UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Bioactive natural products Woods, Katherine B. 2010

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

Item Metadata

Download

Media
24-ubc_2010_fall_woods_katherine.pdf [ 5.57MB ]
Metadata
JSON: 24-1.0059410.json
JSON-LD: 24-1.0059410-ld.json
RDF/XML (Pretty): 24-1.0059410-rdf.xml
RDF/JSON: 24-1.0059410-rdf.json
Turtle: 24-1.0059410-turtle.txt
N-Triples: 24-1.0059410-rdf-ntriples.txt
Original Record: 24-1.0059410-source.json
Full Text
24-1.0059410-fulltext.txt
Citation
24-1.0059410.ris

Full Text

Bioactive Natural Products by Katherine B. Woods B.Sc., University of British Columbia, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) June, 2010 © Katherine B. Woods, 2010  Abstract Montbretins A-E were isolated from the corms of Crocosmia sp., an invasive perennial plant. The montbretins are inhibitors of human pancreatic α-amylase (HPA). Montbretin A (230) is a competitive inhibitor of HPA with a Ki of 1.3 nM. The activity of the other family members varied significantly and provided structure-activity information. Saturation Transfer Difference (STD) NMR spectroscopy was used to determine that the caffeic acid region of the montbretins is important for binding. HPA is involved in the breakdown of complex carbohydrates; inhibition of this enzyme could help with regulation of blood sugar levels after a meal.  In the lungs of cystic fibrosis patients, the activation of Toll-Like Receptor 5 (TLR5) in the presence of flagellin leads to inflammation and obstruction. Girolline (3-1), a known alkaloid, was isolated from a Phonpeian sponge following potent inhibition of the flagellin initiated TLR5 activation. No activity was observed in any synthetic analogues of girolline. The massacreones are a new family of ecdysteroids isolated from an unidentified Dominican cnidarian. The extract of the cnidarian had good TLR5 activity, but the massacreones – namely massacreone A (3-25) and massacreone B (3-26) have only moderate activity and a small window of activity before they are toxic.  ii  The algal pigment caulerpin (4-29) was isolated from Caulerpa sp. as a compound showing good activity in a yeast growth restoration assay designed to identify inhibitors of human indoleamine-2,3-dioxygenase (IDO). Caulerpin did not show any activity in a freeenzyme IDO assay. IDO is involved in immune escape, which prevents the immunological rejection of tumors.  iii  Table of Contents Abstract..........................................................................................................................................ii Table of Contents..........................................................................................................................iv List of Tables...............................................................................................................................viii List of Figures................................................................................................................................ix List of Abbreviations.................................................................................................................xvii Acknowledgments.......................................................................................................................xxi Dedication...................................................................................................................................xxii 1. Introduction...............................................................................................................................1 1.1 Natural Products....................................................................................................................1 1.1.2 Terrestrial Natural Products........................................................................................1 1.1.3 Marine Natural Products.............................................................................................5 1.2 Natural Products as Potential Drugs and Biological Tools.................................................11 1.3 Structure Elucidation...........................................................................................................13 1.3.1 Mass Spectrometry ...................................................................................................13 1.3.2 Nuclear Magnetic Resonance Spectroscopy (NMR)...............................................14 1.4 Scope of Thesis...................................................................................................................15 2. The Montbretins......................................................................................................................18 2.1 Brief Introduction to Plant Natural Products......................................................................18 2.1.1 Flavonoids.................................................................................................................19 2.2 General Introduction to Crocosmia sp................................................................................24 2.2.1 Overview of Known Metabolites From Crocosmia sp.............................................25 2.3 Human Pancreatic α-Amylase.............................................................................................28 2.4 HPA as a Drug Target: Diabetes ........................................................................................31 2.4.1 Current Therapeutics For Diabetes...........................................................................33 2.4.1.1 Insulin Secretagogues.................................................................................33 2.4.1.2 Sensitizers...................................................................................................34 2.4.1.3 α-Glucosidase and α-Amylase Inhibitors...................................................35  iv  2.4.2 HPA Bioassay............................................................................................................37 2.5 Isolation of Inhibitors of HPA from Crocosmia sp.............................................................39 2.5.1 Isolation of Montbretins A, B and C.........................................................................40 2.5.2 Isolation of Montbretins D and E..............................................................................41 2.5.3 Structure Elucidation of Montbretin A......................................................................43 2.5.3.1 Caffeic Acid Residue..................................................................................56 2.5.3.2 Aglycone Portion .......................................................................................57 2.5.3.3 Rhamnose1.................................................................................................59 2.5.3.4 Glucose1.....................................................................................................64 2.5.3.5 Glucose2.....................................................................................................68 2.5.3.6 Xylose ........................................................................................................73 2.5.3.7 Rhamnose2.................................................................................................78 2.5.4 Structure Elucidation of Montbretin B......................................................................83 2.5.5 Structure Elucidation of Montbretin C......................................................................94 2.5.6 Structure Elucidation of Montbretin D...................................................................105 2.5.7 Structure Elucidation of Montbretin E....................................................................123 2.6 Biological Activity of the Montbretins.............................................................................132 2.6.1 STD NMR Binding Studies....................................................................................134 2.6.2 Animal Studies on Montbretin A............................................................................138 2.7 Discussion and Conclusions..............................................................................................143 2.8 Experimental.....................................................................................................................146 2.8.1 General Experimental Procedures...........................................................................146 2.8.2 Isolation of Montbretins A, B and C.......................................................................146 2.8.3 Montbretin A Physical Data....................................................................................148 2.8.4 Montbretin B Physical Data....................................................................................148 2.8.5 Montbretin C Physical Data....................................................................................148 2.8.6 Isolation of Montbretins D and E............................................................................148 2.8.7 Montbretin D Physical Data....................................................................................149 2.8.8 Montbretin E Physical Data....................................................................................149 2.8.9 Preparation of STD NMR Samples.........................................................................149  v  3. Potential Inhibitors of TLR5 – Isolation of Girolline and the Massacreones..................150 3.1 Cystic Fibrosis...................................................................................................................150 3.1.1 Treatment of Cystic Fibrosis...................................................................................151 3.1.2 Treatment of Inflammation.....................................................................................152 3.1.3 TLR5 Bioassay........................................................................................................154 3.2 Isolation of Girolline.........................................................................................................155 3.2.1 Structure Elucidation of Girolline..........................................................................157 3.2.2 Synthesis of Girolline and Analogues.....................................................................167 3.3 Isolation of Massacreones and 20-Hydroxy-Ecdysone.....................................................172 3.3.1 Structure Elucidation Of Massacreone A...............................................................175 3.3.2 Structure Elucidation of Massacreone B................................................................195 3.3.3 Structure Elucidation Of Massacreone C...............................................................207 3.3.4 Structure Elucidation of Massacreone D...............................................................218 3.3.5 Structure Elucidation Of 20-Hydroxy-Ecdysone .................................................230 3.4 Proposed Biosynthesis of the Massacreones.....................................................................231 3.5 Biological Activity............................................................................................................235 3.6 Discussion and Conclusions..............................................................................................237 3.7 Experimental.....................................................................................................................238 3.7.1 General Experimental Procedures...........................................................................238 3.7.2 Isolation of Girolline...............................................................................................239 3.7.3 Girolline Physical Data...........................................................................................240 3.7.4 Preparation of Acylated Oxazolidinone 2...............................................................241 3.7.5 Preparation of Protected Imidazole 3......................................................................242 3.7.6 Preparation of Aldehyde 8.......................................................................................243 3.7.7 Preparation of Amide 9...........................................................................................244 3.7.8 Preparation of Amide 10.........................................................................................245 3.7.9 Preparation of Alcohol 15.......................................................................................246 3.7.10 Preparation of Amide 16.......................................................................................247 3.7.11 Preparation of Amide 17.......................................................................................247 3.7.12 Preparation of Alcohol 18.....................................................................................248 3.7.13 Preparation of Alcohol 19.....................................................................................249  vi  3.7.14 Preparation of Amino Imidazole 22......................................................................249 3.7.15 Isolation of Massacreones A-D.............................................................................250 3.7.16 Massacreone A Physical Data...............................................................................251 3.7.17 Massacreone B Physical Data...............................................................................251 3.7.18 Massacreone C Physical Data...............................................................................252 3.7.19 Massacreone D Physical Data...............................................................................252 3.7.20 20-Hydroxy-Ecdysone Physical Data...................................................................252 4. Isolation of a Potential Inhibitor of IDO from Caulerpa sp. ..........................................253 4.1 Brief Introduction to Algal Metabolites............................................................................253 4.2 Brief Introduction to Indole Alkaloids..............................................................................254 4.3 General Introduction to Algae ..........................................................................................258 4.3.1 Introduction to Caulerpa sp....................................................................................258 4.3.2 Overview of Known Metabolites From Caulerpa sp..............................................259 4.4 Indoleamine-2,3-dioxygenase...........................................................................................262 4.4.1 IDO as a Drug Target..............................................................................................265 4.4.2 Inhibitors of IDO.....................................................................................................268 4.4.3 Yeast Based Screen for IDO Inhibition...................................................................272 4.5 Isolation of Caulerpin from Caulerpa sp..........................................................................273 4.5.2 Structure Elucidation of Caulerpin ......................................................................274 4.6 Discussion and Conclusions..............................................................................................283 4.7 Experimental.....................................................................................................................285 4.7.1 General Experimental Procedures...........................................................................285 4.7.2 Isolation of Caulerpin..............................................................................................286 4.7.3 Caulerpin Physical Data..........................................................................................286 5. Conclusions............................................................................................................................287 References...................................................................................................................................290 Appendix A. NMR Spectra Of Selected Compounds For Chapter 3...................................297 Appendix B. Biological Data For Compounds In Chapter 3................................................308  vii  List of Tables Table 2.1 1D and 2D NMR data for montbretin A (30), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4..............................................................................................................52 Table 2.2 Summary of 1D TOCSY data for montbretin A (30), recorded at 600 MHz in MeOD-d4....................................................................................................................................55 Table 2.3 1D and 2D NMR data for montbretin B (31), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4..............................................................................................................90 Table 2.4 1D and 2D NMR data for montbretin C (51), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4............................................................................................................101 Table 2.5 1D and 2D NMR data for montbretin D (52), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4.............................................................................................................112 Table 2.6 1D TOCSY data for montbretin D (52), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4 .....................................................................................................................115 Table 2.7 1D and 2D NMR data for montbretin E (53), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4............................................................................................................129 Table 2.8 Ki values for the inhibition of HPA by montbretins A-E............................................132 Table 2.9 Residual Activity of various glycosides after exposure to 1μM montbretin A...........133 Table 3.1 1D and 2D NMR data for girolline (1), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD- d4.....................................................................................................................166 Table 3.2 1D and 2D NMR data for massacreone A (25), recorded at 600 MHz (1H) and 150 MHz (13C) in Pyridine-d5........................................................................................................182 Table 3.3 1D and 2D NMR data for massacreone B (26), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4............................................................................................................202 Table 3.4 1D and 2D NMR data for massacreone C (27), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4............................................................................................................215 Table 3.5 1D and 2D NMR data for massacreone D (28), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4............................................................................................................227 Table 4.1 1D and 2D NMR data for caulerpin (29), recorded at 400 MHz (1H) and 150 MHz (13C) in CD2Cl2-d2.....................................................................................................................280  viii  List of Figures Figure 1.1 Some secondary metabolites from the plants that were found in a Neanderthal burial site at Shanidar Caves: ephedrine (l), chlorojanerin (2).............................................................2 Figure 1.2 Secondary metabolites responsible for the biological activity of some medicinal plants: morphine (3), caffeine (4), coniine (5) and quinine (6)...................................................3 Figure 1.3 Penicillin G (7)..............................................................................................................4 Figure 1.4 The origin of all small molecule drugs released onto the market between 1981 and 2006. ...........................................................................................................................................4 Figure 1.5 Marine natural products currently used clinically: Vira-A (8) and Ara-C (9) (based on the structures of spongothymadine (10) and spongouridine (11)), trabectedin (12), and ziconotide (13).............................................................................................................................7 Figure 1.6 Distribution of marine natural products by phylum, according to the MarinLit database (Total: 18,610 structures up to 2006)............................................................................8 Figure 1.7 Marine natural products first attributed to sponges, now attributed to microorganisms: xestodecalactone A (14), swinholide A (15), theopalauamide (16), okadaic acid (17), phenazine-1-carboxylic acid (18) and phenazine-1-carboxamide (19).....................10 Figure 2.1 Biosynthesis of the aromatic amino acids phenylalanine (6) and tyrosine (7) via the Shikimate pathway ...................................................................................................................20 Figure 2.2 Generation of p-coumaric acid (9), caffeic acid (10) and ferulic acid (11) from phenylalanine and tyrosine (7)..................................................................................................22 Figure 2.3 Biosynthetic route to the common plant flavonoids kaempferol (14), quercetin (15) and myricetin (16)..............................................................................................................23 Figure 2.4 Myricetin (16) is commonly found as its 3-O-rhamnoside........................................24 Figure 2.5 Medicagenic acid (17) and polygalic acid (18)...........................................................26 Figure 2.6 Tricrozarin A (19) and B(20).......................................................................................26 Figure 2.7 Saponins of polygalic acid isolated from corms of Crocosmia sp. Cocosmiosides A-I (21-29) and masonosides A-C (32, 33, 34).........................................................................27 Figure 2.8 Montbretin A (30) and B (31), isolated form from Crocosmia montbretia.................28 Figure 2.9 HPA catalyses the degradation of amylose (35) into isomaltose (36), maltotriose(37) and maltose(38) ........................................................................................................................29 Figure 2.10 Proposed double displacement mechanism to catalyze the hydrolysis of α(1→4) linkages in amylose by HPA........................................................................................30 Figure 2.11 Drugs used in the treatment of diabetes: insulin secretagogues such as the sulphonylureas (e.g. tolbutamide (39), acetohexamide (40), glibenclamide (41)) and the melitinides (e.g. repaglinide (42) and nateglinide (43)); sensitizers such as the biguanides (e.g. metformin (44), rosiglitazone (45) and pioglitazone (46))................................................35 Figure 2.12 α-Glucosidase inhibitors miglitol (47), voglibose (48) and acarbose (49). Acarbose is additionally an α-amylase inhibitor........................................................................36 Figure 2.13 2-Chloro-4-nitrophenyl-α-D-maltoside (50) is cleaved by HPA under bioassay conditions, yielding a chloronitrophenyl ion with a high extinction co-efficient (and thus shows strong absorbance) at 405 nm.........................................................................................38 Figure 2.14 Montbretins A-E (30, 31, 51, 52, 53)........................................................................43 Figure 2.15 Montbretin A (30)......................................................................................................44 Figure 2.16 600 MHz 1H NMR spectrum of montbretin A (30), recorded in MeOD-d4..............46 ix  Figure 2.17 150 MHz 13C NMR spectrum of montbretin A (30), recorded in MeOD-d4..............47 Figure 2.18 600 MHz COSY spectrum of montbretin A (30), recorded in MeOD-d4..................48 Figure 2.19 600 MHz HSQC spectrum of montbretin A (30), recorded in MeOD-d4..................49 Figure 2.20 600 MHz HMBC spectrum of montbretin A (30), recorded in MeOD-d4.................50 Figure 2.21 600 MHz TROESY spectrum of montbretin A (30), recorded in MeOD-d4.............51 Figure 2.22 Summary of HMBC and COSY correlations observed in the C1-C9 caffeic acid residue in montbretin A (30)......................................................................................................56 Figure 2.23 COSY and HMBC correlations in the two substructures (C12-C18 and C10&C19-C24) of the aglycone residue in montbretin A (30). .........................................................................58 Figure 2.24 COSY and HMBC correlations in the rhamnose1 residue of montbretin A (30)......60 Figure 2.25 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha1 spin system from Rha1-H-1 to Rha1-H-5, recorded after selective irradiation at δH 4.30 ppm (Rha1-H-2), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4...................61 Figure 2.26 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha1 spin system from Rha1-H-6 to Rha1-H-3, recorded after selective irradiation at δH 1.07 ppm (Rha1-H-6), after a delay of: a) 20 ms b) 40 ms c) 60 ms d) 80 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4..................62 Figure 2.27 Summary of key ROESY correlations in the rhamnose1 residue of montbretin A (30).............................................................................................................................................64 Figure 2.28 COSY and HMBC correlations in the glucose1 residue of montbretin A (30)........65 Figure 2.29 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc1 spin system from Glc1-H-1 to Glc1-H-4, recorded after selective irradiation at δH 4.56 ppm (Glc1-H-1), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4...................66 Figure 2.30 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc1 spin system from Glc1-H-6b to Glc1-H-4, recorded after selective irradiation at δH 4.20 ppm (Glc1-H-6b), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4..................67 Figure 2.31 Summary of ROESY correlations in the glucose1 residue of montbretin A (30)....68 Figure 2.32 HMBC and COSY correlations in the glucose2 residue in montbretin A (30).........69 Figure 2.33 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc2 spin system from Glc2-H-1 to Glc2-H-4, recorded after selective irradiation at δH 4.60 ppm (Glc2-H-1), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4................70 Figure 2.34 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc2 and Rha2 spin systems from Glc2-H-6a to Glc2-H-4 and from Rha2-H-6 to Rha2-H-2, recorded after selective irradiation at δH 3.95 ppm (Glc2-H-6a and Rha2-H-5), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4................................................................................72 Figure 2.35 Summary ROESY correlations in the glucose2 residue of montbretin A (30).........73 Figure 2.36 HMBC and COSY correlations in the xylose residue of montbretin A (30).............74 Figure 2.37 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Xyl spin system from Xyl-H-1 to Xyl-H-3, recorded after selective irradiation at δH 4.85 ppm (Xyl-H-1), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.............................75 Figure 2.38 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization x  along the Xyl spin system from Xyl-H-5a to Xyl-H-1, recorded after selective irradiation at δH 4.14 ppm (Xyl-H-5a), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4................................77 Figure 2.39 Summary of ROESY correlations in the xylose residue of montbretin A (30)........78 Figure 2.40 HMBC and COSY correlations in the rhamnose2 residue of montbretin A (30)......79 Figure 2.41 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha2 spin system from Rha2-H-6 to Rha2-H-3, recorded after selective irradiation at δH 1.28 ppm (Rha2-H-6), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4...................80 Figure 2.42 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha2 spin system from Rha2-H-1 to Rha2-H-2, recorded after selective irradiation at δH 4.82 ppm (Rha2-H-1), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4...................81 Figure 2.43 ROESY correlations in the rhamnose2 residue of montbretin A (30)......................82 Figure 2.44 Montbretin B (31)......................................................................................................83 Figure 2.45 600 MHz 1H NMR spectrum of montbretin B (31), recorded in MeOD-d4..............84 Figure 2.46 150 MHz 13C NMR spectrum of montbretin B (31), recorded in MeOD-d4.............85 Figure 2.47 600 MHz COSY spectrum of montbretin B (31), recorded in MeOD-d4..................86 Figure 2.48 600 MHz HSQC spectrum of montbretin B (31), recorded in MeOD-d4..................87 Figure 2.49 600 MHz HMBC spectrum of montbretin B (31), recorded in MeOD-d4................88 Figure 2.50 600 MHz TROESY spectrum of montbretin B (31), recorded in MeOD-d4.............89 Figure 2.51 COSY and HMBC correlations in the p-coumaric acid reside of montbretin B (31).............................................................................................................................................93 Figure 2.52 Montbretin C (51)......................................................................................................94 Figure 2.53 600 MHz 1H NMR spectrum of montbretin C (51), recorded in MeOD-d4..............95 Figure 2.54 150 MHz 13C NMR spectrum of montbretin C (51), recorded in MeOD-d4.............96 Figure 2.55 600 MHz COSY spectrum of montbretin C (51), recorded in MeOD-d4..................97 Figure 2.56 600 MHz HSQC spectrum of montbretin C (51), recorded in MeOD-d4..................98 Figure 2.57 600 MHz HMBC spectrum of montbretin C (51), recorded in MeOD-d4................99 Figure 2.58 600 MHz TOCSY NMR spectrum of montbretin C (51), recorded in MeOD-d4.. .100 Figure 2.59 Summary of the key COSY, HMBC and ROESY correlations in the ferulic acid ester of montbretin C (51)................................................................................................104 Figure 2.60 Montbretin D (52)...................................................................................................105 Figure 2.61 600 MHz 1H NMR spectrum of montbretin D (52), recorded in MeOD-d4............106 Figure 2.62 150 MHz 13C NMR spectrum of montbretin D (52), recorded in MeOD-d4...........107 Figure 2.63 600 MHz COSY spectrum of montbretin D (52), recorded in MeOD-d4................108 Figure 2.64 600 MHz HSQC spectrum of montbretin D (52), recorded in MeOD-d4...............109 Figure 2.65 600 MHz HMBC spectrum of montbretin D(52), recorded in MeOD-d4...............110 Figure 2.66 600 MHz TROESY spectrum of montbretin D (52), recorded in MeOD-d4...........111 Figure 2.67 HMBC and COSY correlations in the glucose3 residue of montbretin D (52)......118 Figure 2.68 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc3 spin system from Glc3-H-1 to Glc3-H-6a/b, recorded after selective irradiation at δH 5.05 ppm (Glu3-H-1), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. 