllotoxin (3) (Scheme 49).  102  However this could not be  confirmed by co-injections of authentic 138 and repeated analysis of the mixture by HPLC. Due to the large number of products seen in the reaction mixture (TLC and HPLC analysis) further isolation of pure compounds was difficult, and the results as to whether the desired hydroxylation had occurred are inconclusive. Control experiments (substrates in buffer without CFE and without co-factors together with CFE with all of the co-factors only) were conducted to indicate (HPLC) if any interaction occurred. Also the expected products 138 and 139 were incubated with the CFE to determine if they were stable and thereby not metabolized further, prior to isolation. Although the results were not conclusive, in the sense that metabolic products being obtained by enzymatic conversion of 133 and 134 into products other than 138 and 139, were not characterized. Also, generally a poor recovery of starting material was observed. However, in one study utilizing CFE and hydrogen peroxide as the only co-factor, there was a suggestion that both 133 and 134 were being converted into potentially interesting compounds and it was decided to perform larger scale experiments under these conditions. 4'-Demethyldemethylenedeoxypodophyllotoxin (134) (205 mg) was dissolved in ethanol and water and treated with the CFE of Catharanthus roseus cell culture (1600 units per mmol) in the presence of hydrogen peroxide (2.8 equivalents). The reaction was quenched after 70 minutes and after an exhaustive extraction with dichloromethane and then with ethyl acetate a sample of the starting precursor 134 was isolated from the dichloromethane fraction (15%). After preparative TLC, more starting precursor 134 (2.6%) was recovered together with a small sample of impure material (3.5%) which was  confirmed as the hydroxylated product 139 by comparison of the 1H-NMR spectrum (see Appendix, Figure 21) with that of a sample of the authentic material prepared as in Scheme 49.  Scheme 48. The Proposed Hyroxylation of the Precursors 133 and 134  Hydroxylation had been achieved but the yield was very low although obviously other bio-transformation reactions had occurred, since therecoveryof starting material was only 18%. Clearly our reaction conditions require further refinement in order to achieve higher yields of 139. However, therecentlypublishedl02 microbial conversion of  112 deoxypodophyllotoxin (6) to podophyllotoxin (3) (quantitative yield) as shown in Scheme 50 may render the above study irrelevant to any commercial process. OH  OH  1) BC1 , -78°C 3  2) acetone/H 0, CaC0 2  3  CH 0  CH 0  3  3  OCH,  OCHo  138 OH  OH  ^ 0  3>  2) acetone/H 0, CaC0 2  OCH  3  CH3O  OCH,  CH30  HO  1) BC1 -78°C-^0°C  y  "0CH3  OH  A  139 Scheme 49. The Synthesis of 138 and 139 From Podophyllotoxin (3)  1  113  Scheme 50. The Microbial Hydroxylation of Deoxypodophyllotoxin (6) to Podophyllotoxin (3)  CHAPTER 3 INTRODUCTION  3.1  THE ISOLATION OF THE PODOPHYLLOTOXINS FROM PLANT C E L L CULTURES  Podophyllotoxins as mentioned previously, have been isolated from several plant species.  104  The advantages afforded by direct production of podophyllotoxins using plant  cell culture (Section 1.1), together with the possible isolation of new metabolites has led to an interest in the culturing of Podophyllum species. The initial work was carried out by Kadkade  103  in 1981, who developed an  undifferentiated callustissueculture from Podophyllum peltatum. The callus was initiated from the rhizomes of the plant and grown on Murashige and Skoog medium with 2, 4-D and casmino acids added. After 8 weeks the callus was extracted and podophyllotoxin (3), a-peltatin (8) and P-peltatin (7) were isolated. The yield of podophyllotoxin (3) was 0.64% by weight. Recent work by Dewick* has successfully used a callus culture of 04  Podophyllum hexandrum to produce podophyllotoxin (3) and 4'-dememylpodophyllotoxin (5) at levels similar to those they found in the explant material. However, no reports of a successful cell suspension have been made by Kadkade or Dewick.  The plant  Podophyllum hexandrum, is known to produce a higher percentage of podophyllotoxin (3) by dry weight than the other species. However, without data on cell suspensions the relationship between the concentration of compounds in a culture versus plants is impossible to evaluate. A paper by Dutch research workers in 1990 5 described the first isolation of 10  podophyllotoxin (3) from cell suspension cultures. The cell suspension culture was initiated from a callus culture of Podophyllum hexandrum. It was reported that the callus  115 was very difficult to initiate. Podophyllotoxin (3) was isolated and the yield was typically between 0.0 and 0.1 % based on dry weight A cell suspension culture of Linum flavumX^ has been established and the production of 5'-methoxypodophyllotoxin (140) (Figure 12) has been investigated. The lignan 140 was initially isolated in yields of up to 0.004% on a dry weight basis using a medium containing naphthaleneacetic acid. Naphthaleneacetic acid is a plant hormone. With the use of (L)-phenylalanine (52) as a precursor the yield was increased 3-5 fold. When the culture was transferred to a medium not containing naphthaleneacetic acid a 4050 fold increase in the accumulation of 5'-methoxypodophyllotoxin (140) was found. MeO  OH  O  OCH  3  (140) Figure 12. The Structure of 5'-Methoxypodophyllotoxin  Work in our department has led to the development of a cell suspension culture of Podophyllum peltatum. Initially explants of the leaf, stem and rhizome of Podophyllum peltatum were initiated on MSNa (Murashige-Skoog supplemented with naphthalene acetic 3  acid (3 mg/L) agar (10 g/L). Some success was achieved with the stem explants, but extremely heavy fungal contamination prevented development of all but three of several hundred explants. These three, designated as R l , R2 and R3, were not initially stable in suspension culture and were developed through a series of different media, using a  116  qualitative assessment of growth vigor. Finally, the media chosen for development of cell calli cultures were, MS (50% concentration, Nai.oKn^Caioo, naphthalene acetic acid 1.0 mg/L, kinetin 0.2 mg/L, casein hydrolysate 100 mg/L) and MS (50% concentration) for suspension cultures. In general, callus cultures were transferred every 6-8 weeks. The suspensions, initially grown in Erlenmeyer shake flasks, were mostly embryonic in nature, and were transferred approximately every 17 days. We have investigated this culture concerning the production of the podophyllotoxins and the results obtained so far are described in the next section. Subsequently, innoculum from the shake flasks could be transferred into Microferm bioreactors and cells successfully grown (5.5-15 L). These latter fermentations now provided sufficient material for characterization of the metabolites produced (Section 3.2).  RESULTS AND  3.2  DISCUSSION  T H E ISOLATION OF LIGNANS F R O M THE C E L L C U L T U R E OF  PODOPHYLLUM  PELTATUM  The growth of the Podophyllum peltatum cell culture is slow (17 days to reach the stationary phase, compared with 10 days for Catharanthus roseus) and yield of cell material low in comparison to other cultures available to our research group. However, the culture was of interest to our research for two reasons. Firstly, it could serve as a possible culture for the bio-transformation of precursors into intermediates useful in the production of podophyllotoxin (3) and etoposide ( 1 ) . Secondly, it could provide a potential direct source of the podophyllotoxin lignans which, in turn, can then be manipulated to optimize the yields of the desired precursors. The initial experiment involved the harvesting of 22 day old Podophyllum peltatum cell culture (4.0 L ) which was then freeze-dried. Cell material of other ages was also extracted using the same procedure. 107 This material was homogenized and extracted (Scheme 51). Initial analysis of all of the crude fractions shown in Scheme 51, by EI M S showed the presence of a peak at m/z 414 which could have represented the molecular ion peak of podophyllotoxin (3). However, a comparison of the chromatographic characteristics of the fractions, using T L C , revealed that a significant component was P-sitosterol  (141).  presence of this material was confirmed by high resolution EI M S which identified the peak with a m/z value of 414.3903 as having the molecular formula C29H50O, and subsequent comparison with an authentic sample. This sterol is a very common component in living plants and has been isolated from tissue cultures in our laboratory. A portion (239 mg) of the "neutral fraction" was treated with dilute hydrochloric acid and methanol at 40'C for 3 hours in an attempt to hydrolyze any lignans which may be  The  118 present as glycosidic compounds. After hydrolysis the mixture was extracted with chloroform.  SHAKE FLASK CULTURE  HOMOGENATE  I  ) Aqueous layer  CH C1 layer 2  2  Neutral fraction  Aqueous layer  EtOAc layer  EtOAc fraction  CH C1 layer 2  2  Aqueous layer  Acidic fraction  Scheme 51. The Extraction Procedure used for the Initial Experiments for the Isolation of Lignans from the Cell Culture of Podophyllum peltatum  Further analysis by HPLC of the hydfolyzed fraction suggested the presence of podophyllotoxin (3) in the culture and this was confirmed by high resolution EI MS which identified a peak at m/z 414.1336 with a molecular formula of C22H22O8. A sample (118 mg) of this hydrolyzed fraction was separated by. column chromatography using chloroform as the initial eluent. The analysis of the fractions by !H-NMR showed no protons to be present in the region 6 6.0-8.0 ppm and hence no aromatic protons were present, thus indicating that 3 was present in very low concentration so that it could be detected by EI MS but not by 1H-NMR. In a parallel study of a 41 day old culture by other co-workers who utilized the extraction procedure shown in Scheme 51, trace quantities of 3 (EI MS) were found in the "acidic" and the neutral "fraction" after hydrolysis. The detection of podophyllotoxin (3) in a fraction after hydrolysis may indicate that the podophyllotoxin (3) is present in the cells as a glycoside, a situation observed in other studies in our laboratory where metabolites , for example , indole alkaloids occur as glycosides. The above studies were encouraging although the low yields and the apparent instability of the cell culture required further experimentation in terms of developing the essential growth parameters for the cell line which would yield, in a reproducible manner, the desired products. After approximately 2 years of effort in the Biological Services Facility, in which numerous experiments were performed, a stable cell line of Podophyllum peltatum became available. At thistimewe investigated the nature of the metabolites produced in shake flasks and in Microferm bioreactors. For the initial studies, materialfromshake flasks (3.0 L) and a Microferm bioreactor (3.84 L) of 22 days age were combined and the extraction procedure in Scheme 52 was employed. The extracts were carefully exarnined by T L C .  1 0 8  The TLC was carried out with  plastic backed silica plates. After application of the samples and standards to the plate, the  120 plate was developed to height of 6 cm using a solvent rnixture of chloroform: methanol, 9: 1.  Microferm culture Shake flask culture  Broth Liquid-liquid extraction  Cells Homogenization  Freeze drying  Soxhlet extraction  Chromatography |  Chromatography  Scheme 52. The Extraction Procedure for the Isolation of the Podophyllotoxins from the Podophyllum peltatumCell Culture  The plates were then developed a second time, before visualization, with a solvent rnixture of chloroform ; acetone, 65 : 35. The results showed that compounds with the  same Rf value as the podophyllotoxin (3) was visible. No material was visible in the region on the TLC plates corresponding to the Rf values of the glycosides of the podophyllotoxins. The material extracted from the broth (103 mg) was separated using careful chromatography (Scheme 53).  The initial eluent used was a solvent mixture of  dichloromethane ; acetone, 4 : 1 .  From the chromatography of the broth extract  podophyllotoxin (3) (21 mg) was identified. For this study only materials showing the characteristic spectral features of the podophyllotoxin lignans were characterized. The 1H-NMR showed characteristic resonances at 8 3.70 and 3.82 ppm corresponding to the methoxy groups at positions 3', 5' and 4 respectively. The aromatic 1  region showed three sets of singlets; at 8 6.38 ppm due to the protons at positions 2' and 6', at 8 6.51 ppm to the proton at position 8 and, at 8 7.12 ppm attributed to the proton at position 5. Also the signal at 8 4.76 ppm indicated the proton at position 4 was in a trans orientation to the proton in position 3 as the peak was a doublet with a coupling constant of 8.9 Hz. The proton at position 3 appeared as a multiplet at 8 2.73-2.81 ppm. The peak at 8 2.83 ppm corresponds to the proton at position 2 and occurs as a doublet of doublets with coupling constants of 14 and 7 Hz. These two coupling constants are indicative of a trans and cis coupling to this proton which is expected from the stereochemistry of podophyllotoxin (3). The other peaks in the 1H-NMR spectrum matched, as did those mentioned, with the peaks found in the 1H-NMR spectrum of an authentic sample of podophyllotoxin (3). The EI MS, HPLC, TLC, and IR data was identical with the data from the authentic compound. The material extracted from the cells was also purified by column chromatography (Scheme 53), using the solvent system previously mentioned, and further podophyllotoxin (3) (27 mg, 0.058% dry weight) was isolated.  