1H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4................119 Figure 2.69 1D ROESY spectra of montbretin D (52) recorded after irradiation on Glc3-H-1 (top), H-17 (middle) and H-15 (bottom). Spectra recorded at 600 MHz in MeOD-d4...........120 Figure 2.70 2D JRES spectrum of montbretin D (52), recorded at 600 MHz in MeOD-d4. xi  The signal at δH 3.42 ppm is a triplet, J = 7.1 Hz.....................................................................121 Figure 2.71 Summary of ROESY correlations and key coupling constants in the glucose3 residue of montbretin D (52)..................................................................................................122 Figure 2.72 Montbretin E (53)....................................................................................................123 Figure 2.73 600 MHz 1H NMR spectrum of montbretin E (53) recorded in MeOD-d4.............124 Figure 2.74 150 MHz 13C NMR spectrum of montbretin E (53), recorded in MeOD-d4............125 Figure 2.75 600 MHz COSY spectrum of montbretin E (53), recorded in MeOD-d4................126 Figure 2.76 600 MHz HSQC spectrum of montbretin E (53), recorded in MeOD-d4................127 Figure 2.77 600 MHz HMBC spectrum of montbretin E (53), recorded in MeOD-d4...............138 Figure 2.78 Basic schematic of an STD NMR experiment, with shading used to indicate magnetization. a) Protein and ligand are mixed together. b) Protein is selectively saturated, and it transfers magnetization selectively to its ligands as shown in c). The protein will relax quickly. A 1D STD NMR spectrum is thus the difference between the 1H NMR spectra of a) and c).......................................................................................................................................134 Figure 2.79 Areas on the ligand that are close to areas on the protein will become more saturated due to NOE effects than areas that are not. The gray shading indicated the amount of magnetization, with increasing darkness indicating increasing saturation..........................135 Figure 2.80 1H NMR spectra, recorded at 600 MHz in MeOD-d4. a) HPA in 3:2 D2O/H2O with solvent suppression. b) STD spectrum of HPA and no ligand, on resonance at δH 1.7 ppm, off resonance at δH 40 ppm. c) HPA and montbretin B in 3:2 D2O/H2O with solvent suppression. d) STD spectrum of HPA and montbretin B, on resonance at δH 1.7 ppm, off resonance at δH 40 ppm...........................................................................................................137 Figure 2.81 1H NMR spectra, recorded at 600 MHz in MeOD-d4. Expansion of the aromatic portion of a) HPA and montbretin B in 3:2 D2O/H2O with solvent suppression. b) STD spectrum of HPA and montbretin B, on resonance at δH 1.7 ppm, off resonance at δH 40 ppm.................................................................................................................................138 Figure 2.82 Summary of observed biological activity (HPA inhibition) and STD data for the montbretins........................................................................................................................139 Figure 2.83 Plasma glucose levels in diabetic and control rats, with and without acarbose (top graph) or montbretin A (bottom graph)............................................................................141 Figure 2.84 Plasma glucose levels in diabetic rats given different doses of montbretin A.......143 Figure 3.1 Pohnpeian sponge 47643..........................................................................................156 Figure 3.2 Girolline (1)..............................................................................................................157 Figure 3.3 600 MHz 1H NMR spectrum of girolline (1), recorded in in MeOD-d4...................158 Figure 3.4 600 MHz 1H NMR spectrum of girolline (1), recorded in in MeOD-d4...................159 Figure 3.5 150 MHz 13C NMR spectrum of girolline (1), recorded in D2O...............................160 Figure 3.6 600 MHz COSY spectrum of girolline (1), recorded in D2O...................................161 Figure 3.7 600 MHz HSQC spectrum of girolline (1), recorded in D2O...................................162 Figure 3.8 600 MHz HMBC spectrum of girolline (1), recorded in D2O..................................163 Figure 3.9 Key substructure in structure elucidation of girolline (1) with expansion of the key region of the COSY spectrum...........................................................................................164 Figure 3.10 Substructure in structure elucidation of girolline (1) ...........................................165 Figure 3.11 Key HMBC correlations used to solve structure of girolline (1)...........................166 Figure 3.12 The synthesis of girolline(1) performed by Commerçon and Paris.......................167 Figure 3.13 The synthesis of girolline(1). Possible route to the carbon skeleton, with a Cl at the 2-position of the imidazole. ..........................................................................................168 Figure 3.14 The synthesis of girolline(1): possible route to form the amino-imidazole. ........169 xii  Figure 3.15 The synthesis of girolline(1): synthesis of the aldehyde, 8, from imidazole. .......169 Figure 3.16 Synthesis of the amide 10 and the model compound 15........................................170 Figure 3.17 The synthesis of analogues 16-19 of girolline(1). .................................................171 Figure 3.18 Synthesis of the 2-amino imidazole derivative 22.................................................172 Figure 3.19 The massacreones (25-28) and 20-hydroxy-ecdysterone (29), isolated from the Dominican cnidarian RJA06-08..............................................................................................174 Figure 3.20 Massacreone A (25)................................................................................................175 Figure 3.21 600 MHz 1H NMR spectrum of massacreone A (25), recorded in Pyridine-d5.....176 Figure 3.22 150 MHz 13C NMR spectrum of massacreone A (25), recorded in Pyridine-d5.....177 Figure 3.23 600 MHz COSY spectrum of massacreone A (25), recorded in Pyridine-d5..........178 Figure 3.24 600 MHz HSQC spectrum of massacreone A (25)), recorded in Pyridine-d5........179 Figure 3.25 600 MHz HMBC spectrum of massacreone A (25), recorded in Pyridine-d5........180 Figure 3.26 600 MHz TROESY spectrum of massacreone A (25), recorded in Pyridine-d5.....181 Figure 3.27 Substructures of massacreone A (25) determined from analysis of COSY and HSQC data...............................................................................................................................184 Figure 3.28 Structure elucidation of the A ring of massacreone A (25) – key HMBC correlations from the methyl group at position 19..................................................................186 Figure 3.29 Structure elucidation of the B ring of massacreone A (25) – key HMBC correlations from the the protons at positions 5, 9 and 11......................................................187 Figure 3.30 Structure elucidation of the C ring of massacreone A (25) – key HMBC correlations from protons at positions 7, 12 and 18...............................................................188 Figure 3.31 Structure elucidation of the D ring of massacreone A (25) – key HMBC correlations from protons at positions 15, 16, 17 and 18.......................................................188 Figure 3.32 Structure elucidation of the steroidal side chain of massacreone A (25) – key HMBC correlations from protons at positions 16, 21 and 22..................................................189 Figure 3.33 Structure elucidation of the steroidal side chain of massacreone A (25) – key HMBC correlations from protons at positions 23, 26, 27 and 28............................................190 Figure 3.34 Relative configuration of massacreone A (25): key ROESY correlations between the methyl protons H-19 and H-5, as well as between H-4α and H-7, establishing a cis ring fusion. The A ring flips, leading to broadening of the H-3 peak in the 1H NMR spectrum...................................................................................................................................191 Figure 3.35 Relative configuration of massacreone A (25): key ROESY correlations between H-19 and H-11β, as well as between H-2, H-4α and H-9. ......................................192 Figure 3.36 Relative configuration of massacreone A (25): key ROESY correlations between H-18 and H-11β and between H-9 and H-12α..........................................................192 Figure 3.37 Relative configuration of massacreone A (25): key ROESY correlations between H-18 and H-15β, H-18 and H-16α, H-12α and H-17, establishing the C/D ring fusion as trans..........................................................................................................................193 Figure 3.38 Massacreone A (25), and 20-hydroxy-ecdysone (29). The relative configuration at C-24 of massacreone A is unknown..............................................................194 Figure 3.39 Massacreone B (26) ...............................................................................................195 Figure 3.40 600 MHz 1H NMR spectrum of massacreone B (26), recorded in MeOD-d4........196 Figure 3.41 150 MHz 13C NMR spectrum of massacreone B (26), recorded in MeOD-d4.......197 Figure 3.42 600 MHz COSY spectrum of massacreone B (26), recorded in MeOD- d4...........198 Figure 3.43 600 MHz HSQC spectrum of massacreone B (26), recorded in MeOD- d4...........199 Figure 3.44 600 MHz HMBC spectrum of massacreone B (26), recorded in MeOD- d4..........200 Figure 3.45 600 MHz TROESY spectrum of massacreone B (26), recorded in MeOD- d4......201 xiii  Figure 3.46 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations from H-22 and H-30, allowing the placement of the C-29/C-30 acetate at position 21...............................................................................................................204 Figure 3.47 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations from H-21 and H-23. .............................................................................205 Figure 3.48 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations used to position a gem-di substituted double bond in the steroidal side chain. ..............................................................................................................................205 Figure 3.49 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations used to place the C25-C27 group................................................................206 Figure 3.50 Massacreone B (26). The relative configuration at C-25 is unknown..................207 Figure 3.51 Massacreone C (27)................................................................................................208 Figure 3.52 A comparison of the 13C NMR spectra of massacreone A (top) and massacreone C (bottom). Relevant differences in the spectra are marked with *..................209 Figure 3.53 600 MHz 1H NMR spectrum of massacreone C (27), recorded in MeOD-d4........210 Figure 3.54 150 MHz 13C NMR spectrum of massacreone C (27), recorded in MeOD-d4........211 Figure 3.55 600 MHz COSY spectrum of massacreone C (27), recorded in MeOD-d4............212 Figure 3.56 600 MHz HSQC spectrum of massacreone C (27)), recorded in MeOD-d4..........213 Figure 3.57 600 MHz HMBC spectrum of massacreone C (27), recorded in MeOD-d4...........214 Figure 3.58 Structure elucidation of massacreone C (27): key HMBC and COSY correlations in the side chain..................................................................................................................217 Figure 3.59 Massacreone C (27). The relative configuration at C-24 is unknown. ................218 Figure 3.60 Massacreone D (28)................................................................................................218 Figure 3.61 A comparison of the 150 MHz 13C NMR spectra of massacreone A (top) and massacreone C (bottom), recorded in MeOD-d4. Relevant differences in the spectra are marked with *..........................................................................................................................220 Figure 3.62 600 MHz 1H NMR spectrum of massacreone D (28), recorded in MeOD-d4.......221 Figure 3.63 150 MHz 13C NMR spectrum of massacreone D (28), recorded in MeOD-d4.......222 Figure 3.64 600 MHz COSY spectrum of massacreone D (28), recorded in MeOD-d4............223 Figure 3.65 600 MHz HSQC spectrum of massacreone D (28)), recorded in MeOD-d4..........224 Figure 3.66 600 MHz HMBC spectrum of massacreone D (28), recorded in MeOD-d4..........225 Figure 3.67 600 MHz TROESY spectrum of massacreone D (28), recorded in MeOD-d4.......226 Figure 3.68 Structure elucidation of massacreone D (28): key HMBC correlations in the side chain.................................................................................................................................229 Figure 3.69 Massacreone D (28). The relative configuration at C-24 is unknown..................230 Figure 3.70 20-hydroxy-ecdysone (29).....................................................................................230 Figure 3.71 Ecdysteroids from cnidarians: ajugasterone C (30), gerardiasterone (31), 2deoxyecdysterone (32), 4-dehydroecdysterone (33), palythoalones A (34) and (35), zoanthusterone (36).................................................................................................................232 Figure 3.72 Biosynthesis of ecdysone (37) in insects and crustaceans (via cholesterol, (38)), or in plants (via lathosterol (39)). An additional oxidation step at C-20 yields 20hydroxy-ecdysone (29)............................................................................................................233 Figure 3.73 Proposed biosynthetic transformations from 20-hydroxy-ecdysone (29) to the massacreones...........................................................................................................................234 Figure 3.74 Results of the TLR5 bioassay for massacreone A (left) and massacreone B (right). .....................................................................................................................................236 Figure 4.1 Natural products from macro algae: kahalide F (1), a potent anti tumor xiv  compound, and frimbrolide I (2), one of seventeen known frimbrolides, which are antifouling compounds............................................................................................................254 Figure 4.2 Biosynthetic pathway from chorismic acid (4) to tryptophan (3).............................255 Figure 4.3 Various indole alkaloids with biological activity: lysergic acid diethylamide (7), ergine (8), bufotenin (9), psilocybin (10), dimethyltryptamine (11), melatonin (12), serotonin (13), strychnine (14) , vinblastine (15), and yohimbine (16)..................................256 Figure 4.4 Auxins, plant hormones: Indole-3-acetic acid (17), indole-3-butyric acid (18), 4chloroindole-3-acetic acid (19), indole-3-propinoic acid (20), indole-pyruvic acid (21), indole-3-acrylic acid (22)........................................................................................................257 Figure 4.5 Caulerpol (23), flexilin (24) and trifarin (25)............................................................260 Figure 4.6 Caulerpenyne (26) is converted to oxytoxins 1 (27) and 2 (28) in a process that results from wounding, in Caulerpa taxifola........................................................................261 Figure 4.7 Caulerpin (29), caulerpinic acid (30) and caulersin (31)...........................................261 Figure 4.8 The Kynurenine pathway. Tryptophan is degraded into quinolinic acid (37), a precursor to NAD....................................................................................................................263 Figure 4.9 Possible mechanism for formation of N-formyl-kynurenine from Tryptophan, as catalyzed by IDO. ..................................................................................................................265 Figure 4.10 IDO inhibitors norharman (40), 3-butyl-β-carboline (41), 1-methyl tryptophan (42) and necorostatin (43) ...............................................................................................269 Figure 4.11 Basic schematic of bioassay used to identify numerous inhibitors of IDO.............270 Figure 4.12 IDO inhibitors identified from marine organisms: annulin A (46), B (47) and C (48), graveatin A (49), C (50) and E (51), 2-hydroxygraveatin A (52), 2-hydroxygarvin E (53), exiguamine A (54) and B (55), and plectosphaeroic acid A (56), B (57) and C (58).....271 Figure 4.13 Caulerpin (29), with numbering scheme................................................................274 Figure 4.14 400 MHz 1H spectrum of caulerpin (29), recorded in CD2Cl2-d4...........................275 Figure 4.15 100 MHz 13C spectrum of caulerpin (29), recorded in DMSO-d6..........................276 Figure 4.16 400 MHz COSY spectrum of caulerpin (29), recorded in CD2Cl2-d4.....................277 Figure 4.17 400 MHz HSQC spectrum of caulerpin (29), recorded in CD2Cl2-d4, with inset expansion of H-13/C-13 (methoxy) data.................................................................................278 Figure 4.18 400 MHz HMBC spectrum of caulerpin (29), recorded in CD2Cl2-d4, with inset expansion of aromatic region. ...............................................................................................279 Figure 4.19 Substructures of caulerpin (29)..............................................................................281 Figure 4.20 Key COSY and HMBC correlations in caulerpin (29)...........................................282 Figure 4.21 ORTEP diagram of caulerpin (29)..........................................................................283 Figure 4.22 A comparison of the structures of tryptophan (3) and caulerpin (29) ...................284 Figure A.1 1H and 13C NMR of 2, recorded in CDCl3-d3 at 400MHz and 100MHz, respectively. ...........................................................................................................................298 Figure A.2 1H and 13C NMR of 8 recorded in CD2Cl2-d4 at 400MHz and 100MHz, respectively..............................................................................................................................299 Figure A.3 1H and 13C NMR of 9 recorded in (CD3)2CO-d6 at 400MHz and 100MHz, respectively..............................................................................................................................300 Figure A.4 1H and 13C NMR of 10 recorded in CD2Cl2-d4 at 400MHz and 100MHz, respectively..............................................................................................................................301 Figure A.5 1H and 13C NMR of 15 recorded in CD2Cl2-d4 at 400MHz and 100MHz, respectively..............................................................................................................................302 Figure A.6 1H and 13C NMR of 16 recorded in MeOD-d4 at 400MHz and 100MHz, respectively..............................................................................................................................303 xv  Figure A.7 1H and 13C NMR of 17 recorded in MeOD-d4 at 400MHz and 100MHz, respectively..............................................................................................................................304 Figure A.8 1H and 13C NMR of 18 recorded in D2O-d2 at 400MHz and 100MHz, respectively..............................................................................................................................305 Figure A.9 1H and 13C NMR of 19 recorded in D2O-d2 at 400MHz and 100MHz, respectively..............................................................................................................................306 Figure A.10 1H and 13C NMR of 22 recorded in MeOD-d4 at 400MHz and 100MHz, respectively..............................................................................................................................307 Figure B.1 Bioassay data for the natural product fraction from which giroline(1) was isolated..............................................................................................................................309 Figure B.2: Bioassay data, compound (16)................................................................................310 Figure B.3: Bioassay data, compound (17)................................................................................311 Figure B.4: Bioassay data, compound (18)................................................................................312 Figure B.5: Bioassay data, compound (19)................................................................................313 Figure B.6: Bioassay data, crude reaction mixture, compound (22) prior to recrystallization..314 Figure B.7: Bioassay data, pure, recrystallized compound (22)................................................315 Figure B.8: Bioassay data, massacreone C (27)........................................................................316 Figure B.9: Bioassay data, massacreone D (28)........................................................................317 Figure B.10: Bioassay data, 20-hydroxy-ecdysone (29)............................................................318  xvi  List of Abbreviations 1D  - one dimensional  2D  - two dimensional  1-MT  - 1-methyl tryptophan  [α]  t D  - specific rotation at wavelength of sodium D line at temperature t (°C)  A600nm  - the absorbance at 600 nm  Abg  - Agrobacterium sp.  Ac  - acetate  ATP  - adenosine triphosphate  BCE  - before the common era  BH3-Me2S  - borane-dimethyl sulphide  BOM  - benzyloxymethyl  BOMCl  - benzyloxymethyl chloride  br  - broad  Bu  - butyl, C4H9-  Bu2BOTf  - dibutylboron triflate  calcd  - calculated  cDNA  - copy DNA  CE  - common area  CFTR  - cystic fibrosis transmembrane conductance regulator  CoA  - coenzyme A  COSY  - two-dimensional correlation spectroscopy  d  - doublet  D  - dextrorotatory  dd  - doublet of doublets  DMF  - N, N-dimethylformamide  DMSO  - dimethyl sulfoxide  DMSO-d6  - Deuterated dimethyl sulfoxide  dt  - doublet of triplets  EC50  - half maximal effective concentration  Et  - ethyl, C2H5-  ESI  - electrospray ionization xvii  g  - gram(s)  Glc  - glucose  hr  - hour  HMBC  - two-dimensional heteronuclear multiple bond coherence  HPA  - human pancreatic α-amylase  HPLC  - high performance liquid chromatography  HRESI-TOFMS  - high resolution electrospray ionization time of flight mass spectrometry  HRMS  - high resolution mass spectrometry  IC50  - inhibitory concentration (for 50% of the biological sample)  IDO  - indoleamine-2,3-dioxygenase  J  - coupling constant (expressed in hertz)  k  - kilo-  kbp  - kilobase pairs  Ka  - acid dissociation constant  Ki  - dissociation constant for an inhibitor  Km  - Michaelis-Menten coefficient  L  -  μ  - micro-  m  - multiplet  m  - milli-  M  - molar concentration in moles per liter  M  - mega-  m/z  - mass to charge ratio  Me  - methyl, CH3-  MeOD-d4  - Deuterated methanol  MeOH  - methanol  min  - minute(s)  mol  - mole  mRNA  - messenger RNA  MS  - mass spectrometry  mult  - multiplicity  n  - nano-  n-BuLi  - n-butyl-lithium  levorotatory  xviii  NCI  - National Cancer Institute  NCS  - N-chlorosuccinimide  NEt3  - triethylamine  nM  - nanomolar  NMR  - nuclear magnetic resonance  NOE  - nuclear Overhauser enhancement  NOESY  - nuclear Overhauser enhancement spectroscopy  o  C  - degrees Celsius  PDA  - photodiode array  pH  - -log[H+]  pKa  - -log[Ka]  PLP  - Pyridoxal-phosphate  ppm  - parts per million  q  - quartet  Rt  - retention time  Rf  - retention factor  Rha  - rhamnose  ROESY  - rotating frame Overhauser enhancement spectroscopy  RT  - room temperature  s  - singlet  SAM  - S-adenosyl methionine  SCUBA  - self contained underwater breathing apparatus  sep  - septet  SGF  - simulated gastric fluid  SIF  - simulated intestinal fluid  sp.  - species  STD  - saturation transfer difference  t  - triplet  TDO  - tryptophan-2,3-dioxygenase  THF  - tetrahydrofuran  TLC  - thin layer chromatography  TLR  - Toll-like receptor  TOCSY  - Total correlation spectroscopy  xix  UPLC  - ultra performance liquid chromatography  UV  - ultra violet  UV-VIS  - ultra violet and visible  Xyl  - xylose  xx  Acknowledgments I would like to thank my supervisor, Professor Raymond J. Andersen, for his seemingly limitless generosity, for providing numerous opportunities to me in a scientific field so ripe with reward, and for his patience with me through numerous successes and failures. I would also like to thank Dr. David Williams always having valuable input and ideas on how to overcome roadblocks in the lab and in life. Thanks as well to Michael LeBlanc, for helping connect me to resources I needed, for countless diving stories shared, and to bottom time spent chasing yellow fins. I sincerely appreciate the aid and efforts of the guidance of Dr. Nick Burlinson, Dr. Zhicheng (Paul) Xia, and Zorana Danilovic for assistance with various NMR experiments. I am similarly grateful for the biological assessments carried out by my numerous collaborators, including Dr. Christoper Tarling, Ran Zhang and Dr. Stephen Withers (Department of Chemistry, UBC); Rachel Victor, Aaron Hirschfeld and Dr. Stuary Turvery (BC Child and Family Research Institute); Dr. Eduardo Vottero, Leslie Williams and Professor Grant Mauk (Biochemistry and Molecular Biology, UBC); Dr John Coleman, Chris Hii, Pamela Lincez (UBC Center for Disease Control); Dr. Aruna Balgi and Dr. Michael Roberge (Biochemistry and Molecular Biology, UBC). These highly rewarding partnerships have shown me the intrinsic power in collaborative science. To the members of the Andersen lab, past and present, a heartfelt thank you for the amazing friendship and also, for the demonstration of the excellence that may be achieved by scientists at all levels of academia. To my lab dive buddies, thank you for amazing underwater collecting synergy and so awesome underwater moments. Finally, thank you to my husband, Dr. Ryan Philippe, for ridiculous amounts of love and support through all of this.  "A journey is a person in itself; no two are alike. And all plans, safeguards, policing, and coercion are fruitless. We find that after years of struggle that we do not take a trip; the trip takes us.” John Steinbeck  xxi  Dedication  “Somewhere, something incredible is waiting to be known.” - Carl Sagan For my Dad, and for my Mum.  xxii  1. Introduction 1.1 Natural Products Quite simply stated, a natural product is something from a natural origin. This can be anything from a whole organism, a part of a whole organism, an extract from an organism, or the pure compounds that comprise the metabolome of an organism.1,2  A metabolome is the complete set of small molecules, or metabolites, found in a biological sample. A good portion of a metabolome represents the primary metabolites metabolites involved in functions key to an organism's normal survival and fecundity. 1 The remainder of the metabolome is formed by secondary metabolites, which can be thought of as metabolites with more of an ecological purpose - for example defense, competition or attracting a mate. Generally, when one speaks of “natural products” the assumed meaning is pure secondary metabolites isolated from a natural source.3  The diversity of work done in natural products chemistry is a testament to the diverse backgrounds of the scientists who study it. Natural products can be potential drugs4, or leads for new drugs5; they can be molecules of interest in defining an ecological interaction6,7; they can be the answers to questions posed by ethnobotanists8,9; and they can provide a tool for understanding biosynthesis1,10 or the workings of various enzymes.11-13  1.1.2 Terrestrial Natural Products Humanity has a long and rich history of using natural products for therapeutic and/or 1  recreational purposes. The earliest record of what are presumably medicinal plants dates back to 60,000 BCE, to a burial site in Shanidar Caves (located in present day Iraq). One of the grave sites in this archaeological site was found to contain the skeletal remains of a Neanderthal man, laid atop a cocktail of plant pollens.2 Among these plants were Ephedra altissima (source of ephedrine14 (1)), Centaurea solstitialis (source of neurotoxins and anti-ulcerogenics such as chlorojanerin15 (2)), as well as other known medicinal plants such as Althea sp. (Mallow root), that have therapeutic use not directly associated with a particular small molecule natural product.2  OH  H N  CH3  H O  HO HO Cl  H  OH  H O O O  1  2  Figure 1.1 Some secondary metabolites from the plants that were found in a Neanderthal burial  site at Shanidar Caves: ephedrine (l), chlorojanerin (2).  Numerous preserved ancient texts from a number of cultures worldwide describe the use of natural products for medicinal purposes. The oldest known is the “Ebers Papyrus”, an Egyptian medical handbook detailing hundreds of treatments for various diseases and ailments, dating back to 1550 BCE.16 Other notable examples include the Greek Corpus Hippocraticum (ca. 400 BCE)17 and the Chinese Shen nong ben cao jing (ca. 300 BCE to 200 AD).18 By the middle ages, many of these texts had been adopted by the Arabic empire, and then passed back to Europe through the hands of monks, who diligently copied and preserved this knowledge, using  2  it to cultivate medicinal plant gardens for learning and for curing. In the 17th and 18th centuries there was significant interest in distilling out the active ingredients from plants, but it wasn't until the early 19th century that it became clear that the properties of some medicinal plants could be attributed to specific molecules that could be isolated and characterized.2 Early compounds isolated include morphine (3)19,20, an analgesic from Papaver somniferum, caffeine (4)21, a stimulant from the coffee shrub, coniine (5)22, a potent toxin from the bark of Conium sp., and quinine (6)23, an anti-inflammatory analgesic from Cinchona sp. Although isolated and understood to be the active components from their respective plants, the structures of all of these early isolated natural products were not elucidated untill the end of the 19th and early 20th centuries.  HO N O  CH3  O H3C O  H  N  CH3 N  N N CH3  OMe H N H  OH N  HO 3  4  N  5  6  Figure 1.2 Secondary metabolites responsible for the biological activity of some medicinal plants: morphine (3), caffeine (4), coniine (5) and quinine (6).  When the antibiotic fungal metabolite penicillin G (7) was discovered serendipitously by Florey and Fleming in 192824, the field of natural products grew to accommodate the concept of finding drugs or potential drugs from sources other than traditional medicinal plants. Moreover, the field grew to accommodate the concept that natural products may have uses that the producing organism is not known for – a removal of the bias built into isolating the active 3  constituents from medicinal plants. This, along with advances in synthetic organic chemistry, allowed natural products to become a cornerstone of the drug discovery industry.2 H N O  H  S  N OH 7  O  Figure 1.3 Penicillin G (7).  As shown in Figure 1.4, over 50% of the drugs currently available on the market are a natural product or are derived from or inspired by a natural product.3  Figure 1.4 The origin of all small molecule drugs released onto the market between 1981 and 2006.  4  1.1.3 Marine Natural Products Before the mid 20th century, the marine environment was only really accessible via surface supplied diving or skin diving, limiting the potential of exploration to specific tasks and/or specific depths. With the invention of the demand regulator driven Aqua Lung by Jacques-Yves Cousteau and Emile Gagnon in 1943, exploration of this new frontier became possible and advantageous in a number of fields, including natural products chemistry.25  Despite a long and rich list of successes from the terrestrial environment, the search for new and more potent drugs continues. Advances in molecular biology and biochemistry continue to unveil new potential targets for drugs, as well as illustrating the need for more specific tools to probe the functions of these new targets. Furthermore, issues like antibiotic resistance necessitate a constant lineup of newer drugs in order to keep one step ahead of certain infections. And while the terrestrial realm has been probed for centuries, the marine environment represents an uncharted frontier, and one that continues to expand as submersible technology evolves.26  Marine organisms live in an environment that occupies over 70% of the Earth's surface. Looking at volume, the oceanic environment represents an area many times more inhabitable than land, particularly when considering marine microorganisms. It has been estimated that a single drop of sea water contains over 100,000 microorganisms. Moreover, the diversity in many phyla of marine organisms is without compare to anything seen in the terrestrial environment. In particular, invertebrates such as Colenterata, Porifera, Bryozoa and Echinodermata are completely aquatic (and largely marine). Since there is a rough correlation between species 5  diversity and chemical diversity in secondary metabolites, one would imagine that the marine environment is a rich storehouse for untapped potential in natural products chemistry.26  Many marine organisms are sessile or slow moving, and have no hard protective outer shell or claws to provide physical protection. And yet, despite being oftentimes brightly colored, they survive and flourish, seemingly immune to predation. Alternative forms of defense employed by such organisms are largely chemical in nature - chemical warfare is rampant on the sea floor, with soft corals and sponges producing anti-feedants, marine bacteria producing powerful antibiotics, and nudibranchs sequestering defense chemicals to make themselves unpalatable to predators. As one might imagine, such a rich and varied mosaic of secondary metabolites might afford compounds with medicinal or commercial properties.25  In fact, a surprisingly wide range of chemical structural classes have been observed in extracts of marine organisms, some completely unique to the marine environment. Of these compounds, many have been found to have interesting bioactivity, and some have been found to be active enough to warrant their attempted development (or development of a closely related compound) into pharmaceuticals.27 Early marine natural product drug successes include Vira-A (8), an antiviral drug used clinically for the treatment of herpes, and Ara-C (9), a chemotherapy drug used clinically in the treatment of leukemia.26,28 Vira-A and Ara-C are related to the marine sponge compounds spongothymadine (10) and spongouridine (11).29,30 Ziconotide (12), which is sold under the brand name Prialt®, is an ω-conotoxin MVIIA found in the marine fish-hunting cone snail Conus magus31, and is currently in use as a treatment for chronic pain due to spinal cord injury.32 Trabectedin (13), sold under the brand name Yondelis, was isolated from the sea squirt Ecteinascidia turbinata33,34 and has recently been approved as an orphan drug in the EU 6  and US for treatment of soft tissue sarcomas and ovarian cancer.32 NH2  NH2  N  N  N HO  N  N  O HO  HO  N  O NH HO  O  N  8  NH HO  O  OH  OH  9  10  OH 11 Cys Lys  HO  Gly Ala  NH O O O  O  OH  12  Gly  Lys  Cys  N O  Lys  Cys  O N  O  Cys  HO  S  O  O HO  H2N  OO  N  O HO  O HO  OH  O  H2 N  Thr Lys  Gly  Asp Ser Cys Ser  Gly  Ser  Arg  Tyr  Arg  Leu Met  Cys  13  Figure 1.5 Marine natural products currently used clinically: Vira-A (8) and Ara-C (9) (based on the structures of spongothymadine (10) and spongouridine (11)), trabectedin (12), and ziconotide (13).  An investigation into the literature published up to the start of 200935 shows some interesting trends in compounds being reported in marine natural products research (see Figure 1.6). The phylum Porifera (sponges) has historically been the most prolific (and that remains the case today) but increasingly marine natural products are being reported from marine microorganisms. Furthermore, a growing number of studies are being published that look into  7  whether marine natural products attributed to sponges are sponge natural products or, in fact, attributable to endobiotic bacterial symbionts. This question will likely continue to be one of high interest in the next decade, particularly as culturing techniques mature. Perhaps we will eventually find that most, if not all, sponge natural products are actually from microbial symbionts. Similarly, a good portion of marine natural products reported are from cnidarians, which are also known to be hosts for microbial symbionts, and from algae, which is a wide ranging group of organisms encompassing everything from unicellular to multicellular organisms that often boast epibiotic symbionts.  Figure 1.6 Distribution of marine natural products by phylum, according to the MarinLit database (Total: 18,610 structures up to 2006).  Indeed, examples of marine natural products first attributed to sponges that were later isolated from microorganisms certainly exist. Xestodecalactone A (14) has been shown to originate from a Penicillium fungus found in Xestospongia exigua36, and both swinholide A (15)  8  and theopalauamide (16) have been shown to originate from bacterial and cyanobacterial cells found in Theonella swinhoei.37 Okadaic acid (17), originally isolated from Halichondria okadai has been shown to be from a symbiotic dinoflagellate.38 Pseudomonas aeruginosa, an endosymbiont of the antarctic sponge Isodictya setifera, was successfully cultured outside of its host sponge and subsequently found to be the organism responsible for the interesting bioactivity found in a crude extract of the sponge (namely, two antibiotic phenazine alkaloids, phenazine-1carboxylic acid (18) and phenazine-1-carboxamide (19)).39 With rapidly advancing culturing and genetic techniques, it is likely that many future studies into invertebrates with interesting biological activities will result in isolation of natural products that can be credited to endosymbiotic microbes.40  9  OMe OH O O HO  14  O O  CH3  OH OH  OH O  -  OOC O NH O OH HO OH O OH HO HN O HN N O H HN OH O OH N HN O O NH N HN2OC O Ph NH O H H OH N N NH O N H O O HN2OC HO HO  Br  OMe O  O  OH  OH O O  MeO O  HO  OH OH  O  OMe  15 O  OH  O  NH2  N  16 N  Ph  18  OH  O O  O  HO OH  O OH  H  O  O O OH  17  N N 19  Figure 1.7 Marine natural products first attributed to sponges, now attributed to microorganisms: xestodecalactone A (14), swinholide A (15), theopalauamide (16), okadaic acid (17), phenazine-1carboxylic acid (18) and phenazine-1-carboxamide (19).  10  1.2 Natural Products as Potential Drugs and Biological Tools While it is true that not every secondary metabolite isolated has a known biological activity, an overwhelming number are discovered via the use of some sort of directed bioassay or biological test that looks for a particular cellular response that is believed to have some sort of potential pharmaceutical application. These bioassays can range in complexity from very simple cytotoxicity tests such as the brine shrimp assay, to more specific and expensive in vitro screens that use purified receptors, to cell based bioassays that utilize genetically engineered eukaryotic cells or microorganisms.  In the majority of pharmaceutical based research, bioassays are designed with a specific therapeutic goal in mind – occasionally, however, bioassays identify potential drug leads that are not appropriate for development into a drug. Often, issues of availability pose a significant problem – rare is the natural product that is a potential drug that may be isolated reliably and in a sustainable manner from its host organism. Moreover, many natural products identified in screens for potential pharmaceuticals are not optimally designed for use as an inhibitor (or activator) of a target. Issues such as solubility in water (the medium of the human body), pose a significant challenge to overcome in drug development.41 As well, the natural product that is found to interact with a particular target is rarely optimized to that target, with synthetic modifications potentially improving upon any observed biological activity, and certainly illuminating the structural motifs that result in the observed biological activity. 42 As such, natural products often become lead structures to improve upon and/or synthesize in order to allow further progress towards pharmaceutical development. Additionally, many natural products identified as having interesting biological activity may be inappropriate for development into a  11  drug (due to undesirable side effects, for example) but are important tools for use in understanding the biology behind whatever condition the bioassay target is involved with. Finally, it should be noted that bioactive natural products identified and purified from a complex mixture such as a crude extract are occasionally discovered to not actually be the active compound, with potent 'tag along' molecules being responsible for observed activity, or also in some cases, a fundamental flaw in the logic of the bioassay resulting in the identification of a compound that is not actually bioactive.43  Recently, many drug discovery efforts have embraced the use of robotics combined with target based bioassays in a process that is called high throughput screening. High throughput screening allows large numbers of compounds (or extracts, or mixtures of compounds) to be tested in an automated fashion as inhibitors (or alternatively, activators) of particular molecular targets. A dedicated high throughput screening program can allow 3000 enzyme inhibition or 4800 cell based bioassays per day, which in turn allows for evaluation of up to 100,000 chemical entities in as little as a month.44 Thus, a high throughput screen can easily allow for rapid testing of chemical libraries, such as the National Cancer Institute (NCI)'s open plant repository (a collection of crude plant extracts), or any in-house natural extract library, such as the collection of extracts of marine organisms from the Andersen laboratory.  All three of the studies discussed herein focus on the isolation and characterization of natural products – both from marine origin and from plant origin – based on their biological activity, as identified by high throughput screening efforts and also in newly developed bioassays not yet adapted to a high throughput process. Each study demonstrates a different result in the process of discovering a biologically active natural product. 12  1.3 Structure Elucidation Once a natural product is isolated, its structure must be determined. Historically, this task was far from trivial, with complex series of degradation experiments and analysis of changes in easily observed properties giving clues as to the complete structure, which could then be confirmed by total synthesis. In some cases, isolated natural products may be crystallized and thus their structure elucidated by X-ray crystallography – however, this presupposes that good quality crystals may be grown from the natural product. This is often the limiting factor with this method. Currently, there are two analysis tools that are most vital to structure elucidation of natural products – mass spectrometry (MS) and nuclear magnetic resonance (NMR).  1.3.1 Mass Spectrometry MS is often one of the first key steps in structure elucidation. As well, it provides a good tool to aid in isolation efforts and also in dereplication. Sometimes a particular mass (or family of masses) may be followed during the isolation process, either for reisolation of a compound or for simple chemical prospecting. Additionally, simple database searches for compounds with a particular mass from a particular organism can sometimes allow for rapid identification of molecules in an extract. For an unknown compound, MS can provide clues about the presence (or absence) of atoms that naturally occur as mixtures of isotopes and the possible number of heteroatoms in a molecule. In conjunction with NMR data, high resolution MS (HRMS) allows the determination of a molecular formula, which in turn allows for the resolution of degrees of unsaturation in a molecule. Although most of the structure elucidation can be accomplished without MS data, it is not possible to propose a structure without confirmation of its mass, and thus, molecular formula.45  13  1.3.2 Nuclear Magnetic Resonance Spectroscopy (NMR) Since its inception in the 1940's, NMR has been a tool that provides information into the structure of molecules.46 With the production of increasingly powerful instruments and newer, more sophisticated, pulse sequences, NMR has become a powerful tool of increasing importance in structure elucidation. NMR based structure elucidation has almost completely replaced degradation studies47, and in some cases has removed the necessity of using total synthesis to prove a structure.45  Basic 1D NMR techniques – such as a simple 1H NMR spectrum or a proton decoupled 13C spectrum - have benefited from instrument improvements such as increasingly powerful magnets and sophisticated probes capable of handling smaller samples. These improvements have allowed for increased dispersion of 1D NMR spectra, and have decreased the amount of sample (or alternatively, of time) needed to collect a spectrum. Additionally, improvements in instrumentation, plus the advent of increasingly sophisticated pulse sequences, have allowed NMR based investigations into the binding of small molecules to receptors.  To answer complex questions about structure, or to access information about multi-bond coupling, advanced (and more powerful) 2D NMR techniques have been created. 48 Some of the most commonly used 2D NMR techniques are heteronuclear single quantum coherence (HSQC) - which gives information about one bond coupling between carbon and hydrogen nuclei (which protons are bound to which carbons), correlation spectroscopy (COSY) – which gives information about protonproton coupling (identifying the protons that are neighbors of one another), heteronuclear multiple bond correlation (HMBC) – which gives information about two, three and four bond couplings between hydrogen and carbon nuclei (identifying the carbons which are nearest neighbors to a given  14  proton), and nuclear Overhauser enhancement spectroscopy (NOESY and its rotational frame cousin, ROESY) – which gives information about through space proton to proton correlation (which protons are close in proximity to which protons). Thus, through use of these techniques, the structure of an unknown molecule may be elucidated – HSQC and COSY providing fragments that identify the carbons with attached protons, and also the neighboring protons, HMBC allowing for the connection of those fragments to each other and also to quaternary carbons, and finally NOESY and/or ROESY providing information as to the orientation of protons in space with respect to one another, providing insight into the relative stereochemistry.  1.4 Scope of Thesis Research done in the Andersen laboratory focuses on the isolation and structure elucidation of marine natural products, due to a large in house library of marine extracts from locations all over the world. Access to the NCI Open Plant Repository provides a valuable resource that allows for investigations into terrestrial natural products with biological activities. The scope of projects undertaken ranges from chemical prospecting to biosynthetic studies to total synthesis to chemical ecology. The majority of the studies undertaken, however, begin with the isolation of a natural product with interesting biological activity - as identified by activity observed in a particular bioassay – with the aim of identifying a lead compound for development into a drug for the treatment of a human disease. Once a potential new lead compound is found, there are a number of possible outcomes that may result from an investigation into its biological activity. This thesis will describe three projects that have resulted in the identification of potential new lead compounds, and the results of continued investigation into these biological activities.  15  The first project, described in chapter 2, details how a family of plant derived flavonoids – the montbretins – from the corms of Crocosmia sp. are capable of inhibiting the human pancreatic α-amylase. Human pancreatic α-amylase is a digestive enzyme involved in the cleavage of larger carbohydrate chains into smaller (two or three sugar residue) units, which is a target whose inhibition would aid in the management of diabetes. The montbretins are particularly interesting for two reasons. First, the observed inhibition of human pancreatic αamylase is potent enough for one of the montbretins – montbretin A – to be further developed into a drug. Secondly, the source – a plant that grows from underground corms – is readily available and grows in a weed-like manner, with the montbretins being easily isolable and decently plentiful in the source extract.  The next project, described in chapter 3, details the isolation of a known alkaloid girolline, from a Pohnpeian sponge, followed by the isolation of a new family of ecdysteroids – the massacreones – from a Dominican cnidarian. In both cases, these compounds were identified as being inhibitors of the activation of Toll-Like Receptor 5 (TLR5), a membrane protein whose activation by a bacterial motif (specifically, flagellin) is known to be significant in the inflammation response in the lungs of a cystic fibrosis patient. It is not known how this TLR5 activation results in inflammation.  Finally, the project described in chapter 4 details the isolation of another known alkaloid, caulerpin, found in a number of different species of the marine algae Caulerpa sp., as a potential inhibitor of human indoleamine 2,3-dioxygenase, an enzyme involved in the ability of tumors to evade the host immune system. Caulerpin shows potent activity in a high throughput cell based bioassay, but does not demonstrate any activity in a free enzyme based assay – thus 16  demonstrating that the bioassay used to identify caulerpin as an inhibitor of IDO needs further improvement.  In summary, the selection of bioactive natural products presented in the following chapters act to illustrate the diverse results possible from the identification of a bioactive natural product, such as:  •  Identification of bioactive natural products that may be collected from a natural source and developed into pharmaceutical therapeutics, potentially without the need for synthesis (Isolation of the montbretins, Chapter 2).  •  Identification of a difficult to isolate natural product, and its subsequent attempted synthesis, to the end of identifying the source of the observed biological activity and/or any structural motifs on the isolated molecule that contribute to the observed activity (Isolation of girolline and synthesis of analogues, Chapter 3).  •  Identification of a family of bioactive natural products whose activity is not appropriate for further pharmaceutical development (Isolation of the massacreones, Chapter3).  •  Identification of a bioactive natural product whose biological activity is only observed under the conditions of the bioassay used to identify it (Isolation of caulerpin, Chapter 4).  17  2. The Montbretins 2.1 Brief Introduction to Plant Natural Products It is estimated that there are approximately 300,000 species of higher plants (vascular plants, not including mosses and algae)49, although estimates range from 250,000 to 500,000, depending upon the differences in philosophies of systematic biologists and estimations based upon 'undiscovered' plants in unusual environments. However, only about 1% of these plants – or approximately 3000 – are utilized for food, with only about 150 or so representing commercially cultivated crops. While current trends in marketplaces world wide reflect a desire to consume more exotic foods, the vast majority of caloric intake worldwide derives from just 20 species of plants. In contrast, approximately 10,000 of the world's plants have a documented medical use, with approximately 150-200 plant species being utilized in western medicine. Clearly, despite a long history of human use, there is a lot of untapped potential in plant natural products for sustenance and for pharmaceutical applications.50  Plants produce a number of different types of chemical compounds as part of their primary and secondary metabolism – lipids, nucleic acids, amino acids, proteins – much like any organism.51,1 Certain themes and trends, however, are observed in plant natural products. Terpenes are a diverse group of compounds seen in plants that are generally believed to play an ecological or physiological role – acting as insecticides and also insect attractants, as inhibitors of growth of competing plants, and as plant hormones. Alkaloids, which are common in flowering plants (but also found in microorganisms, and some vertebrates), are also believed to be mainly involved in secondary metabolism, acting as insecticides and plant hormones. 18  Carbohydrates, the primary products of photosynthesis, are an important energy source for plants and are often stored as starch and fructans, as cellulose in plant cell walls, as sucrose, and also linked to other classes of natural products (alkaloids, terpenes and aromatics) as glycosides. The acetogenins are plant polyketides that are also believed to play a largely ecological role. Finally, plants make a number of aromatic and phenolic natural products that are derived from metabolites in the shikimate pathway, which is a non-polyketide route to phenolics that is only utilized by plants and microorganisms.52  2.1.1 Flavonoids Flavonoids are naturally occurring aromatic compounds derived from the aromatic amino acids tyrosine and phenylalanine, which come from the shikimate pathway in microorganisms and plants (Figure 2.1).52 Phenylalanine and tyrosine are formed starting with the coupling of phosphoenolpyruvate (PEP, 1) and D-erythrose-4-phosphate to form D-arabino-heptulsonic acid7-phosphate (DAHP, 2), in a reaction that is an aldol-type condensation. Elimination of the phosphate, followed by an intermolecular aldol condensation yields 3-dehydroquinic acid (3), which after dehydration becomes 3-dehydroshikimic acid, and finally with a reduction step becomes shikimic acid. After phosphorylation to shikimic acid 3-phosphate by ATP, another molecule of phosphoenolpyruvate is added via nucleophilic attack on its protonated double bond, followed by 1,2 elimination of phosphoric acid to yield 3-enolpyruvylshikimic acid 3-phosphate, which undergoes a 1,4 elimination of phosphoric acid to yield chorismic acid (4).1  19  OH P O O  HO  CO2H  CO2H  H+ O  PiO O H HO  OH HO P O O HO  H  OH  HO  OH  OH 2  H+ O  HO  CO2H  O  OH  OH  OH 3 -H2O  OH 1  NADPH ATP H+  CO2H  CO2H  - PiOH O  H  CO2H  PiO  OH 4  OH  H HO2C  O  O  H  CO2H PiO  - PiOH  H  CO2H OPi  PiO  CO2H  OH OH  O CO2H  O O  CO2H  CO2H  O -CO2 -H2O  OH+ H 5  OH  CO2H  NH2 PLP  6 +  NAD  CO2H HO  O  CO2H  PLP  NH2  HO 7  Figure 2.1 Biosynthesis of the aromatic amino acids phenylalanine (6) and tyrosine (7) via the Shikimate pathway.  