Broth extract (103 mg) Column chromatography Si0 CH C1 : acetone, 4:1 CH,OH 2  Fractions  2-5  6-7  8  9-11  12  2  2  13-19 20-22 23-25 26-36  Podophyllotoxin (3)  Cell extract (1.99 g) Column chromatography Si0 CH C1 : acetone, 4 :1 CH OH 2  2  2  3  Fractions  1-2  3-5  6-10  11-18 19-24 25-30 127 mg Column chromatography Si0 CH C1 : acetone, 4:1 CH C1 : acetone, 4 :1  Fractions  1-7  2  2  2  2  2  8-11 12-13 14-22 22-28  Podophyllotoxin (3)  Scheme 53. Chromatographic Separation of the Combined Shake Flask and Microferm Bioreactor Extracts  The combined yield of isolated podophyllotoxin (3) from the culture (48 mg) amounts to 7.0 mg/L of cell culture. Of this material 27 mg were isolated from freeze dried cell material (46.8 g) which corresponds, as previously mentioned to 0.058%. This value can be compared with the 0.3% literature value recorded for podophyllotoxin (3) content in plant cell culture by van Uden but it does not include the material which was extracted from the broth as it was not freeze-dried. If this is included with the weight of podophyllotoxin (3) then the yield is 0.1%. It should be further emphasized that no. studies to optimize the yield of 3 in the cell culture have been performed as yet, so that higher yields are certainly possible.  OCH  3  Podophyllotoxin (3)  As material from both Microferm bioreactor and shake flasks was used in the extraction procedure we could not determine from these results the source of the podophyllotoxin (3). Therefore, a large scale extraction was carried out on material collected from a single 5.5 L Microferm bioreactor. Podophyllum peltatum cell culture (5.5 L) was incubated in a Microferm bioreactor for 21 days after which the cells and broth were separated by filtration. The isolation procedure was different from the previous experiment, as the cell material was not freeze-  124 dried but was extracted directly and is shown in Scheme 54. This change in procedure was made because of concerns about the possible loss of extractable material caused by the freeze drying procedure. Column chromatography was used to separate the extracts (Scheme 55). Once again, apart from the isolation of ^-sitosterol (141) only compounds identified as podophyllotoxin lignans were identified. The extract from the culture broth was separated using a solvent mixture of dichloromethane: acetone, 4:1.  Microferm culture Filter  Broth Liquid-liquid extraction  Cells Homogenization withCH Cl 2  2  Liquid-liquid extraction  Chromatography  Chromatography  Scheme 54. TheModified Large Scale Extraction Procedure  Podophyllotoxin (3) was isolated (22 mg, 4.0 mg/L) and was identified as discussed previously by comparison of the spectral data with the data from a sample of authentic material, and by comparison using TLC. Broth extract (284 mg) Column chromatography Si0 CH C1 : acetone, 4: 1 acetone 2  Fractions 1-3  4-5  6-8  9-12  2  2  13-22 23-29  I  1  141  3  30  Cell extract (435 mg) Column chromatography Si0 CH C1 : acetone, 4: 1 2  2  2  CH3OH  Fractions  1-2  3-4  8  6 60  9-16  18-30  3  5  \  I  31  32  Scheme 55. Chromatographic Separation of the Extracts From the 5.5 L Microferm Bioreactor Podophyllotoxin (3) was also isolated from the cell material and in larger amounts than from the broth (127 mg 23.1 mg/L). This gave a total yield of podophyllotoxin (3)  126 from the Microferm bioreactor of 149 mg, which corresponds to a yield of 26.7 mg/L of cell culture. A comparison with the values previously discussed is difficult because the material was not freeze dried in the extraction procedure. However, in terms of mg/L of cell culture the yield from the bioreactor is approximately fourtimesthe previous value (extraction from shake flask material and bioreactor material). This may indicate a higher yield of podophyllotoxin (3) from the Microferm bioreactor of or a deficiency in the extraction procedure when freeze drying is performed. From the cell material three other Podophyllum lignans were isolated apart from podophyllotoxin (3). The compounds were isolated by column chromatography using the solvent system previously discussed, they were eluted in the following order; deoxypodophyllotoxin (6) and podophyllotoxone (60), followed by podophyllotoxin (3), followed by 4'-demethylpodophyllotoxin (5). A mixture of deoxypodophyllotoxin (6) and podophyllotoxone (60) was isolated and was inseparable by chromatography with silica. The two components of the mixture were, however, clearly identified by comparison of the 1H-NMR with the data from authentic compounds and by HPLC analysis which involved co-injection of the mixture with authentic samples. Podophyllotoxone (60) was previously synthesized in our laboratory from podophyllotoxin (3) by oxidation. The 1H-NMR of podophyllotoxone 98  (60) corresponded with the peaks assigned to the compound in the !H-NMR of the mixture. The 1H-NMR of deoxypodophyllotoxin (6) was also found to correspond very well with the material identified in the mixture. The EI MS of the mixture showed the molecular ion peaks of both compounds, with peaks at 398 and 412 corresponding to deoxypodophyllotoxin (6) and podophyllotoxone (60), respectively. The molecular formulae were confirmed by high resolution EI MS. Dewick had previously reported the isolation of these two compounds also as a mixture and had separated the components eventually using Sephadex chromatography.  127 Finally, a sample of 4'-demethylpodophyllotoxin (5) was isolated from the cell material. The yield of this important potential etoposide (1) intermediate was 36 mg (6.5 mg/L of cell culture). The compound was identified by analysis using 1H-NMR, EI MS and melting point. All data collected were consistent with the literature values. The *HNMR showed three characteristic singlet peaks. The two proton singlet at 8 6.34 ppm corresponds to the protons at position 2' and 6'. The one proton singlets at 8 6.42 and 7.04 ppm correspond to the protons at positions 8 and 5 respectively. The major difference between the iH-NMR spectra of podophyllotoxin (3) and 4'-demethylpodophyllotoxin (5) concerns the peaks attributed to the methoxy protons. Podophyllotoxin (3) as previously noted has two peaks corresponding to methoxy protons at 8 3.76 and 3.82 ppm, however the spectrum of 4'-demethylpodophyllotoxin (5) shows only one six proton singlet at 8 3.68 ppm indicating the presence of the methoxy groups at positions 3' and 5'. This is the first time that podophyllotoxin lignans have been isolated from Podophyllum peltatum cell cultures and only the secondtime,according to the literature, that podophyllotoxins have been isolated from cell suspension cultures. The only other example was from cultures of Podophyllum hexandrum and this was reported in 1990. The production of substantial quantities of 4'-demethylpodophyllotoxin (5) from our cell culture is of great interest as the compound is a very useful intermediate in the production of etoposide (1). The culture also produces reasonable quantities of podophyllotoxin (3) which at present is isolated commercially from whole plants and converted into etoposide (1).  In conclusion, it is clear that our developed cell culture of Podophyllum peltatum is not only an excellent potential source of the desired podophyllotoxins, but can also serve as a "medium" for the bio-transformation of appropriate synthetic precursors to the desired end products. The enzymes inherent in this cell culture should be able to convert such precursors to end products with the correct stereochemistry. Study in his direction are presently underway in our laboratory.  4'-Demethylpodophyllotoxin (5)  CHAPTER 4 EXPERIMENTAL  4.0  GENERAL INTRODUCTION  All 1H-NMR spectra were recorded on Bruker AC-200, Bruker WH-400 or Varian XL-300 spectrometers, with chemical shift values recorded in ppm relative to tetramethylsilane as an internal standard. C-NMR spectra were recorded on a Varian XL13  300 spectrometer at 75.3 MHz or a Brucker AC-200 spectrometer at 50.2 MHz. Infrared spectra were recorded on Perkin Elmer 1710,710 or 710B spectrophotometers, using thin film or Nujol mull. Polystyrene was used as a standard (1601 cm- ). Mass spectra were 1  obtained from AEI-MS-9 or KRATOS-MS-50 instruments for low or high resolution spectra respectively, using electron impact, chemical ionisation or fast atom bombardment methods. Melting points were recorded on a Reichert melting point apparatus and are all uncorrected. Column chromatography was carried out using silica gel (230-400 mesh). Thin layer chromatography was carried out using alurninium or plastic-backed silica gel plates, visualization was initially carried out by UV detection followed by treatment with a 5% ammonium molybdate in 10% aqueous sulfuric acid spray followed by heat treatment. Elemental analyses were performed using combustion analysis by Mr. P. Borda, Microanalytical Laboratory, University of British Columbia. Cell suspension culture of  Catharanthus roseus, Podophyllum peltatum and Tripterygium wilfordii were obtain from Biological Services of the Department of Chemistry, University of British Columbia. High-pressure liquid chromatography was performed using a Waters C-i8 "Radical Pak" liquid chromatography cartridge, a Waters 440 Absorbance Detector set at 280 nm and a methanol/water eluent. Anhydrous ether and tetrahydrofuran were prepared by distillation from sodium and benzophenone. Anhydrous dichloromethane and di-wo-propylamine  were prepared by distillation from calcuim hydride. Anhydrous methanol was prepared by distillation from magnesium. Boron trichloride was used as a dichloromethane solution preparedfromthe condensed gas. Hydrogen peroxide was prepared as a 147 mM solution or a 81.6 mM solutionfroma stock solution (30%w/v). The solutions were standardized by iodmetric titration with sodium thiosulphate.  4.1  SYNTHESIS OF PRECURSORS  4.1.1  3-HYDROXY-4-iso-PROPOXYBENZALDEHYDE (93) O  To a solution of 3,4-dihydroxybenzaldehyde (89) (20 g, 0.14 mol) in DMSO was added anhydrous potassium carbonate (11.0 g), zra-propyl iodide (8.0 g, distilled) and the solution was stirred in the dark at room temperature. The reaction was monitored by TLC and when complete (usually after several hours) was quenched by pouring into brine, followed by acidification (IM HCl). The mixture was extracted with dichloromethane (3 x 300 mL), dried with anhydrous sodium sulphate, filtered and evaporated in vacuo to yield the crude product Column chromatography was carried out using benzene : ethyl acetate : acetic acid, 90 : 5 : 5 as the initial eluent which was then graded to 60 : 20 : 20. The following compounds (including Sections 4.1.2 and 4.1.3) were isolated and characterized in order of elution.  131 The 4-wo-propyl ether (93) (9.81 g, 7 7 % ) as an oil. IR v  : 1705 cm-1. Hl  m a x  NMR: (200 MHz, C D C 1 ) 8: 1.40 (6 H, d, CH(CH3) , J = 6 Hz), 4.75 (1 H, heptet, 3  2  CE(CH ) , J = 6 Hz), 5.92 (1 H, br s, OH), 6.95 (1 H, d, H5, J = 8 Hz), 7.42 (2 H, m, 3  2  H 2 H6), 9.81 (1 H, s, CHO). EI MS m/z: 180 (M+), 137, 109. Exact mass calculated for C10H12O3:  4.1.2  180.0786; found: 180.0784.  4-HYDROXY-3-ISO-PROPOXYBENZALDEHYDE  (94)  O  The 4-hydroxy-3-z5o-propoxybenzaldehyde (94) was isolated as a minor constituent (0.56 g, 3.5%) from the chromatographic separation described above. IR Vmax:  1705 cm-1. lR-NMR: (200 MHz, CDCI3) 8: 1.38 (6 H, d, CH(CH3) , J = 6 Hz), 2  4.75 (1 H, heptet, CH(CH ) , J = 6 Hz), 6.35 (1 H, br s, OH), 7.05 (1 H, d, H5, J = 8 3  2  Hz), 7.42 (2 H, m, H2 H6), 9.83 (1 H, s, CHO). EI MS m/z: 180 (M+), 137, 109. Exact mass calculated for CioHi 03:180.0786; found: 180.0783. 2  4.1.3  3,4-DI-iso-PROPOXYBENZALDEHYDE  (95)  The 3,4-m-J5o-propoxybenzaldehyde (95) (0.77 g, 4.9%) was isolated as a minor product from the chromatographic separation described above. IR Vmax: 1705 cm-1. *H-NMR: (200 MHz, CDCI3) 8: 1.42 (12 H, m, CH(CH3) ), 4.55 (1 H, heptet, 2  CH(CH ) , J = 6 Hz), 4.65 (1 H, heptet, CH(CH ) , J = 6 Hz), 6.98 (1 H, d, H5, J = 8 3  2  3  2  Hz), 7.44 (2 H, m, H2 H6), 9.84 (1 H, s, CHO). EI MS m/z: 222 (M+), 180, 138. Exact mass calculated for Ci Hi803: 222.1256; found: 222.1258. 3  4.1.4  3 - B E N Z Y L O X Y - 4 - W O - P R O P O X Y B E N Z A L D E H Y D E (96) O  4-wo-Propoxy-3-hydroxybenzaldehyde (93) (9.8 g, 0.054 mmol) was dissolved in ethanol (50 mL). Potassium carbonate (8.85 g), sodium iodide (0.25 g) and benzyl chloride (8.5 g) were added and the mixture refluxed for 3 hours. Water (30 mL) was added and the ethanol evaporated in vacuo. The resulting solution was then extracted with dichloromethane (3 x 100 mL). Organic extracts were washed with NaOH (l.M) and brine before drying with anhydrous sodium sulphate, filtration and evaporation in vacuo.  133  Column chromatography with hexane : ethyl acetate, 85 : 15 gave the benzyl ether 96 (11.53 g, 78%) as a clear oil. IR v  m a x  : 1710 cm-1. lH-NMR: (200 MHz, CDC1 ) 8: 3  1.40 (6 H, d, CH(CH3)2, J = 6 Hz), 4.70 (1 H, heptet, CH.(CH ) , J = 6 Hz), 5.20 (2 H, 3  2  s, Cl&Ph), 7.0 (1 H, d, H5, J = 8 Hz), 7.25-7.50 (7 H, m, aromatic), 9.80 (1 H, s, CHO). EI MS m/z: 270 (M+), 242, 228, 179. Exact mass calculated for C17H18O3: 270.1256; found: 270.1259.  