Chorismic acid undergoes a [3, 3]-sigmatropic Claisen rearrangement to form prephenic acid (5), transferring the phosphylenoyl pyruvate-derived side chain onto the carbocycle and establishing the basic carbon skeleton of tyrosine and/or phenylalanine. Decarboxylation and aromatization of prephenic acid yields phenylpyruvic acid, the keto acid of phenylalanine, which 20  is then converted to phenylalanine (6) via PLP-dependent transamination. In plants, tyrosine is typically biosynthesized separately from phenylalanine (rather than being simply derived from the oxidation of phenylalanine). During decarboxylation and aromatization of prephenic acid, there is an additional oxidation of the carbocycle alcohol into a ketone, meaning that the alcohol is retained resulting in 4-hydroxyphenyl pyruvic acid, which is transaminated with PLP into tyrosine (7).50-52  L-phenylalanine (6) and L-tyrosine (7) are precursors to a wide range of plant natural products. The frequent first step the E2 elimination of ammonia, to yield (E) cinnamic acid (8) in the case of phenylalanine and (E) p-coumaric acid (9) in the case of tyrosine (Figure 2.2). pCoumaric acid may also be derived from the oxidation of cinnamic acid in a cytochrome P450 dependent oxidation reaction. Additional oxidation ortho to the phenol yields caffeic acid (10), which in turn may be methylated by SAM to form ferulic acid (11). These three acids are common in plant natural products.51  21  CO2H  CO2H NH2 PAL 8  6  O2 NADPH CO2H  CO2H  CO2H  CO2H  NH2  PAL OH 6  OH 9  O2 NADPH  SAM  HO OH 10  MeO OH 11  Figure 2.2 Generation of p-coumaric acid (9), caffeic acid (10) and ferulic cid (11) from phenylalanine and tyrosine (7).  Cinnamic and p-coumaric acids are standard building blocks for more complex groups of plant natural products, including the flavonoids (Figure 2.3). From the coenzyme A thioester of p-coumaric acid, polyketide chain extension with three units of malonyl CoA gives a polyketide that can be folded to generate the chalcone (12) carbon skeleton through a Claisen condensation that generates an aromatic ring. The oxygenation patterns of these natural products are characteristic of plant natural products formed by the shikimate and acetate pathways. Chalcones (for example, naringenin-chalcone (12)) can undergo a Michael type nucleophilic attack of one of the phenolic -OH groups onto the α,β- unsaturated ketone linking the two aromatic rings, to form the bicyclic flavone skeleton – in this case, naringenin (13). This conversion can actually happen in chemically – acidic conditions favoring the formation of naringenin (13), basic conditions favoring the chalcone 12. 22  OH  OH CoASH  HO  OH O  CoAS  O  SCoA O  3x MalonylCoA  O  O  O  OH HO  OH  O  HO  OH H+  O2 AKG  OH O 13  OH O 12 OH  OH OH HO  O  HO OH  OH O 14  OH  OH HO  O  [O]  OH  O  [O]  OH O 15  OH OH  OH O 16  Figure 2.3 Biosynthetic route to the common plant flavonoids kaempferol (14), quercetin (15) and myricetin (16).  Oxidation of 13 with molecular oxygen and α-ketoglutarate to form an alcohol at the 3 position of the flavone, followed by another oxidation to form a double bond yields the very common plant flavonol, kaempferol (14). Additional oxidation of the phenyl ring of naringenin in ortho positions will yield, instead, the flavonols quercetin and myrecetin. These three flavonols (kaempferol (14), quercetin (15) and myrecetin (16)) are known to have strong antioxidative properties, chelate metals, scavenge free radicals, and prevent the oxidation of low density lipoprotein. They are commonly found as water soluble glycosides. Myricetin, specifically, is often found as its 3-O-rhamnoside53 (Figure 2.4). 23  OH OH HO  O  OH  OH O OH  O  O CH3  OH OH Figure 2.4 Myricetin (16) is commonly found as its 3-O-rhamnoside.  2.2 General Introduction to Crocosmia sp. Crocosmia is a genus of perennial plants in the iris family Iridaceae, which is native to the grasslands of South Africa. Crocosmia sp. can be evergreen or deciduous, and typically have long (up to one meter), erect, sword-shaped leaves with distinct, parallel veining and pleating. When flowering, tall arching stems will carry sprays of funnel shaped flowers in bright shades of reds and oranges, although through breeding yellows and browns are also now available.54  Crocosmia sp. foliage grows from underground structures called corms, which are short, vertical, swollen underground plant stems that serve as a storage organ for plants that need to be able to survive adverse weather conditions like drought, summer heat, or winter. Crocosmia sp. corms form vertical chains, with the youngest at the top and the oldest (and largest) deepest in the soil. The lowermost corm has contractile roots, which act to drag the corm deeper into the ground (as conditions permit), in response to soil temperatures and light. Once the corm is deep in the soil where there is no longer light and the soil temperature is uniform, it no longer digs deeper with its contractile roots.55 The chains of corms are fragile and easily separated, with 24  each corm then able to become a blooming Crocosmia sp. plant. This gives some varieties of Crocosmia sp. the ability to be invasive and difficult to control in the garden, and can cause Crocosmia sp. plants to grow in areas where garden soil is dumped or washed down waterways. As a result, certain species of Crocosmia sp. are classified as a noxious weed and are illegal to sell and grow in Australia56, and are considered a noxious weed, but are not illegal, in parts of the United States such as Hawaii.57 Still, Crocosmia sp. is a popular addition to many gardens, and is widely available for sale in North America and Europe. Of the varieties of Crocosmia sp. available, one of the most common is a hybrid called Crocosmia x crocosmiiflora or Crocosmia montbretia or Tritonia crocosmiflora (also, Emily Mckenzie, or simply Montbretia), a perennial Crocosmia sp. with bright orange flowers.54,55  2.2.1 Overview of Known Metabolites From Crocosmia sp. Several interesting compounds have been reported from a few different varieties of Crocosmia sp. Nagamoto and coworkers found that a water soluble fraction from the methanolic extract of corms of Crocosmia montbretia had strong antitumor activity. The metabolites responsible for that activity were not identified, although the primary constituents of the antitumor fraction were identified as being a mixture of saponins of medicagenic acid (17) and polygalic acid (18).58  H H  HO HO HOOC 17  H  H COOH  H  HO HO HOOC  H  COOH OR  OH  18  Figure 2.5 Medicagenic acid (17) and polygalic acid (18).  25  From the water soluble fractions of the methanolic extract of Tritonia crocosmia (another name for Montbretia), Masuda et al. identified tricrozarins A59 and B,60 (19,20), which are napthazarin derivatives possessing broad spectrum antimicrobial activity against certain gram positive bacteria (Bacillus subtillis, Mimoccus luteus, Aspergillus niger, Mucor ramus, Candida albicans and Saccharomyces sake). OH O O O OH O 19  OH O OMe  H  OMe  OMe  MeO  OMe OH O 20  Fig 2.6. Tricrozarin A (19)59 and B(20)60  Asada and colleagues investigated the methanolic extract of Crocosmia montbretia corms in an effort to find the compounds responsible for the antitumour activity reported by Nagamoto et al, and were able to identify the crocosmiosides A-I61-63 (21-29), which were saponins of polygalic acid. None of these compounds displayed any of the antitumour activity seen in the parent fraction.  26  R1  R2  R3  R4  R5  Crocosmioside A (21)  Rha  OH, H  Xyl  H  Api  Crocosmioside B (22)  H  OH, H  Xyl  H  Api  Crocosmioside C (23)  H  O  Xyl  Glc  Api  Crocosmioside D (24)  H  OH, H  Xyl  Glc  Api  Crocosmioside E (25)  Rha  OH, H  H  Glc  Api  Crocosmioside F (26)  Rha  OH, H  H  Glc  Api  Crocosmioside G (27)  Rha  O  Xyl  Glc  Api  Crocosmioside H (28)  Rha  OH, H  Xyl  H  H  Crocosmioside I (29)  Rha  OH, H  H  Glc  H  Masonoside A (32)  Rha  O  H  Glc  Api  Masonoside B (33)  H  O  H  Glc  Api  Masonoside C (34)  H  O  H  Glc  H  H  O OH O OH OH  OH OH  OH  (H2C)5  HO O  OH OH  O  OH OH  O  HO  OH  O H HOOC OH O  R1O  OH  O  H  HO  OR3 (H2C)5 R2  O  O OR4  O  O OH  O  O  O  OR5 OH  O  O OH OH  OH OH OH OH OH OH Glc Api Xyl Rha Figure 2.7 Saponins of polygalic acid isolated from corms of Crocosmia sp. Crocosmiosides A-I 61-63(2129) and masonosides A-C (32, 33, 34).64  Asada and colleagues also identified two acylated flavonols present in the corms of Crocosmia montbretia – montbretins A and B (30, 31)65, which as well displayed no antitumour activity. From the corms of another strain of Crocosmia, Crocosmia masonorum, Asada and coworkers identified a family of triterpene saponins related to the crocosmiosides called the masonosides A-C (32, 33, 34)64, and desacylmasonosides A-C66 (the masonosides without the fatty acyl chain). None of these compounds has been reported to display any of the antitumour activity found by Nagamoto.58 Nevertheless, as a result of the antitumour findings by Nagmoto, a fermented beverage of Crocosmia montbretia corms is used as a folk remedy in Japan. Its applications range from drinking for antitumour properties67 to skin cleansing.68 27  O OH  R2  O  OH O  OH  OH O OH OH O R  O  CH3 O  HO  O  OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O  O  O  R=OH; Montbretin A (30) R=H; Montbretin B (31)  OH  Figure 2.8 Montbretin A (30) and B (31), isolated form from Crocosmia montbretia..65  2.3 Human Pancreatic α-Amylase α-Amylases are widely distributed across bacteria, plants and animals. Human α-amylase is a single peptidic chain of 496 amino acids that is coded for on chromosome one as part of a multigene family. These genes are regulated such that the different isozymes of α-amylase are synthesized in either the salivary glands or the pancreas. Salivary and pancreatic α-amylases have high degrees of identity – 97% – in terms of their primary sequence, but they do exhibit different cleavage patterns. This is due to the 15 amino acid substitutions between the sequences, a number of which occur in the active site region.69  Starch, the primary source of glucose in the diet, is made up of two components:  28  amylopectin and amylose. Amylose (35), which is soluble in water, is a long chain of glucose molecules linked together in series of α(1→4) linkages. Breakdown of amylose occurs in several stages (Figure 2.9). First, salivary α-amylase provides a partial digestion into several oligomers that then continue on to the gut. Upon reaching the gut, this partially digested starch is extensively hydrolyzed into smaller oligosaccharides by the α-amylase synthesized by the pancreas (HPA, human pancreatic α-amylase). The resultant mixture of oligosaccharides isomaltose (36), maltotriose (37), predominantly maltose (38), passes through the mucosal layer into the brush border membrane of the small intestine, where a number of α-glucosidases function to degrade the oligosaccharides into glucose, which is then absorbed and enters the bloodstream through a specific transport system69.  OH OH  O OH  OH  OH  OH  OH O  O  OH  OH  O n  O  OH  OH HPA  35 OH OH HO  OH 36  OH  O OH OH  O  OH  O OH  OH  OH  OH  O  OH O  O OH  OH O  37  O OH  OH  OH OH  OH HO  O OH 38  OH O  O  O H  OH  OH  Maltase OH OH  O  OH  OH  OH Glucose  Figure 2.9 HPA catalyses the degradation of amylose (35) into isomaltose(36), maltotriose(37), maltose(38).  29  Evidence from NMR, mutagenic, and structural studies suggests that α-amylases use a double displacement mechanism to catalyze the hydrolysis of α(1→4) linkages, which leads to retention of anomeric configuration (Figure 2.10).70  HO  OH  -  O  O  OH  OH O Glc  O  O  Glc O  O  Glc O  HO  HO  O  Glc  O  H O  H O  O  O H2O Glucose OH  HO Glc O  O  O  O  OH H O H O-  O  OH Glc  OH -  O  O  O HO  O  Glc  O HO  OH O H O  H O  O  O HO  O  O  H O  O  H  Figure 2.10 Proposed double displacement mechanism to catalyze the hydrolysis of α(1→4) linkages in amylose by HPA.  30  First, a covalent glycosyl-enzyme intermediate is formed as a result of the nucleophilic displacement of the downstream aglycon by a catalytic carboxylic acid. This covalent intermediate is then hydrolyzed via another nucleophilic displacement, but this time with a water molecule. The result is the release of the product with net retention of configuration at the anomeric carbon. Both the glycosylation and deglycosylation steps are believed to proceed via oxocarbenium ion-like transition states, which are stabilized by another active site carboxylic acid that first acts as a general acid, then as a general base catalyst.71  2.4 HPA as a Drug Target: Diabetes Diabetes mellitus is quickly becoming one of the world's major threats to human health in the 21st century.72 The past two decades have seen a marked increase – or rather, an explosion – of the number of people suffering from diabetes mellitus worldwide. This is occurring both in developed nations and also in the developing world, due to changes in human behaviors and lifestyle that have come as a result of globalization.73  There are two main forms of diabetes – type 1, or juvenile, and type 2, or adult onset. These two diseases, while similar, have different causes and thus, different treatment requirements. When a meal is eaten, the carbohydrates and starches from the food are broken down by enzymes in the saliva and in the stomach. Starch breaks down into amylose, which in turn is digested further by human pancreatic α-amylase into glucose that then enters the bloodstream. Insulin is the principal hormone that regulates uptake of glucose from the blood into most cells such as primarily muscle and fat cells, but not central nervous system cells. Glucose is required by a number of cells for normal function, and is additionally used by the 31  liver to build glycogen, a long-term glucose storage molecule. If the amount of insulin available is insufficient, if cells respond poorly to the effects of insulin, or if the insulin itself is defective, then glucose will not be absorbed properly by those body cells that require it nor will it be stored appropriately in the liver and muscles. The net effect is high levels of blood glucose, poor protein synthesis, and a whole host of other complications. In the case of type 1 diabetes, the amount of insulin available is insufficient due to loss of insulin-producing beta cells in the pancreas (commonly via T-cell mediated autoimmune attack), resulting in elevated blood glucose levels. However, in the case of type 2 diabetes, the problem is reduced insulin sensitivity, resulting in elevated levels of both insulin and glucose in the blood.74  Estimates of the global figure of people with diabetes are expected to rise from the current 150 million to 220 million in 2010, to 300 million in 2025.75 Type 2 diabetes currently accounts for approximately 90% of all diabetes cases worldwide, and is strongly associated with a sedentary lifestyle and obesity.76 As well, due to advances in public health in the 20th century, people are now living longer due to the elimination of many communicable diseases worldwide. As such, non communicable diseases such as diabetes are quickly becoming the main public health challenges of the 21st century.73  While type 1 diabetes can be effectively managed by monitoring blood glucose levels and delivery of artificial insulin, type 2 diabetes is harder to treat, with treatment focusing on improving insulin sensitivity and also slowing the release of glucose into the bloodstream after a meal or from glycogen stores in the liver. Obesity, particularly with an increase in the amount of fat stored abdominally, and physical inactivity are major contributing factors to type 2 diabetes. Fat concentrated around the waist in relation to abdominal organs is known to predispose 32  individuals to insulin resistance. Abdominal fat is especially active hormonally, secreting a group of hormones called adipokines that may possibly impair glucose tolerance. Obesity is found in approximately 55% of patients diagnosed with type 2 diabetes.72 Thus, often the first method of treatment, particularly when the type 2 diabetes is only moderate, is a lifestyle change that includes diet and exercise as well as stress reduction. In many cases, however, type 2 diabetes can go unnoticed for years before diagnosis, since symptoms are typically milder than those seen with type 1 diabetes and can be sporadic. As a result, pharmaceutical treatment is often a necessary complement to a lifestyle overhaul for proper management of blood glucose levels.74  Because type 2 diabetes is characterized by insulin resistance, treatment solely via introduction of artificial insulin is not feasible; however, artificial insulin may be necessary if other methods of management prove to be insufficient. There are three main forms of treatment for type 2 diabetes: drugs that increase the amount of insulin secreted by the pancreas, drugs that improve the sensitivity of target organs to insulin, and finally, drugs that decrease the rate at which glucose is released after a meal, or is absorbed by the digestive tract.74  2.4.1 Current Therapeutics For Diabetes 2.4.1.1 Insulin Secretagogues  A secretagogue is a substance that causes another substance to be secreted. Sulphonylureas are a class of insulin secretagogues that trigger the release of insulin by pancreatic beta cells via direct action on the KATP channel of those cells, and were the first drugs other than artificial insulin to come out in the diabetes market. Sulphonylureas77 inhibit the  33  outflux of potassium in beta cells, causing the electric potential across that membrane to become more positive. In response to this, voltage-gated calcium channels open, with the rise in intracellular calcium levels leading to increased secretion of insulin74. Early examples of this class include tolbutamide (39)78 and acetohexamide (40)73, while second generation drugs in this class such as glibenclamide (41)79 are now more commonly used as they have milder and fewer side effects. Another class of insulin secretagogues are the meglitinides.80 Meglitinides, such as repaglinide (42) and nateglinide (43), bind to the KATP channel on the cell membrane of pancreatic beta cells in a manner similar to sulphonylureas, but at a different binding site. 2.4.1.2 Sensitizers  Biguanides, particularly metformin (44)81, are very commonly prescribed antidiabetic drugs, useful for treatment of both type 1 (in conjunction with insulin) and type 2 diabetes. Their mechanism of action is poorly understood; however it is known that they reduce glucanogenesis in the liver, thereby slowing the release of glucose by the liver and reducing the amount of it present in blood. Also, they tend to make cells more willing to absorb glucose from the blood, also reducing the amount of glucose present in the bloodstream. Thiazolidinediones or “glitazones” – such as rosiglitazone (45) and pioglitazone (46)82 – work to increase insulin sensitivity by stimulating the up regulation of mRNAs of insulin dependent enzymes, resulting in more efficient use of glucose by the cells thus affected.83  34  O O S O HN O  Cl  O S O HN O  O  O S O HN O  N H OMe  HN  HN  HN  40  39  41 COOH  O N H  HO  O  OEt  O  N H  N 42  43  O NH N  NH  N H 44  NH2  O  HN S  O  O  H N  N  45  HN S O  O  N  46  Figure 2.11 Drugs used in the treatment of diabetes: insulin secretagogues such as the sulphonylureas (e.g. tolbutamide (39), acetohexamide (40), glibenclamide (41)) and the melitinides (e.g. repaglinide (42) and nateglinide (43)); sensitizers such as the biguanides (e.g. metformin (44), rosiglitazone (45) and pioglitazone (46)).  2.4.1.3 α-Glucosidase and α-Amylase Inhibitors  Another approach to helping a diabetic patient manage their blood glucose levels is to slow the release of glucose from starch after a meal.83 This may be done by inhibiting the actions of either α-glucosidases or α-amylase. α-Amylase, which is found in humans as both salivary αamylase and pancreatic α-amylase, is responsible for breaking down starches into smaller starches and oligosaccharides, respectively. Human α-glucosidase is an enzyme found in the brush border of the small intestine that is responsible for the hydrolysis of oligosaccharides, trisaccharides and disaccharides into glucose and other monosaccharides. By inhibiting either of these enzymes, the rate at which free monosaccharides are released in the small intestine and 35  absorbed into the bloodstream is reduced. The overall effect of these types of drugs is reduced blood sugar levels, which allows diabetics to more easily control sugar spikes after meals, working with their reduced insulin sensitivity in order to maintain a healthy balance.80  There are three α-glucosidase inhibitors currently available – miglitol (47), voglibose (48) and acarbose (49).84 Of those three, only one – acarbose – inhibits human pancreatic α-amylase. In fact, acarbose is the only inhibitor of HPA currently available on the market, highlighting the potential for new therapeutics that inhibit HPA.85  Because α-glucosidase and α-amylase  inhibitors prevent the degradation of complex carbohydrates into glucose, carbohydrates will remain in the intestine longer than they would normally, leading to some unfortunate side effects. Bacteria in the colon will digest complex carbohydrates, leading to gastrointestinal effects such as flatulence and diarrhea. These side effects are dose related, so typically patients given miglitol, voglibose or acarbose are started at a low dosage that is slowly increased until the desired amount is being taken. However, these side effects hinder the widespread use of miglitol, voglibose and acarbose as primary therapeutics for the control of diabetes.  α-  Glucosidase inhibitors are typically prescribed in addition to other hypoglycemic medications such as insulin sensitizers and insulin secretalogues.85 OH  OH HO HO  OH N  HO OH  OH  HO HO OH HN  OH  47  OH  HO  HN  O  HO  OH  O HO  OH 48  OH O OH O HO  OH O OH OH  49  Figure 2.12 α-Glucosidase inhibitors miglitol (47), voglibose (48) and acarbose (49). Acarbose is additionally an α-amylase inhibitor. 36  As a drug used in the management of diabetes mellitus, an inhibitor of HPA would be taken with the first bite of a meal, resulting in glucose being released slower than it would be without an HPA inhibitor, and allowing the impaired insulin response to clear the blood sugars. This happens without requiring a perturbation of the natural insulin secretion of the pancreas, or by interacting with the cellular machinery responsible for the adsorption of glucose, minimizing the potential for undesirable side effects. Thus, of the drug targets currently identified as candidates for control of diabetes mellitus, inhibition of HPA is an attractive option due to the relatively benign (although potentially uncomfortable) side effects of its inhibition, as well as for the simplicity of the potential dosing.  2.4.2 HPA Bioassay In order to identify compounds that inhibit HPA, a high throughput screening methodology was developed based upon the idea of observing a chromophoric product cleaved from an aryl-glycoside substrate of HPA.84 This assay was developed as a continuous assay, which has benefits over a stop point assay in that it is less affected by the background readings due to extract compositions or other molecules with absorption at the same wavelength. Furthermore, a system that works based on observing wavelengths in the UV-VIS spectrum was chosen over a fluorescence based assay, to diminish the possibility of false positives due to fluorescence quenchers.  2-Chloro-4-nitrophenyl-α-D-maltoside (50) (a substrate for HPA) was chosen for this assay because of its commercial availability, and also, because the pKa of the chloronitrophenyl leaving group (pKa = 5.45) is lower than the pH value of the assay, yielding a high extinction coefficient for the chromophore obtained under assay conditions (Figure 2.13). Assay conditions 37  also included Triton X-100, which was used to minimize the detection of promiscuous inhibitors, and solutions of test compounds and biological extracts in DMSO, to a final DMSO concentration of 0.1%. The enzymatic activity of HPA was found to be completely unaffected by this concentration of DMSO and of Triton X-100 (0.01%). OH  OH OH HO  O OH  OH O  Cl  O OH  O  OH  OH  O  O  HPA NO2  OH  pH=7.0 HO  OH  OH O  -  OH  O  NO2 Cl  OH detect at 405nm  50  Figure 2.13 2-Chloro-4-nitrophenyl-α-D-maltoside (50) is cleaved by HPA under bioassy conditions, yielding a chloronitrophenyl ion with a high extinction co-efficient (and thus shows strong absorbance) at 405 nm.  In initial screening, biological extracts were tested at a final assay concentration of 5 μg/mL, and the tests were done in duplicate to ensure the robustness of the data collected. Additionally, two test plates with serial dilutions of the known HPA inhibitor acarbose were used as positive controls and were run at the start and at the end of the experimental runs to ensure the robustness of the data. Inhibitors (potential, known, or biological extracts at 5 μg/mL) and enzyme (1 μg/mL) were allowed to incubate together at room temperature for ten minutes to allow the detection of slower inhibitors (“slow-on” inhibitors) before the assay was initiated by the addition of substrate (1 mM) at a concentration below the Km value for the substrate (Km = 3.6 mM). The subsequent release of the chloronitrophenolate anion was detected by observing 405 nm wavelength continuously for seven minutes. These were then used to calculate the rates of the reaction in each well, which were then normalized versus controls with no inhibitor that were present on each plate, and expressed in terms of % residual activity in each well. The two replicates were then plotted against each other, and those that showed consistent data between 38  the two replicates, and also activity within the hit window (set to three standard deviations from the mean of the whole sample set, or 81% residual activity) were then validated by retesting. Samples that showed reproducible activity were deemed to be 'true hits' and were subjected to further investigation.84.  2.5 Isolation of Inhibitors of HPA from Crocosmia sp. The initial high throughput screen for inhibitors of HPA tested 30,000 NCI (National Cancer Institute, USA) plant and marine extracts, and of those only 25 had residual activity below 81% that was reproducible when retested. The most active of all the extracts, with only 2% residual activity, was chosen for further investigation. Before this was done the extract was subjected to a kinetic analysis by testing HPA inhibition on a series of dilutions. This series showed a semilogarithmic sigmoidal shaped dose response curve, characteristic of a well behaved inhibitor, with a very low IC50 value for the crude extract of 2.7 μg/mL. In addition to this, there was no time-dependent inactivation of HPA, as shown by the fact that the level of inhibition remained constant over four hours. These results indicated that the active components of this extract did not have modes of action that would be undesirable for an HPA inhibitor, namely, denaturing or covalent modification of HPA.84  The crude extract with 2% residual activity was a plant extract obtained from the NCI open plant repository. The sample that was obtained was a dried methanolic extract of the corms from Crocosmia sp. In order to isolate the principal bioactive components from the complex mixture of the extract, a series of bioassay-guided purification steps were performed. At each step in the purification, subfractions and/or column fractions were bioassayed for HPA inhibition, with active fractions being grouped together for further purification. 39  2.5.1 Isolation of Montbretins A, B and C The crude methanolic extract was tested and validated as active upon arrival to confirm the activity. The extract was suspended in water and partitioned with ethyl acetate, and the aqueous fraction was then partitioned with butanol. The majority of the activity was seen in the butanolic fraction, which was then applied to lipophilic size exclusion chromatography (Sephadex LH-20) in methanol and eluted over a series of fractions. An aliquot of each of these column fractions was transferred into a well of a 96 well plate, and then allowed to evaporate to dryness, only to be resuspended in 10µL of water and tested for HPA activity using the same protocol used for the initial high throughput screening. The fractions showing the clear 'peak' in inhibitory activity were then grouped together and evaporated to dryness, to afford a slightly purified mixture containing the montbretins. This active fraction was then subjected to an isocratic, polarity based separation on reversed phase HPLC, eluting with 22% aqueous acetonitrile and monitored for absorbance at 254 nm and 210 nm. From the HPLC run, a series of fractions was obtained. The fraction that showed potent HPA activity was a major peak on the HPLC trace and it appeared to be the first of a family of three or more compounds eluting sequentially as progressively smaller peaks. An analysis of the 1H NMR spectra of the compounds in these peaks showed that they were similar and presumably a family. In total, there were three peaks that showed some HPA activity as well as similarities in their 1H NMR spectra – most notably, intense doublets at around 1-2 ppm. The first and most potent of these fractions was further purified by reversed phase HPLC with a gradient from 30% to 40% aqueous acetonitrile over 30 min. This afforded montbretin A (30, Rt=15 min, 8.4 mg) as a yellow powder. The second fraction was purified via reversed phase HPLC with a gradient from 30% to 70% aqueous acetonitrile over 30 min to afford montbretin B (31) as a yellow powder (Rt=16  40  min, 0.9 mg), and finally the third fraction was purified by reversed phase HPLC with gradient from 20% to 30% aqueous acetonitrile to afford montbretin C (51, Rt=20 min, 1.6 mg) as a yellow powder.  2.5.2 Isolation of Montbretins D and E After initial studies on montbretins A, B and C showed enough potential for further investigation, a large scale effort was undertaken to obtain gram quantities of the montbretins – most specifically, montbretin A. In the first published report of montbretins A and B65, Asada and co-workers isolated the montbretins from Crocosmia montbretia. The NCI extract that yielded montbretins A, B and C was simply identified as Crocosmia sp. Thus, to understand which varieties of Crocosmia sp. corms would yield montbretins, all major cultivars of available Crocosmia sp. were tested. It was found that in every case, montbretins A and B were produced as confirmed by testing for HPA inhibitory activity as well as by analytical HPLC, with montbretin A consistently the predominant montbretin found. The cultivars that were investigated were Crocosmia x crocosmiiflora (Crocosmia pottsii x Crocosmia aurea; Crocosmia x crocosmiiflora “Emily Mackenzie”; Crocosmia montbretia), Crocosmia cv. “Emberglow” (Crocosmia pottsii x Crocosmia paniculata), and Crocosmia cv. “Lucifer” (Crocosmia masoniorum x Crocosmia paniculata). Based on these findings, a local source of Crocosmia “Emily Mackenzie” was found and those corms were then collected and extracted on a large scale86.  In a simplified isolation procedure performed by Dr. John Coleman and coworkers at the Center For Drug Research and Development at UBC, the Crocosmia corms were minced and extracted two times with methanol overnight, after which the methanol extract was dried by 41  rotary evaporation. The crude extract was resuspended in water and adsorbed onto diaion HP20, after which fractions are collected by eluting with 30% aqueous acetone, 50% aqueous acetone and finally 100% acetone. Using this isolation procedure, the montbretins mainly eluted in the 50% aqueous acetone fraction, reducing the amount of material to be purified down to about 1%, by mass of the crude extract. The material from the HP-20 was then subjected to purification by reversed phase UPLC using a gradient from 5% to 90% aqueous acetonitrile over 30 minutes, with monitoring done by UV absorption at 266 nm. As a result, a number of peaks were collected – including montbretin A (at Rt=20.2 min) and montbretin B (at Rt=21.0 min), as identified by their molecular weights and 1H NMR spectra. Interestingly, montbretin C was not isolated in this large scale effort, which is likely due to the fact that different strains of Crocosmia sp. have slightly different compositions of minor montbretin analogues, in addition to the seemingly ubiquitous nature of montbretins A and B.  In addition to the known montbretins A and B, a number of the peaks collected in the large scale isolation of montbretin A showed 1H NMR resonances and masses characteristic of montbretins. Of particular interest were two more polar peaks at Rt=17.6 minutes and Rt=18.7 minutes. While not completely purified, they appeared to be two new montbretins. Montbretin D (52) was purified from the crude 18.7 minute fraction by further reversed phase HPLC using 15% aqueous acetonitrile (Rt=25 min, 4.7 mg) as a yellow/brown powder. Montbretin E (53) was purified as a brown powder from the 17.6 minute fraction, again using 15% aqueous acteonitrile (Rt=17 min, 0.9 mg).  42  Montbretin A (30)  R1  R2  OH  OH  Montbretin B (31) H  OH  Montbretin C (51) OMe  OH  Montbretin D (52) H  Glu  Montbretin E (53) OH  Glu  R2  O  OH O OH O  OH OH OH  O  R1  O  O  OH  O OH O  CH3 O  = Glu  OH  HO  O  O OH  CH3  OH OH  OH O OH  OH  O  OH  O  OH OH O OH OH  O OH  Figure 2.14 Montbretins A-E (30, 31, 51, 52, 53).  2.5.3 Structure Elucidation of Montbretin A Montbretin A was isolated as yellow powder that gave an [M+Na]+ pseudomolecular ion at m/z 1251.3226 in its high resolution electrospray ionization time-of-flight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C53H64O33 (C53H64O33Na calculated for m/z 1251.3228), indicating 22 degrees of unsaturation. Analysis of 1H and 13C NMR, as well as 2D NMR (COSY, HSQC, HMBC, ROESY) and 1D selective NOE and TOCSY NMR data identified several sugar and aromatic fragments, numbered below to aid in discussion of the structure elucidation. Each sugar is also numbered individually for ease of discussion.  43  O OH 23 22 24 17 21 HO 18 O 10 19 16 20 OH 11 12 15 O 14 13 OH O OH HO 6  5  HO  2  4 3  7  O CH3  O 1  O  8  OH Rha1  OH  O  O O  CH3  O  Xyl  OH Rha2  OH OH  OH O  9  OH  Glc1  OH OH O OH OH  O  Glc2  OH  30 Figure 2.15 Montbretin A (30).  Analysis of the NMR data suggested that this molecule was a glycoside, owing to the readily observable and characteristic anomeric proton-carbon correlations in the HSQC spectrum linking proton resonances at δH 4.5-5.7 ppm to carbon resonances at δC 100-110 ppm. Furthermore, it was clear from initial observation that the sugars in the glycoside were linked to one or more aromatic moieties based upon the preponderance of sp2 hybridized carbon resonances between δC 115-180 ppm, typical of double bond, aromatic or carbonyl carbons. Discussion of the spectra of montbretin A and its elucidation will focus first on the elucidation of the aromatic portions of the molecule, followed by a stepwise analysis through each sugar residue and the connection to its neighbor or neighbors.  44  Furthermore, because of the structural similarities between all of the montbretins, particularly in the linking sugar residues and oxygenation pattern in the aglycone portion of the molecule, a detailed elucidation of montbretin A will be presented, followed by a discussion on montbretins B, C, D and E building off of the established core structure of montbretin A.  Although montbretin A and B have previously been reported in the literature, no two dimensional NMR data was reported, and moreover, there are some errors in the reported proton and carbon resonances.  45  O OH HO  O  OH OH  O OH O OH O HO  OH O  CH3 O  HO  OH O OH  O  O CH3  O OH  OH OH  O  OH OH O OH OH  O OH  Figure 2.16 600 MHz 1H NMR spectrum of montbretin A (30), recorded in MeOD-d4. 46  O OH HO  O  OH OH  O OH O OH  O CH3  O HO  O  HO  OH O OH  OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.17 150 MHz 13C NMR spectrum of montbretin A (30), recorded in MeOD-d4.  47  O OH HO  O  OH OH  O OH O OH O HO  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.18 600 MHz COSY spectrum of montbretin A (30), recorded in MeOD-d4.  48  O OH HO  O  OH OH  O OH O OH O HO  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.19 600 MHz HSQC spectrum of montbretin A (30), recorded in MeOD-d4.  49  O OH HO  O  OH OH  O OH O OH O HO  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.20 600 MHz HMBC spectrum of montbretin A (30), recorded in MeOD-d4.  50  O OH HO  O  OH OH  O OH O OH O HO  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.21 600 MHz TROESY spectrum of montbretin A (30), recorded in MeOD-d4.  51  Table 2.1 1D and 2D NMR data for montbretin A (30), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4 O OH 23 22 24 17 21 HO 18 O 10 19 20 OH 16 11 12 15 O 14 13 OH O OH HO 6  5  2  4 3  7 HO  O CH3  O 1  O  8  OH  OH  O  O O  O  Xyl  OH  CH3  Rha1  Rha2  OH OH  OH O  9  OH  Glc1  OH OH O OH OH  O  Glc2  OH  30  C#  13  1 b C H δ δ (ppm) (ppm)a mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Caffeic Acid  a  1  169.1  -  -  -  -  2  114.8  6.04 d (16)  C-1, C-4  H-3  H-3, H-5, H-9  3  147.3  7.4 d (16)  C-1, C-2, C-4, C-5, C-9  H-2  H-2, H-5, H-9  4  127.7  -  -  -  -  5  115.1  6.86 d (1.9)  C-6, C-7, C-9  H-9  H-2, H-3  6  149.7  -  -  -  -  7  146.8  -  -  -  -  8  116.4  6.66 d (7.9)  C-4, C-7  H-9  H-9  9  123.1  6.75 dd (7.9, 1.9)  C-5, C-7, C-8  H-5, H-8  H-2, H-3  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. b  c  Recorded at 600 MHz.  52  Table 2.1 (continued) 1D and 2D NMR data for montbretin A (30), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4 C#  13  1 b C H δ δ (ppm) (ppm)a mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Aglycone 10  158  -  -  -  -  11  137.5  -  -  -  -  12  179.8  -  -  -  -  13  106  -  -  -  -  14  163.3  -  -  -  -  15  100.4  6.17 d (1.9)  C-14, C-16, C-17  -  H-17  16  166  -  -  -  -  17  95.3  6.21 d (1.9)  C-12, C-13, C-15, C-16, C-18  -  H-15  18  158.6  -  -  -  -  19  128.8  -  -  -  -  20/24  109.9  6.91 s, 2H  C-10, C-19, C-20/24, C-21/23, C-22  -  Rha1-H-1  21/23  152.1  -  -  -  -  22  136.2  -  -  -  -  Rhamnose-1, Rha1 Rha1-1  102.8  5.65 s  C-11, Rha1-C-2, Rha1-C-3  Rha1-H-2  Rha1-H-2, H-20/24  Rha1-2  84.4  4.3 d (3.3)  Rha1-C-3, Rha1-C-4, Glc1-C-1  Rha1-H-1, Rha1-H-3  Rha1-H-1, Rha1-H-3, Glc1-H-1  Rha1-3  72.2  3.73 m  -  Rha1-H-2, Rha1-H-4  Rha1-H-5  Rha1-4  74  3.37 m  Rha1-C-6  Rha1-H-3, Rha1-H-5  Rha1-H-6  Rha1-5  72  3.55 m  -  Rha1-H-4, Rha1-H-6  Rha1-H-3, Rha1-H-6  Rha1-6  17.9  1.07 d (6), 3H  Rha1-C-4, Rha1-C-5  Rha1-H-5  Rha1-H-4  Glc1-1  105.6  4.56 d (7.8)  Rha1-C-2, Glc1-C-2  Glc1-H-2  Glc1-H-3, Glc1-H-5  Glc1-2  84.8  3.47 m  Glc1-C-1, Glc1-C-3, Glc2-C-1  Glc1-H-1, Glc1-H-3  Glc1-H-6a  Glc1-3  77.8  3.63 m  Glc1-C-2, Glc1-C-4  Glc1-H-2, Glc1-H-4  Glc1-H-1  Glc1-4  71.4  3.38 m  Glc1-C-3, Glc1-C-5, Glc1-C-6  Glc1-H-3, Glc1-H-5  Glc1-H-6b  Glc1-5  74.5  3.55 m  -  Glc1-H-4, Glc1-H-6  Glc1-H-1  Glc1-6  64.2  H-6a: 4.48 d (12)  C-1, Glc1-C-5  Glc1-H-6b  Glc1-H-2, Glc1-H-6b  H-6b: 4.2 dd (12, 5)  C-1, Glc1-C-5  Glc1-H-6a, Glc1-H-5  Glc1-H-4, Glc1-H-6a  Glucose-1, Glc1  a  b  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz.  c  Recorded at 600 MHz.  53  Table 2.1 (continued). 1D and 2D NMR data for montbretin A (30), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4 C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Glucose-2, Glc2 Glc2-1  106.6  4.6 d (7.8)  Glc1-C-2  Glc2-H-2  Glc2-H-3, Glc2-H-5  Glc2-2  76.1  3.28 m  Glc2-C-1, Glc2-C-3  Glc2-H-1  -  Glc2-3  79.3  3.4 m  -  -  Glc2-H-1  Glc2-4  71.1  3.32 m  Glc2-C-3, Glc2-C-5  -  Glc2-H-6a  Glc2-5  78  3.34 m  Glc2-C-4, Glc2-C-6  Glc2-H-6  Glc2-H-1  Glc2-6  62.5  H-6a: 3.95 m  -  Glc2-H-6b  Glc2-H-4, Glc2-H-6b  H-6b: 3.74 m  -  Glc2-H-5, Glc2-H-6a  Glc2-H-6a  Xylose, Xyl Xyl-1  107.3  4.85 d (7.4)  C-22, Xyl-C-5,  Xyl-H-2  Xyl-H-3, Xyl-H-5b  Xyl-2  75  3.6 m  Xyl-C-1, Xyl-C-3  Xyl-H-1  Xyl-H-4  Xyl-3  75.7  3.56 m  Xyl-C-2  Xyl-H-4  Xyl-H-1  Xyl-4  75.3  3.72 m  Xyl-C-5, Xyl-C-3, Rha2-C-1  Xyl-H-3, Xyl-H-5a, Xyl-H-5b  Xyl-H-5a  Xyl-5  64.6  H-5a: 4.14 dd (12, 5)  Xyl-C-1, Xyl-C-4  Xyl-H-4, Xyl-H-5b  Xyl-H-4, Xyl-H-5b, Rha2-H-1  H-5b: 3.34 m  Xyl-C-1, Xyl-C-4  Xyl-H-4, Xyl-H-5a  Xyl-H-1, Xyl-H-3, Xyl-H-5a  Rhamnose-2, Rha2  a  Rha2-1  100  4.82 d (1.7)  Xyl-C-4, Rha2-C-2, Rha2-C-5  Rha2-H-2  Rha2-H-2, Xyl-H-5a  Rha2-2  72.3  3.76 dd (3.3, 1.7)  Rha2-C-1  Rha2-H-1  Rha2-H-1  Rha2-3  72.5  3.69 dd (9.5, 3.3)  Rha2-C-4  Rha2-H-4  Rha2-H-5  Rha2-4  74.1  3.38 m  Rha2-C-3, Rha2-C-5  Rha2-H-3, Rha2-H-5  Rha2-H-6  Rha2-5  70.3  3.96 m  Rha2-C-4, Rha2-C-6  Rha2-H-4, Rha2-H-6  Rha2-H-3  Rha2-6  18.1  1.28 d (6)  Rha2-C-4, Rha2-C-5  Rha2-H-5  Rha2-H-4  b  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz.  c  Recorded at 600 MHz.  54  Table 2.2 Summary of 1D TOCSY data for montbretin A (30), recorded at 600 MHz in MeOD-d4 Sugar  Position  TOCSYa (H→H)  Rha1-H-1 Rha1-H-2  Rhamnose-1  Rha1-H-1; Rha1-H-3 → Rha1-H-4 → Rha1-H-5  Rha1-H-3 Rha1-H-4 Rha1-H-5 Rha1-H-6  Rha1-H-5 → Rha1-H-4 → Rha1-3  Glc1-H-1  Glc1-H-2 → Glc1-H-3 → Glc1-H-4  Glc1-H-2  Glucose-1  Glc1-H-3 Glc1-H-4 Glc1-H-5 Glc1-H-6a Glc1-H-6b  Glc1-H-6a; Glc1-H-5 → Glc1-H-4  Glc2-H-1  Glc2-H-2 → Glc2-H-3 → Glc2-H-4  Glc2-H-2  Glucose-2  Glc2-H-3 Glc2-H-4 Glc2-H-5 Glc2-H-6a  Glc2-H-6b; Glc2-H-5 → Glc2-H-4  Glc2-H-6b Xyl-H-1  Xyl-H-2 → Xyl-H-3  Xyl-H-2  Xylose  Xyl-H-3 Xyl-H-4 Xyl-H-5a  Xyl-H-5b; Xyl-H-4 →Xyl-H-3 → Xyl2 → Xyl-H-1  Xyl-H-5b Rha2-H-1  Rha2-H-2  Rha2-H-2  Rhamnose-2  Rha2-H-3 Rha2-H-4  a  Rha2-H-5  Rha2-H-6; Rha2-H-4 → Rha2-H-3  Rha2-H-6  Rha2-H-5 → Rha2-H-4 → Rha2-H-3  Recorded at 600 MHz.  55  2.5.3.1 Caffeic Acid Residue  A proton that resonates at δH 6.04 ppm (H-2) in the 1H NMR spectrum of montbretin A (30) showed a distinct COSY correlation to a second proton resonance at δH 7.40 ppm (H-3) (Figure 2.22). Moreover, these two protons share a coupling constant of 16 Hz, indicative of a trans relationship across a double bond. HMBC correlations between these two protons and a carbon that resonates at δC 169.1 ppm (C-1) established that this double bond was connected to an ester functionality. Further HMBC correlations from these two protons into a carbon resonance at δC 127.7 ppm (C-4) established that the other functionality attached to the double bond was an aromatic ring. O COSY  5  HO 6  HMBC  7  3  4 8  2  1 O  9  HO  Figure 2.22 Summary of HMBC and COSY correlations observed in the C 1-C9 caffeic acid residue in montbretin A (30).  Another pair of proton resonances, at δH 6.66 ppm (H-8) and δH 6.75 ppm (H-9), showed a distinct COSY correlation to one another and a coupling constant of 7.9 Hz, indicative of a pair of ortho coupled aromatic protons. Additionally, a proton singlet at δH 6.86 ppm (H-5) showed a weak COSY correlation to the proton H-9, suggesting a meta coupling between the two (J = 1.9 Hz). An HMBC correlation between the resonance at H-5 (δH 6.86 ppm) and δC 123.1 (C-9) as well as another between the resonances at δH 6.75 ppm (H-9) and δC 116.4 ppm (C-5), confirmed 56  that these two positions were meta to one another. HMBC correlations from δH 6.86 ppm (H-5) and δH 6.66 ppm (H-8) to carbon resonances at δC 146.8 ppm (C-6) and δC 149.7 ppm (C-7) suggested that these two oxygenated aromatic carbons (C-6 and C-7) are situated between C-5 and C-8. An HMBC correlation observed between δH 6.75 ppm (H-9) and δC 149.7 ppm (C-7) confirmed this placement. Finally, an HMBC correlation between δH 6.66 ppm (H-8) and the carbon resonance at δC 127.7 ppm (C-4) allowed for the assignment of this substructure as a caffeic acid residue, as shown in Figure 2.22. 2.5.3.2 Aglycone Portion  After identifying the caffeic acid residue, the remaining resonances in the aromatic region of the 1H NMR spectrum were three singlets at δH 6.17 ppm (1H), δH 6.21 ppm (1H) and δH 6.91 ppm (2H). The proton resonance at δH 6.17 ppm (H-15) and the proton resonance at δH 6.21 ppm (H-17) showed a weak COSY correlation to one another, suggesting a meta coupling between the two (J = 1.9 Hz). This is confirmed by the presence of strong HMBC correlations from δH 6.17 ppm (H-15) to δC 95.3 ppm (C-17) and from δH 6.21 ppm (H-17) to δC 100.4 ppm (C-15), shown in Figure 2.23. The H-15 resonance at δH 6.17 ppm had only two additional HMBC correlations – to resonances at δC 163.3 ppm (C-14) and δC 166 ppm (C-16), assigned to oxygenated aromatic carbons. The H-17 resonance at δH 6.21 ppm had HMBC correlations to C-16 (δC 166 ppm), as well as to another oxygenated aromatic carbon at δC 158.6 ppm (C-18), an aromatic carbon at δC 106.1 ppm (C-13) and a very weak four bond correlation to a carbonyl carbon at δC 179.8pm (C12), allowing the substructure i in Figure 2.23 to be proposed.  57  O 17 O  O 18  16  24  COSY HMBC  15  14  13  O  12  23  22  O  21  O 10  19  20  O  O i  ii  Figure 2.23 COSY and HMBC correlations in the two substructures (C 12-C18 and C10&C19-C24) of the aglycone residue in montbretin A (30).  The two proton resonance at δH 6.91 ppm (H-20 and H-24) showed a very distinct HMBC correlation to the shared carbon resonance at δC 109.8 ppm (C-20 and C-24) – indicative of symmetric meta positioning on an aromatic ring (Figure 2.23). Further HMBC correlations from H-20/H-24 to an oxygenated aromatic carbon resonance at δC 152.0 ppm (C-21 and C-23), plus one to another oxygenated aromatic carbon resonance at δC 136.2 ppm (C-22) and finally, an aromatic carbon resonance at δC 128.8 ppm (C-19) allowed for the assignment of C19-C24 as an aromatic ring. A further correlation from H-20/H-24 to a resonance at δC 158.6 ppm (C-10) provided substructure ii in Figure 2.23. When used for a literature search, the two substructures from Figure 2.23, in combination with the caffeic acid residue, resulted in identification of a probable match from the literature – montbretin A (30) – and the identification of this aglycone portion as being a myricetin residue.  Comparison of the remaining proton and carbon resonances from each of the five sugars (to be discussed shortly) with the values reported in the literature revealed a strong match to the  58  values reported for montbretin A. 2.5.3.3 Rhamnose1  Attached to position C-11 of the aglycone, via an ether linkage to its anomeric position, is a rhamnose sugar residue. Rhamnose is a common sugar found in flavoring glycosides, and is also a 6-deoxy sugar, having a characteristic doublet at around 1.0 ppm that integrates to three in its 1H NMR spectrum. The montbretins all display two such doublets in their 1H NMR spectra, indicating the presence of two rhamnose residues.  A proton resonance at δH 5.65 ppm (Rha1-H-1) appeared as a broad singlet and had an HMBC correlation to a carbon resonance at δC 137.5 ppm, which was previously assigned as the oxygenated aromatic carbon C-11 in the aglycone portion. This resonance at δH 5.65 ppm showed a characteristic chemical shift for a proton that is attached to a carbon that is doubly oxygenated, and was assigned as the proton at the anomeric position of rhamnose1. Rha1-H-1 (δH 5.65 ppm) also showed HMBC correlations to other carbon resonances within the sugar at δC 84.4 ppm (Rha1-C-2) and at δC 72.2 ppm (Rha1-C-3). The proton resonance at δH 4.30 ppm (Rha1-H-2) was identified as being directly attached to the carbon that resonates at δC 84.4 ppm (Rha1-C-2), and showed HMBC correlations to carbon resonances at δC 102.8 ppm (Rha1-C-1) and δC 72.2 ppm (Rha1-C-3), and then also to a resonance at δC 105.6 ppm (Glc1-C-1), assigned as the anomeric carbon of another sugar residue (glucose1). A complete set of COSY correlations between all adjacent protons identified in this rhamnose residue, plus many additional HMBC correlations, allowed for the complete assignment. This data is shown in Figure 2.24 and is summarized in Table 2.1.  59  O  OH  4  5  C-11  O O 1  6  O  COSY  3  2  OH  O  HMBC  Glc1-1 O Figure 2.24 COSY and HMBC correlations in the rhamnose1 residue of montbretin A (30).  The assignment of the Rha1-H-1 to Rha1-H-6 spin system was confirmed with 1D TOCSY NMR spectroscopy (Table 2.2). Saturation of the proton resonance at δH 4.30 ppm (Rha1-H-2), followed by observation of the resulting 1H spectrum after a delay showed progressive transfer of the magnetization down the rhamnose spin system as delay time increased (Figure 2.25). With a 10 ms delay, magnetization is transferred to the proton resonance at δH 3.73 ppm (Rha1-H-3), with a 40 ms delay to the proton resonance at δH 5.65 ppm (Rha1-H-1) and to the proton resonance at δH 3.37 ppm (Rha1-H-4), and with an 80 ms delay to the proton resonance at δH 3.55 ppm (Rha1-H-5).  60  O  Rha1-H-5 3.55ppm Rha1-H-6 1.07ppm  Rha1-H-1 5.65ppm  O  TOCSY  HO Rha1-H-3 3.73ppm  Rha1-H-4 3.37ppm  HO  Rha1-H-2 4.30ppm  O  Figure 2.25 The 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha1 spin system from Rha1-H-1 to Rha1-H-5, recorded after selective irradiation at δH 4.30 1  ppm (Rha1-H-2), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  Saturation of the proton resonance at δH 1.07 ppm (Rha1-H-6) followed by a 20 ms delay showed transfer of magnetization to proton resonances at δH 3.55 ppm (Rha1-H-5) and 3.37 ppm (Rha1-H-4), while a delay of 40 ms showed transfer to the resonance at δH 3.73 ppm (Rha1-H-3) (Figure 2.26). 61  O  Rha1-H-5 3.55ppm Rha1-H-6 1.07ppm  Rha1-H-1 5.65ppm  O  TOCSY  HO Rha1-H-3 3.73ppm  Rha1-H-4 3.37ppm  HO  Rha1-H-2 4.30ppm  O  Figure 2.26 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha1 spin system from Rha1-H-6 to Rha1-H-3, recorded after selective irradiation at δH 1.07 ppm (Rha1-H-6), 1  after a delay of: a) 20 ms b) 40 ms c) 60 ms d) 80 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  2D ROESY data supported the identification of this sugar residue as α-Lrhamnopyranoside (as previously reported), as well as confirming its connection to the aglycone portion of the molecule and to its neighboring glucosyl residue. The proton resonating at δH 62  5.65 ppm (Rha1-H-1) appeared as a singlet, and showed a ROESY correlation to the aromatic proton resonance at δH 6.91 ppm (H-20 and H-24), as well as to the resonance at δH 4.30 ppm (Rha1-H-2). The coupling constant between Rha1-H-1 and Rha1-H-2 was zero, establishing that both are situated in equatorial positions. The proton resonance at δH 4.30 ppm (Rha1-H-2) appeared as a doublet with a coupling constant of 3.3 Hz, and ROESY correlations to both its neighbors (Rha1-H-1 and Rha1-H-3) as well as to the proton resonance at δH 4.56 ppm, Glc2-H-1 (supporting the established O-glycoside linkage at this point). The coupling constant of 3.3 Hz indicated a cis axial-equatorial relationship between Rha1-H-2 and Rha1-H-3. Rha1-H-3 (at δH 3.73 ppm), assigned as axial, showed a ROESY correlation to the proton resonance at δH 3.55 ppm (Rha1-H-5), indicating that Rha1-H-5 was also positioned axial on the same face as Rha1H-3. Finally, the protons resonating at δH 1.07 ppm (Rha1-H-6) show a ROESY correlation to the proton resonance at δH 3.37 ppm (Rha1-H-4). This data is summarized in Figure 2.27.  63  OH  O H  H H-20  O  OH  H 1  5  HO  H  O 6  4  3 OH  H  H 2  O  O Glc1-1  H  Figure 2.27 Summary of key ROESY correlations in the rhamnose1 residue of montbretin A (30).  2.5.3.4 Glucose1  An HMBC correlation between the proton resonance at δH 4.30 ppm (Rha1-H-2) and the carbon resonance at δC 105.6 ppm (Glc1-C-1) established that the rhamnose1 sugar residue is attached, via an ether linkage at it's second carbon (Rha1-C-2) to the anomeric carbon on a glucose1 residue (Glc1-C-1). This is supported by the corollary HMBC correlation observed between the resonance at δH 4.56 ppm (Glc1-H-1) and the resonance at δC 84.4 ppm (Rha1-C-2) Figure 2.28). The proton resonance at δH 3.47 ppm (Glc1-H-2) showed HMBC correlations to resonances at δC 105.6 ppm (Glc1-C-1) and δC 77.8 ppm (Glc1-C-3), as well as an additional HMBC correlation to another resonance assigned as the anomeric carbon on a different sugar 64  residue (glucose2) at δC 106.4 ppm (Glc2-C-1). The proton resonance at δH 4.48 ppm (Glc1-H6a) showed HMBC correlations to the carbon resonance at δC 74.5 ppm (Glc1-C-5) and also to a resonance assigned as the carbonyl carbon, C-1 (δC 169.0 ppm), which established that the caffeoyl moiety was linked, via its terminal ester, to Glc1-C-6. This was supported by an additional HMBC correlation between the proton resonance at δH 4.20 ppm (Glc1-H-6b) and that same carbon, C-1. A complete set of COSY correlations between all adjacent protons resonances identified in this glucose residue, plus many additional HMBC correlations between proton resonances and carbon resonances in this glucose residue allowed for the complete assignment, as shown in Figure 2.28 and summarized in Table 2.1. O C-1  Rha1-2  6  O  5  O O 1  4  OH  OH  3  2 O  COSY HMBC  Glc2-1 O  Figure 2.28 COSY and HMBC correlations in the glucose1 residue of montbretin A (30).  The assignment of the Glc1-H-1 to Glc1-H-6a/b spin system was confirmed with 1D TOCSY NMR spectroscopy (Table 2.2). Saturation of the proton resonance at δH 4.56 ppm (Glc1-H-1), followed by observation of the resulting 1H spectrum after a delay showed  65  progressive transfer of the magnetization down the glucose spin system as delay time increased. With a 10 ms delay, magnetization transferred to the proton resonance at δH 3.47 ppm (Glc1-H2), with a 40 ms delay to the proton resonance at δH 3.63 ppm (Glc1-H-3), and with an 80 ms delay to the proton resonance at δH 3.38 ppm (Glc1-H-4) (Figure 2.29). Glc1-H-6b O Glc1-H-6a 4.20ppm 4.48ppm Glc1-H-5 3.55ppm  TOCSY HO  O  Glc1-H-4 3.38ppm  HO  Glc1-H-3 3.63ppm  O Glc1-H-1 4.56ppm  Glc1-H-2 3.47ppm  O  Figure 2.29 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc1 spin system from Glc1-H-1 to Glc1-H-4, recorded after selective irradiation at δH 4.56 ppm (Glc1-H-1), 1  after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  66  Selective saturation of the resonance at δH 4.20 ppm (Glc1-H-6b) followed by a delay of 10 ms showed a transfer of magnetization to the resonances at δH 4.48 ppm (Glc1-H-6a) and at δH 3.55 ppm (Glc1-H-5), while a delay of 40 ms showed a transfer of magnetization to the resonance at δH 3.38 ppm (Glc1-H-4) (Figure 2.30). Glc1-H-6b O Glc1-H-6a 4.20ppm 4.48ppm Glc1-H-5 3.55ppm  TOCSY HO  O  Glc1-H-4 3.38ppm  HO  Glc1-H-3 3.63ppm  O Glc1-H-1 4.56ppm  Glc1-H-2 3.47ppm  O  Figure 2.30 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc1 spin system from Glc1-H-6b to Glc1-H-4, recorded after selective irradiation at δH 4.20 ppm (Glc1-H-6b), 1  after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  67  2D ROESY data, plus analysis of the coupling constants between protons in this sugar residue support its established identity as being a β-D-glucopyranosyl residue. The resonance at δH 4.56 ppm (Glc1-H-1) appeared as a doublet, with a coupling constant of 7.8 Hz – indicative of an axially oriented anomeric proton with an axially oriented neighboring proton at the two position. ROESY correlations were observed between Glc1-H-1 and the proton resonances at δH 3.63 ppm (Glc1-H-3) and δH 3.55 ppm (Glc1-H-5). The resonance at δH 3.47 ppm (Glc1-H-2) showed a ROESY correlation to the resonance at δH 4.48 ppm (Glc1-H-6a), and the resonance at δH 3.38 ppm (Glc1-H-4) showed a ROESY correlation to the proton at δH 4.20 ppm (Glc1-H6b). This data is summarized in the Figure 2.31.  Ha  Hb  6  Caffeic-O  H  H  4 HO  HO  5 2  3  O-Rha1  O 1  H  ROESY H  H  O-Glc2  Figure 2.31 Summary of ROESY correlations in the glucose1 residue of montbretin A (30).  2.5.3.5 Glucose2  A previously discussed HMBC correlation between the proton resonance at δH 3.47 ppm (Glc1-H-2) and the anomeric carbon resonance at δC 106.6 ppm (Glc2-C-1) established that the second glucosyl residue was attached, via ether linkage at its anomeric position, to Glc1-C-2. 68  This is confirmed by observation of the corollary HMBC correlation between the proton resonance at δH 4.60 ppm (Glc2-H-1) and the carbon resonance at δC 84.8 ppm (Glc1-C-2). Additionally, a sufficient number of COSY and HMBC correlations were seen within the spin system of the second glucose residue (glucose2) to establish connectivity and allow for complete assignment, as shown in Figure 2.32 and summarized in Table 2.1. HO 6 O 5 4  O 1  OH 3  2  OH  COSY HMBC  OH Figure 2.32 HMBC and COSY correlations in the glucose2 residue in montbretin A (30).  The assignment of the Glc2-H-1 to Glc2-H-6a/b spin system was confirmed with 1D TOCSY NMR spectroscopy (Table 2.2). Saturation of the resonance at δH 4.60 ppm (Glc2-H-1), followed by observation of the resulting 1H spectrum after a delay showed progressive transfer of the magnetization down the glucose spin system as delay time increased. With a 10 ms delay, magnetization transferred to the proton resonance at δH 3.28 ppm (Glc2-H-2), with a 40 ms delay to the resonance at δH 3.40 ppm (Glc2-H-3), and with an 80 ms delay to the resonance at δH 3.32 (Glc2-H-4) (Figure 2.33).  69  HO Glc2-H-6b 3.74ppm  TOCSY  HO  Glc2-H-6a 3.95ppm Glc2-H-5 3.34ppm  O  Glc2-H-4 3.32ppm  HO  Glc2-H-3 3.40ppm  O Glc2-H-1 4.60ppm  Glc2-H-2 3.28ppm  OH  Figure 2.33 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc2 spin system from Glc2-H-1 to Glc2-H-4, recorded after selective irradiation at δH 4.60 ppm (Glc2-H-1), 1  after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  In order to observe the remainder of the spin system via 1D TOCSY NMR, selective saturation was done on a complex peak that corresponds to two different proton resonances (δH  70  3.95 ppm, which is Glc2-H-6a and Rha2-H-5). Saturation of the complex multiplet at δH 3.95 ppm, followed by a 10 ms delay showed transfer of magnetization to the resonance at δH 3.74 (Glc2-H-6b) as well as to resonances at δH 1.28 ppm (Rha2-H-6) and δH 3.38 ppm (Rha2-H-4). After a 40 ms delay, a transfer of magnetization to the resonance at δH 3.34 ppm (Glc2-H-5) was observed, as well as to the resonance at δH 3.69 ppm (Rha2-H-3). After an 80 ms delay, a transfer of magnetization to the resonance at δH 3.32 ppm (Glc2-H-4) was observed, as well as to the resonance at δH 3.74 ppm (Rha2-H-2) (Figure 2.34).  