4.1.5  3-BENZYLOXY-l-BIS(PHENYLTHIO)METHYL-4-ta>PROPOXYBENZENE  (86)  2  The substituted benzaldehyde 96 (6.3 g, 0.023 mmol) was dissolved in dichloromethane (50 mL) and cooled to -40°C under a dry argon atmosphere. Thiophenol (5.0 mL) was added followed by dropwise addition of boron trifluoride-etherate (6.6 mL). After 30 minutes the reaction was quenched by pouring into ice water (50 mL). The mixture was extracted with dichloromethane (3 x 100 mL) and the organic phase was washed with KOH (7%), water and brine and dried with anhydrous potassium carbonate. Evaporation in vacuo gave the crude material which was then subjected to column chromatography using a mixture of hexane : diethyl ether, 9 : 1 as the eluent, to yield the dithioketal 86 as the product (5.1 g, 46%). IR v ax: 2895,1480 cm" . !H-NMR: (200 1  m  MHz, CDCI3) 8: 1.25 (6 H, d, CH(CH3) , J = 6 Hz), 4.40 (1 H, heptet, CH(CH ) , J = 6 2  3  2  Hz), 4.95 (2 H, s, CH^Ph), 5.30 (1 H, s, CH(SPh) ), 6.75 (2 H, m, H 2* H 6'), 6.90 (1 2  H, d, H 5', J = 2 Hz), 7.10-7.42 (15 H, m, aromatic). EI MS m/z: 363 (M+-SPh), 321.  134 Exact mass calculated for C29H28S2O2: 472.1531; found: 472.1523. Elemental analysis: calculated for C29H28S2O2: C (73.69), H (5.97); found: C (73.81), H (5.92).  4.1.6  3-((3 -BENZYLOXY-4 -wo-PROPOXY)-a,a,  ,  BIS(PHENYLTHIO)BENZYL)BUTANOLIDE (97) PhS  SPh  The dithioketal 86 (0.635 g, 1.34 mmol) was dissolved in dry THF (10 mL) and cooled to -78"C under an argon atmosphere. n-Butyl-Uthium (0.95 mL, 1.47 mmol, 1.1 equivalents) was added by syringe and the resulting yellow solution stirred at -78*C for 1 hour. The butenolide 87 (0.1 mL, 1.47 mmol, 1.1 equivalents) was then added in THF (5.0 mL) dropwise over a period of 30 minutes. The solution was stirred for a further 1 hour before being warmed to room temperature, then dilute HCl (1 M, 6.0 mL) was then added. The mixture was extracted with dichloromethane (3 x 50 mL). The organic extract was then washed with water (50 mL), saturated sodium bicarbonate solution (50 mL) and brine (50 mL) before being dried with anhydrous sodium sulphate, filtered and evaporated in vacuo. The crude mixture was then separated by column chromatography using hexane: ethyl acetate, 2 :1 as the eluent. The butanolide 97 was isolated in poor yield as an oil (90 mg, 15%). IR v  : 1770 cm" . H-NMR: (300 MHz, CDCI3) 8: 1.38. (6 H, d, 1  m a x  J  CH(CH3) , J = 6 Hz), 2.53 (1 H, dd, H(2), J = 12, 8 Hz), 2.73 (1 H, dd, H(2), J = 12, 8 2  Hz), 3.12 (1 H, m, H(3)), 4.27 (2 H, m, H(4)), 4.57 (1 H, heptet, CH(CH ) , J = 6 Hz), 3  2  135 5.02 (2 H , s, CH Ph), 6.82 (1 H, d, H(5'), J = 8 Hz), 7.05 (1 H, dd, H(6'), J = 8, 2 2  Hz), 7.18-7.60 (16 H, m, aromatic). EI MS m/z: 447 (M+- SPh), 405, 337.  4.1.7  rUAA S-2-(4 -BENZYLOXY-3",5"-DIMETHOXYBENZYL) ?  ,,  -3-((3^BENZYLOXY-4Wso-PROPOXY)-a,aBIS(PHENYLTHIO)BENZYL)BUTANOLlDE (98) PhS  SPh  The lactone 97 (200 mg, 0.36 mmol) was dissolved in dry THF (1.8 mL) and cooled to -78"C under an argon atmosphere. HMDS (0.095 mL, 0.45 mmol) was added slowly and the solution stirred for 40 minutes after which n-butyllithium (0.3 mL, 1.1 equivalents) was added. The solution was then stirred at -78°C for 2 hours. Then a solution of the bromide 88 (128 mg, 0.047 mmol) and HMPA (0.08 mL, 0.44 mmol) in dry THF (1.3 mL) was added. This solution was stirred for a further 4 hours at -78°C before it was allowed to warm to room temperature and was quenched with saturated ammonium chloride solution (5 mL). The mixture was then extracted with ethyl acetate (3 x 25 mL) and the organic phase washed with HCl (10%, 10 mL), water (10 mL), saturated sodium bicarbonate solution (10 mL) and finally brine (10 mL). The solution was dried with anhydrous sodium sulphate before being filtered and evaporated in vacuo.  Purification of the crude extract by column chromatography using petroleum ether: ethyl acetate, 9 :1 as the initial eluent, gave the dibenzylbutanolide 98 (166 mg, 62%). The full characterization of this material is given in section 4.1.8.  4.1.8  7 /?A^S-2-(4 -BENZYLOXY-3 ,5 -DIMETHOXYBENZYL),  , ,  M  , ,  S^O'-BENZYLOXY^'-wo-PROPOXYJ-a.aBIS(PHENYLTHIO)BENZYL)BUTANOLIDE PhS  (98)  SPh  n-Butyllithium (5.19 mL, 1.6 M in hexanes, 8.3 mmol) was added to a stirring solution, in THF (30 mL) of the dithioketal 86 (3.56 g, 7.5 mmol) under rigorous anhydrous conditions at -78°C. The solution was stirred for 25 minutes before the butenolide 87 (716 mg, 8.5 mmol, 1.3 equivalents) in THF (6.0 mL) was added. The mixture was then stirred for another 40 minutes before the bromide 88 (2.977 g, 10.5 mmol, 1.5 equivalents) was added in THF (7 mL). This solution was then stirred at -78°C for 1 hour before allowing it to warm to room temperature and stir for a further hour. Work up consisted of quenching with brine followed by extraction with ethyl acetate (3 x 100 mL). The organic extract was washed with brine and dried with anhydrous sodium sulphate before filtration and evaporation in vacuo. Column chromatography using hexane  137  : ethyl acetate, 5 : 1 as the eluent gave the butanolide product 98 (3.95 g, 65% yield. Mp 48-50-C. JR v  m a x  : 1770 cm-1. lH-NMR: (400 MHz, CDC1 ) 5: 1.38 (6 H, d, 3  CH(CH3) , J = 6 Hz), 2.25 (1 H, dd, H(7'), J = 14, 5 Hz), 2.90-2.96 (1 H, m, H(2)), 2  3.15 (1 H, dd, H(7'), J = 14, 4 Hz), 3.20-3.35 (1 H, m, H(3)), 3.39 (1 H, dd, H(4), J = 11, 8 Hz), 3.73 (6 H, s, OCH3), 4.27 (1 H, dd, H(4), J = 11, 3 Hz), 4.59 (1 H, heptet,  Ca(CH ) , J = 6 Hz), 5.03 (2 H , s, CH2), 5.05 (2 H, s, CH2PI1), 6.20 (2 H, s, H(2")), 3  2  6.83 (1 H, d, H(5'), J - 8 Hz), 7.05 (1 H, dd, H(6'), J = 8, 2 Hz), 7.15-7.52 (21 H, m, aromatic). EI MS m/z: 595 (M+-2 SPh), 505. Exact mass calculated for C43H43SO7 (M+ -SPh): 703.2738; found: 703.2739. Exact mass calculated for C37H39O7 (M+ -2 SPh): 595.2696; found: 595.2699. Elemental analysis: calculated for C48H48S2O7: C (71.97), H (6.04), S (8.00); found: C (72.18), H (5.95), S (7.83).  4.1.9  ^flAA'S-2-(3 ,5 -DIMETHOXYBENZYL-4 ,,  ,,  ,,  HYDROXYBENZYLJ-S-O'-HYDROXY^'-isoPROPOXYBENZYL)BUTANOLIDE (85) HO.  2'  7  4  OCH3  OH  The dithioketalbutanolide 98 (0.13 g, 0.17 mmol) was dissolved in ethanol (20 mL) and excess Raney-nickel was added. The resulting suspension was refluxed for 40 minutes and worked up by filtration and evaporation to yield the butanolide 85 as crude  138  product (61 mg). The pure product 85 (41 mg, 84%) was obtained by column chromatography using petroleum ether: ethyl acetate, 4:6 as the eluent. Mp 50-52'C. IR Vmax: 3540,1775 cm-1. ljj-NMR: (200 MHz, CDCI3) 5: 1.35 (6 H, d, CH(CH3) , J = 6 2  Hz), 2.45-2.68 (4 H, m, H(2), H(3), H(7")), 2.87 (1 H, dd, H(7'), J = 14, 5 Hz), 2.95 (1 H, dd, H(7'), J = 14, 6 Hz), 3.78-3.90 (7 H, m, H(4), -OCH3), 4.12 (1 H, dd, H(4), J = 8, 6 Hz), 4.51 (1 H, heptet, CH(CH )2, J = 6 Hz), 5.40 (1 H, s, OH), 5.65 (1 H, s, 3  OH), 6.37 (1 H, s, H(2")), 6.45 (1 H, dd, H(6'), J = 8.5, 2.0 Hz), 6.60 (1 H, d, H(2'), J = 2.0 Hz), 6.73 (1 H, d, H(5') J = 8.0 Hz). EI MS m/z: 416 (M+), 374, 167, 137, 123. Exact mass calculated for C23H28O7: 416.1835; found: 416.1829. Elemental analysis: calculated for C23H28O7.H2O: C (63.58), H (6.96); found: C (63.58), H (6.49).  4.1.10  r*AA'.S-2-(3 \5"-DIMETHOXY-4"-HYDROXYBENZYL)-3,  (3',4'-METHYLENEDIOXYBENZYL)BUTANOLIDE  (62)  The dithioketal 114 (200 mg, 0.289 mmol) was dissolved in ethanol (3 mL) and methanol (3 mL). A suspension of excess Raney-nickel in ethanol was added and the mixture stirred at room temperature. The reaction was monitored by TLC and stopped after 1 hour. The rnixture was cooled and filtered. The filtrate was washed with ethanol and the combined ethanol fractions evaporated in vacuo. The rnixture was purified by column chromatography using ethyl acetate : petroleum ether, 1 : 1, to give the  139  monohydroxydibenzylbutanolide 62 (80.2 mg, 70%). Mp 38-41°C. IR v  m a x  : 3550,  1775 cm-1. lH-NMR: (400 MHz, CDCI3) 8: 2.88 (4 H, m, H(2), H(3), H(7")), 2.88 (2 H, m, H(7')), 3.86 (6 H, s, -OCH3), 3.88 (1 H, dd, H(4), J = 9, 7 Hz), 4.12 (1 H, dd, H(4), J = 9, 7 Hz), 5.40 (1 H, s, OH), 5.94 (2 H, m, -OCH2O-), 6.39 (1 H, s, H(2")),  6.45 (1 H, dd, H(6'), J = 8.5, 2.0 Hz), 6.60 (1 H, d, H(2'), J = 2.0 Hz), 6.73 (1 H, d, H(5') J = 8.0 Hz). EI MS m/z: 386 (M+), 220, 184, 167, 135. Exact mass calculated for C21H22O7:  4.1.11  386.1365; found: 386.1368.  77?AA'S-2-(3 ,5 -DIMETHOXY-4"-HYDROXYBENZYL)-3,,  ,,  (3',4'-DIHYDROXYBENZYL)BUTANOLIDE  (64)  A solution of boron trichloride (1.0 M, 0.8 mL 0.8 mmol) in dichloromethane was cooled to -78°C in a dry argon atmosphere.  Over a period of 10 minutes  monohydroxydibenzylbutanolide 62 (80 mg, 0.2 mmol), in dichloromethane (3.0 mL), was added to the boron trichloride solution. The rnixture was stirred for 2.5 hours at 78"C. The reaction was then quenched by pouring the mixture into a saturated solution of potassium hydrogen carbonate (5 mL) and ice (5 g). This mixture was left to stir at room temperature for 30 minutes. The mixture was then extracted with ethyl acetate (3 x 25 mL) and the combined organic extracts were washed with brine (25 mL) and dried with anhydrous magnesium sulphate before filtration and evaporation in vacuo. This residue  140 was then refluxed with calcium carbonate (150 mg), acetone (5 mL) and water (5 mL) for 2 hours. The mixture was then cooled and carefully acidified with HCl (6 M). The acetone was then evaporated in vacuo and the residue extracted with ethyl acetate (3 x 25 mL). The combined organic extracts were washed with brine (25 mL) and then dried over anhydrous magnesium sulphate before filtration and evaporation in vacuo. The product was purified by column chromatography using acetone : dichloromethane, 1 : 5 to give the trihydroxydibenzylbutanolide 64 (55 mg, 73%). Mp 71-74'C. IR v  m a x  : 3550,1780 cm"  1. 1H-NMR: (400 MHz, acetone d6) 8: 2.48 -2.97 (6 H, m, H(2), H(3), H(7"), H(7')), 3.80 (6 H, s, -OCH ), 3.88 -4.05(2 H, m, H(4)), 6.48 (1 H, dd, H(6'), J = 8, 2 Hz), 3  6.51 (2 H, s, H(2", 6")), 6.64 (1 H, d, H(2'), J = 2.0 Hz), 6.73 (1 H, d, H(5*), J = 8.0 Hz), 7.05 (1 H, br s, OH), 7.64, (2 H, br s, OH). EI MS m/z: 374 (M+), 167, 123.  4.1.12  DEMETHYLENEDEOXYPODOPHYLLOTOXIN  (133)  O  OCH3  A solution of boron trichloride was prepared by bubbling gaseous boron trichloride through dichloromethane at 0°C under nitrogen. The amount of boron trichloride present was determined by weight. Deoxypodophyllotoxin (6) (1.1 g, 2.7 mmol) was dissolved in dichloromethane (5.0 mL), and this solution was then added dropwise to a solution of boron trichloride (1.25 g, 11 mmol) in dichloromethane (20 mL) at -78"C. The mixture  was then stirred at -78°C under an argon atmosphere for 2 hours. After thistimethe mixture was quenched with saturated potassium hydrogen carbonate solution (20 mL) and ice (20 g). The suspension formed was stirred at room temperature for 30 minutes. The mixture was then extracted with ethyl acetate (3 x 100 mL) and the organic extract washed with brine (23 mL) before being dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo. This crude product was then refluxed with acetone (IS mL), water (IS mL) and calcium carbonate (1.2S g) for 2 hours. The suspension was then cooled and quenched with HCl (1 M). The rnixture was then extracted with ethyl acetate (3 x 100 mL) and the organic extracts washed with brine (50 mL) before being dried with anhydrous magnesium sulphate, filtered and evaporated. The product was purified by column chromatography  using  acetone  : dichloromethane,  1  : S,  demethylenedeoxpodophyllotoxin (133) (0.71 g, 67%). Mp 238-240°C. IR v  to m a x  give  : 3370,  3320, 1730 cm-1. lH-NMR: (400 MHz, acetone-d6) 8: 2.77 (3 H, m, H(4), H(4), H(3)), 3.08 (1 H, m, H(3)), 3.69 (3 H, s, 4'-OCH ), 3.72 (6 H, s, 3', 5'-OCH ), 3.96 (1 H, 3  3  dd, H(ll), J = 8, 10 Hz), 4.42 (1 H, dd, H(ll), J = 8, 6 Hz), 4.51 (1 H, d, H(l), J = 4 Hz), 6.42 (2 H, s, H(2')), 6.51 (1 H, s, H(8)), 6.73 (1 H, s, H(5)), 7.73 (1 H, s, OH), 7.79 (1 H, s, OH). EI MS m/z: 386 (M+), 371, 181, 168.  4.2  THE PROPAGATION OF THE CELL CULTURES AND THE PREPARATION OF CELL FREE EXTRACTS  4.2.1  PROPAGATION OF T H E CATHARANTHUS  ROSEUS  PLANT C E L L CULTURE  Catharanthus roseus cell suspension culture is propagated using 1-B5 medium (see Appendix Table 31) with an inoculum volume of 10% and a period between sub-cdturing of 10 days.  4.2.2  PREPARATION OF THE CELL FREE EXTRACT FROM PLANT C E L L CULTURES OF CATHARANTHUS  ROSEUS.  Plant cells were harvested using the following general procedure from shake flasks at the required age. The cell age chosen was 11-13 days. The suspension was filtered through Miracloth, which is a coarse cloth which retains only cells, and then washed with cold distilled water (75 mL per 500 mL of culture) before the cells were quickly cooled in an ice-water bath. The wet weight of the cells was recorded. During the remainder of the process care was taken to keep the cells between 0-4°C. A portion of the spent medium was kept for pH measurement, Rl measurement and to check for microbial contamination. Potassium phosphate buffer (0.1 M, pH 6.3, 0.50 mL/g wet weight of cells) was then added to the cells. The cells were then homogenized using the Ultraturax T-25 disperser at 20,000 rpm for four 30 second periods with a 30 second rest period between each homogenization. The homogenate was then centrifuged, using a Sorvall RC-5B centrifuge and Sorvall GSA rotor, at 10,000 G for 30 rninutes. The supernatant was decanted from the cell debris. This supernatant was defined as the cell free extract (CFE) and was kept at 0-4°C until used in bio-transformations. Samples of the CFE were analyzed for peroxidase  143 activity and protein concentration using a Bausch and Lomb Spectronic 20 spectrophotometer.  4.2.3  PROPAGATION OF T H E PODOPHYLLUM  PELTATUM  PLANT C E L L CULTURE.  Initially explants of the leaf, stem and rhizome of Podophyllum peltatum were initiated on MSNa3 (Murashige-Skoog supplemented with naphthalene acetic acid (3 mg/L) agar (10 g/L).  Some success was achieved with the stem explants, but extremely heavy fungal  contamination prevented development of all but three of several hundred explants. These three, designated as R l , R2 and R3, were not initially stable in suspension culture and were developed through a series of different media, using a qualitative assessment of growth vigor. Finally, the media chosen for development of cell calli cultures were, MS (50% concentration, Nai.oKo.2Caioo, naphthalene acetic acid 1.0 mg/L, kinetin 0.2 mg/L, casein hydrolysate 100 mg/L) and MS (50% concentration) for suspension cultures. In general, callus cultures were transferred every 6-8 weeks. The suspensions, initially grown in Erlenmeyer shake flasks, c  were mostly embryonic in nature, and were transferred approximately every 17 days. Subsequently, innoculum from the shake flasks could be transferred into Microferm bioreactors and cells successfully grown (5.5-15 L). Thus at its present stage Podophyllum peltatum cell culture is propagated with MS medium (50% strength) with 1.5% sucrose added (see Appendix Table 30). The inoculum volume is 15 mL of harvested aggregates and embryos per 250 mL of medium. The period between sub-cultures is 17 days.  4.2.4  PREPARATION OF THE CELL FREE EXTRACT FROM PLANT CELL CULTURES OF PODOPHYLLUM  PELTATUM  The CFE of Podophyllum peltatum is prepared as in the procedure for the preparation of CFE from Catharanthus roseus cell culture with one difference; the cell suspension is first homogenized with a Waring blender before the use of the Ultra-turrax T25. This is necessary because of the physical strength of the cell material.  4.2.5  PROPAGATION OF THE TRIPTERYGIUM WILFORDII PLANT C E L L CULTURE  Tripterygium wilfordii cell culture is propagated with PRD2 C0100 medium (see 108  Appendix Table 32). The inoculum volume is 10% and the period between sub-cultures is 14-17 days.  145  4.3  A T T E M P T E D BIO-TRANSFORMATION O F D I B E N Z Y L B U T A N O L I D E S (62, 63, 64) USING T H E W H O L E C E L L SUSPENSION C U L T U R E O F  CATHARANTHUS  ROSEUS  Catharanthus roseus  W h o l e cells  OH  62 R,R' = - C H 63 R ' = C H , R = H 64 R,R' = H 2  3  68 R,R' = - C H 69 R = C H , R ' = H 70 R, R' = H 2  3  Catharanthus roseus cells were propagated using the procedure outlined (Section 4.2.1).  The three dibenzylbutanolide precursors 62, 63, 64 (300 mg each) were  dissolved separately in ethanol (10 mL). For each precursor 6 x 500 mL flasks of seven day old culture were used for the bio-transformation. The precursors were added under sterile conditions. Two control flasks were also used, one containing only cells and the second containing cells plus ethanol (10 mL). The flasks were left on shaker beds at room temperature and samples were removed (2 mL) for each precursor, plus the controls, after 24 and 48 hours for HPLC analysis. The precursor 64 was left for 72 hours. After 48 or 72 hours the flasks were taken from the shaker beds and the cells filtered from the broth using Miracloth. The broth was saturated with NaCl (20 g per 100 mL of broth) and extracted with dichloromethane (3 x 500 mL). The aqueous layer was then acidified (2 M HCl) and extracted with dichloromethane (3 x 500 mL). The organic extracts were dried  146 with anhydrous magnesium sulphate, filtered and evaporated. The aqueous layer was then extracted with ethyl acetate (3 x 300 mL). The combined ethyl acetate extracts were then dried with anhydrous magnesium sulphate, filtered and evaporated. The cells were homogenized, using an Ultra-Turrax T-25 disperser in dichloromethane (200 mL) and then filtered. The extract was washed with HCl (0.5 M, 100 mL). This extract was then dried with anhydrous magnesium sulphate, filtered and evaporated. Theremainingcell material was then acidified with HCl (0.5 M, 300 mL) and extracted with dichloromethane (3 x 500 mL). This extract was then washed with HCl (0.5 M, 100 mL) and the organic phase dried with anhydrous magnesium sulphate, filtered and evaporated. HPLC ANALYSIS. Methanol (1.0 mL) was added to the samples which were then shaken and filtered through a cotton wool plug before analysis by HPLC. Results: 62- Some precursorremainedbut a large portion was converted to more polar product. 63- Some precursorremainedbut some conversion had occurred to more polar products. 64- During the 72 hours the peak for (64) decreased and several new peaks appeared both more and less polar than the precursor.  Extraction weights for 64; Broth  Cells  dichloromethane neutral.  91.5 mg  dichloromethane acidic.  50.5 mg  ethyl acetate  31.2 mg  dichloromethane neutral  106.1 mg  dichloromethane acidic  204.9 mg  Total Recovery  484.2 mg  147 Column chromatography was carried out on 228 mg of the combined extract using chloroform as eluent Total recovery was 181.2 mg. These fractions were analyzed by IR and the major fractions by NMR and MS. One compound was identified 7 1 ; IR v x : ma  3540, 1768 cm-1, lR-NMR :(300 MHz, CDC1 ) S: 2.40-2.70 (4 H, m, H(2), H(3), 3  H(7")), 2.85-2.95 (2 H, m, H(7')), 3.82 (3 H, s, OCH3), 3.86 (6 H, s, OCH3), 3.883.95 (1 H, dd, H(4), J = 6, 11 Hz), 5.42 (1 H , br s, OH), 5.54 (1 H, br s, OH), 6.33 (2 H, s, H(2"), H(6")), 6.42 (1 H, d, H(2'), J = 3 Hz), 6.53 (1 H, dd, H6', J = 3, 6 Hz), 6.82 (IH, d, H(5'), J = 6 Hz). EI MS m/z: 388 (M+) 167, 137. Exact mass calculated for C21H24O7:  4.4  388.1522; found: 388.1524.  BIO-TRANSFORMATIONS USING HORSERADISH PEROXIDASE (HRP)  4.4.1  INITIAL SMALL SCALE EXPERIMENT  OCH3 OH  63  OH  69  The precursor 63 (150 mg, 0.39 mmol) was dissolved in a mixture of ethanol (15 mL) and water (135 mL) to give a concentration of 1 mg/mL. HRP (1 mg) was dissolved  148 in phosphate buffer (50 mL, 0.1 M, pH 6.3), (1 mg = 300 units) to give a concentration of 6 units/mL. This value gave 415 units of peroxidase per mmol of precursor. Hydrogen peroxide was used as a co-factor (1.1 equivalents). Flasks were set up stirring at room temperature for the times indicated (see Table 14) and samples (1.0 mL) were removed for HPLC analysis. Aliquots (40 mL) were removed from the reaction flask at 30 and 60 minutes.  TABLE 14 Conditions for the Bio-transformation of 63 Using HRP Reaction flask  Control  Control  No enzyme  No hydrogen peroxide  Solution of 63  90 mL  30 mL  30 mL  HRP solution  27 mL  9 mL (buffer only)  9mL  3 mL  1 mL  1 mL  Hydrogen peroxide  (water only)  After final incubation (90 minutes) the reaction was quenched with dichloromethane (40 mL) and the contents of each flask extracted with dichloromethane (3 x 40 mL). The aqueous layer was then extracted with ethyl acetate (3 x 40 mL), followed by acidification with HCl (1 M) and extraction with dichloromethane (3 x 40 mL). Finally, the aqueous layer was extracted a secondtimewith ethyl acetate, the organic extracts were dried with anhydrous magnesium sulphate, filtered and evaporated. After extraction, the fractions for the 60 minute and 90 minute aliquots were combined. The crude extracts were analyzed by MS and NMR.  149 30 minute sample. Dichloromethane EI MS m/z: 386 (M+,69), 167, 154, 137. Ethyl acetate EI MS m/z: 386 (M+, 69), 169,154. 60 minute plus 90 minute sample. Dichloromethane EI MS m/z: 386 (M , 69), +  167, 154, 137. Exact mass calculated for 69 C21H22Q7: 386.1366; found: 386.1363. Ethyl acetate EI MS m/z 386 (M+, 69), 186,149. IR v : m!a  (thin film): 3400,2925,1775  cm . -1  Control without enzyme sample. Dichloromethane EI MS m/z: 388 (M , 63), 167, +  137. Control without hydrogen peroxide sample. Dichloromethane EI MS m/z: 386 (M+, 63), 167, 137. The combined dichloromethane fractions after 60 minutes and 90 minutes (48 mg) were purified using column chromatography with ethyl acetate as eluent Compound 69 (2.9 mg) was identified. iH-NMR (400 MHz, CDCI3) 8: 2.50 (1 H, dd, H(2), J = 14,10 Hz), 2.60 (1 H, m, H(3)), 2.94 (1 H, br dd, H(4), J = 15, 14 Hz), 2.98 (1 H, dd, H(4), J = 15.5, 4.5 Hz), 3.65 (3 H, s, -OCH (7)), 3.86, (6 H, s, -OCH (3')), 3.98 (1 H, dd 3  3  H(ll), J = 11, 7 Hz), 4.10 (1 H, d, H(l), J = 11 Hz), 4.53 (1 H, dd, H(ll), J = 11, 6 Hz), 5.47 (1 H, s, OH), 5.54 (1 H, s, OH), 6.34 (1 H, s, aromatic), 6.47 (2 H, s, H(2')), 6.73 (1H, s, aromatic). Complete characterization was carried out in Section 4.4.2.  4.4.2  BIO-TRANSFORMATION OF 63 WITH HRP.  LARGE  SCALE, EXPERIMENT 1  Two flasks were prepared each containing 63 (150 mg, 0.39 mmol) dissolved in ethanol (15 mL) and water (135 mL)), hydrogen peroxide (5 mL, 1.9 equivalents) and HRP (45 mL of phosphate buffer 0.1 M, pH 6.3, containing 0.9 mg of enzyme which equals 690 units/mmol of precursor). Flask A was incubated for 15 minutes at room  temperature and Flask B for 60 minutes at room temperature. Samples (1.0 mL) were removed for HPLC analysis. The flasks were quenched by addition of dichloromethane (200 mL). Extraction was carried out with dichloromethane (3 x 200 mL) followed by ethyl acetate (3 x 200 mL). The organic extracts were then dried with anhydrous magnesium sulphate and filtered before evaporation in vacuo Table 15.  TABLE 15. Recovery From the Bio-transformation of 63 with HRP RECOVERY  FLASK A  FLASK B  Dichloromethane  150 mg  131 mg  Ethyl acetate  5 mg  19 mg  TOTAL  155 mg  150 mg  The dichloromethane fraction from Flask A was separated by column chromatography using dichloromethane: ethyl acetate 4:1 gradient to 1: 1 as eluent The compound 69 (28.8 mg 19%) was identified. Mp 238-240°C (chloroform). IR v  mSK  :  3400, 1775 cm-1. lH-NMR (400 MHz, CDC1 ) 6: 2.50 (1 H, dd, H(2), J = 14, 10 Hz), 3  2.60 (1 H, m, H(3)), 2.94 (1 H, br dd, H(4), J = 15, 14 Hz), 2.98 (1 H, dd, H(4), J = 15.5, 4.5 Hz), 3.65 (3 H, s, -OCH (7)), 3.86 (6 H, s, -OCH3(3)) 3.98 (1 H, dd H(ll), ,  3  )  J = 11, 7 Hz), 4.10 (1 H, d, H(l), J = 11 Hz), 4.53 (1 H, dd, H(ll), J = 11, 6 Hz), 5.47 (1 H, s, OH), 5.54 (1 H, s, OH), 6.34 (1 H, s, aromatic), 6.47 (2 H, s, H(2*)), 6.73 (1 H, s, aromatic). EI MS m/z: 386 (M , 69), 167, 154, 137. Exact mass calculated for +  C2iH2207:386.1366; found: 386.1363. Elemental analysis: calculated for C21H22O7: C (65.28), H (5.74); found: C (65.12), H (5.72).  151 4.4.3  BIO-TRANSFORMATION OF 63 WITH HRP. LARGE SCALE EXPERIMENT 2  The precursor 63 (600 mg, 1.5 mmol) was dissolved in ethanol (30 mL) and water (570 mL). Hydrogen peroxide (40 mL, 3.9 equivalents) was added together with HRP (180 mL of phosphate buffer (0.1 M) at pH 6.3 containing 3.6 mg of enzyme which equals 720 units/mmol of precursor) The mixture was split into two portions and stirred at room temperature in two 1 L erlenmeyer flasks. After 15 minutes the reaction was stopped by addition of dichloromethane (600 mL). Extraction was carried out with dichloromethane (3 x 500 mL) followed by ethyl acetate (3 x 500 mL). The organic extracts were dried with anhydrous magnesium sulphate and filtered before evaporation in vacuo. The dichloromethane extract weighed 553 mg and the ethyl acetate extract 17 mg to give a total recovery of 570 mg (96.5%). The dichloromethane extract was purified by column chromatography using chloroform as initial eluent, and chloroform with 50% ethyl acetate as the final eluent. The product 69 (91.4 mg, 15%) was isolated. Some impure product was also isolated (15.5 mg). The spectral data was consistent with an authentic sample.  4.5  THE EVALUATION OF pH DEPENDENCY ON THE RING CLOSURE OF 63 BY THE CFE OF  CATHARANTHUS  ROSEUS  Catharanthus roseus cell culture (age 11 days) was harvested and filtered through Miracloth. The cells were washed with water (600 mL) and then split into five portions (Table 16). To each portion was added phosphate buffer and the cells were homogenized using the standard procedure (Section 4.2.2).  152  T A B L E 16. Biotransformations of 63 Using Catharanthus roseus at Different pH Values  pH  Volume (mL) of buffer added  Flask 1  5.95  50  Flask 2  6.49  75  Flask 3  7.15  50  Flask 4  7.78  75  Flask 5  8.17  75  The homogenates were then centrifuged for 30 minutes at 10,000 G. For each CFE the protein concentration and peroxidase activity was measured and the specific activity calculated. The units of active protein were calculated for 40 mL of CFE (Table 17).  153 TABLE 17. Peroxidase Activity of the CFE of Catharanthus roseus Prepared at Different pH Values Flask #  1  2  3  4  5  pH  5.95  6.49  7.15  7.78  8.17  1.26  1.38  1.89  2.3  1.54  Peroxidase activity, units 1.7  2.1  1.95  2.26  2.8  Specific activity  1.35  1.53  1.03  0.98  1.82  68  84  78  90  112  Protein concentration mg/mL  Peroxidase uints/40 mL CFE  The percursor 63 (100 mg, 0.26 mmol) was dissolved in ethanol (10 mL) and water (17 mL). Hydrogen peroxide (1.0 mL, 1.8 equivalents) was added followed by the CFE and the mixture was stirred at room temperature for 160 minutes. Samples were removed at 5,10, 20, 30, 60,90, and 160 minutes for HPLC analysis. pH 5.95. No precursor remained after 5 minutes. 69 formed plus unknown. pH 6.49. No precursor remained after 5 minutes. 69 formed plus unknown. pH 7.15. 63 :69 1:3 after 10 minutes. Trace of 63 remains at 160 minutes pH 7.78. 63 :69 1:2 after 5 minutes and remained at that ratio. pH 8.17.63 :69 2:1 after 5 minutes and remained at that ratio.  4.6  ATTEMPTED RING CLOSURE OF 62 USING THE CFE OF  CATHARANTHUSROSEUS  62  79  Catharathus roseus cells, age 11 days were used to prepare CFE (section 4.2.2) using 0.1 M phosphate buffer (pH 7.35; peroxidase activity 1.32 units). Two flasks were set up as outlined below using 396 units of peroxidase per mmol of precursor and 3.7 equivalents of hydrogen peroxide (Table 18).  