71  HO Glc2-H-6b 3.74ppm  TOCSY  HO  Glc2-H-6a 3.95ppm Glc2-H-5 3.34ppm  O  Glc2-H-4 3.32ppm  HO  Glc2-H-3 3.40ppm  O Glc2-H-1 4.60ppm  Glc2-H-2 3.28ppm  OH  Rha2-H-6 1.28ppm  O  Rha2-H-5 3.96ppm  Rha2-H-1 4.82ppm  O  HO Rha2-H-2 3.76ppm  Rha2-H-3 3.69ppm  Rha2-H-4 3.38ppm  HO  O  Figure 2.34 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc2 and Rha2 spin systems from Glc2-H-6a to Glc2-H-4 and from Rha2-H-6 to Rha2-H-2, recorded after selective irradiation at δH 3.95 ppm (Glc2-H-6a and Rha2-H-5), after a delay of: a) 10 ms b) 40 ms c) 80 1  ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  72  2D ROESY data plus observable coupling constants for the protons of this glucose residue supported its assignment as β-D-glucopyranosyl (as was reported in the literature). The proton resonance at δH 4.60 ppm (Glc2-H-1), with a coupling constant of 7.8 Hz, was assigned as being axial with its neighbor at the 2 position also being axial due to the size of the coupling constant between them. ROESY correlations between Glc2-H-1 and the resonances at δH 3.40 ppm (Glc2-H-3) and δH 3.34 ppm (Glc2-H-5), as well as a ROESY correlation between the resonance at δH 3.32 ppm (Glc2-H-4) and the resonance at δH 3.95 ppm (Glc2-H-6a) supported the assignment of this sugar residue as being β-D-glucopyranosyl. This data is summarized in Figure 2.35 below. H  H  6  HO  H  H  4 HO  HO  5 2  3  O-Glc1  O OH 1  H  ROESY H  H  Figure 2.35 Summary ROESY correlations in the glucose2 residue of montbretin A (30).  2.5.3.6 Xylose  A proton, resonating at δH 4.85 ppm (Xyl-H-1) was attached to a carbon resonance at δC 107.3 ppm, characteristic of a proton at the anomeric position of a sugar residue. This proton, Xyl-H-1, showed an HMBC correlation to a resonance at δC 136.2 ppm (C-22), which is assigned as the oxygenated aromatic carbon C-22, establishing that the xylose residue was attached at C-  73  22 via ether linkage at Xyl-C-1. Xyl-H-1 (δH 4.85 ppm) displayed a COSY correlation to a resonance at δH 3.60 ppm (Xyl-H-2), which in turn displayed an HMBC correlation back to the anomeric carbon resonance at δC 107.3 ppm (Xyl-C-1) as well as a resonance at δC 75.8 (Xyl-C3). The proton resonating at δH 3.56 ppm (Xyl-H-3) was attached to this carbon (Xyl-C-3), and showed a COSY correlation to a resonance at δH 3.38 ppm (Xyl-H-4). The resonance at δH 3.38 ppm (Xyl-H-4) displayed HMBC correlations to carbon resonances at δC 64.62 ppm (Xyl-C-5) and δC 75.73 (Xyl-C-3), and also to an anomeric carbon resonance belonging to another sugar residue (rhamnose2) at δC 100.0 ppm. In addition to these correlations, a sufficient number of other COSY and HMBC correlations were seen within the spin system of this xylose residue to establish connectivity and allow for complete assignment, as shown in Figure 2.36 and summarized in Table 2.1.  5  4  C-22  O 1  OH  O 3 Rha2-1 O  O  2 OH  Figure 2.36 HMBC and COSY correlations in the xylose residue of montbretin A (30).  The assignment of the Xyl-H-1 to Xyl-H-5a/b spin system was confirmed with 1D TOCSY NMR spectroscopy (Table 2.2). Saturation of the proton resonance at δH 4.85 ppm (XylH-1), followed by observation of the resulting 1H spectrum after a delay showed progressive transfer of the magnetization down the xylose spin system as delay time increased. With a 10 ms delay, magnetization transferred to the proton resonance at δH 3.60 ppm (Xyl-H-2), and with a 40 74  ms delay to the resonance at δH 3.56 ppm (Xyl-H-3) (Figure 2.37). Xyl-H-5a Xyl-H-5b 4.14ppm 3.34ppm  O  TOCSY  O Xyl-H-4 3.72ppm  HO  Xyl-H-3 3.56ppm  O Xyl-H-1 4.85ppm  Xyl-H-2 3.60ppm  OH  Figure 2.37 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Xyl spin system from Xyl-H-1 to Xyl-H-3, recorded after selective irradiation at δH 4.85 ppm (Xyl-H-1), after 1  a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  75  Selective saturation of the resonance at δH 4.14 ppm (Xyl-H-5a) followed by a delay of 10 ms showed a transfer of magnetization to the resonances at δH 3.34 ppm (Xyl-H-5b) and δH 3.72 ppm (Xyl-H-4), while a delay of 40 ms showed a transfer of magnetization to the resonances at δH 3.56 ppm (Xyl-H-3) and δH 3.60 ppm (Xyl-H-2), and finally, a delay of 80 ms showed a transfer of magnetization to the resonance at δH 4.85 ppm (Xyl-H-1) (Figure 2.38). These observed 1D TOCSY correlations, in conjunction with the observed HMBC and COSY correlations within the Xyl residue, allowed the complete assignment of that sugar, as summarized in Table 2.1.  76  Xyl-H-5a Xyl-H-5b 4.14ppm 3.34ppm  O  O Xyl-H-4 3.72ppm  HO  Xyl-H-3 3.56ppm  O Xyl-H-1 4.85ppm  Xyl-H-2 3.60ppm  TOCSY  OH  Figure 2.38 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Xyl spin system from Xyl-H-5a to Xyl-H-1, recorded after selective irradiation at δH 4.14 ppm (Xyl-H-5a), 1  after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  2D ROESY correlations and analysis of observable coupling constants of the protons in the xylose residue supported its established literature assignment as β-D-xylopyranoside. The 77  proton resonating at δH 4.85 ppm (Xyl-H-1) appeared as a doublet with a coupling constant of 7.38 Hz, allowing to it be assigned axial with its nearest neighbor (Xyl-H-2) also in an axial position. Xyl-H-1 showed a ROESY correlation to the resonances at δH 3.56 ppm (Xyl-H-3) and δH 3.34 ppm (Xyl-H-5b). The proton resonance at δH 3.60 ppm (Xyl-H-2) displayed a ROESY correlation to the resonance at δH 3.72 ppm (Xyl-H-4). Finally, the resonance at δH 4.14 (Xyl-H5a) showed a ROESY correlation to the resonance at δH 4.82 ppm (Rha2-H-1), supporting the connection to the rhamnose2 residue at this position (Xyl-C-4). This data is summarized in Figure 2.39. Ha  H  H H 4  5  HO O  Rha2-1 O  O 2  3  O-Aglycone 1  OH Hb  ROESY  H  H  Figure 2.39 Summary of ROESY correlations in the xylose residue of montbretin A (30).  2.5.3.7 Rhamnose2  As noted previously, an HMBC correlation between the proton resonance at δH 3.72 ppm (Xyl-H-4) to a carbon resonance at δC 100.0 ppm (Rha2-C-1) indicated that a rhamnose2 was connected via ether linkage at its anomeric position (Rha2-C-1) to Xyl-C-4. This was confirmed by observation of the corollary HMBC correlation between the proton resonance at δH 4.82 ppm (Rha2-H-1) and the carbon resonance at δC 75.3 ppm (Xyl-C-4). A sufficient number of COSY  78  and HMBC correlations were additionally seen within the spin system to establish connectivity and allow for complete assignment of rhamnose2, as shown in Figure 2.40 and summarized in Table 2.1.  OH  O O  COSY OH  OH  HMBC  Figure 2.40 HMBC and COSY correlations in the rhamnose2 residue of montbretin A (30).  The assignment of the Rha2-H-1 to Rha2-H-6 spin system was confirmed with 1D TOCSY NMR spectroscopy. Saturation of the proton resonance at δH 1.28 ppm (Rha2-H-6), followed by observation of the resulting 1H spectrum after a delay showed progressive transfer of the magnetization down the rhamnose spin system as delay time increased. With a 20 ms delay, magnetization transferred to the proton resonance at δH 3.96 ppm (Rha2-H-2), with a 40 ms delay to the resonance at δH 3.38 ppm (Rha2-H-3), and with a 60 ms delay to the resonance at δH 3.69 ppm (Rha2-H-3) (Figure 2.41).  79  Rha2-H-6 1.28ppm  O  Rha2-H-5 3.96ppm  Rha2-H-1 4.82ppm  O  TOCSY  HO Rha2-H-2 3.76ppm  Rha2-H-3 3.69ppm  Rha2-H-4 3.38ppm  HO  O  Figure 2.41 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha2 spin system from Rha2-H-6 to Rha2-H-3, recorded after selective irradiation at δH 1.28 ppm (Rha2-H-6), 1  after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  Selective saturation of the resonance at δH 4.82 ppm (Rha2-H-1) followed by a delay of 10 ms showed a transfer of magnetization to the resonance at δH 3.76 ppm (Rha2-H-2) (Figure 2.42). As previously stated in the discussion of the 1D TOCSY of the glucose2 residue, selective 80  saturation of the complex multiplet at 3.95 ppm (Rha2-H-5) showed progressive transfer of magnetization to Rha2-H-4 and Rha2-H-3, as well as transfer of magnetization to Rha2-H-6 (Figure 2.34). These observed 1D TOCSY correlations, along with the observed HMBC and COSY correlations within the rhamnose2 residue, allowed the complete assignment of the sugar as summarized in Table 2.1.  Rha2-H-6 1.28ppm  O  Rha2-H-5 3.96ppm  Rha2-H-1 4.82ppm  O  TOCSY HO Rha2-H-2 3.76ppm  Rha2-H-3 3.69ppm  Rha2-H-4 3.38ppm  HO  O  Figure 2.42 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Rha2 spin system from Rha2-H-1 to Rha2-H-2, recorded after selective irradiation at δH 4.82 ppm (Rha2-H-1), 1  after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  81  2D ROESY data plus the analysis of observable coupling constants confirms the literature assignment of this sugar residue as α-L-rhamnopyroside. The proton resonance at δH 4.82 ppm (Rha2-H-1) appeared as a doublet with a coupling constant of 1.7 Hz, with a ROESY correlation to its nearest neighbor Rha2-H-2 at δH 3.76 ppm. The size of this coupling constant indicated a trans equatorial-equatorial positioning for these two protons. The proton resonance at δH 3.76 ppm (Rha2-H-2) appeared a doublet of doublets with a second coupling constant of 3.3 Hz, which it shares with its other neighbor, the proton resonance at δH 3.69 ppm (Rha2-H-3). The magnitude of this coupling constant is consistent with a cis axial-equatorial relationship between the two. The proton resonance at δH 3.69 ppm (Rha2-H-3) appears as a doublet of doublets, with a second coupling constant of 9.5 Hz, indicative of a trans axial-axial relationship with its neighbor, Rha2-H-4. Rha2-H-3 also displayed a ROESY correlation to a proton resonance at δH 3.96 ppm (Rha2-H-5), while the proton resonance at 3.38 ppm (Rha2-H-4) displayed a ROESY correlation to the three-proton doublet resonating at δH 1.28 ppm (Rha2-H-6). This data confirmed the assignment of this residue as α-L-rhamnopyroside, and is summarized in the Figure 2.43. H  O-Xyl H  5  HO 6  1 O  4 H  3 OH  H  H  2 OH ROESY  Figure 2.43 ROESY correlations in the rhamnose2 residue of montbretin A (30).  82  2.5.4 Structure Elucidation of Montbretin B O  OH  HO  O  OH O OH O  OH O OH O  CH3 O  HO  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  31  Figure 2.44 Montbretin B (31).  Montbretin B (Figure 2.44) was isolated as yellow powder that gave an [M+Na]+ pseudomolecular ion at m/z 1235.3184 in its high resolution electrospray ionization time-offlight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C53H64O32 (C53H64O32 Na calculated for m/z 1251.3278), indicating 22 degrees of unsaturation, and differing from montbretin A by having one less oxygen atom. Analysis of 1H and 13C NMR, as well as 2D NMR (COSY, HSQC, HMBC, ROESY) and 1D TOCSY NMR data revealed that the compound matched the published structure of montbretin B.  83  O  OH  HO  O  OH O OH  O CH3  O O HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  31  Figure 2.45 600 MHz 1H NMR spectrum of montbretin B (31), recorded in MeOD-d4.  84  O  OH  HO  O  OH O OH  O CH3  O O HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  31  13  Figure 2.46 150 MHz C NMR spectrum of montbretin B (31), recorded in MeOD-d4.  85  O  OH  HO  O  OH O OH  O CH3  O O HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  31  Figure 2.47 600 MHz COSY spectrum of montbretin B (31), recorded in MeOD-d4.  86  O  OH  HO  O  OH O OH  O CH3  O O HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  31  Figure 2.48 600 MHz HSQC spectrum of montbretin B (31), recorded in MeOD-d4.  87  O  OH  HO  O  OH O OH  O CH3  O O HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  31  Figure 2.49 600 MHz HMBC spectrum of montbretin B (31), recorded in MeOD-d4.  88  O  OH  HO  O  OH O OH  O CH3  O O HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  31  Figure 2.50 600 MHz TROESY spectrum of montbretin B (31), recorded in MeOD-d4.  89  1  Table 2.3 1D and 2D NMR data for montbretin B (31), recorded at 600 MHz ( H) and 150 MHz  (13C) in MeOD-d4 O OH 23 22 24 17 21 HO 18 O 10 19 20 OH 16 11 12 15 O 14 13 OH O OH 6  5  2  4 3  7  1  O  8  Rha1  OH  O  O  O  Xyl  OH  O  CH3  Rha2  OH OH  OH O  9  HO  OH  O CH3  O  OH  Glc1  OH OH O OH OH  O  Glc2  OH  31  C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  p-Coumaric Acid  a  1  169  -  -  -  -  2  114.9  6.07 d (16)  C-1, C-4  H-3  H-3  3  146.8  7.44 d (16)  C-1, C-2, C-4, C-5/9  H-2  H-2, H-5, H-6  4  127.1  -  -  -  -  5  131.2  7.22 d (7.9)  C-4, C-6, C-7, C-9  H-6  H-2, H-6  6  116.8  6.68 d (7.9)  C-3, C-5, C-7  H-5  7  161.3  -  -  -  -  8  116.8  6.68 d (7.9)  C-3, C-7, C-9  H-9  H-3, H-9  9  131.2  7.22 d (7.9)  C-4, C-5, C-7, C-8  H-8  H-2, H-6  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz. b  c  90  1  Table 2.3 (Continued) 1D and 2D NMR data for montbretin B (31), recorded at 600 MHz ( H)  and 150 MHz (13C) in MeOD-d4 C#  13  C δ  (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Aglycone 10  157.9  -  -  -  -  11  136.2  -  -  -  -  12  179.7  -  -  -  -  13  105.8  -  -  -  -  14  163.3  -  -  -  -  15  100.5  6.16 d (1.9)  C-13, C-14, C-16, C-17  -  H-17  16  167  -  -  -  -  17  95.3  6.15 d (1.9)  C-13, C-15, C-16, C-18  18  158.6  -  -  -  -  19  128.6  -  -  -  -  20/24  109.8  6.89 s, 2H  C-10, C-18, C-19, C-20/24, C-21/23, C-22  -  Rha1-H-1  21/23  152.1  -  -  -  -  22  137.4  -  -  -  -  H-15  Rhamnose-1, Rha1 Rha1-1  102.7  5.68 s  C-11, Rha1-C-2, Rha1-C-3  Rha1-H-2  Rha1-H-2, H-20/24  Rha1-2  84.5  4.31 d (3.3)  Rha1-C-3, Rha1-C-4, Glc1-C-1  Rha1-H-1, Rha1-H-3  Rha1-H-1, Rha1-H-3, Glc1-H-1  Rha1-3  72.2  3.75 m  -  Rha1-H-2, Rha1-H-4  Rha1-H-5  Rha1-4  74.1  3.38 m  Rha1-C-6  Rha1-H-3, Rha1-H-5  Rha1-H-6  Rha1-5  72  3.56 m  -  Rha1-H-4, Rha1-H-6  Rha1-H-3, Rha1-H-6  Rha1-6  17.9  1.08 d (6)  Rha1-C-4, Rha1-C-5  Rha1-H-5  Rha1-H-4  Glc1-1  105.5  4.56 d (7.8)  Rha1-C-2, Glc1-C-2  Glc1-H-2  Glc1-H-3, Glc1-H-5  Glc1-2  84.7  3.49 m  Glc1-C-1, Glc1-C-3, Glc2-C-1  Glc1-H-1, Glc1-H-3  Glc1-H-6a  Glc1-3  77.9  3.64 m  Glc1-C-2, Glc1-C-4  Glc1-H-2, Glc1-H-4  Glc1-H-1  Glc1-4  71.7  3.38 m  Glc1-C-3, Glc1-C-5, Glc1-C-6  Glc1-H-3, Glc1-H-5  Glc1-H-6b  Glc1-5  75  3.53 m  -  Glc1-H-4, Glc1-H-6  Glc1-H-1  Glc1-6  64.4  H-6a: 4.5 d (12)  C-1, Glc1-C-5  Glc1-H-6b  Glc1-H-2, Glc1-H-6b  H-6b: 4.17 dd (12, 5)  C-1, Glc1-C-5  Glc1-H-6a, Glc1-H-5  Glc1-H-4, Glc1-H-6a  Glucose-1, Glc1  a  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  91  1  Table 2.3 (Continued) 1D and 2D NMR data for montbretin B (31), recorded at 600 MHz ( H)  and 150 MHz (13C) in MeOD-d4 C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Glc2-H-3, Glc2-H-5  Glucose-2, Glc2 Glc2-1  106.6  4.6 d (7.8)  Glc1-C-2  Glc2-H-2  Glc2-2  76  3.28 m  Glc2-C-1, Glc2-C-3  Glc2-H-1  Glc2-3  79.3  3.34 m  -  -  Glc2-H-1  Glc2-4  71.1  3.32 m  Glc2-C-3, Glc2-C-5  -  Glc2-H-6a  Glc2-5  78  3.38 m  Glc2-C-4, Glc2-C-6  Glc2-H-6  Glc2-H-1  Glc2-6  62.5  H-6a: 3.94 m  -  Glc2-H-6b  Glc2-H-4, Glc2-H-6b  H-6b: 3.75 m  -  Glc2-H-5, Glc2-H-6a  Glc2-H-6a  Xylose, Xyl Xyl-1  107.2  4.84 d (7.4)  C-22, Xyl-C-5,  Xyl-H-2  Xyl-H-3, Xyl-H-5b  Xyl-2  75.3  3.61 m  Xyl-C-1, Xyl-C-3  Xyl-H-1  Xyl-H-4  Xyl-3  75.8  3.56 m  Xyl-C-2  Xyl-H-4  Xyl-H-1  Xyl-4  75.4  3.72 m  Xyl-C-5, Xyl-C-3, Rha2-C-1  Xyl-H-3, Xyl-H-5a, Xyl-H-5b  Xyl-H-5a  Xyl-5  64.6  H-5a: 4.15 dd (12, 5)  Xyl-C-1, Xyl-C-4  Xyl-H-4, Xyl-H-5b  Xyl-H-4, Xyl-H-5b, Rha2-H-1  H-5b: 3.37 m  Xyl-C-1, Xyl-C-4  Xyl-H-4, Xyl-H-5a  Xyl-H-1, Xyl-H-3, Xyl-H-5a  Rhamnose-2, Rha2  a  Rha2-1  100  4.82 d  Xyl-C-4, Rha2-C-2, Rha2-C-5  Rha2-H-2  Rha2-H-2, Xyl-H-5a  Rha2-2  72.3  3.78 dd (3.3, 1.7)  Rha2-C-1  Rha2-H-1  Rha2-H-1  Rha2-3  72.5  3.69 dd (9.5, 3.3)  Rha2-C-4  Rha2-H-4  Rha2-H-5  Rha2-4  74  3.39 m  Rha2-C-3, Rha2-C-5  Rha2-H-3, Rha2-H-5  Rha2-H-6  Rha2-5  70.3  3.95 m  Rha2-C-4, Rha2-C-6  Rha2-H-4, Rha2-H-6  Rha2-H-3  Rha2-6  18.1  1.28 d (6)  Rha2-C-4, Rha2-C-5  Rha2-H-5  Rha2-H-4  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  92  Analysis of the aromatic region of the 1H NMR spectrum of montbretin B showed that as compared to montbretin A, the caffeic acid ester was replaced by a p-coumaric acid ester. Specifically, this was seen by the absence of the three caffeic acid aromatic proton resonances (δH 6.86 ppm, δH 6.66 ppm and δH 6.75 ppm) and the appearance of two doublets that integrated to two protons each at δH 7.22 ppm (H-5 and H-9) and δH 6.68 ppm (H-6 and H-8), and also share a coupling constant of 8.0 Hz. The p-coumaric acid structure was confirmed by the relevant HMBC correlations within the aromatic system and into the trans double bond, summarized in Figure 2.51. O 5  COSY 6  HMBC  7  4 8  3  2  1 O  9  HO  Figure 2.51 COSY and HMBC correlations in the p-coumaric acid reside of montbretin B (31).  The remainder of the molecule (the aglycone portion and the sugar residues) was confirmed to match montbretin B by analysis of the COSY, HMBC, HSQC, ROESY and 1D TOCSY data, and comparison to the values reported in literature65. This data is summarized in Table 2.3.  93  2.5.5 Structure Elucidation of Montbretin C O  OH  HO  O  OH O OH O MeO  OH O OH O  CH3 O  HO  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  51  Figure 2.52 Montbretin C (51).  Montbretin C (51) (Figure 2.52) was isolated as a yellow powder that gave an [M+Na]+ pseudomolecular ion at m/z 1265.3381 in its high resolution electrospray ionization time-offlight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C54H66O33 (C54H66O33 Na calculated for m/z 1265.3384), indicating 22 degrees of unsaturation and differing from montbretin A by the addition of a -CH2-. Analysis of 1H and 13C NMR, as well as 2D NMR (COSY, HSQC, HMBC, ROESY) and 1D selective NOE and TOCSY NMR data revealed that the compound was exceedingly similar to montbretin A, with the exception being a pronounced singlet at δH 3.71 ppm in the 1H NMR spectrum, and a new carbon resonance at δC 55.8 ppm in the 13C NMR spectrum. This information lead to the conclusion that montbretin C was a methyl ether of montbretin A. 94  O  OH  HO  O  OH O OH  O CH3  O MeO  O  HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  1  Figure 2.53 600 MHz H NMR spectrum of montbretin C (51), recorded in MeOD-d4.  95  O  OH  HO  O  OH O OH  O CH3  O MeO  O  HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  13  Figure 2.54 150 MHz C NMR spectrum of montbretin C (51), recorded in MeOD-d4.  96  O  OH  HO  O  OH O OH  O CH3  O MeO  O  HO  OH O OH  O  O CH3  OH  O OH  OH OH  OH O OH  O  OH OH O OH OH  O OH  Figure 2.55 600 MHz COSY spectrum of montbretin C (51), recorded in MeOD-d4.  97  O  OH  HO  O  OH O OH  O CH3  O MeO  O  HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  Figure 2.56 600 MHz HSQC spectrum of montbretin C (51), recorded in MeOD-d4.  98  O  OH  HO  O  OH O OH  O CH3  O MeO  O  HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  Figure 2.57 600 MHz HMBC spectrum of montbretin C (51), recorded in MeOD-d4.  99  O  OH  HO  O  OH O OH  O CH3  O MeO  O  HO  OH O OH  O O  OH  O OH  CH3  OH OH  OH O OH  O  OH OH O OH OH  O OH  Figure 2.58 600 MHz TOCSY NMR spectrum of montbretin C (51), recorded in MeOD-d4.  100  Table 2.4 1D and 2D NMR data for montbretin C (51), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. O OH 23 22 24 17 21 HO 18 O 10 19 16 20 OH 11 12 15 O 14 13 OH O OH MeO 6  5  3  7 HO  2  4  1  OH  O CH3  O O  Rha1  OH  O  O  Xyl  OH  O  CH3  Rha2  OH OH  OH O  9 8  OH  O  Glc1  OH OH O OH OH  O  Glc2  OH  51  C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Caffeic Acid  a  1  168.9  -  -  -  -  2  115.1  6.1 d (16)  C-1, C-4  H-3  H-5  3  147.3  7.48 d (16)  C-1, C-2, C-5, C-9  H-2  4  127.5  -  -  -  -  5  110.9  6.92 d (1.9)  C-3, C-6, C-7, C-9  H-9  H-2, H-6'  6  149.8  -  -  -  -  7  150.7  -  -  -  -  8  116.9  6.68 d (7.9)  C-4, C-6, C-7  H-9  -  9  123.7  6.85 dd (7.9, 1.9)  C-3, C-5, C-7  H-5, H-8  -  6'  55.8  3.78 s  C-6  -  H-5  Recorded at 150 MHz. b Assigned according to HSQC recorded at 600 MHz. c Recorded at 600 MHz.  101  Table 2.4 (Continued) 1D and 2D NMR data for montbretin C (51), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4 C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Aglycone 10  157.9  -  -  -  -  11  137.1  -  -  -  -  12  178.4  -  -  -  -  13  104.9  -  -  -  -  14  163.4  -  -  -  -  15  101.5  6.07 d (1.9)  C-13, C-14, C-17  -  -  16  164.1  -  -  -  -  17  95.5  6.05 d (1.9)  C-13, C-18  -  -  18  158.9  -  -  -  -  19  128.8  -  -  -  -  20/24  109.3  6.83 s, 2H  C-10, C-19, C-20/24, C-21/23, C-22  -  -  21/23  152.7  -  -  -  -  22  136.2  -  -  -  -  Rhamnose-1, Rha1 Rha1-1  103  5.69 s  C-11, Rha1-C-2, Rha1-C-3  Rha1-H-2  Rha1-2  83.92  4.3 d (3.3)  -  Rha1-H-1, Rha1-H-3  Rha1-H-3, Glc1-H-1  Rha1-3  72.22  3.74 m  -  Rha1-H-2, Rha1-H-4  Rha1-H-2  Rha1-4  74.04  3.37 m  Rha1-C-6  Rha1-H-3, Rha1-H-5  -  Rha1-5  71.98  3.55 m  -  Rha1-H-4, Rha1-H-6  -  Rha1-6  17.8  1.07 d )6)  Rha1-C-4, Rha1-C-5  Rha1-H-5  -  Glc1-1  105.5  4.56 d(7.8)  Rha1-C-2, Glc1-C-2  Glc1-H-2  Rha1-H-2  Glc1-2  84.7  3.47 m  Glc1-C-1  Glc1-H-1, Glc1-H-3  Glc2-H-H-1  Glc1-3  77.9  3.63 m  Glc1-C-2  Glc1-H-2, Glc1-H-4  Glc1-4  71.7  3.38 m  Glc1-H-3, Glc1-H-5  Glc1-5  75  3.55 m  Glc1-H-4, Glc1-H-6a, Glc1-H-6b  Glc1-6  64.4  H-6a: 4.56 d(7.8)  Glc1-H-5, Glc1-H-6b  Glc1-H-6b  Glc1-H-5, Glc1-H-6a  Glc1-H-6a  Glucose-1, Glc1  H-6b: 3.47 m a  b  C-1 c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  102  Table 2.4 (Continued) 1D and 2D NMR data for montbretin C (51), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4  C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Glucose-2, Glc2 Glc2-1  107.1  4.59 d (7.8)  Glc1-C-2  Glc2-H-2  Glc1-H-2  Glc2-2  76  3.27 m  Glc2-C-1, Glc2-C-3  Glc2-H-1, Glc2-H-3  -  Glc2-3  78.5  3.33 m  -  Glc2-H-2  -  Glc2-4  71.07  3.32 m  -  -  -  Glc2-5  77.9  3.37 m  -  Glc2-H-6a, Glc2-H-6b  -  Glc2-6  62.5  H-6a: 3.93 m  -  Glc2-H-5, Glc2-H-6b  Glc2-H-6b  H-6b: 3.74 m  -  Glc2-H-5, Glc2-H-6a  Glc2-H-6a  Xylose, Xyl Xyl-1  107.4  4.82 d (7.4)  C-22  Xyl-H-2  -  Xyl-2  74.99  3.6 m  Xyl-C-1, Xyl-C-3  Xyl-H-1, Xyl-H-3  -  Xyl-3  75.73  3.56 m  -  Xyl-H-2, Xyl-H-4  -  Xyl-4  75.29  3.72 m  -  Xyl-H-3, Xyl-H-5a, Xyl-H-5b  Xyl-H-5a  Xyl-5  64.6  H-5a: 4.14 dd (12, 5)  -  Xyl-H-4, Xyl-H-5b  Xyl-H-4  H-5b: 3.36 m  -  Xyl-H-4, Xyl-H-5a  -  Rhamnose-2, Rha2  a  Rha2-1  99.6  4.8 d (1.7)  Xyl-C-4, Rha2-C-2, Rha2-C-5  Rha2-H-2  -  Rha2-2  72.33  3.77 dd (3.3, 1.7)  --  Rha2-H-1, Rha2-H-3  -  Rha2-3  72.53  3.69 dd (9.5, 3.3)  -  Rha2-H-2, Rha2-H-4  -  Rha2-4  74  3.37 m  Rha2-C-6  Rha2-H-3, Rha2-H-5  -  Rha2-5  70.2  3.96 m  -  Rha2-H-4, Rha2-H-6  Rha2-H-6  Rha2-6  17.9  1.27 d (6)  Rha2-C-4, Rha2-C-5  Rha2-H-5  Rha2-H-5  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  103  Analysis of the 1H, 13C, HSQC, HMBC, COSY and ROESY spectra of montbretin C confirmed that the core structure of montbretin C (specifically the aglycone and sugar residues) was identical to montbretin A. Analysis of the 2D HMBC spectrum quickly revealed a correlation from the proton resonance assigned to the methyl ether (δH 3.71 ppm) to a resonance assigned as an oxygenated aromatic carbon at δC 149.8 ppm (C-6). HMBC correlations to this carbon resonance from the proton resonances at δH 6.93 ppm (H-5) and δH 6.69 ppm (H-8) did not clarify which of the two phenols was methylated, especially given the fact that the resonances corresponding to H-5 and H-8 also showed an HMBC correlation to the other phenolic carbon at δC 150.7 ppm (C-7). The proton resonance at δH 3.71 ppm (H-6') displayed a ROESY correlation the the proton resonance at δH 6.93 ppm (H-5). This correlation was only possible if the methyl ether were placed on C-6. Thus, montbretin C has a ferulic acid ester instead of a caffeic or pcoumaric acid ester. All three of these acids are common components of plant natural products. This NMR data is summarized in Table 2.4 and in the Figure 2.59. H  6'  O 3  O 6  2  4  5  1 O  9 HO  7  H 8 H  COSY HMBC ROESY  Figure 2.59 Summary of the key COSY, HMBC and ROESY correlations in the ferulic acid ester of montbretin C (51).  104  2.5.6 Structure Elucidation of Montbretin D O OH O  HO Glc3 OH OH  O  O  OH OH  O OH O OH  OH  O CH3  O O HO  OH Rha1  O  O  CH3  Xyl  OH Rha2  OH OH  OH O OH  O  O  Glc1  OH OH O OH OH  O  Glc2  OH  52  Figure 2.60 Montbretin D (52).  Montbretin D (52) (Figure 2.60) was isolated as a brown powder that gave an [M+Na]+ pseudomolecular ion at m/z 1397.3792 in its high resolution electrospray ionization time-offlight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C59H74O37 (C59H74O37Na calculated for m/z 1397.3807), indicating 23 degrees of unsaturation.  105  O OH O  HO OH OH  O  OH  O  OH  O OH O OH  OH O  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.61 600 MHz 1H NMR spectrum of montbretin D (52), recorded in MeOD-d4.  106  O OH O  HO OH OH  O  OH  O  OH  O OH O OH  OH O  OH  O CH3  O  CH3 O  HO  OH  OH OH  OH O OH  O  O  O  OH OH O OH OH  O OH  13  Figure 2.62 150 MHz C NMR spectrum of montbretin D (52), recorded in MeOD-d4.  107  O OH O  HO OH OH  O  OH  O  OH  O OH O OH  OH O  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.63 600 MHz COSY spectrum of montbretin D (52), recorded in MeOD-d4.  108  O OH O  HO OH OH  O  OH  O  OH  O OH O OH  OH O  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.64 600 MHz HSQC spectrum of montbretin D (52), recorded in MeOD-d4.  109  O OH O  HO OH OH  O  O  OH OH  O OH O OH  OH  O CH3  O O HO  OH O OH  OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.65 600 MHz HMBC spectrum of montbretin D(52), recorded in MeOD-d4.  110  O OH O  HO OH OH  O  O  OH OH  O OH O OH  OH  O CH3  O O HO  OH O OH  OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.66 600 MHz TROESY spectrum of montbretin D (52), recorded in MeOD-d4.  111  Table 2.5 1D and 2D NMR data for montbretin D (52), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. O OH 23 22 24 17 21 O HO 18 O 10 19 20 OH 16 O Glc3 11 OH O 15 12 OH O 14 13 O OH OH OH O OH O OH OH O CH3 O Rha1 CH 3 5 2 6 4 OH OH O OH O 3 1 7 9 HO O 8 OH Glc1 OH OH O OH OH  O  Xyl  Rha2  Glc2  OH  52  C#  13  1 b C H δ δ (ppm) (ppm)a mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  p-Coumaric Acid  a  1  168.9  -  -  -  -  2  115  6.06 d (16)  C-1, C-4  H-3  H-3  3  146.8  7.41 d (16)  C-1, C-2, C-4, C-5/9  H-2  H-2  4  127.1  -  -  -  -  5  131.1  7.19 d (7.9)  C-4, C-7, C-6/8  H-6  -  6  116.9  6.75 d (7.9)  C-3, C-5, C-7, C-6/8  H-5  -  7  161.1  -  -  -  -  8  116.9  6.75 d (7.9)  C-3, C-5, C-7, C-6/8  H-9  -  9  131.1  7.19 d (7.9)  C-4, C-7, C-6/8  H-8  -  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz. b  c  112  Table 2.5 (Continued) 1D and 2D NMR data for montbretin D (52), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4.  