TABLE 18 The Bio-transformation of 62 FLASK 1  FLASK 2  PRECURSOR  CONTROL  62 (384 mg, 1.0 mmol) Ethanol (40 mL)  Ethanol (4.0 mL)  Water (80 mL)  Water (8.0 mL)  H 02(25mL)  H O2(3.0mL)  CFE (300 mL)  CFE (30 mL)  2  2  155 Samples (1.0 mL) were taken for HPLC analysis and the reaction was quenched with dichloromethane (Flask 1,500 mL; Flask 2,50 mL) after one hour. Flask 1 was then extracted, after addition of celite and filtration, with dichloromethane (2 x 500 mL) followed by extraction with ethyl acetate (3 x 500 mL). The aqueous layer was then acidified with HCl (1 M) and extracted with ethyl acetate (3 x 500 mL). Flask 2 was extracted using the same procedure as for Flask 1. All organic extracts were then dried with anhydrous magnesium sulphate,filteredand evaporated in vacuo (Table 19).  TABLE 19 Extraction from the bio-transformation of 62 Dichloromethane extract  368 mg  96%  Ethyl acetate extract  29 mg  8%  Ethyl acetate./ acidic extract  96 mg  25%  Total  493 mg  The dichloromethane extract was separated by column chromatography using 1:1 ethyl acetate, petroleum ether as eluent. The starting precursor 62 (235 mg, 61%) was isolated and identified by comparison with the authentic starting material. The unsaturated compound 79 (36 mg, 9%) was also isolated. IR v  m a x  : 3565, 1745 cm-1. lH-NMR  :(300 MHz, CDC1 ) 8: 2.63 (1 H, dd, H(7'), J = 8, 12 Hz), 3.25 (1 H, dd, H(7'), J = 3, 3  13 Hz), 3.77-3.86 (1 H, m, H(3)), 3.90 (3 H, s, -OCH3), 4.20-4.32 (2 H, m, H(4)), 5.85-5.97 (3 H, m, -OCH 0-, OH), 6.56-6.65 (2 H, m, H(2'), H(6')), 6.70 (1 H, d, 2  H(5'), J = 7, Hz), 6.78 (2 H, s, H(2")), 7.48 (1 H, d, H(7"), J = 2 Hz). 13C-NMR (75.3 MHz, CDC1 )8: 37.579, 39.511, 56.436, 69.636, 101.094, 107.259, 108.444, 3  108.981, 121.816, 125.398, 125.555, 131.378, 136.984, 137.928, 146.578, 147.226, 147.985, 172.985. EI MS m/z: 384 (M+), 249, 135. Exact mass calculated for  C21H20O7: 384.1209; found: 384.1220. Other fractions contain mixtures of 62 and 79 or unidentified material.  4.7  A T T E M P T E D BIO-TRANSFORMATION O F 63 WITH C F E F R O M PODOPHYLLUM  PELTATUM  The Podophyllum peltatum cell culture (propagated as described in section 4.2.3) was harvested after 29 days (4.7 L , pH 5.71) and the cells collected by filtration through cheesecloth and Miracloth (wet weight 540 g). The cells were mixed with Polyclar AT (50 g) and phosphate buffer (0.1 M , pH 6.3,500 mL) and homogenized with the Ultra-Turrax T-25 disperser whilst being cooled in an ice bath. The homogenate was then centrifuged (10,000 G, 30 minutes) and the CFE (600 mL) monitored for protein concentration and peroxidase activity. Spent medium CFE  0.19 mg/mL protein 2.00 mg/mL protein  specific activity 0.16.  The precursor 63 (150 mg, 0.39 mmol) was dissolved in ethanol (15 mL) and water (135 mL). To this solution was added hydrogen peroxide (10 mL, 3.7 equivalents) and the prepared CFE (45 mL). The mixture was stirred at room temperature and samples (1.0 mL) were removed for HPLC analysis. After 10 minutes more hydrogen peroxide (10 mL) was added. The reaction was quenched after 30 minutes by the addition of dichloromethane (200 mL). HPLC profiles showed no disappearance of precursor and no formation of new peaks over the 30 rninute reaction period. Extraction was carried out with dichloromethane (3 x 200 mL), followed by ethyl acetate (3 x 200 mL). A control flask containing only CFE (300 mL) and hydrogen peroxide was also quenched after 30 minutes and extracted in the same way. All organic extracts were dried with anhydrous magnesium sulphate, filtered and evaporated.  157  Recovery:-  sample flask, dichloromethane fraction weight:-  237 mg  ethyl acetate fraction weight:control flask, dichloromethane fraction weight:ethyl acetate fraction weight:-  22 mg 42 mg 91 mg  The fractions were analyzed by 400 MHz 1H-NMR and HRMS The sample flask's dichloromethane fraction gave a product with an NMR identical with that of authentic starting material 63 with some unidentified impurity peaks. HRMS: exact mass calculated for C21H24O7:388.1522; found: 388.1523. No change was seen in the control experiment by HPLC and TLC analysis,  4.8  BIO-TRANSFORMATION OF 77fAA'S-2-(3 ,5"-DIMETHOXYM  4"-HYDROXYBENZYL)-3-(3 -HYDROXY-4 -wo,  ,  PROPOXYBENZYL)BUTANOLIDE (85) WITH THE CFE OF CATHARANTHUS  85  ROSEUS  115  4.8.1  EXPERIMENT 1 SMALL SCALE  The dibenzylbutanolide 85 (26 mg 0.0628 mmol) was dissolved in a mixture of ethanol and water (8:17) (4.8 mL). Hydrogen peroxide-(1.53 mL, 0.125 mmol, 2.0 equivalents) and water (6.4 mL) were added to the substrate before the CFE (10 mL, peroxide activity 1.62,256 units per mmol of precursor) together with buffer (phosphate 0.1 M, 19 mL) was added at zero time. Samples (1.0 mL) were removed for HPLC analysis and the reaction was stirred at room temperature for 30 minutes. A control reaction was incubated containing appropiate amounts of buffer, CFE, ethanol, water and hydrogen peroxide. The pH of the reaction flask was monitored during the reaction. The pH remained at 6.38-6.40. The reaction flask and control flask were quenched after 30 minutes by the addition of dichloromethane (50 mL). The mixtures were shaken with celite and then filtered (Buchner filtration). The filtrates were extracted with dichloromethane (3 x 50 mL) and then ethyl acetate (3 x 50 mL). The celite was placed in an ultrasonicator with dichloromethane (100 mL) for 30 minutes, then filtered and combined with the other dichloromethane fractions. The organic fractions were dried with anhydrous magnesium sulphate, filtered and evaporated. The dichloromethane fraction yielded crude material (26 mg) which was purified by column chromatography using petroleum ether: ethyl acetate, 1 : 1,. The product 115 (14 mg, 54%, 74% based on recovered starting precursor) was isolated. Mp 180-183'C. IR v  3540, 1775 cm-1. lR-NMR (300 MHz, CDC1 ). 8:  m a x  3  1.15 (3 H, d, CH(CH3)2, J = 6 Hz), 1.24 (3 H, d, CH(CH3) , J = 5 Hz), 2.47 (1 H, dd, 2  H(2), J = 11.5, 13.5 Hz), 2.53-2.70 (1 H, m, H(3)), 2.88 (1 H, br dd, H(4), J = 11, 15 Hz), 2.98 (1 H, dd, H(4), J = 4.5,15 Hz), 3.84 (6 H, s,  -OCH3), 3.98  (1 H, dd, H(ll),  J = 14, 9 Hz), 4.06 (1 H, d, H(l), J = 11.5 Hz), 4.26 (1 H, heptet, Ca(CH ) , J = 6 3  2  Hz), 4.52 (1 H, dd, H(ll), J = 9, 6 Hz), 5.45 (1 H , s, OH), 5.63 (1 H, s, OH), 6.34 (1 H, s, aromatic), 6.45 (2 H, s, H(2')), 6.72 (IH, s, aromatic). EI MS m/z: 414 (M+), 372, 154. Exact mass calculated for C 3H 607: 414.1678; found: 414.1685. The starting 2  2  159 material 85 (7 mg) was also recovered and characterized by comparison with authentic material.  4.8.2  EXPERIMENT 2 LARGE SCALE  The dibenzylbutanolide 85 (148 mg, 0.36) was dissolved in ethanol/water (8 : 17) (2.7 mL) and to this solution was added hydrogen peroxide (8.7 mL, 2 equivalents), water (27.4 mL), phosphate buffer (0.1 M, pH 6.3,97.5 mL) and, at zero time the prepared CFE (65 mL, 11 days old, 250 units per mmol of precursor) was added. The control flask contained ethanol and water (2.7 mL, 8:17), hydrogen peroxide (0.87 mL), phosphate buffer (0.1 M phosphate pH 6.3, 9.75 mL) and CFE (6.5 mL). Samples were taken at intervals for HPLC (1.0 mL) and the reaction was stopped after 30 minutes by the addition of dichloromethane (225 mL). The mixture was shaken with celite and then filtered (Buchner filtration). The resulting solution was extracted further with dichloromethane (3 x 250 mL) and then ethyl acetate (3 x 250 mL), the celite was placed in an ultrasonicator under dichloromethane for 30 minutes, then filtered and combined with the other dichloromethane fractions. The organic fractions were dried with anhydrous magnesium sulfate, filtered and evaporated. The dichloromethane fraction was found after column chromatography using petroleum ether: ethyl acetate, 1 : 1, to contain 115 (70 mg, 47%, 52% allowing for recovered starting precursor 15 mg). Identification was made by comparison of chromatographic and spectral data with authentic compounds.  4.8.3  EXPERIMENT 3 LARGE SCALE  The dibenzylbutanolide 85 (190 mg, 0.46 mmol) was dissolved a mixture of ethanol and water (8:17) (36.5 mL). The CFE (age 11 days, peroxidase activity 1.24) was prepared. To the precursor solution was added hydrogen peroxide (11.8 mL, 2 equivalents), water (37.0 mL), phosphate buffer (122.8 mL, 0.1 M, pH 6.3) and at zero time the CFE (96.8 mL, 211 units per mmol of precursor). Samples were taken at intervals for HPLC (1.0 mL) and the reaction was stopped after 30 minutes by the addition of dichloromethane (300 mL). The mixture was shaken with celite and then filtered (Buchner filtration). The resulting solution was extracted further with dichloromethane (3 x 300 mL) and then ethyl acetate (3 x 300 mL). The filtered celite was placed in an ultrasonicator under dichloromethane for 1 hour, then filtered and combined with the other dichloromethane fractions. The organic fractions were dried with anhydrous magnesium sulphate, filtered and evaporated. The dichloromethane fraction was found after column chromatography using petroleum ether: ethyl acetate, 1: 1, to contain 115 (102 mg, 54%, 75% based on recovered starting precursor 85 (56 mg)). Compounds were characterized by comparison of chromatographic and spectral data with authentic compounds.  161  4.9  THE RING CLOSURE OF r/?A/V5-2-(3",5"-DIMETHOXY4"-HYDROXYBENZYL)-3-(3 -BENZYLOXY)-4 -w0,  PROPOXYBENZYL)BUTANOLIDE (85) FREE EXTRACT OF PODOPHYLLUM  ,  WITH THE C E L L PELTATUM CELL  CULTURE  a  Cells from the Podophyllum peltatum culture (age 25 days) were harvested and used to produce a CFE using the procedure given in Section 4.2.4 (specfic activity 1.2 units/mL). The CFE (3.0 mL, 257 units per mmol of substrate) was added to a solution containing 85 (6 mg in ethanol: water 8:17, 1.1 mL), hydrogen peroxide (0.4 mL, 2.3 equivalents), water (1.1 mL) and phosphate buffer (0.1 M, pH 6.3, 3.0 mL). The mixture was stirred at room temperture for 30 minutes and samples (1.0 mL) were removed for HPLC analysis. The control flask was also incubated for 30 minutes and contained the ingredients as listed for the reaction flask without the substrate. The reaction was quenched with dichloromethane (10 mL). The mixture was shaken with celite and then filtered (Buchner filtration). The resulting solution was extracted further with dichloromethane (3 x 50 mL) and then ethyl acetate (3 x 50 mL), the filtered celite was placed in an ultrasonicator with dichloromethane for 1 hour, then filtered and combined with the other  dichloromethane fractions. The organic fractions were dried with anhydrous magnesium sulphate, filtered and evaporated. The control flask was extracted using the same procedure. HPLC and TLC analysis indicated a yield of approximately 20% of the product 115 with some starting material still present. The extract from the control flask showed no change.  4.10  ATTEMPTED HYDROXYLATION OF 69 WITH THE CFE  OF CATHARANTHUS  ROSEUS OH  4.10.1  INITIAL EXPERIMENT  Cell culture of Catharanthus roseus was harvested at age 10 days and the CFE prepared using the method outlined (Section 4.2.2). The protein concentration was 1.45 mg/mL and the peroxidase activity 1.57 units. The precursor 69 (150 mg) was dissolved in ethanol (10 mL) and hydrogen peroxide added (5 mL, 1.9 equivalents). On addition of the prepared CFE (50 mL, 200 units per mmol of precursor), at zero time, some precipitation occurred. The mixture was stirred at room temperature for 20 hours and  163  samples (1.0 mL) were removed for HPLC analysis peroidically. The mixture was quenched by addition of dichloromethane (60 mL) and then extracted with dichloromethane (3 x 60 mL) and ethyl acetate (3 x 60 mL). The organic extracts were dried with anhydrous magnesium sulphate before filtration and evaporation.  Spectral analysis of the  dichloromethane extract showed recovery of 69, by comparison with an authentic material.  4.10.2  CO-FACTORS EXPERIMENT 1  Cell free extract was prepared (Section 4.2.2) from 15 day old cell culture. The precursor 69 (200 mg) was dissolved in DMSO (5 mL) and water (50 mL) added and some precipitation occurred. This solution was then split into four separate portions and placed in Erlenmeyer flasks (25 mL, 0.13 mmol of precursor per flask). To each of these flasks was added a co-factor as follows: Flask 1. FMN (50 mg) Flask 2. Ascorbic acid (100 mg) Flask 3. NADPH (25 mg) Flask 4. Hydrogen peroxide (5 mL) Cell free extract (20 mL, peroxidase activity 1.59 units, 245 units per mmol of precursor) was then added and the mixtures stirred at room temperature for 28 hours. A control experiment with only cell free extract present was also prepared. Samples (1.0 mL) were taken at intervals for HPLC analysis. The reactions were quenched by the addition of dichloromethane (50 mL) followed by rapid stirring. Samples, for HPLC analysis, were mixed well with methanol (1.0 mL) and filtered through cotton wool.  164  Results:Ascorbic acid: No decrease in precursor. NADPH: No decrease in precursor. H2O2: No decrease in precursor. FMN: Precursor peak masked. The FMN reaction was extracted with dichloromethane (3 x 50 mL) and ethyl acetate (3 x 50 mL). The dichloromethane fraction, after drying, filtration and evaporation gave 29.6 mg and the ethyl acetate fraction yielded 9.4 mg, this corresponds to a recovery of 78%. The 29.6 mg was confirmed by *H-NMR to be identical to an authentic sample of the precursor 69. The ethyl acetate fraction showed no compounds of interest by 1H-NMR and TLC.  4.10.3  CO-FACTORS EXPERIMENT 2  Cell free extract of 11 day old cell culture was prepared (Section 4.2.2) (peroxidase activity 1.2) and the precursor 69 (60 mg) dissolved in acetone (5 mL) and water (50 mL). This solution was divided into three portions and placed in three flasks (0.052 mmol of precursor per flask and 450 peroxidase units per mmol of precursor). Co-factors were then added (14 equivalents of hydrogen peroxide were used) (Table 20).  