C#  13  1 b C H δ δ (ppm) (ppm)a mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Aglycone 10  158.6  -  -  -  -  11  136.3  -  -  -  -  12  180.1  -  -  -  -  13  105.5  -  -  -  -  14  162  -  -  -  -  15  101.1  6.45 d (1.9)  C-13, C-14, C-16, C-17  -  H-17  16  164.9  -  -  -  -  17  96  6.54 d (1.9)  C-12, C-13, C-15, C-16, C-18  -  H-15  18  158  -  -  -  -  19  128.5  -  -  -  -  20/24  109.9  6.7 s, 2H  C-10, C-19, C-20/24, C-21/23, C-22  -  Rha1-H-1  21/23  152.1  -  -  -  -  22  137.7  -  -  -  -  Rhamnose-1, Rha1 Rha1-1  102.7  5.69 s  C-11, Rha1-C-2, Rha1-C-3  Rha1-H-2  Rha1-H-2, 20/24  Rha1-2  84.4  4.29 s (3.3)  Rha1-C-3, Rha1-C-4, Glc1-C-1  Rha1-H-1, Rha1-H-3  Rha1-H-1, Rha1-H-3, Glc1-H-1  Rha1-3  72.2  3.74 m  Rha1-C-4, Rha1-C-5  Rha1-H-2, Rha1-H-4  Rha1-H-2  Rha1-4  74  3.38 m  Rha1-C-3, Rha1-C-5, Rha1-C-6  Rha1-H-3, Rha1-H-5  Rha1-H-6  Rha1-5  72  3.58 m  Rha1-C-1, Rha1-C-4, Rha1-C-6  Rha1-H-4, Rha1-H-6  Rha1-H-6  Rha1-6  17.9  1.1 d (6)  Rha1-C-4, Rha1-C-5  Rha1-H-5  Rha1-H-4, Rha1-H-5  Glucose-1, Glc1  a  Glc1-1  106.6  4.55 d (7.8)  Rha1-C-2, Glc1-C-2  Glc1-H-2  Glc1-H-3  Glc1-2  84.7  3.49 m  Glc1-C-1, Glc1-C-3, Glc1-C-4, Glc2-C-1  Glc1-H-1, Glc1-H-3  Glc1-H-4  Glc1-3  77.9  3.64 m  Glc1-C-2, Glc1-C-4  Glc1-H-2, Glc1-H-4  Glc1-H-1  Glc1-4  71.3  3.38 m  Glc1-C-3, Glc1-C-5  Glc1-H-3, Glc1-H-5  Glc1-H-2, Glc1-H-6a  Glc1-5  74.9  3.5 m  Glc1-C-3, Glc1-C-4, Glc1-C-6  Glc1-H-4, Glc1-H-6a/b  Glc1-6  64.4  H6a: 4.54 d (12)  C-1  Glc1-H-6b, Glc1-H-5  Glc1-H-6b  H6b: 4.13 m  C-1, Glc1-C-4, Glc1-C-5  Glc1-H-6a, Glc1-H-5  Glc1-H-6a, Glc2-H-1  Recorded at 150 MHz. b Assigned according to HSQC recorded at 600 MHz. c Recorded at 600 MHz.  113  Table 2.5 (Continued) 1D and 2D NMR data for montbretin D (52), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. C#  13  1 b C H δ δ (ppm) (ppm)a mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  ROESYc (H→H)  Glc2-H-3, Glc2-H-5, Glc1-H-6b  Glucose-2, Glc2 Glc2-1  107.2  4.62 d (7.8)  Glc1-C-2  Glc2-H-2  Glc2-2  76  3.31 m  Glc2-C-1, Glc2-C-3  Glc2-H-1, Glc2-H-3  Glc2-3  79.3  3.27 m  Glc2-C-4, Glc2-C-5  Glc2-H-2, Glc2-H-4  Glc2-H-1  Glc2-4  71.1  3.52 m  Glc2-C-3, Glc2-C-5  Glc2-H-3, Glc2-H-5  Glc2-H-6b  Glc2-5  77.9  3.40 m  Glc2-C-1, Glc2-C-3, Glc2-C-4  Glc2-H-4, Glc2-H-6a  Glc2-H-1  Glc2-6  62.5  H-6a: 3.95 m  Glc2-C-4, Glc2-C-5  Glc2-H-5, Glc2-H-6b  Glc2-H-6b  H-6b: 3.75 m  Glc2-C-3, Glc2-C-4, Glc2-C-5  Glc2-H-5, Glc2-H-6a  Glc2-H-4, Glc2-H-6a  Xylose, Xyl Xyl-1  108  4.85 d (7.4)  C-22, Xyl-C-5,  Xyl-H-2  Xyl-H-3, Xyl-H-5b  Xyl-2  75.5  3.61 m  Xyl-C-1, Xyl-C-3, Xyl-C-4  Xyl-H-1, Xyl-H-3  Xyl-H-4  Xyl-3  75.7  3.58 m  Xyl-C-2, Xyl-C-4  Xyl-H-2, Xyl-H-4  Xyl-H-1  Xyl-4  75.3  3.73 m  Xyl-C-2, Xyl-C-3, Xyl-C-5, Rha2-C-1  Xyl-H-3, Xyl-H-5b  Xyl-H-2  Xyl-5  64.6  H-5a: 4.15 m  Xyl-C-3, Xyl-C-4  Xyl-H-4, Xyl-H-5b  Xyl-H-4, Rha2-H-1  H-5b: 3.36 m  Xyl-C-1, Xyl-C-3, Xyl-C-4  Xyl-H-5a  Xyl-H-1, Xyl-H-3  Rhamnose-2, Rha2 Rha2-1  99.9  4.83 d (1.7)  Xyl-C-4, Rha2-C-2, Rha2-C-3, Rha2-C-4  Rha2-H-2  Rha2-H-2, Xyl-H-5a  Rha2-2  72.3  3.78 m  Rha2-C-1, Rha2-C-3  Rha2-H-1, Rha2-H-3  Rha2-H-1  Rha2-3  72.5  3.68 dd (9.5, 3.3)  Rha2-C-2, Rha2-C-4  Rha2-H-2, Rha2-H-4  Rha2-4  74  3.4 m  Rha2-C-3, Rha2-C-5, Rha2-C-6  Rha2-H-3, Rha2-H-5  Rha2-H-6  Rha2-5  70.6  3.95 m  Rha2-C-6  Rha2-H-4, Rha2-H-6  Rha2-H-6  Rha2-6  18.1  1.3 d (6)  Rha2-C-4, Rha2-C-5  Rha2-H-5  Rha2-H-4, Rha2-H-5  Glucose-3, Glu3  a  Glc3-1  101.7  5.05 d (7.6)  C-16  Glc3-H-2  15, 17, Glc3-H-3, Glc3-H-5  Glc3-2  78  3.52 m  Glc3-C-3, Glc3-C-4, Glc3-C-5  Glc3-H-1  Glc3-H-4  Glc3-3  78.5  3.52 m  Glc3-C-2, Glc3-C-4  Glc3-H-4  Glc3-H-1  Glc3-4  71.8  3.42 m  Glc3-C-2, Glc3-C-3, Glc3-C-5  Glc3-H-3, Glc3-H-5  Glc3-H-2  Glc3-5  75.4  3.55 m  Glc3-C-3, Glc3-C-4, Glc3-C-6  Glc3-H-4, Glc3-H-6a/b  Glc3-H-1  Glc3-6  62.6  H-6a: 3.74 m  Glc3-C-4, Glc3-C-5  Glc3-H-5, Glc3-H-6b  Glc3-H-6b  H-6b: 3.95 m  Glc3-C-4  Glc3-H-5, Glc3-H-6a  Glc3-H-6a  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  114  Table 2.6 1D TOCSY data for montbretin D (52), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. Sugar  Position  TOCSYa (H→H)  Rha1-H-1 Rha1-H-2  Rhamnose-1  Rha1-H-1; Rha1-H-3 --> Rha1-H-4 --> Rha1-H-5 → Rha1-H-6  Rha1-H-3 Rha1-H-4 Rha1-H-5 Rha1-H-6 Glc1-H-1  Glc1-H-2 → Glc1-H-3 → Glc1-H-4 → Glc1-H-5 → Glc1-H-6a/b  Glc1-H-2  Glucose-1  Glc1-H-3 Glc1-H-4 Glc1-H-5 Glc1-H-6a Glc1-H-6b Glc2-H-1  Glc2-H-2 --> Glc2-H-3 --> Glc2-H-4 → Glc2-H-5 → Glc2-H-6a/b  Glc2-H-2  Glucose-2  Glc2-H-3 Glc2-H-4 Glc2-H-5 Glc2-H-6a Glc2-H-6b Xyl-H-1  Xyl-H-2 → Xyl-H-3 → Xyl4 → Xyl-H-5a/b  Xyl-H-2  Xylose  Xyl-H-3 Xyl-H-4 Xyl-H-5a Xyl-H-5b Rha2-H-1  Rha2-H-2 → Rha2-H-3  Rha2-H-2  Rhamnose-2  Rha2-H-3 Rha2-H-4 Rha2-H-5 Rha2-H-6  a  Rha2-H-5 → Rha2-H-4 → Rha2-H-3 → Rha2-H-2  Recorded at 600 MHz.  115  Table 2.6 (Continued) 1D TOCSY data for montbretin D (52), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. Sugar  TOCSYa (H→H)  Position Glc3-H-1  Glc3-H-2 → Glc3-H-3 → Glc3-H-4 → Glc3-H-5 → Glc3H-6a/b  Glc3-H-2  Glucose-3  Glc3-H-3 Glc3-H-4 Glc3-H-5 Glc3-H-6a  a  Recorded at 600 MHz.  116  Examination of the 1H-NMR and HSQC spectrum of montbretin D quickly revealed that it was an analogue of montbretin B with what appeared to be an extra sugar residue attached somewhere in the molecule. This was indicated by a clear doublet at δH 5.04 ppm (which correlated via HSQC to a carbon resonance at δC 99.9 ppm, assigned as an anomeric carbon), plus the addition of four other proton signals in the δH 3.4 ppm- 3.6 ppm region that displayed HSQC correlations to four individual carbon resonances in the δC 70 ppm-80 ppm region of the 13C spectrum. Finally, two proton resonances slightly more downfield at δH 3.74 ppm and δH 3.95 ppm were observed to have HSQC correlations to the same carbon resonance at δC 63.7 ppm. These six carbon and six proton signals were the only differences observed between the 1D 1H and 13C spectra of montbretins B and D. Additionally, the MS data showed that the difference in mass between montbretin D and montbretin B was 162 mass units, which is the mass of a monosaccharide residue (C6O5H10).  Differences in the 1H spectra of montbretins B and D suggested that the location of the glucose residue was likely in the aglycone portion of the structure, as the two protons belonging to H-15 and H-17 were significantly shifted between the respective 1H NMR spectra of the two molecules. Examination of the 2D HMBC spectrum showed a clear correlation from the anomeric proton resonance, Glc3-H-1 (at δH 5.04 ppm), to a carbon at δC 164.9 ppm (C-16).  Glc3-H-1 (δH 5.04 ppm) displayed a COSY correlation to a resonance at δH 3.50 ppm (Glc3-H-2). Glc3-H-2 (δH 3.52 ppm), showed HMBC correlations to resonances at δC 78.5 ppm (Glc3-C-3) and δC 71.8 ppm (Glc3-C-4). A resonance at δH 3.74 ppm (Glc3-H-6a) showed a COSY correlation to a resonance at δH 3.53 ppm (Glc3-H-5), as well as HMBC correlations to carbon resonances at δC 75.4 ppm (Glc3-C-5) and δC 71.8 ppm (Glc3-C-4). In addition to these 117  key HMBC and COSY correlations, other HMBC and COSY correlations were seen within this sugar residue that support this established connectivity, as summarized in Table 2.5 and shown in Figure 2.67. HO C-16 O O OH COSY  OH OH  HMBC  Figure 2.67 HMBC and COSY correlations in the glucose3 residue of montbretin D (52).  The assignment of this Glu3-H-1 to Glu3-H-6a/b spin system was confirmed by 1D TOCSY NMR (Table 2.6). Saturation of the resonance at δH 5.04 ppm (Glc3-H-1), followed by observation of the resulting 1H spectrum after a delay showed progressive transfer of the magnetization down the glucose spin system as delay time increased (Figure 2.68). With a 10 ms delay, magnetization transferred to the proton resonance at δH 3.50 ppm (Glc3-H-2), with a 40 ms delay to the resonances at δH 3.51 ppm (Glc3-H-3) and δH 3.42 ppm (Glc3-H-4), with a 80 ms delay to the resonance at δH 3.53 ppm (Glc3-H-5), and finally at 120 ms to the resonances at δH 3.74 ppm (Glc3-H-6a) and δH 3.95 ppm (Glc3-H-6b).  118  HO Glc3-H-6a 3.95ppm  TOCSY HO  Glc3-H-6b 3.74ppm Glc3-H-5 3.55ppm  O  Glc3-H-4 3.42ppm  HO  Glc3-H-3 3.52ppm  O Glc3-H-2 3.52ppm  Glc3-H-1 5.05ppm  OH  Figure 2.68 1D TOCSY spectra of montbretin A (30) showing transfer of magnetization along the Glc3 spin system from Glc3-H-1 to Glc3-H-6a/b, recorded after selective irradiation at δH 5.05 ppm (Glu3-H1  1), after a delay of: a) 10 ms b) 40 ms c) 80 ms d) 120 ms. H NMR spectrum shown in e). Spectra recorded at 600 MHz in MeOD-d4.  119  1D ROESY correlations between the Glc3-H-1 and the proton resonances at δH 6.45 ppm (H-15), as well as δH 6.54 ppm (H-17), support the placement of this new glucosyl residue via an ether linkage on C-16 (Figure 2.69).  Figure 2.69 1D ROESY spectra of montbretin D (52) recorded after irradiation on Glc3-H-1 (top), H-17 (middle) and H-15 (bottom). Spectra recorded at 600 MHz in MeOD-d4.  Identification of this new six carbon sugar residue as β-glucopyranose was achieved by careful examination of the ROESY correlations between the protons in the spin system, 1D TOCSY experiments, and coupling constant analysis. Because the proton at δH 5.04 ppm (Glc3-  120  H-1) appears as a clear doublet, J = 7 Hz, it was established as being axial, with the proton on the neighboring carbon (Glc3-H-2, δH 3.52 ppm) also being positioned axial. Examination of the resonance at δH 3.42 ppm (Glc3-H-4) in 1D TOCSY and ROESY spectra showed that it was a triplet, which suggested that its coupling constant to both its nearest neighbors is equivalent, or close to equivalent. Moreover, careful examination of the J-resolved spectrum of montbretin D in the δH 3.35 ppm to δH 3.45 ppm region confirms that the signal at δH 3.43 ppm is a triplet, J = 7.1 Hz (Figure 2.70).  Figure 2.70 2D JRES spectrum of montbretin D (52), recorded at 600 MHz in MeOD-d4. The signal at δH 3.42 ppm is a triplet, J = 7.1 Hz.  121  For the Glc3-H-3 resonance to appear as a triplet with J = 7.1 Hz, it must be axial with its two nearest neighbors also both axial. Thus, Glc3-H-3, Glc3-H-4 and Glc3-H-5 are all axially oriented, which means the residue is indeed a β-glucopyranoside. This result is confirmed by observed ROESY correlations between Glc3-H-5 and Glc3-H-1, as well as a number of other ROESY correlations with in the sugar residue that are less diagnostic and not as clearly resolved. This data is summarized in Figure 2.71. H OH  6 t, J=7.1Hz HO  O C-17  H 4  HO  5 3 H  ROESY  2 H  O  H  O OH  C-16 C-15  1 H  OH  H d, J=7Hz  Figure 2.71 Summary of ROESY correlations and key coupling constants in the glucose3 residue of montbretin D (52).  122  2.5.7 Structure Elucidation of Montbretin E O OH O  HO Glc3 OH OH  O  O  OH  O OH O OH  OH  O CH3  O HO  OH  O  HO  OH Rha1  O  O  CH3  Xyl  OH Rha2  OH OH  OH O OH  O  O  Glc1  OH OH O OH OH  O  Glc2  OH  53  Figure 2.72 Montbretin E (53).  Montbretin E (53) was isolated as brown powder that gave an [M+Na]+ pseudomolecular ion at m/z 1414.3788 in its high resolution electrospray ionization time-of-flight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C59H74O38 (C59H74O38 Na calculated for m/z 1414.3756), indicating 23 degrees of unsaturation. Analysis of the 1H spectrum of montbretin E, particularly in the δH 5-9 ppm region, quickly revealed it to be an analogue of montbretin A with what appeared to be an added β-glucopyranosyl residue – like the one seen for montbretin D. Investigation of the HMBC, COSY, and 1D TOCSY spectra of montbretin E confirmed it to be an analogue of montbretin A, with the difference being the addition of a βglucopyranosyl residue at C-16 in the aglycone portion. This data is summarized in Table 2.7.  123  O OH O  HO OH OH  O  OH  O OH O OH  OH O  HO  OH  O  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.73 600 MHz 1H NMR spectrum of montbretin E (53) recorded in MeOD-d4.  124  O OH O  HO OH OH  O  OH  O OH O OH  OH O  HO  OH  O  OH O  HO  OH OH  OH O OH  OH  CH3  CH3 O  O O  O  O  OH OH O OH OH  O OH  13  Figure 2.74 150 MHz C NMR spectrum of montbretin E (53), recorded in MeOD-d4.  125  O OH O  HO OH OH  O  OH  O OH O OH  OH O  HO  OH  O  OH O  CH3 O  HO  OH O OH  O  O CH3  O OH  OH OH  O  OH OH O OH OH  O OH  Figure 2.75 600 MHz COSY spectrum of montbretin E (53), recorded in MeOD-d4.  126  O OH O  HO OH OH  O  OH  O OH O OH  OH O  HO  OH  O  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.76 600 MHz HSQC spectrum of montbretin E (53), recorded in MeOD-d4.  127  O OH O  HO OH OH  O  OH  O OH O OH  OH O  HO  OH  O  OH O  CH3 O  HO  OH O OH  O O  O OH  CH3  OH OH  O  OH OH O OH OH  O OH  Figure 2.77 600 MHz HMBC spectrum of montbretin E (53), recorded in MeOD-d4.  128  Table 2.7 1D and 2D NMR data for montbretin E (53), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. O OH 23 22 24 17 21 O HO 18 O 10 19 16 20 OH O Glc3 11 OH O 15 12 OH O 14 13 O OH OH OH O OH O OH OH O CH3 O Rha1 CH3 5 2 HO 6 4 OH OH O OH O 3 1 7 9 HO O 8 OH Glc1 OH OH O OH OH  O  Xyl  Rha2  Glc2  OH  53  C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  Caffeic Acid  a  1  166.9  -  -  -  2  112.9  6.02 d (16)  C-1, C-4  H-3  3  145.1  7.35 d (16)  C-1, C-2, C-4, C-5, C-9  H-2  4  125.6  -  -  -  5  112.9  6.83 d (1.9)  C-6, C-7, C-9  H-9  6  142.5  -  -  -  7  144.7  -  -  -  8  114.1  6.61 d (7.9)  C-4, C-7  H-9  9 121.2 6.69 dd (7.9, 1.9) C-5, C-6, C-7 Recorded at 150 MHz. b Assigned according to HSQC recorded at 600 MHz. c Recorded at 600 MHz.  H-5, H-8  129  Table 2.7 (Continued) 1D and 2D NMR data for montbretin E (53), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  Aglycone 10  158  -  -  -  11  137.5  -  -  -  12  179.8  -  -  -  13  103.5  -  -  -  14  162.9  -  -  -  15  99  6.53 d (1.9)  C-14, C-16, C-17  -  16  160.9  -  -  -  17  93.7  6.44 d (1.9)  C-12, C-13, C-15, C-16, C-18  -  18  156.6  -  -  -  19  126.5  -  -  -  20/24  107.9  6.92 s, 2H  C-10, C-19, C-20/24, C-21/23, C-22  -  21/23  150  -  -  -  22  135.7  -  -  -  Rhamnose-1, Rha1 Rha1-1  100.7  5.65 s  C-11, Rha1-C-2, Rha1-C-3  Rha1-H-2  Rha1-2  82.4  4.27 d (3.3)  Rha1-C-3, Rha1-C-4, Glc1-C-1  Rha1-H-1, Rha1-H-3  Rha1-3  70.2  3.73 m  -  Rha1-H-2, Rha1-H-4  Rha1-4  72  3.37 m  Rha1-C-6  Rha1-H-3, Rha1-H-5  Rha1-5  70  3.55 m  -  Rha1-H-4, Rha1-H-6  Rha1-6  15.9  1.07 d (6)  Rha1-C-4, Rha1-C-5  Rha1-H-5  Glucose-1, Glc1  a  Glc1-1  104.6  4.51 d (7.8)  Rha1-C-2, Glc1-C-2  Glc1-H-2  Glc1-2  82.6  3.47 m  Glc1-C-1, Glc1-C-3, Glc2-C-1  Glc1-H-1, Glc1-H-3  Glc1-3  75.8  3.64 m  Glc1-C-2, Glc1-C-4  Glc1-H-2, Glc1-H-4  Glc1-4  69.3  3.38 m  Glc1-C-3, Glc1-C-5, Glc1-C-6  Glc1-H-3, Glc1-H-5  Glc1-5  72.9  3.55 m  -  Glc1-H-4, Glc1-H-6  Glc1-6a  62.3  4.5 m  C-1, Glc1-C-5  Glc1-H-6b  Glc1-6b 62.3 4.15 m C-1, Glc1-C-5 Recorded at 150 MHz. b Assigned according to HSQC recorded at 600 MHz. c Recorded at 600 MHz.  Glc1-H-6a, Glc1-H-5  130  Table 2.7 (Continued) 1D and 2D NMR data for montbretin E (53), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  HMBCc (H→C)  COSYc (H→H)  Glucose-2, Glc2 Glc2-1  105.1  4.59 d (7.8)  Glc1-C-2  Glc2-H-2  Glc2-2  74  3.30 m  Glc2-C-1, Glc2-C-3  Glc2-H-1  Glc2-3  77.3  3.42 m  -  -  Glc2-4  69.1  3.32 m  Glc2-C-3, Glc2-C-5  -  Glc2-5  75.9  3.34 m  Glc2-C-4, Glc2-C-6  Glc2-H-6  Glc2-6a  60.5  3.95 m  -  Glc2-H-6b  Glc2-6b  60.5  3.76 m  -  Glc2-H-5, Glc2-H-6a  Xylose, Xyl Xyl-1  105.9  4.84 d (7.4)  C-22, Xyl-C-5,  Xyl-H-2  Xyl-2  73  3.6 m  Xyl-C-1, Xyl-C-3  Xyl-H-1  Xyl-3  73.7  3.57 m  Xyl-C-2  Xyl-H-4  Xyl-4  73.2  3.75 m  Xyl-C-5, Xyl-C-3, Rha2-C-1  Xyl-H-3, Xyl-H-5a, Xyl-H-5b  Xyl-5a  62.6  4.13 m  Xyl-C-1, Xyl-C-4  Xyl-H-4, Xyl-H-5b  Xyl-5b  62.6  3.37 m  Xyl-C-1, Xyl-C-4  Xyl-H-4, Xyl-H-5a  Rhamnose-2, Rha2 Rha2-1  98  4.81 s  Xyl-C-4, Rha2-C-2, Rha2-C-5  Rha2-H-2  Rha2-2  70.3  3.76 m  Rha2-C-1  Rha2-H-1  Rha2-3  70.5  3.69 m  Rha2-C-4  Rha2-H-4  Rha2-4  72  3.40 m  Rha2-C-3, Rha2-C-5  Rha2-H-3, Rha2-H-5  Rha2-5  68.2  3.96 m  Rha2-C-4, Rha2-C-6  Rha2-H-4, Rha2-H-6  Rha2-6  16.1  1.28 d (6)  Rha2-C-4, Rha2-C-5  Rha2-H-5  Glucose-3, Glc3  a  Glc3-1  99.7  5.05 d (7.6)  C-16  Glc3-H-2  Glc3-2  75.9  3.52 m  Glc3-C-3, Glc3-C-4, Glc3-C-5  Glc3-H-1, Glc3-H-3  Glc3-3  76.4  3.50 m  Glc3-C-2, Glc3-C-4  Glc3-H-2, Glc3-H-4  Glc3-4  69.7  3.44 m  Glc3-C-2, Glc3-C-3, Glc3-C-5  Glc3-H-3, Glc3-H-5  Glc3-5  73.4  3.53 m  Glc3-C-3  Glc3-H-4, Glc3-H-6b  Glc3-6a  60.5  3.95 m  -  Glc3-H-6b  Glc3-6b 60.5 3.76 m Glc3-C-4, Glc3-C-5 Recorded at 150 MHz. b Assigned according to HSQC recorded at 600 MHz. c Recorded at 600 MHz.  Glc3-H-6a, Glc3-H-5  131  2.6 Biological Activity of the Montbretins Kinetic analysis of the montbretin family members revealed that montbretin A (30) was by far the most potent member of the family, displaying a Ki for HPA 1000+ fold smaller than those found for montbretins B (31), C (51), D (52) and E (53). This, coupled with the fact that montbretin A is by far the most abundant of the montbretins found in a variety of different Crocosmia sp., made it a good candidate for further biological testing and studies. The Ki's found for the montbretins are summarized in Table 2.8.  Table 2.8 Ki values for the inhibition of HPA by montbretins A-E.87,88,84  Ki (nM) Montbretin A (30)  1.3  Montbretin B (31)  3600  Montbretin C (51)  6100  Montbretin D (52)  890  Montbretin E (53)  12000  Montbretin A was demonstrated to be a tight-binding, competitive inhibitor of HPA. In addition to this, montbretin A displayed a high degree of selectivity for HPA over other glycosidases69, as summarized in the Table 2.9, which is a highly desirable trait of an inhibitor of HPA, as inhibition of bacterial glycosidases in the gut will contribute to the negative side effects noted with HPA inhibitors such as acarbose.  132  Table 2.9 Residual Activity of various glycosides after exposure to 1μM montbretin A  Glycoside  Residual Activity (%)  α-amylase (HPA)  11%  β- glucosidase (Agrobacterium sp.)  100%  β- galactosidase (E. coli)  98%  β- hexosaminidase (Jack Bean)  99%  α- mannosidase (Jack Bean)  100%  α- galactosidase (Green Coffee Beans)  100%  α- glucosidase (Brewers Yeast)  97%  An interesting observation made during the testing of montbretin A against the glycosidases listed in Table 2.9 was that montbretin A seemed to show inhibition of βglucosidase from Agrobacterium sp. (Abg) that decreased over time. This was found to be the result of montbretin A acting as a substrate for Abg, resulting in the cleavage of its terminal β-Dglucose residue. However, further testing of this truncated montbretin A showed no reduction in its ability to inhibit HPA, showing that this terminal glucose residue is not required for HPA inhibition.  In an effort to further probe the structural motifs of montbretin A required for HPA inhibition, commercially available compounds that correspond to the two aromatic fragments – myricetin and ethyl caffeate – were independently tested for HPA inhibition. It was found that myricetin was an inhibitor of HPA, albeit one many orders of magnitude less potent (Ki 110 μM), and also a competitive inhibitor, indicating that its inhibition stems from binding to the active site of HPA. Ethyl caffeate, while also an inhibitor of HPA, was weaker still, and also a noncompetitive inhibitor (Ki = 1.3 mM), suggesting that its inhibition arises through interactions  133  away from the active site.84  2.6.1 STD NMR Binding Studies In an effort to further probe the structural motifs responsible for HPA inhibition by the montbretins, a technique called Saturation Transfer Difference (STD) NMR was employed. STD NMR is a technique that allows the observation of NOE effects between a receptor protein and a ligand (in this instance, between HPA and the montbretins), and can be applied to a variety of standard NMR experiments.89,90  The simple 1D STD NMR experiment is a difference experiment between a 1H NMR spectrum of the receptor and ligand, recorded with a receptor that has been selectively saturated with magnetization (with resultant NOE effects occurring in the ligand), and one where the receptor is not saturated with magnetization. The difference between these two will then be the NOE effects transferred to the ligand. This experiment is summarized in Figure 2.78.  Figure 2.78 Basic scheme of an STD NMR experiment, with shading used to indicate magnetization. a) Protein and ligand are mixed together. b) Protein is selectively saturated, and it transfers magnetization selectively to its ligands as shown in c). The protein will relax quickly. A 1D STD NMR spectrum is thus the difference between the 1H NMR spectra of a) and c).  134  As organic macromolecules, proteins feature a large system of protons tightly coupled by spin-spin coupling. Selective saturation of a single proton in a protein will result in rapid transfer to the rest of the macromolecule via spin diffusion, effectively and rapidly saturating the whole protein. As depicted in Figure 2.79, intermolecular transfer from protein to ligand then leads to progressive saturation of the ligand, with varying amounts of transfer dependent on the proximity of individual protons on the ligand to the protein. As such, it is possible to gain insight from an STD NMR spectrum into the structural features of a ligand closest to the protein.90  Figure 2.79 Areas on the ligand that are close to areas on the protein will become more saturated due to NOE effects than areas that are not. The gray shading indicates the amount of magnetization, with increasing darkness indicating increasing saturation.  There are, however, some limitations to the efficacy of the 1D STD NMR experiment. If a ligand binds too strongly to the protein, the resulting STD effects will be diminished because of relaxation occurring due to spin diffusion when in the bound state. Generally, one wishes to have a system that has a dissociation constant in the range of 10-3 – 10-8, or a so called 'lowaffinity ligand'.89 Montbretin A, with a Ki of 1.3 nM, is a high affinity ligand and thus does not show any significant STD NMR effects. The same is true with montbretin D, and insufficient amounts of pure montbretin C and montbretin E were available for 1D STD NMR experiments to be a success. However, montbretin B, which has a Ki value in the μM range, is a much  135  stronger candidate for successful 1D STD NMR, and in fact, does show some STD NMR effects on a few key protons.  Another limitation of the 1D STD NMR experiment is the fact that it is performed in water, so compounds like myricetin and ethyl caffeate that are not water soluble could not be studied. Additionally challenging is the presence of H2O in the protein-ligand mixture. HPA was supplied in an aqueous buffer solution as concentrated as possible, but the presence of H2O still means that the 1D spectrum of the HPA is primarily dominated by the water resonance. This issue was overcome by use of solvent (water) suppression using excitation sculpting with gradients, however, the region from 4.20 ppm – 5.20 ppm remained unusable for the purposes of determining the presence or absence of any STD NMR effects (Figure 2.80a).  1  Figure 2.80 H NMR spectra, recorded at 600 MHz in MeOD-d4. a) HPA in 3:2 D2O/H2O with  136  solvent suppression. b) STD spectrum of HPA and no ligand, on resonance at δH 1.7 ppm, off resonance at δH 40 ppm. c) HPA and montbretin B in 3:2 D2O/H2O with solvent suppression d) STD spectrum of HPA and montbretin B, on resonance at δH 1.7 ppm, off resonance at δH 40 ppm.  A clear STD effect can be seen in the 1D STD NMR spectrum of montbretin B (Figure 2.80d), showing that NOE enhancement was observed in the protons of the p-coumaric acid moiety: the ortho-coupled protons from the phenyl ring, and the two protons from the trans double bond (Figure 2.81).  1  Figure 2.81 H NMR spectra, recorded at 600 MHz in MeOD-d4. Expansion of the aromatic portion of a) HPA and montbretin B in 3:2 D2O/H2O with solvent suppression. b) STD spectrum of HPA and montbretin B, on resonance at δH 1.7 ppm, off resonance at δH 40 ppm.  While this data does indeed show that this region is in close proximity to the HPA, it only serves to indicate that the caffeic/courmaric acid moiety is a key structural motif involved in the 137  efficacy of the montbretins as HPA inhibitors. The results of the 1D STD NMR and of the bioassay studies are summarized in Figure 2.82. When Present AND R=H: Improves activity by a factor of ~4  Structurally similar to Myricetin, a competitive inhibitor of HPA  W hen Present AND R=OH: Kills activity by four orders of magnitude  OH O  Glc3  O  O OH  Xyl O  O OH  O  Rha2  Rha1  Area showing STD enchancement O O O  Glc1  O  Glc2  HO  R R=OH : nM Ki  Not required for activity Structurally similar to Ethyl Caffeate (when R=OH), a non-competitive inhibitor of HPA  R=H or OM e: Kills activity by three orders of magnitude  Figure 2.82 Summary of observed biological activity (HPA inhibition) and STD data for the montbretins.  2.6.2 Animal Studies on Montbretin A Montbretin A has been the subject of a number of studies investigating its potential efficacy as an inhibitor of HPA, preformed in Dr. John McNeil's laboratory at UBC. As an HPA  138  inhibitor, montbretin A would be consumed with a meal and would act in the gastrointestinal tract to inhibit HPA and slow the release of glucose after a meal. Thus, it is important that montbretin A not degrade in the environment of the stomach and intestine. As such, its stability in simulated intestinal fluid (SIF) and simulated gastric fluid (SGF) was investigated.86 Montbretin A showed no significant amount of degradation after one hour in SIF at 37oC, and only an approximate 10% degradation after one hour in SGF, with no additional degradation observed at three hours. Thus, montbretin A remains an attractive candidate for use as an HPA inhibitor delivered via oral consumption.  To investigate the efficacy of montbretin A, as compared to the currently prescribed acarbose, an acute starch tolerance test was performed on adult rats with either acarbose or montbretin A (Figure 2.83). A group of rats was divided into control and diabetic, with the diabetic rats having diabetes induced via a single intravenous tail injection of streptozotocin (60 mg/kg in 0.9% saline). Streptozotocin is a natural product found in the late 1950's as an antibiotic, which brings about type 1 diabetes by way of destruction of the pancreatic beta cells in mammals.91 The animals were then subdivided again and fed 17.5% suspension of starch in distilled water at a dose of 2 g/kg, and where appropriate (and in different experiments for each drug), an aqueous solution of either acarbose or montbretin A (concentration 10 mg/mL, dose 1 mL/kg). Blood samples were taken from the animals before feeding, and at 30, 60, 90 and 120 minutes after feeding, with blood glucose levels then determined. A significant decrease in plasma glucose was seen at 30 and 60 minutes in both acarbose treated rats, and then again in the montbretin A treated rats in a separate experiment. Thus, both acarbose and montbretin A are effective at preventing the increase in plasma glucose following a starch challenge at a dose of 10 mg/kg. Acarbose, however, has approximately half the molecular weight of montbretin A 139  (645.6 g/mol versus 1228 g/mol), so these preliminary findings encouragingly show that montbretin A is effective at a lower molar dosage than acarbose in diabetic rats.  140  Figure 2.83 Plasma glucose levels in diabetic and control rats, with and without acarbose (top graph) or montbretin A (bottom graph).  141  With the knowledge that montbretin A is effective in preventing an increase in blood plasma glucose levels at a concentration of 10 mg/kg, a further study was undertaken to investigate the effect of varying levels of montbretin A in diabetic rats, feeding them varying amounts (by body weight) of montbretin A.  It was found that montbretin A was effective at preventing an increase in blood plasma glucose levels following starch challenge at 5 mg/kg, but not significantly effective at the lower concentrations (Figure 2.84). Methanolic and ethanolic extracts of Crocosmia sp. were also tested, and showed efficacy at preventing an increase in blood plasma glucose levels, however, moderately severe diarrhea was noted as a side effect. This was never observed with rats given pure montbretin A. Further to these findings, it was also found that there was no taste aversion to montbretin A in the rats given dosage in their water, as shown by the careful monitoring of food and water consumption (compared to previously determined normal consumption) in rats over a three day period.  142  Figure 2.84 Plasma glucose levels in diabetic rats given different doses of montbretin A.  These results suggest that montbretin A is an attractive candidate for use as an orally delivered therapeutic for inhibition of human pancreatic α-amylase, for aiding in the control of the release of glucose in diabetic patients after a meal (much like the currently prescribed acarbose). More work will have to be done to determine its safety and efficacy in longer animal trials, as well as in humans.86  2.7 Discussion and Conclusions The isolation of montbretins A and B from a number of different cultivars of Crocosmia shows that the montbretin family are most likely common glycosylated flavonoids made across the entire species. However, between different varieties of Crocosmia sp., there exist differences in overall composition, especially with respect to the minor montbretins seen. The NCI Crocosmia sp. extract, for example, contained montbretin C, which was easily as plentiful as montbretin B in the extract. Montbretin C was not reported in the published isolation of 143  montbretins A and B – however it is possible that it was simply overlooked during the isolation. In the large scale isolation of montbretin A from locally obtained corms of Crocosmia “Emily Mackenzie”, no montbretin C was observed, despite the fact that it was actively looked for in the extract. In this larger scale isolation, two other montbretins – montbretin D and montbretin E – were present in significant amounts. Granted, the isolation conditions were different between the NCI extract and the large scale extract, but there was no hint of the presence of montbretins D and E in the isolation from the NCI extract. Thus, it appears that although montbretins A and B are ubiquitous across Crocosmia sp., the exact composition of the mixture of montbretins made by a particular cultivar does vary.  Structurally, the montbretins show a number of motifs common in plant flavonoids. Myricetin, common in plants, is often found as a glycoside with rhamnose – a sugar present twice in the montbretins. The cinnamic acid derivatives seen in the family (caffeic acid, pcoumaric acid and ferulic acid) are all common building blocks found in a variety of plant derived natural products. Finally, β-D-glucopyranose is also a common sugar residue seen in plant natural products.  Montbretin A does not show broad activity towards other glucosidase enzymes, which is an important feature for an HPA inhibitor. The bioassay results on montbretin A and also on the individual aromatic portions (tested on their own) present an attractive and suggestive model as to the way montbretin A binds to HPA for inhibition. Montbretin A itself is a tight binding competitive inhibitor, meaning that it binds, somehow, to the active site of HPA. Myricetin, the flavonoid portion of montbretin A, is also a competitive inhibitor, with activity far below that seen for montbretin A. Ethyl caffeiate is a non competitive inhibitor, meaning it binds 144  somewhere other than the active site of HPA. It is also far less active than montbretin A. Together, these pieces of information suggest that the two aromatic regions of montbretin A work together to give its overall potent inhibitory effect, with the flavonoid core fitting into the active site, and the caffeic acid moiety (linked by two sugar residues) binding to some other site on HPA. 1D STD NMR data shows that these caffeic/p-coumaric acid residue is involved in the biding to HPA. The bioassay data for montbretins A, B and C also suggests that the metahydroxy group of the caffeic acid group is likely to play a critical role in correctly orienting the portion of montbretin A binds into the active site (or plays a direct role in the binding to the active site), as deviations from this structure significantly reduce activity. Further studies are needed to decipher what that role is, specifically, and work looking into the structure of the enzyme-inhibitor complex is currently underway and will hopefully provide some insight. Interestingly, the addition of a β-D-glucopyranosyl residue to the aglycone (flavonoid) portion of montbretin B results in a more potent inhibitor of HPA (montbretin D), but drastically decreases activity when added to the montbretin A structural core.  