165 TABLE 20. Co-factor Experiment 2 FLASK 2  FLASK 3  Precursor (69)20 mg  20 mg  20 mg  CFE Hydrogen  20 mL  20 mL  20 mL  peroxide  S mL  5 mL  FLASK 1  Ascorbic acid 50 mg NADPH  80 mg  80 mg  EDTA  20 mg  20 mg  DTT (113)  10 mg  10 mg  Samples (1.0 mL) were removed at intervals for HPLC analysis. The HPLC analysis showed no new peaks. After 20 hours the reaction was quenched and the major peak present was found to co-inject with the starting precursor 69.  4.10.4  CO-FACTORS EXPERIMENT 3  Cell free extract of 12 day old Catharanthus roseus cell culture was prepared (Section 4.2.2) and used in the following experiments. FLASK 1. The precursor 69 (200 mg) was dissolved in acetone (13 mL) and water (87 mL). To this solution was added NADPH (40 mg) and, at zero time, the prepared CFE (300 mL). The solution was stirred at room temperature and samples (1.0 mL) were removed at intervals for HPLC analysis. The reaction was quenched after 25 hours by the addition of dichloromethane (400 mL).  166  FLASK 2. The precursor 69 (104 mg) was dissolved in acetone (13 mL) and water (87 mL). To this solution was added ascorbic acid (600 mg) and, at zerotime,the prepared CFE (300 mL). The solution was stirred at room temperature and samples (1.0 mL) removed at intervals for HPLC analysis. The reaction was quenched after 25 hours by the addition of dichloromethane (400 mL). FLASK 3. The precursor 69 (100 mg) was dissolved in acetone (9 mL) and water (60 mL). To this solution was added ferrous sulphate (100 mg) and hydrogen peroxide (10 mL) followed by CFE (300 mL) at zero time. Samples (1.0 mL) were removed during the reaction for HPLC analysis and the reaction quenched after 25 hours by the addition of dichloromethane (400 mL). FLASK 4. This flask contained only the prepared CFE. Samples (1.0 mL) were removed over the 25 hours for analysis by HPLC. After 25 hours the mixture was quenched by addition of dichloromethane (400 mL). Each sample was treated with celite, filtered, then extracted with dichloromethane (3 x 400 mL) and ethyl acetate (3 x 400 mL). The organic extracts were dried with anhydrous magnesium sulphate and filtered before evaporation (Table 21). For all of the flasks 1,2 and 3 the dichloromethane fractions were found, by TLC and 1H-NMR, to correspond with the starting precursor 69, with small amounts of material from the plant cells as impurities. HPLC results showed no loss of precursor during reaction and no formation of new peaks. The material isolated as a dichloromethane fraction from the control flask contained P-sitosterol (141). Exact mass calculated for C29H50O1:  414.3862; found: 414.3866.  Flask 1 ethyl acetate fraction showed only cell material by 1H-NMR. HRMS gave a peak matched to product 123: Exact mass calculated for C21H22O8: 402.1315; found: 402.1338. Flask 2 ethyl acetate fraction showed aromatic peaks by !H-NMR and MS analysis. Flask 3 ethyl acetate fraction contained cell material by !H-NMR and MS analysis.  TABLE 21. Extraction From Co-factor Experiment 3  RECOVERY: Flask 1  Dichloromethane  210.7 mg  Ethyl acetate  18.4 mg Total 229.1 mg  Flask 2  Dichloromethane  165.0 mg  Ethyl acetate  21.0 mg Total  Flask 3  Dichloromethane  108.0 mg  Ethyl acetate  9.0 mg Total  Control  186.0 mg  117.0 mg  Dichloromethane  13.9 mg  Ethyl acetate  4.4 mg  Total 18.3 mg  168 4.11  A T T E M P T E D BIO-TRANSFORMATIONS O F D E O X Y P O D O P H Y L L O T O X I N (6) W I T H T H E C F E O F  CATHARANTHUS  OCH  ROSEUS .  3  6  OCH3  3 or 37  Cell free extract of 14 day old cell culture was prepared (section 4.2.2) (peroxidase activity 1.45). Deoxypodophyllotoxin (6) was dissolved in acetone and then the other components were added to the reaction flask, some precipitation occurred (9 equivalents of hydrogen peroxide were added) (Table 22). The CFE (1760 units per mmol of the precursor) was added at zerotimeand the mixtures stirred at room temperature. Over the 18 hours samples (1.0 mL) were taken at intervals for HPLC analysis. The mixture was then quenched by addition of dichloromethane (150 mL) and extracted with dichloromethane (3 x 150 mL) followed by extraction with ethyl acetate (3 x 150 mL). The organic extracts were then dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo. Chromatography gave recovery of starting material 6, 27 mg (82%) for Flask A, and 22 mg (73%) for Flask B. Identification was carried out by 1H-NMR and comparison with an authentic sample of starting material 6.  169 TABLE 22. Bio-transformation of 6 FLASK 1  FLASK 2  Precursor  33 mg  30 mg  CFE  100 mL  100 mL  50 mL  Water  30 mL  30 mL  15 mL  Co-factor  Hydrogen  Hydroquinone  Hydroquinone  peroxide 5 mL  150 mg  75 mg  Acetone  5 mL  5 mL  2.5 mL  4.12  THE USE OF ENZYME STABILIZERS  FLASK 3  Two buffers were prepared. Buffer A was a phosphate buffer (0.1 M, pH 6.3). Buffer B was also a phosphate buffer (0.1 M, pH 6.3) but also containing the additives; DTT (113) (1 mM), sucrose (0.4 M) and EDTA (2 mM). Cell culture (1.5 L) from Catharanthus roseus cell culture (Section 4.2.1) was harvested at age 12 days and split into two portions. Each portion was filtered through Miracloth and the cells washed with ice cold distilled water (150 mL). The cells were then homogenized in the presence of buffer A (50 mL) using the Ultra-Turrax T-25 disperser. The second portion of cell culture was filtered and the cells washed with buffer B (2 x 150 mL) These cells were then homogenized in the presence of buffer B (50 mL) using the Ultra-turrax T-25 disperser. To this homogenate was added phenylmethylsulphonyl fluoride (PMSF; 0.23 g in a acetone) to give a 1 mM solution, and pepstatin A in DMSO to give a 0.1 p.M solution. Both homogenates were then centrifuged (10,000 G) for 30 minutes. The resulting CFE were collected by decantation and used for measurement of  protein concentration, peroxidase activity and CO difference spectra. Results are shown in Tables 23 and 24.  TABLE 23. BUFFER A Time past homogenization  90 minutes  260 minutes  Protein concentration (mg/mL)  1.1  1.0  Peroxidase activity (units)  1.2  1.2  Specific activity  1.1  1.1  TABLE 24. BUFFER B Time past homogenization  90 minutes  260 minutes  Protein concentration (mg/mL)  1.5  1.2  Peroxidase activity (units)  1.5  1.4  Specific activity  1.0  1.1  The CO difference spectra of the two CFE's were measured using the procedure in Section 4.1.3.  4.13  CHARACTERIZATION OF THE CELL EXTRACT FROM CATHARANTHUS  ROSEUS  Catharanthus roseus cell culture (6 x 500 mL), age 11 days, was cooled and filtered through Miracloth. The cells were then washed with the cooled extraction buffer (500 mL). Extraction buffen-  220mMD-Mannitol 70 mM sucrose 20 mM potassium phosphate pH 7.4  The wet weight of the cells was measured (750 g) and extraction buffer added (600 mL). This mixture was homogenized using the Ultra-Turrax T-25 disperser for four separate 30 second periods with 30 second rest periods between, the whole process being carried out at 0-4°C. The homogenate was then centrifuged at 1,000 G for 10 minutes to give a pellet (PI), which was discarded, and a supernatent (SI, 850 mL). The supernatent (SI) was then centrifuged at 10,000 G for 10 rninutes to give pellet (P10) and supernatent (S10, 800 mL) which was decanted off and further centrifuged at 100,000 G for 75 minutes. The supernatent (S100) was then separated from the pellet (P100). The pellets were resuspended in resuspension buffer (50 mL). Again all procedures were carried out quickly at 0-4°C. Resuspension buffer-  1 mM EDTA 0.1 mM DTT 124 10% w/v glycerol 50 mM potassium phosphate pH 7.5  The samples, S10, P10, S100 and P100 were measured for peroxidase activity, protein concentration and the specific activities were then calculated (Table 25).  T A B L E 25. The Specific Activities and Protein Concentration of the Centrifugation Fractions  Fraction  [Protein] mg/mL  Specific activity  SI  1.48  0.99  S10  1.46  0.90  S100  0.86  1.65  P10  1.52  0.16  P100  1.72  0.00  The CO difference absorption spectra were measured using the following procedure. The appropriate buffer (extraction or resuspension) was placed in both sample and reference UV cells to obtain a baseline. Then the fraction was placed in the sample cell and the absorption measured (530 nm-390 nm). A few mgs of fresh sodium dithionate were then added to the sample cell, effervesence was observed, and the absorption spectrum measured. The sample cell was then quickly taken into a fume hood and CO was passed through the suspension for 60 seconds, then the CO-difference spectrum was recorded.  4.14  ATTEMPTED BIO-TRANSFORMATION OF DEOXYPODOPHYLLOTOXIN (6) WITH THE S10 FRACTION OF THE CATHARANTHUS ROSEUS CELL EXTRACT OH  <7  Cells of the Catharanthus roseus culture (age 10 days) were harvested and centrifugation and resuspension methods were carried out as described in section 4.1.3. Deoxypodophyllotoxin (6) (98.6 mg) was dissolved in acetone (30 mL) and FAD (500 mg) was added (ratio of co-factor to precursor was 2.44 : 1). At zero time the prepared S10 fraction (640 mL) was added at pH 7.9 and the mixture was stirred at room temperature for 24 hours, before quenching with dichloromethane (500 mL). Extraction was then carried out with dichloromethane (3 x 500 mL) followed by ethyl acetate (3 x 500 mL). The organic extracts were then dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo.  Column chromatography of the fractions gave  deoxypodophyllotoxin (6) (94.3 mg, 95.6% recovery) characterized by comparison of chromatographic and spectral data with the authentic starting precursor. No other products were identified.  174  4.15  EVALUATION OF HYDROXYLATION OF DEOXYPODOPHYLLOTOXIN (6) AND 69 WITH THE S10 FRACTION OF CATHARANTHUS CO-FACTORS.  ROSEUS  WITH VARIOUS  SMALL SCALE  Cell culture of Catharanthus roseus (age 11 days) (10 x 500 mL) was harvested and the S10 fraction prepared as described in Section 4.1.3. The protein concentration of the S10 fraction was measured and found to be 1.2 mg/mL. The two substrates 6 and 69 were dissolved in acetone at a concentration to provide 1 mg of substrate in 0.2 mL of acetone. Enough of the S10 fraction was used to provide 24 mg of protein. Ten flasks were prepared (Table 26) and stirred at room temperature.  TABLE 26  The S10 Fraction Using Co-factors  Flask  S10  A  21 mL  B  21mL  0.2 mL  C  21 mL  0.2 mL  D  21 mL  0.2 mL  E  21 mL  0.2 mL  F  21 mL  0.2 mL  G  21 mL  0.2 mL  H  21 mL  I  21 mL buffer  J  21 mL buffer  6  69  FAD  NADPH  16 mg 17 mg  16 mg  16 mg 17 mg  16 mg  17 mg  16 mg  0.2 mL  17 mg  16 mg  0.2 mL  17 mg  16 mg  Samples (1.0 mL) were removed at zerotime,30 minutes, 100 minutes and 17 hours for HPLC analysis. The reactions were quenched after 17 hours by the addition of ethyl acetate (20 mL). Samples of podophyllotoxin (3) were used to 'spike' the sample from the flasks containing 6 to check for any hydroxylation. No podophyllotoxin (3) or epipodophyllotoxin (37) was found in any of the flasks. For the reactions involving the precursor 69 no new peaks were found by HPLC analysis.  4.16  ATTEMPTED HYDROXYLATION OF D E O X Y P O D O P H Y L L O T O X I N (6) WITH W H O L E C E L L S O F  T H E TRIPTERYGIUM WILFORDII C E L L SUSPENSION CULTURE  Deoxypodophyllotoxin (6) (100 mg) was dissolved in acetone (20 mL) and this solution split into two portions. Each portion was added to resuspended Tripterygium wilfordii cells (Section 4.2.5) (500 mL), under aseptic conditions. The resuspension buffer used was tris-HCl (pH 7.5, with 8% sucrose w/v). Samples (15 mL) were removed after 24, 48 and 120 hours. A portion (1.0 mL) of these samples was used for HPLC analysis. The remaining samples were extracted with ethyl acetate (2 x 50 mL). This extract was dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo. One flask was taken from the shaker beds after 120 hours and the cells filtered through Miracloth. The second flask was found to be contaminated with bacteria and so was not harvested. The broth from the 120 hour experiment was extracted with ethyl acetate (3 x 300 mL), the extract dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo. A portion (70 mg) of the crude extract (184 mg) was separated by preparative TLC using chloroform : methanol, 25 : 1 as the developing solvent. Only starting precursor 6  was identified by !H-NMR, EI MS and comparison using HPLC, with authentic material 6.  4.17  A T T E M P T E D H Y D R O X Y L A T I O N O F 69 AND 6 WITH T H E  C F E PREPARED F R O M T H E TRIPTERYGIUM  WILFORDII  C E L L SUSPENSION C U L T U R E  The cell culture (2 L) was harvested after 21 days and the cells filtered through Miracloth and washed with ice-cold water. Phosphate buffer (0.1 M, pH 7.3, 270 mL) was added and the cells homogenized with the Ultra-Turrax T-25. The homogenate was then centrifuged (10,000 G) for 30 minutes. The CFE was then decantedfromthe cell debris and samples taken for protein concentration measurement (0.71 mg/mL) and peroxidase activity measurement (specific activity 7.3 units). Three experiments were carried out as described in Table 27, zerotimewas taken as being at the addition of the CFE. Samples (1.0 mL) were removed at intervals over a 14 hour period for HPLC analysis before the reactions were quenched by addition of dichloromethane (100 mL). The HPLC analysis of the samples indicated that no reaction had occurred.  177  TABLE 27 Attempted Hydroxylation of 69 and 6 with the CFE of Tripterygium wilfordii Flask #  Precursor  CFE  Solvent 1  69(50mg)  Hydrogen peroxide  150 mL  10 mL  150 mL  10 mL  150 mL  10 mL  acetone (2.0 mL) 6(50mg) acetone (1.0 mL) none acetone (1.5 mL)  4.18  ATTEMPTED HYDROXYLATION OF THE PRECURSOR 133 WITH THE WHOLE CELLS OF THE TRIPTERYGIUM WILFORDII C E L L CULTURE. OH  133  138  178 Cell culture of Tripterygium wilfordii (17 days old) was resuspended using phosphate buffer (pH 6.4) with 8 % w/v sucrose. Six flasks were prepared as shown below in Table 28.  T A B L E 28 Tripterygium wilfordii Whole C e l l attempted B i o transformation  Flask 1  500 mL  control  Flask 2  500 mL  control  Flask 3  500 mL  1 3 3 (50 mg) (ethanol water)  Flask 4  500 mL  1 3 3 (50 mg) (ethanol water)  Flask 5  500 mL  1 3 3 (50 mg) DMSO  Flask 6  500 mL  1 3 3 (50 mg) DMSO  DMSO  Samples were taken (1.0 mL) for HPLC and each flask quenched after 100 hours by addition of dichloromethane with vigorous stiiring. Flasks 3 and 4 were combined as were flasks 5 and 6. Each suspension was homogenized with the Ultra-Turrax T-25 disperser for five minutes. Celite was added and the mixture was filtered. The filtrate was extracted with dichloromethane (3 x 500 mL) followed by extraction with ethyl acetate (3 x 500 mL). The organic extracts were dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo. Flasks 3 and 4 yielded from the dichloromethane fraction 93.3 mg of extract, from ethyl acetate fraction 21 mg of extract. The acidified ethyl acetate fraction gave 123.3 mg of extract, providing a total extract of 237.6 mg. Only a trace of the starting precursor 1 3 3 was identified but no other product could be characterized. The Flasks 5 and 6 were not extracted and the controls showed no change by HPLC analysis.  179  4.19  ATTEMPTED HYDROXYLATIONS OF 4'-DEMETHYL-6,7DEMETHYLENEDEOXYPODOPHYLLOTOXIN (134) AND 6,7-DEMETHYLENEDEOXYPODOPHYLLOTOXIN (133) USING THE CFE OF CATHARANTHUS ROSEUS  WITH VARIOUS CO-FACTORS  Catharanthus roseus  CFE Co-factors  —  HO  CH O'  0CH3  3  OCH3  138  Catharanthus roseus  CFE Co-factors CH3O"  "OCH3  y  "  CH3O'  0  C  H  3  OH OH  134  139  Cell culture (Section 4.2.1) (age 10 days) was used to prepare the CFE (Section 4.2.2) at pH 5.88. The precursor 134 (50 mg) was dissolved in ethanol (18 mL) and water (32 mL) to provide a stock solution. The precursor 133 (50 mg) was dissolved in ethanol (20 mL) and water (30 mL) to provide a stock solution. Flasks were prepared  180 containing the precursors and co-factors (Table 29). A stock solution of manganese chloride was used (10 mg in 10 mL of water). Samples (1.0 mL) were removed at intervals for HPLC analysis over a period of 20 hours. The reactions were initiated by addition of the CFE. Flasks number 18 and 19 contain a small sample of the authentic expected products (138,139) in order to check their stability under the reaction conditions.  181 TABLE 29. Small Scale Co-factor Experiment with Precursor 133 Flask CFE  134  133  NADPH  FAD FMN MnCl  2mL  10 mg  10 mg 10 mg 1 mL O.f  2  H2O2  #  30 mL  2mL  1  <8>  <g>  2  <8»  3  ®  <8> (8)  4  ®  5  <8>  6  <8>  7  <8>  8  <8>  9  <8>  10  <8>  11  ®  <8> ®  <8>  ® ® ®  <8» ® <8>  <8>  ®  <8>  12  <g>  13  ®  14  B  15  B  <8>  16  <8>  <S>  17  <8»  18  <8>  19  <8>  <8> <S>  <8>  ®  n  <8»  <8>  • <8» ®  <S>  ®  ®  <8> <8>  ®  ®  <8> <8>  n  <8> = reagent added. B = buffer added. II = expected product added.  ®  182  RESULTS: Compound 134. HPLC results showed no reaction for Flasks 1, 3, 5, 9, 14. Precursor 134 was masked in Flasks 7, 14 and 16 by the FMN co-factor, no new peaks were visible. Flask 18 showed no reduction in the amount of compound 134. Flask 12 did show the dissappearence of starting material 134 but no product was visible. RESULTS: Compound 133. Flasks 2, 4, 8, 10, 15 and 17 showed no loss of starting material 133 and no formation of product Flask 19 showed no loss of compound 133. Flask 13 showed some loss of starting material 133 and Flask 6 showed little loss of starting material 133 but the appearence of a new peak was observed at a similar retentiontimeto the expected product 138.  4.20  A T T E M P T E D H Y D R O X Y L A T I O N O F 4'-DEMETHYL-6,7D E M E T H Y L E N E D E O X Y P O D O P H Y L L O T O X I N (134) W I T H  T H E C F E O F CATHARANTHUS  ROSEUS OH  134  139  183 Cell free extract from the Catharanthus roseus cell culture (age 10 days) was prepared (specific activity 1.42). The precursor 134 (205 mg, 0.53 mmol) was dissolved in ethanol (20.5 mL) and water (41 mL) and hydrogen peroxide (10 mL, 2.8 equivalents) were added. The reaction was started by addition of the CFE (600 mL, 1600 units per mmol of the precursor) and was stirred at room temperature for 70 minutes before quenching with dichloromethane (600 mL). Samples were taken during the reaction for HPLC monitoring. Celite was added to the emulsion and further dichloromethane (500 mL) was added before filtration and extraction with dichloromethane (2 x 500 mL). The aqueous layer was then extracted with ethyl acetate (3 x 500 mL). Organic fractions were dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo. The remaining aqueous solution was freeze-dried. The dichloromethane extract (114 mg) was separated  by  column  chromatography  to  give  4'-demethyl-6,7-  demethylenedeoxypodophyllotoxin (134) (30 mg, 15%) which was confirmed by comparison with starting precursor. The ethyl acetate extract (33 mg) was separated by preparative TLC (two 0.25 mm silica plates were used) using an eluent mixture of chloroform  : acetone  : methanol,  8 : 2 : 3 .  4'-Demethyl-6,7-  demethylenedeoxypodophyllotoxin (134) was recovered (5.4 mg, 2.6%) and 4'-demethyl -6,7-demethylenepodophyllotoxin (139) (7.3 mg, 3.5%) was isolated and confirmed by comparison of spectral data with that of authentic material.  4.21  T H E ISOLATION O F M E T A B O L I T E S F R O M  PODOPHYLLUM PELTATUM C E L L C U L T U R E  4.21.1  INITIAL E X P E R I M E N T S .  The cell culture was propagated as outlined previously. After 22 days of growth the culture was harvested (4.0 L) and freeze-dried. The freeze-dried material was then homogenized using an Ultra-Turrax T-25 disperser and extracted with dichloromethane (3 x 300 mL), dried with anhydrous magnesium sulphate, filtered and this 'neutral' fraction was then evaporated in vacuo. The aqueous layer was further extracted with ethyl acetate (3 x 300 mL) to yield an ethyl acetate fraction which was dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo. The aqueous fraction was then acidified (HCl 1 M) and extracted with dichloromethane to give an 'acidic' fraction which was dried with anhydrous magnesium sulphate, filtered and evaporated in vacuo (Table 30).  T A B L E 30. Extraction of the Initial Isolation Experiment from  Podophylum peltatum 'Neutral' fraction  474 mg  Ethyl acetate fraction  218 mg  'Acidic' fraction  1853 mg  Total extract  2544 mg  A portion of the 'neutral' fraction (239 mg) was treated with HCl (5 mL, 1 M) and methanol (5 mL) at 40°C for 3 hours. The mixture was extracted with chloroform and evaporated to give the crude product (218 mg). Samples of this crude hydrolyzed mixture  together with crude samples of the other fractions, were analyzed by TLC, HPLC, IR, 1HNMR and HRMS. 'Neutral' fraction: IR v , ^ : 3300,1730 cm-1. 'Acidic' fraction: IR v axJ 3300,1700 cm-1. Exact mass calculated for C21H20O7: m  384.1207; found: 384.1210. Ethyl acetate fraction: IR Vmax: 3350 cm-l. 'Neutral' hydrolyzed fraction: IR v for  : 3300,1730 cnr . Exact mass calculated 1  max  C22H22O8: 414.1315; found: 414.1336. The crude products (118 mg) of the hydrolyzed fraction were separated by column  chromatography using chloroform as initial eluent. 1H-NMR of fractions showed no identifiable products and no aromatic protons. EI MS data indicated sterol material was present.  4.21.2  INVESTIGATION OF THE MAJOR CONSTITUENTS OF THE PODOPHYLLUM  PELTATUM CELL CULTURE.  EXPERIMENT 1. SHAKE FLASK CULTURE AND MICROFERM CULTURE OH  Podophyllotoxin (3)  186  Cell culture (3.84 L) from a bioreactor (5.5 L) was harvested after 22 days of growth and combined with cell culture (3.0 L) from shake flasks also of age 22 days. The cells were separated from the broth by filtration using Miracloth. The broth was divided into two portions and extracted with dichloromethane (3.0 L per portion) in a continous liquid-liquid extraction apparatus for 116 hours. The organic phase was dried with anhydrous magnesium sulphate, filtered and evaporated to give 103 mg of crude extract. This extract was purified by column chromatography using dichloromethane : acetone, 4:1 as the eluent. Podophyllotoxin (3) (21 mg) was characterized. Mp 180-183°C. IR v  m a  x  3500, 1770 cm' . 1H-NMR (400 MHz, CDC1 ). 8: 2.73-2.81 (1 H, m, H(3)), 2.83 (1 1  3  H, dd, H(2), J = 14, 7 Hz), 3.76 (6 H, s, - O C H 3 ) , 3.82 (3 H, s, - O C H 3 ) , 4.07 (1 H, t,  H(ll), J = 9.2 Hz), 4.59 (2 H, m, H(l), H(ll)), 4.76 (1 H , d, H(4), J = 8.9 Hz), 5.97 (2 H, d, - O C H 2 O - , J = 3.0 Hz), 6.38 (2 H, s, H(2', 6')), 6.51 (1 H, s, H(8)), 7.12 (1 H, s, H(5)). EI MS m/z: 414 (M+), 396, 181,168,153. The compound was also compared to a sample of authentic pc)dophyllotoxin by TLC. It had the same Rf value, and by HPLC it had the same retention time as the authentic material. The cells were homogenized in a blender for 5 minutes and then homogenized using the Ultra-Turrax T-25 disperser. The homogenate was then freeze-dried. After freezedrying, the brown solid (46.8g) was divided into two portions and extracted with dichloromethane in a soxhlet extractor for 116 hours. After extraction the solution was dried with anhydrous magnesium sulphate, filtered and evaporated. The extracted material (1.99 g) was purified by column chromatography using dichloromethane : acetone, 4 : 1 as the initial eluent. Column chromatography was repeated with the podophyllotoxin (3) containing fractions to give podophyllotoxin (3), (27 mg, 0.058% dry weight). Podophyllotoxin (3) was characterized by comparison of the 1H-NMR spectrum, MS data, TLC Rf value and HPLC retentiontimewith a sample of authentic podophyllotoxin (3).  4.21.3  E X P E R I M E N T 2. L A R G E S C A L E E X T R A C T I O N O F 5.5 L OF PODOPHYLLUM PELTATUM M I C R O F E R M C U L T U R E O  OH  4'-Demethylpodophyllotoxin (5) The culture was incubated in a microferm for 21 days and then separated by filtration with Miracloth Both the cells and broth were extracted immediately or frozen with liquid nitrogen and stored in a freezer until required for extraction. The broth (5.5 L) was divided into two portions and each portion was extracted with dichloromethane (3.0 L) in a liquid-liquid continous extraction apparatus for 100 hours. The extract was dried with anhydrous magnesium sulphate, then evaporated to give  188 weight of 284 mg. The extract was then purified by column chromatography using dichloromethane : acetone 4 : 1 as the initial eluent. The following compounds were characterized in order of elution:- ^-sitosterol, (22 mg) was characterized by comparison of !H-NMR and EI MS with authentic material. Podophyllotoxin (3), (22 mg) was identified by comparison of the 1H-NMR and EI MS data with an authentic sample. The cell material was mixed with dichloromethane (300 mL) and water (100 mL) and homogenized in a blender for 5 minutes, then homogenized with the Ultra-Turrax T-25 disperser for two minutes at 20,000 rpm with the coarse head, then for 2 minutes at 20,000 rpm with the fine head. Celite was added to this emulsion which was then filtered and extracted with dichloromethane (2 x 500 mL). The organic phase was dried with anhydrous magnesium sulphate, and filtered before evaporation. This crude extract (826 mg) was washed with hexanes (50 mL) and the insoluble material (435 mg) was then purified by column chromatography. The initial eluent was dichloromethane : acetone, 4: 1. The final fraction was collected using methanol as eluent The material recovery from the column was 312 mg (72%). The following compounds were characterized in order of elution:- Deoxypodophyllotoxin (6), (17 mg with podophyllotoxone (60)). !H-NMR: (400 MHz, CDC1 ) 8: 2.66-2.85 (3 H, m, H(2), H(3), H(4p)), 3.08 (1 H, m, H(4a), 3  3.74 (6H, s, 3', 5' -OCH3), 3.80 (3H, s, 4* -OC&3), 3.92 (1 H, m, H(ll)), 4.45 (1 H, m, H(ll)), 4.59 (1 H, d, H(l), J = 3 Hz), 5.94 (1 H , d, -OCH 0-, J = 1 Hz), 5.95 2  (1  H , d, -OCH2O-, J = 1 Hz), 6.35 (2 H, s, H(2'), H(6')), 6.53 (1 H, s, H(8)), 6.69 (1 H, s, H(5)). EI MS m/z: 398 (M+), 383, 367, 339, 308, 181. Exact mass calculated for C22H22O7: 398.1365;  found: 398.1363.  Podophyllotoxone (60), (17 mg with deoxypodophyllotoxin (6)). IR v cm-1-  IH-NMR  m a x  : 1730  :(400 MHz, CDCI3) !H-NMR: (400 MHz, CDCI3) 8: 3.27 (1 H, dd,  H(2), J = 16, 5 Hz), 3.53 (1 H, ddd, H(3), J = 15, 10, 6 Hz), 3.74 (6H, s, 3', 5' OCH3), 3.81 (3H, s, 4' -OCH3), 4.36 (1 H,dd 10, 10 Hz), H(llp)), 4.56 (1 H, dd,  H(llct), J = 8, 7.5 Hz), 6.09 ( l H . d , -OCH 0-, J = 1 Hz), 6.11 (1 H , d, -OCH 0-, J = 2  2  189  1 Hz), 6.39 (2 H, s, H(2'), H(6')), 6.66 (1 H, s, H(8)),  7.56 (IH, s, H(5)). EI M S m/z:  412 (M+), 383, 367, 339, 308, 181. Exact mass calculated for C22H20O8: 412.1158; found: 412.1149. Podophyllotoxin (3), (127 mg) was characterized by comparison of the 1H-NMR spectrum, EI MS data, TLC, Rf value and HPLC retentiontimewith a sample of authentic podophyllotoxin (3). ^-Demethylpodophyllotoxin (5), (36 mg). Mp 250-253°C. IR v  cm-1.  IR-NMR  OCH3), 4.01  m a x  : 3540, 1750  :(200 MHz, CDCI3). 8: 2.60-2.80 (2 H, m, H(2), H(3)), 3.68 (6H, s, -  (IH, m, H(llp)), 4.50 (2 H, m, H(l), H(lla)),  Hz), 5.42 (1 H, br s, OH), 5.90  (2H, d, -OCH2O-,  H(6')), 6.42 (1 H, s, H(8), 7.04 (IH, s, H(5)).  4.66 (1 H, d, H(40), J = 8  J = 3 Hz), 6.34  (2 H, s, H(2'),  EI MS m/z: 400 (M+), 167, 154. Exact  mass calculated for Q21H20O8: 400.1158; found: 400.1151.  Figure 13. The Major Fragments Produced in the EI MS of Compound 63  191  +  m/z 167 100  m/z 153  Figure 14. The Major Fragments Produced in the EIMS of Compound 69  Figure 15. Single crystal X-ray structure of 69.  Figure 16. ' H - N M R spectrum and partial NOEDIFF spectra of 93 (400 M H z , CDCI3).  Figure 17. EI MS spectrum of 85.  I  I  I 4.6  I I I I  I I I I  I  I 4.4  I I I 1  I I I I  I  I 4.2  I I I I  I  T I I I  I 4.0  I I I I  I I I I  I  I '  I I [ 1 1 '  I  3.8  JJU*L T T T - r r n - r r T T T  7  f r n  r  |  6  i T  I  r  |  r n  r  j r r n  jrr  r r | ;  r r r  r|-ii  .1 i  i  |  i  i  i  i  |  5 4 PP 3 Figure 19. H-NMR spectrum of 115 (300 MHz, m  !  i  i  i  i  |  2  A_ i  CDCI3).  i  i  i  |  i  i  i  i  |  1  i  i  i  i  j  i  i  i  .  ,  0  ON  VO  198  Figure 21. The Major Fragments Produced in the EI MS of Compound 133  TABLE 31 Murashige and Skoog (M.S.) Medium (1962) for the Propagation of the Podophyllum peltatum Cell Culture Ingredients  mg/L  N H 4 N 0 3  1,650  K N 0 3  1,900  CaCl .2H 0  440  MgS0 .7H 0  370  KH P04  170  Na^TA  7.3  2  2  2  4  2  FeS04.7H 0  27.8  H 3 B O 3  6.2  2  MnS0 .H 0  16.9  ZnS04.7H 0  8.6  KI  0.83  4  2  2  Na MoC»4.2H 0  0.25  CuS0 .5H 0  0.025  CoCl2.6H 0  0.025  TWamineHCl  0.4  Inositol (meso.)  100  Sucrose  30,000  Agar (Bacto.)  7g - lOg  2  2  4  2  2  TABLE 32. 1-B5 Medium used for the Propagation of Catharanthus roseus Ingredient  mg/L  NaH P04.H 0  90  Na HP04  30  KCI  300  (NrLteSCH  200  2  2  2  MgS0 .7H 0  250  KNO3  1,000  4  2  CaCl .2H 0  150  KI  0.75  Iron (Sequestrene 330 Fe)  28  2  2  1.0 mL  Micronutrients  mg/lOOmL MnS0 .H 0  1,000  H3BO3  300  4  2  ZnS0 .7H 0 4  300  2  Na Mo04.2H 0  25  CuS04.5H 0  25  CoCl .6H 0  25  2  2  2  2  2  Vitamins  10.0 mL mg/lOOmL  Nicotinic acid  10  Thiamine.HCl  100  Pyridoxine.HCl  10  Myoinositol  1,000  202 TABLE 33. PRD2C0100 Medium for the propagation oTTripterygium wilfordii Cell Culture mg/L  Ingredient NaH2P04.H 0  150  KNO3  2,500  2  134  (NH4)2S0  4  MgS0 .7H 0 2  250  CaCl .2H 0  150  Iron  28  KI  1 ml (75 mg/100 ml)  Micronutrients  1.0 ml  4  2  2  mg/lOOml MnS0 .H 0  1,000  H3BO3  300  4  2  200  ZnS0 .7H 0 4  2  Na Mo0 .2H 0  25  CuS0 .5H 0  2.5  CoCl .6H 0  2.5  2  4  2  4  2  2  2  Vitamins  10.0 ml mg/lOOml  Nicotinic acid  100  Thiamine.Ha  1,000  Pyridoxine.HCl  100  Myo-Inositol  10,000  Sucrose  20.0  2,4-D (2.2 mM)  2ml  203 REFERENCES  1)  Plant Tissue Culture and its Bio-technological Application, Ed. Barz, W; Zenke, M. H., Springer-Verlag, New York, 1977.  2)  Plant Tissue and Cell Culture, Botanical Monographs, Vol. Ul, Ed. Street, H. E.,University Press, 1977.  3)  Plant Tissue Culture Methods Ed. Welter, L.R.; Constabel, F. 1982.  4)  JJceda, T.; Matsumoto, T.; Noguchi, M. Phytochemistry, 1976,15, 568.  5)  Plant Tissue Culture as a Source of Biochemicals. Staba, E. J, C.R.C. press, 1982.  6)  Studies in Secondary Metabolism with Plant Tissue Cultures, Overton, K. H.; Picken, DJ. 1977, 284.  7)  Plant Cell Biotechnology. Ed. Pais, M. S. S.; Mavituna, F.; Novais, J. M. Springer-Verlog New York 1987:  8)  Biotechnology, Microbial Products, Ed Pape, H.; Rehm, HJ. 1986  9)  Reubgardm E.; Kreis, W.; Barthlen, U.; Helmbold, U. Biotechnol., Bioeng. 1989, 34, 502.  10)  Tuominen, U.; Toivonen, L.; Kauppinen, V.; Markkanen, P.; Bjork, L. Biotechnol. Bioeng. 1989,33, 558.  11)  Enzymes in Metabolic Pathways, Milyon, H. S. Jr., Harper and Row, New York 1987.  12)  Smith, J.E. Biotechlnology Principles, 1985.  13)  Issell, B. F.; Maggia, F. M.; Carter, S. K., Etoposide, 1984.  14)  Stahelin, H.; von Wartburg, A. Progress in Drug Research, 1989,33, 169-266  15)  King, L. S.; Sullivan, M., Science 1946,104, 244.  16)  Lehninger, A. L. Principles of Biochemistry 1982, 123.  17)  Hartwell, J. L.; Schrecker, A. W., Fortschr. Chem. Org. Natstqffe. 1958,15, 83.  18)  a) Stoll, A.; Renaz, J.; vonWartburg, A., / . Am. Chem. Soc. 1954, 76, 3103.  204 b) Stoll, A.; von Wartburg, A.; Angliker, E.; Renz, J., / . Am. Chem. Soc. 1954, 76,5004. c) Von Wartburg, A.; Angliker, E.; Renz, E., Helv. Chim. Acta 1957,40,1331. d) Stoll, A.; von Wartburg, A.; Angliker, E.; Renz,J., / . Am. Chem. Soc. 1954,76, 6413. e) Stoll, A.; von Wartburg,.A.; Renz, J., / . Am. Chem. Soc. 1955, 77, 1710. 19)  Kuhm,M.; von Wartburg,A., Helv. Chim. Acfa.,1968,51, 1631.  20)  Keller-Juslen, C ; Kuhn, M.; von Wartburg, A.; Stahelin, H., / . Med. Chem. 1971, 14, 936.  21)  Holthuis, J. J. M., Pharm.Weekbl (Sci). 1988,10, 101.  22)  a) Inamori, Y.; Kubu, M.; Tsujibo, H.; Ogawa, M.; Baba, K.; Kozawa, M.; Fujita, E., Chem. Pharm. Bull. 1986,34, 3928. b) Bedows, E.; Hatfield, G.M., / . Nat. Prod. 1982,45, 725.  23)  a) Thurston, L. S.; Imakura, Y.; Haruna, M.; Li, D. H.; Liu, Z. C ; L. Y.; Cheng, Y. C ; Lee, K. H., / . Med. Chem.l9$9,32, 604. b) Stahelin, H.; Von Wartburg, A., Progress in Drug Research 1989,33,169266. c) Osheroff, N., Biochemistry, 1989,28, 6157.  24)  Gensler, W. J.; Gatsonis, C. D., / . Org. Chem. 1966,31, 3224.  25)  Gensler, W. J.; Samour, C. M.; Wang, S. Y.; Johnson, F., / . Am. Chem. Soc. 1960, 82, 1714.  26)  Gensler, W. J.; Gatsonis, CD., / . Org. Chem. 1966,31, 4004.  27)  (a) Macdonald, D.I.; Durst, T., / . Org. Chem. 1988,53, 3663. (b) Macdonald, D.L; Durst, T., / . Org. Chem. 1986,51, 4749.  28)  Rajapaksa, D.; Rodrigo, R., / . Am. Chem. Soc. 1981,103, 6208.  29)  a) Kende, A. S.; Liebeskind, L. S.; Mills, J. E.; Rutdedge, P. S.; Curran, D. P., /. Am. Chem. Soc. 1977, 99, 7082. b) Kaneko, T.; Wong, H., Tetrahedron Lett. 1987,28 , 517.  30)  (a)  Vandewalle, M.; van der Eycken, J.; De Clercq, P., TetrahedronLett.  1985,26, 3871. (b)  Vandewalle, M.; van der Eycken, J.; De Clercq, P., Tetrahedron 1986,  42,285-4297.  205 (c)  Jones, D.W.; Thompson, A. M., J.Chem. Soc, Chem. Commun. 1987,  1797. 31)  Meyers, A. I.; Andrews, R. C ; Teaque, S. J., / . Am. Chem. Soc., 1988,110, 7854.  32)  Morimoto, T.; Chilba, M.; Achiwa, K., TetrahedronLett.. 1990,31, 261.  33)  Vyas, D. M.; Skonezny, P. M.; Jenks, T. A.; Doyle, T. W., Tetrahedron , 1986, 31, 3099.  34)  Whiting, D.A., Natural Product Reports, 1990, 7, 349.  35)  Kende, A. S.; King, M. L.; Curran, D. P., / . Org. Chem. 1981,46, 2826.  36)  (a) Kuhn, M.; von Wartburg, A.; Keller-Juslen, C , Helv. Chim. Acta, 1969,52,  944. (b) Kuhn, M.; von Wartburg, A., Helv. Chim. Acta, 1969,52 , 948.  (c) Keller-Juslen, C ; Kuhn, M.; von Wartburg, A.; Stahelin, H., / . Med. Chem., 1971,14, 936. 37)  (a) Ward, R. S., Chem. Soc. Rev. 1984,23, 1207. (b) MacRae, W. D.; Towers, G. H. N., Phytochemistry 1984,23, 1207.  38)  (a) Dewick, P. ML; Jackson, D. E., Phytochemistry 1981,9, 2277. (b) Dewick, P. M.; Jackson, D. E., Phytochemistry 1984,5, 1147.  39)  a) Widad H. K.; Dewick,J>., Phytochemistry 1986,9, 2089. b) Widad H. K.; Dewick,.P., Phytochemistry 1986,9, 2093.  40)  Peroxidase, Saunders, B. C ; Holmes-Siedle, A. G.; Stark B. P., Butterworths, London, 1964  41)  Ebermann, R.; Pichomer, H., Phytochemistry 1989,28, 711.  42)  Metodiewa, D.; Danford, H. B., Arch. Biochem. Biophys. 1989,272, 245.  43)  Paul, K.G. In The Enzymes, Vol. 8, 2nd ed.; Ed. Boyer, P.D.; Lardy, H.; Myrback, K., Academic, New York, 1963; 227  44)  Noble, R. W.; Gibson, Q. H., / . Bio. Chem., 1970, 245, 2409.  45)  Yamazaki, H.; Yamazaki, I., Arch. Biochem. Biophys., 1973,154, 147.  46)  Wilterberg, J. B.; Noble, R. W.; Witterberg, B. A., / . Biol. Chem., 1967,242,  626.  206 47)  Goodbody, A.; Endo, TV, Vukovic, J.; Kutney, J. P.; Choi, L. S. L.; Misawa, M., Planta Med., 1988, 136.  48)  Kutney, J. P.; Boulet, C. A.; Choi, L. S. L.; Gustowdki, W.; McHugh, M ; Nakano, J.; Nikaido, T ; Tsudamoto, H.; Hewitt, G. M.; Suen, R., Heterocycles, 1988,27, 1827.  49)  Kutney, J. P.; Nakanom, J.; Boulet, C ; Tan, K.; Tui, S., unpublished results.  50)  Palaty, J., M. Sc., University of British Columbia, July 1990, Pg 30.  51)  Kutney, J. P.; Suen, R, unpublished results.  52)  Willstatter, R.; Pollings, A., Ann. Chem., 1923,430, 269.  53)  Kohler, H.; Jenzer, H., Free Rad. Biol. Med., 1989,6, 323.  54)  Kutney, J. P.; Gustowski, W.; Suen, R, unpublished results.  55)  Hewitt, G., Biological Services Department, UBC.  56)  Palaty, J., Master Thesis, University of British Columbia, July 1990, Pg 36.  57)  Brown, B. R., in Oxidative coupling of phenols, Eds. Taylor, W. I.; Battersby, A. R., Marcel Dekker, New York, 1967, Pg 167-201.  58)  Kalyanaraman, B.; Nemec, J.; Sinha, B. K., Biochemistry , 1989,28, 4839.  59)  Palaty, J., Master Thesis, University of British Columbia, July 1990, Pg 41.  60)  Ayres, D. C ; Farrow, A.; Carpenter, B. G., / . Chem. Soc. Perkin 1,1981,  2134. 61)  Palaty, J., Master Thesis, University of British Columbia, July 1990, Pg 34.  62)  Ziegler, F. E.; Schwartz, J.A., / . Org. Chem., 1978,43, 985.  63)  Jenzer, H.; Kohler, H.; Broger, C , Arch. Biochem. Biophys., 1987,258, 381.  64)  Schmidt, H. W.; Haemmerli, S. D.; Schoemaker, H. E.; Leisola, M. S., Biochemistry, 1989,28, 1176.  65)  Ziegler, F. E.; Schwartz, J. A., Tetrahedron Lett., 1975,52,4643.  66)  Kessar, S. V.; Gupta, Y. P.; Mohammad, T.; Goyal, M.; Sawal, K. K.J.Chem Soc.Chem. Commun., 1983, 400.  67)  Price, C. C ; Judge, J. M., Org. Synth., 1973, Coll. Vol. E, 255.  207 68)  Seebach, D.; Corey, E. J., / . Org. Chem., 1975,40, 231.  69)  Brown, E.; Daugan, A., Heterocycles, 1987,26, 1169.  70)  Pelter, A.; Ward, R. S.; Satyanarayana, P.; Collins, P., / . Chem. Soc. Per/an Trans. 1., 1983, 643.  71)  Tanoguchi, M.; Kashima, T.; Saika, H.; Inoue, T.; Arimoto, M.; Yamaguchi, H., Chem. Pharm. Bull, 1989, 37, 68.  72)  Kutney, J. P.; Nakano, J.; Boulet, C ; Tan, K.; Tiu, S., unpublished results.  73)  Sala, T.; Sargent, M.V., J. Chem. Soc. Perkin Trans., 1., 1979, 2593.  74)  Clark, J. H.; Holland, H. L.; Miller, J. M., Tetrahedron Lett., 1976, 3361.  75)  Hewitt G., Biological Services Department, UBC.  76)  Gow, S.; Ansell, S.; Kumey, J. P., unpublished results.  77)  a) Klingenberg, M., Arch. Biochem. Biophys., 1958, 75, 376. b) Garfinkel, D., Arch. Biochem. Biophys., 1958, 77, 493.  78)  Omura, T.; Sato, R., / . Biol. Chem., 1964, 7, 2370.  79)  Poulos, T. L., / . Biol. Chem., 1985,250, 1612.  80)  Guengerich, F. P.; Macdonald, T. L., Acc. Chem. Res., 1984,17, 9.  81)  Progress in Clinical and Biological Research., Ed. King, T. E.; Mason, H. S.; Morrison, M., 1987,274, Pg 297.  82)  Murphy, P. J.; West, C. A., Arch. Biochem. Biophys., 1969,133, 395.  83)  Rowell, P.; Potts, M.; Weklych, R.; Conn, E. E., / . Biol. Chem., 1974,249, 5019.  84)  Mansag, D.; Leclaire, J.; Fontecaue, M.; Momenteau, M., Biochem. Biophys. Res. Comm., 1984,119, 319.  85)  Mahler, H. R.; Cordes, E. H., Biological Chemistry, Harper and Row, New York, 1971.  86)  Trevan, M. D., Immobilized Enzymes, John Wiley and Sons, New York, 1982.  87)  Cytochrome P-450: Biochemistry and Biophysics. Ed. Schuster I., Pg 339, Taylor and Francis, London, 1989.  88)  Oxidases and Related Redox Sytems, Vol. II.,Eds. King, T.E.; Masar, H.S.;  Morrison, M., United Park Press, 1973. 89)  Scopes, R., Protein Purification, Ed. Cantor, C. R., Springer-Verlag, New York, 1982  90)  Cytochrome P-450; Biochemistry and Biophysics. Ed. Schuster, I., Pg 21, Taylor and Francis, London, 1989.  91)  Cleland, W. W., Biochemistry, 1964, 3, 480  92)  Tanoguchi, M.; Arimoto, M.; Saika, H.; Yamaguchi, H., Chem . Pharm. Bull., 1987, 35, 4162.  93)  Meehan, T. D.; Cosia, C. J., Biochem. Biophys. Res. Comm., 1973,53, 1043.  94)  Madgastha, K. M.; Meehan, T. D.; Cosia, C. J., Biochem., 1976,15, 1097.  95)  Spitsberg, V.; Cosia, C. J.; Krueger, R. J., Plant Cell Reports, 1981,1, 43.  96)  Gekko, K.; Timasheff, S. N., Biochemistry., 1981,20, 4677.  97)  Chu, A.; Kutney, J. P., unpublished results.  98)  Arimoto, M.; Kutney, J. P., unpublished results.  99)  Pugh, C. E. M.; Raper, H. S., Biochem. J., 1927,21, 1370.  100)  Schmidt, H. W.; Haemmerli, S. D.; Schoemaker, H. E.; Leisola, M. S., Biochemistry 1989,28, 1776.  101)  Mahelr, H. R.; Cordes, E. H., Biological Chemistry., Harper and Row, New York, 1971.  102)  Kondo, K.; Ogara, M.; Midorikawa, Y.; Kozawa, M.; Tsjibo, H.; Baba, K.; Inormori, Y., Agric. Biol. Chem., 1989,53, 111.  103)  a) Kadkade, P. G., Naturwissenschaften, 1981,88, 481. b) Kadkade, P. G., Proc. Plant Growth Regul. Soc. Am., 8th, 1981, 115, c) Kadkade, P. G., Plant Sci. Lett., 1982,25, 107.  104)  Hegenga, G. A.; Lucas, J. A.; Dewick, P. M., Plant Cell Reports, 1990,9, 382.  105)  van Uden, W.; Pras, N.; Visser, J. F.; Malingre, T. M., Plant Cell Reports, 1989, 8, 165  106)  Berlin, J.; Bedorf, N.; Mollenschott, C ; Wray, V.; Sasse, F.; Hofle, G.,Arens, PlantaMed, 1988,204.  107)  Kutney, J. P.; Gustowski, W., unpublished results.  Kutney, J. P.; Choi, L. S. L.; Duffin, R.; Hewitt, G.; Kawamuro, N.; Kurihara, T.; Salisbury, P.; Sindelar, R.; Stuart, K. L.; Townsley, P. M.; Chalmers, W. T.; Webster, R; Jacoli, G. G., Planta. Med, 1983,48, 158.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059760/manifest

Comment

Related Items