Montbretin A does act to prevent the increase of blood plasma glucose levels after a meal of starch in diabetic rats, and seems to do so at lower doses than the currently prescribed HPA inhibitor, acarbose. Moreover, it does this without the side effect of diarrhea, as was seen when simple crude extracts of Crocosmia sp. corms were tested instead. Additionally, montbretin A did not show significant decomposition in simulated gastrointestinal fluids, and was not met with any taste aversion when given to rats in their water. These results are very encouraging and suggest that montbretin A may be an excellent candidate for development as a therapeutic to aid in the management of blood glucose levels in diabetic patients. Deeper investigation into animal models for the purpose of studying the metabolism of montbretin A, as well as structural studies 145  on HPA with bound Montbretin A (to determine the mechanism of action), are still needed.  2.8 Experimental 2.8.1 General Experimental Procedures 1  H, 13C, COSY, HSQC, HMBC, TOCSY and ROESY spectra were recorded on a Bruker  AV600 NMR spectrophotometer equipped with a cryoprobe. 1D STD NMR spectra were recorded on a Bruker AV400 NMR spectrophotometer equipped with a direct probe. 1H chemical shifts were referenced to the residual MeOD-d4 signal (δH 3.31 ppm) or the residual D2O signal (δH 4.80 ppm), and 13C chemical shifts were referenced to the MeOD-d4 solvent peak (δC 49.15 ppm). All NMR solvents were obtained from Cambridge Isotope Laboratories. All NMR data was processed using Bruker XWINNMR® software.  All chromatography was performed using HPLC grade solvents from Fisher Scientific with no additional purification. Water was purified by use of a Millipore MQ filter system. HPLC separations were performed using a Waters 600 pump and a Waters PDA 900 detector, using an Inertsil C18 column, 9.4 x 250 mm, flow 1 mL/min unless otherwise specified. All solvents were filtered prior to use, then sparged with helium. Optical rotations were determined by using a JASCO J-1010 polarimeter equipped with a halogen lamp (589 nm) and a 10 mm micro cell. UV spectra were recorded on a Waters 2487 spectrophotometer. ESI mass spectra were recorded using a Micromass LCT mass spectrometer.  2.8.2 Isolation of Montbretins A, B and C For a general isolation scheme see section 2.5.1. A dried, crude methanolic extract of  146  Crocosmia sp. corms was obtained from the NCI open plant repository and came in the form of a slightly sticky brown powder. This crude material (2 g) was partitioned, 1 g at a time between ethyl acetate and water (150 mL each, water re-extracted three times with 50 mL ethyl acetate. The aqueous fraction was extracted with butanol (3x50 mL). All three fractions were dried down and submitted for HPA bioassay, with the butanolic fraction (a brown glass, 472 mg) showing the majority of the activity. The butanolic fraction was resuspended in a minimal amount of methanol and applied to a Sephadex LH-20 column preswollen in methanol for size exclusion chromatography. 150 mg were applied to the column at a time in approximately 2 mL methanol. Fractions collected off of the column were sampled out into 96 well plates and dried down before being submitted for HPA bioassay. Active fractions were pooled together and dried down to give a yellow-brown oil. This oil (150 mg) was resuspended in 22% aqueous acetonitrile and subjected to isocratic reversed phase HPLC using 22% aqueous acetonitrile (flow 2mL/min) approximately 3 mg at a time. From this HPLC run, a series of peaks and fractions were collected, and then subjected to HPA bioassay. The major peak that showed HPA inhibitory activity had a similar proton NMR spectrum to two other peaks. These turned out to be montbretin A, B and C respectively. Montbretin A was purified using gradient from 30% to 40% aqueous acetonitrile on reversed phase HPLC over 30 min (2 mL/min Rt = 15 min, 8.4 mg) as a yellow powder. Montbretin B was purified via reversed phase HPLC using a gradient from 30% to 70% aqueous acetonitrile over 30 min (2 mL/min Rt = 16 min, 0.9 mg), and finally montbretin C was purified by reversed phase HPLC using a gradient from 20% to 30% aqueous acetonitrile over 30minutes (2 mL/min, Rt = 20 min, 1.6 mg) as a yellow powder.  147  2.8.3 Montbretin A Physical Data Yellow powder, [α]20 D = -41º (c=0.20, H2O), UV λmax MeOH nm: 193, 269, 319. HRESIMS:[M+Na]+ m/z 1251.3226 (calculated 1251.3228 for C53H64O33Na). For 1D and 2D NMR data please refer to Table 2.1 and Table 2.2.  2.8.4 Montbretin B Physical Data Yellow powder, [α]20 D = 70º (c=0.10, H2O), UV λmax MeOH nm: 190, 269,313. HRESIMS: [M+Na]+ m/z 1235.3184 (calculated 1251.3278 for C53H64O32Na). For 1D and 2D NMR data please refer to Table 2.3.  2.8.5 Montbretin C Physical Data Yellow powder, [α]20 D = 68º (c=0.12, H2O), UV λmax MeOH nm: 204, 267, 323. HRESIMS:[M+Na]+ m/z 1265.3381 (calculated 1265.3384 for C54H66O33Na). For 1D and 2D NMR data please refer to Table 2.4.  2.8.6 Isolation of Montbretins D and E The isolation scheme is discussed in section 2.5.2. From the large scale (1 kg) isolation effort done at the Center for Drug Research and Development in Vancouver, a series of fractions collected based on minor peaks from the large scale HPLC run were generated. These fractions were obtained as brown-yellow oils. Two of those fractions – from 17.6 and 18.7 minutes, respectively – afforded montbretins E and D. The 18.7 minute fraction, a brown glass, was resuspended in a minimal amount of 15% aqueous acetonitrile, and purified to yield montbretin D by reversed phase HPLC using 15% acetonitrile (2 mL/min, Rt=25 min, 4.5 mg) as a yellow/brown powder. The 17.6 minute fraction, a brown oil, was resuspended in 15% aqueous 148  acetonitrile and purified to yield montbretin E by reversed phase HPLC using 15% acetonitrile (2 mL/min, Rt=17 min, 0.9 mg) as a yellow/brown powder.  2.8.7 Montbretin D Physical Data Brown powder, [a]D 20 = 55º (c=0.12, H2O), UV λmax MeOH nm: 203, 268, 320 . HRESIMS: [M+Na]+ m/z 1397.3792 (calculated 1397.3807 for C59H74O37Na). For 1D and 2D NMR data please refer to Table 2.5 and Table 2.6.  2.8.8 Montbretin E Physical Data Brown powder, [a]D 20 = -15º (c=0.12, H2O), UV λmax MeOH nm: 203, 270, 320. HRESIMS: [M+Na]+ m/z 1414.3788 (calculated 1414.3756 for C59H74O38Na). For 1D and 2D NMR data please refer to Table 2.7.  2.8.9 Preparation of STD NMR Samples Samples were prepared from stock solutions with varying concentrations of HPA in a 20 mM phosphate buffer (pH 6.9) with 25 mM NaCl, such that the final concentration of HPA in the NMR sample (volume = 0.7 mL) was 20 µM. The montbretins were added to the HPA solution in a D2O solution such that the final concentration of ligand was 2 mM. NMR experiments were recorded in an AVANCE Bruker instrument operating at 400 MHz. STD experiments were performed at 278 K with solvent (water) suppression using excitation sculpting with gradients. On-resonance irradiation was performed at 1.5 ppm, off-resonance irradiation was at 40.0 ppm; appropriate blank experiments were also performed to assure the absence of direct irradiation on the ligand.  149  3. Potential Inhibitors of TLR5 – Isolation of Girolline and the Massacreones 3.1 Cystic Fibrosis Cystic fibrosis is caused by a mutation in the gene called the cystic fibrosis transmembrane conductance regulator (CFTR), which is involved in making sweat, digestive juices and mucus. Cystic fibrosis is an autosomal recessive disease; individuals with only one working copy of the CFTR gene are fine, it is those with two abnormal copies of this gene that show symptoms of the disease. One out of approximately 3500 live births in North America is an individual affected by cystic fibrosis.92,93  Cystic fibrosis is a disease that affects the entire body, causing progressive disability and in many cases, early death. The name cystic fibrosis refers to characteristic scarring, otherwise known as fibrosis, and cysts found in the pancreas. Of the symptoms of cystic fibrosis the most serious is difficulty in breathing, a result of frequent lung infections. Other symptoms include recurring sinus infections, poor growth as compared to peers, diarrhea and infertility. There is no cure for cystic fibrosis, and most patients with cystic fibrosis die in their twenties and thirties.93,94  The CFTR protein is a transmembrane protein found in the cells of the sweat glands, lungs, and pancreas, and is primarily responsible for controlling the movement of chloride ions out of the cell. In sweat ducts, however, it facilitates transport of chlorine ions from the sweat into the cytoplasm. When this protein does not function properly, the result is a buildup of the negatively charged chloride, causing positively charged sodium ions to cross the affected 150  membranes. For this reason, the sweat of individuals with cystic fibrosis is unusually salty.92,94  How this malfunction translates into the clinical manifestations of cystic fibrosis is poorly understood, but the end result is that mucous in the body becomes unusually thick. This thickened mucous clogs the narrow passages in affected organs including the lungs, resulting in difficulties breathing. The lungs of individuals with cystic fibrosis are colonized and infected by bacteria from an early age, and these bacteria thrive in the altered mucus that collects in the small airways of the lungs. This mucus leads to the formation of bacterial microenvironments known as biofilms that are difficult for immune cells and antibiotics to penetrate, resulting in chronic and serious lung infections.95-98  3.1.1 Treatment of Cystic Fibrosis One of the most common treatments for cystic fibrosis is the physical clearance of mucous from the lungs via so-called “airway clearance techniques”. These range from coughing/'huffing' to a variety of techniques that oscillate the chest wall to loosen mucous, or rely on patient positioning changes to get at the blockages. These techniques, although helpful in clearing out the lung to allow easier breathing, only work to help the cystic fibrosis patient control and maintain their condition. Ultimately, a lung transplant can be performed99, but in many cases, this thickened mucous and its associated infections result in an early death for the cystic fibrosis patient.  There are a number of inhaled medications that are used in cystic fibrosis care that are normally administered through aerosol. These include mucolytics, which thin out mucous making it easier to cough out, antibiotics to treat infections, and also hypertonic saline (a 7% salt 151  solution), which draws more water into the airways, making it easier to cough out mucous. Due to the chronic nature of lung and sinus infections in the cystic fibrosis patient, antibiotics are often also administered through oral dosing and intravenously, as needed.100,101  Clearly, the need for additional understanding into the root cause of cystic fibrosis, and treatments that get at the actual generation of this thickened mucous are needed. Additionally, current treatments for the symptoms of cystic fibrosis leave much to be desired and have limited success for allowing lifelong management of the disease. Additional therapeutics that help to manage the symptoms of cystic fibrosis could mean the addition of years to the expected life span of a cystic fibrosis patient.  3.1.2 Treatment of Inflammation The self-sustaining cycle of lung infection in cystic fibrosis patients is one of obstruction, then infection, then inflammation.101 Cystic fibrosis lung disease is characterized by airway inflammation and increased expression of proinflammatory cytokines, as well as infections by a select group of bacterial pathogens – most significantly, Pseudomonas aeruginosa and Burkholderia cepacia. As noted, current treatments revolve around administration of antibiotics as well as airway clearing techniques.97,102-104 Recently, anti-inflammatory therapy has been investigated for use with patients with cystic fibrosis. Clinical trials on corticosteroids 105 and ibuprofen106, which are non specific noninflammatory, have demonstrated that targeting excess inflammation is of use in improving the lung disease of cystic fibrosis patients. However, these drugs are not used for routine treatment of cystic fibrosis patients due to serious complicating side effects of long term administration.  152  Current research into inflammation in cystic fibrosis patients does not clarify whether the CFTR mutation promotes an environment prone to inflammation or whether the inflammation follows from infection. However, this distinction is not overly relevant to cystic fibrosis patients themselves - the cycle of obstruction/infection/inflammation in cystic fibrosis lung disease remains a troubling reality, regardless of the origin of the inflammation.96,98 Whatever the source of the inflammation, it is universal in all cystic fibrosis lung disease, and reduction of inflammation is thus likely to aid and augment current treatments and improve the clinical outcome for a cystic fibrosis patient. Instead of studying the underlying cause and source of the inflammation, simply identifying an immunological pathway that may be blocked to reduce cystic fibrosis airway inflammation is a practical goal in the quest for cystic fibrosis treatment options.  Targets within the signaling cascade for the overproduction of inflammatory cytokines, which are signaling molecules involved in inflammation, present an attractive treatment approach to reducing the amount of inflammation seen in cystic fibrosis.107,108 There is currently little consensus on which molecular pathways are important and critical for the increased inflammation seen in cystic fibrosis. However, work done by Turvey and coworkers has demonstrated that the Toll Like Receptor (TLR) mediated innate immune response is increased in respiratory epithelial cells as well as fresh blood cells in cystic fibrosis patients.109 TLRs are pattern recognition receptors that allow for the recognition of a number of microbial epitopes – for example TLRs 1/2/6 recognize lipoproteins, TLR4 recognizes lipopolysaccharides, TLR5 recognizes flagellin. TLR activation induces a cascade of effector responses within minutes107,110112  , and functional TLRs are expressed by cystic fibrosis airway epithelial cells and also by  hemopoietic cells in the lung. TLR signaling generates the cytokines known to be unregulated in 153  the cystic fibrosis lung.109 Turvey and coworkers specifically found that TLR5 is exclusively relied upon to detect and respond to the presence of the predominant cystic fibrosis pathogen P. aeruginosa, via its flagellin. Inhibition of the TLR5-flagellin interaction significantly reduced the pro-inflammatory response of cystic fibrosis respiratory epithelial cells following exposure to P. aeruginosa.109,113 Thus, specific inhibition of TLR5's activation in the presence of flagellin is an attractive goal (although further studies are needed) in the search for novel therapeutics to reduce inflammation in the lungs of cystic fibrosis patients.  3.1.3 TLR5 Bioassay With the goal of identifying small molecules that may disrupt the flagellin initiated TLR5 activation, Turvey and coworkers developed a bioassay to test and observe the activation of TLR5, with and without added flagellin.  Cells that contain recombinant luciferase that is covalently linked to TLR5 (CHO-K1 cells) are suspended in wells of a 96-well plate. To these wells, varying concentrations of the extract or compound, dissolved in DMSO, are added. Flagellin is then added to some of the wells to a final concentration of 1 μg/mL, and the assay plate is left to incubate for four hours. Each assay plate will thus contain a series of wells containing cells and extract/compound in a step wise series of concentrations, with and without flagellin. After incubation, the cells are lysed, luciferin is added, and the amount of light released as a result of the luciferase catalysed oxidation of luciferin is measured. Because the luciferase is linked to TLR5, the response as a result of the luciferase activity is directly correlated to the amount of TLR5 activation. This data is then used to calculate the 'fold increase' for the cells stimulated with flagellin as compared to control, unstimulated cells. Any toxicity due to the extract/compound will result in a less-than154  one fold increase in the unstimulated cells. If an extract/compound has an effect on the stimulation of TLR5 in the presence of flagellin, cells treated with flagellin will show a reduction in the 'fold increase' as compared with control cells stimulated with flagellin. Thus, an inhibitor of the TLR5-flagellin interaction would ideally show no change in TLR5 activation without flagellin, and would result in a markedly decreased amount of TLR5 activation in the presence of flagellin as compared with a control.114  3.2 Isolation of Girolline The TRL5 bioassay identified a strong hit from a marine natural product extract library, identified as extract #47643. This extract was identified as being derived from an orange rope sponge with branching fingers that was collected by hand using SCUBA on August 14, 1993, at Mont Pass in Pohnpei, at coordinates 6o 52 N, 158o 18 E, from a depth of -12 m. This sponge is shown in Figure 3.1. Powdered, freeze-dried sponge 47643 was immersed in methanol and extracted for twenty four hours. This was repeated two more times, and then a further two times with dichloromethane. These combined extracts were concentrated in vacuo, then resuspended in water. This aqueous suspension was partitioned against ethyl acetate (twice), and then against butanol (twice) to yield three fractions based on solubility. Bioassay guided fractionation indicated that the activity was in the aqueous fraction.  155  Figure 3.1 Pohnpeian sponge 47643.  This active aqueous fraction was then subjected to size exclusion chromatography on a column containing Sephadex LH-20 preswollen in methanol, resulting in three subfractions grouped based upon their TLC profile. The first of these LH-20 fractions, which was the first 50 mL or so eluted off of the column after the void volume, was found to display the biological activity. By TLC this fraction contained mainly very polar material that did not move off of the TLC baseline even when developed in a very polar mixture (4:2:1 butanol : water : acetic acid).  To understand how best to proceed with the isolation, a small sample of this LH-20 fraction was passed through a reversed phase Sep-pak eluted stepwise with 25% increases in methanol mixed with water. A similar test was done with a normal phase Sep-pak. The fractions that displayed the biological activity were the 10% water fraction from the reversed phase Seppak and the most polar of the fractions taken from the normal phase Sep-pak, 1:1 ethyl acetate:methanol. Thus, the remainder of the LH-20 fraction was put through a reversed phase Sep-pak and eluted into three fractions, the first of which eluted with 100% water and showed 156  biological activity. This active fraction was then subjected to reversed phase HPLC using a gradient of water to 40% aqueous acetonitrile over 40 minutes. The majority of the material eluted in the first ten minutes. This HPLC run yielded five fractions, with fractions one to four representing a peak or peaks; the most massive was peak number two (fraction name: 47643.3.1.1.2), which also turned out to display the biological activity. This active fraction was subjected to a second round of HPLC on C18, run with water as an eluent. This HPLC run was monitored at 210 nm and 254 nm, with a major peak eluting at 6.5 minutes with a strong absorbance at 210 nm, and negligible absorbance at 254 nm. A number of peaks with more substantial absorbance at 254 nm were collected as well for a total of seven fractions. Bioassay results showed that the first fraction, with the strong absorbance at 210 nm, was the active fraction (1.9 mg). This fraction, when developed on normal phase TLC using 4:2:1 butanol : water : acetic acid, showed a spot at Rf = 0.1 that charred yellow with a p-anisaldehyde stain. This spot was eventually identified as girolline (1).  3.2.1 Structure Elucidation of Girolline OH HH  H  NH2  N NH H2 N  Cl 1  Figure 3.2 Girolline (1).  Isolated as a brownish-yellow oil, girolline (54) was identified based on comparison of its molecular mass, 1H NMR data and 13C data with those reported in the literature.115 Firstly, it should be noted that this sample of girolline was not 100% pure, and in fact contained impurity  157  signals in the 1H NMR spectrum that suggested the presence of a few compounds, or perhaps one very complex compound, that appeared to be polyketide or alkaloid in nature due to the fact that a significant number of minor resonances could be seen in the δH 3.0-5.2 ppm region of the spectrum (Figure 3.3). In addition, there were a number of small impurity peaks in the region from δH 1.0-2.5 ppm, but no obvious singlets, suggesting that these impurities were likely not terpenoid. However, there is insufficient data to really identify what species are present. All that said, girolline is clearly the most abundant species in this fraction.  Figure 3.3 600 MHz 1H NMR spectrum of girolline (1), recorded in MeOD-d4.  158  OH NH2  N NH  Cl  H2 N  Figure 3.4 600 MHz 1H NMR spectrum of girolline (1), recorded in D2O.  159  OH NH2  N NH  Cl  H2N  Figure 3.5 150 MHz 13C NMR spectrum of girolline (1), recorded in D2O.  160  OH NH2  N NH  Cl  H2N  Figure 3.6 600 MHz COSY spectrum of girolline (1), recorded in D2O.  161  OH NH2  N NH  Cl  H2N  Figure 3.7 600 MHz HSQC spectrum of girolline (1), recorded in D2O.  162  OH NH2  N NH  Cl  H2N  Figure 3.8 600 MHz HMBC spectrum of girolline (1), recorded in D2O.  163  Girolline gave a [M+Na]+ pseudomolecular ion at m/z 191.0611 in its high resolution electrospray ionization time-of-flight mass spectrum (HRESI-TOFMS), with a second peak at approximately 1/3 the height at 193, suggesting the presence of a chlorine atom in the structure. This mass corresponds to a molecular formula of C6H11ON4Cl (C6H12ON4Cl calculated for m/z 191.0700), indicating three degrees of unsaturation. Analysis of 1H and 13C NMR spectra, as well as 2D COSY and HSQC spectra, identified one key substructure, shown in Figure 3.9.  Y 2 3  1 Y  Y Y = Halogen or Heteroatom  Figure 3.9 Key substructure in structure elucidation of girolline (1) with expansion of the key region of the COSY spectrum.  This structure was pieced together as follows. A proton resonance at δH 5.09 ppm (H-3), which was on a carbon that resonated at δC 64.7 ppm, showed a COSY correlation to a proton resonance at δH 4.52 ppm (H-2), which is attached to a carbon at δC 58.5 ppm (C-2). In turn, this proton H-2 shows two COSY correlations to a set of protons at δH 3.39 ppm and δH 3.54 ppm (H-  164  1a and H-1b), which are attached to the same carbon, at δC 41.4 ppm (C-1). The chemical shift values for these three carbons were indicative of carbons with attached heteroatoms or halogens.  This four proton spin system was backed up by the presence of HMBC correlations between the two protons H-2 and H-3 (δH 4.52 ppm and δH 5.09 ppm) and the carbon at δC 41.4 ppm (C-1), as well as HMBC correlations from H-1b (δH 3.39 ppm) to C-2 (δC 58.5 ppm), and from both H-1a and H-1b (δH 3.39 ppm and δH 3.54 ppm) to C-3 (δC 64.7 ppm). External to this spin system, the proton at δH 5.09 ppm showed HMBC correlations into a protonated carbon at δC 109.5 ppm (C-5) and a quaternary carbon at δC 123.5 ppm (C-4). H-5, resonating at δH 6.83 ppm, appears as a doublet with a fine coupling constant of 0.8 Hz, shared with the proton at δH 5.09 ppm that appears as a doublet of fine doublets (J1 = 3.3 Hz, J2 = 0.8 Hz). The magnitude of this shared coupling constant is in keeping with a four bond coupling between the two protons at δH 5.09 ppm and δH 6.83 ppm, allowing the expansion of the substructure to what is shown below in Figure 3.10.  HMBC  H 5  H OH 2 NH2 4 3 1 Cl  Figure 3.10 Substructure in structure elucidation of girolline (1).  Further to this, the proton at δH 6.83 ppm (H-5) displayed HMBC correlations to C-4 (δC 123.5 ppm) and to another non-protonated carbon at δC 145.5 ppm. This information, along with the molecular mass, allowed for identification of this compound as being girolline (Figure 3.11) based upon a search in the literature of compounds isolated from marine sponges using the substructure in Figure 3.10 and the molecular mass.  165  H OH 2 5 4 N 3 1 NH2 NH Cl 6 H2N  HMBC  Figure 3.11 Key HMBC correlations used to solve structure of girolline (1).  1  13  Table 3.1 1D and 2D NMR data for girolline (1), recorded at 600 MHz ( H) and 150 MHz ( C) in  MeOD-d4  H OH 2 5 4 N 3 1 NH2 NH Cl 6 H2N 1  a  C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  1  42.8  H-1a: 3.54 dd (13.8, 3.2)  H-2, H-1b  C-3  42.8  H-1b: 3.4 dd (13.8, 9.6)  H-2, H-1a  C-2, C-3  2  59.8  4.53 dt (9.6, 3.3)  H-3, H-1a, H-1b  C-3, C-1  3  66  5.09 dd (3.3, 0.8)  H-2, H-4  C-2, C-4, C-4  4  124.2  -  -  -  5  110.9  6.83 d (0.8)  H-3  C-3, C-4, C-6  6  147.1  -  -  -  Recorded at 150 MHz. b Assigned according to HSQC recorded at 600 MHz. c Recorded at 600 MHz.  166  3.2.2 Synthesis of Girolline and Analogues Because of the difficulty in isolating pure girolline and the fact that published syntheses of girolline were available in the literature116,117, its synthesis was undertaken. This was done so that its activity could be confirmed and also so that structural analogues could be prepared in order to probe the motifs required for its activity.  Initially, the synthesis of girolline was envisioned to be best accessed via a route similar to the one utilized by Commerçon and Paris (Figure 3.12)116 – the key step being an aldol condensation118 stereospecifically controlled119 via use of an Evans oxazolidinone auxiliary.120,121 O  O N  Cl  O  1) Bu2BOTf, Et3N, CH2Cl2 -78oC to 20oC, 2 hr 2) CHO N N C(C6H5)3  2  O  OH O N  N  N H2N  Cl  N H  1  OH O N  NH2 Cl  Cl N C(C6H5)3  N C(C6H5)3 5  4  3  OH  O  NH4OH, THF, 20oC, 30 min  1) BH3-Me2S, THF, reflux, 30 min 2) 6N HCL(aq), 20oC 3) NaOH(aq) (30%), 5oC  1) Na2CO3-H2O, MeOH, 2oC OH 2) p-ClC6H4N2+Cl-, NH2 o H2O, 2 C, 20 min N 3) H2(1 atm), PtO2, MeOH, Cl HCl (2eq), 20oC, 24 hr N H 4) HCl 7  NH BOC  OH 1) (BOC)2O, CH2Cl2, o 20 C, 1 hr N 2) n-PrOH, AcOH, Cl reflux, 24 hr N C(C6H5)3  NH2  6  Figure 3.12 The synthesis of girolline(1) performed by Commerçon and Paris.  Instead of exactly repeating this method, however, the aldehyde component would be a 167  protected imidazole with a chlorine at the 2-position (8) (Figure 3.13), assembled using methodology developed by Eriksen and co-workers.122,123  O  O N  Cl  O  2)  N Cl  2  N CHO BOM  O  OH O  1) Bu2BOTF, Et3N, CH2Cl2 -78oC to 20oC, 2h  N Cl  N  O  NH4OH, THF, 20oC, 30min  N Cl BOM  OH O N Cl  N Cl BOM 10  9  8  NH2  1) BH3-Me2S, THF, reflux, 30min 2) 6N HCL(aq), 20oC 3) NaOH(aq) (30%), 5oC  OH N Cl  N Cl BOM  NH2  11 Figure 3.13 The synthesis of girolline(1). Possible route to the carbon skeleton, with a Cl at the 2position of the imidazole.  The 2-chloroimidazole could then be converted (Figure 3.14), as one of the final steps, into a 2-aminoimidazole by substitution of the chlorine with an azide124, followed by a reduction.125  168  OH N Cl  N BOM  OH  1) NaN3 in DMSO 2) Reduction 3) Deprotection  NH2 Cl  N H2N  NH2 Cl  N H 1  11  Figure 3.14 The synthesis of girolline(1): possible route to form the amino-imidazole.  The 2-chloroaldehyde 8 was synthesized in three steps in a good yield from imidazole, by the reaction scheme shown in the Figure 3.15.  BOM-Cl, K2CO3 DMF, RT, 24 hr  N N H  12  N N O  13  1) nBuLi, THF, -78oC, 10 min 2) C2Cl6, THF, Cl -78oC, 1hr  N N BOM  14  1) nBuLi, THF, -78oC, 10 min 2) DMF, THF, Cl -78oC, 1hr  N N BOM O  8  Figure 3.15 The synthesis of girolline(1): synthesis of the aldehyde, 8, from imidazole.  This aldehyde 8 crystallized nicely as colorless plates, and provided a good base for exploration into carbon skeleta similar to girolline, as shown in Figure 3.16. Coupling of this aldehyde 8 to the acylated oxazolidinone 2 via standard conditions119 yielded the coupled compound 9 in moderate yield – as well as unreacted aldehyde (8). With this coupled compound 9 in hand, removal of the oxazolidinone to yield the amide 10 went in good yield.116 Reaction of the amide, 10, with BH3-Me2S, did yield the desired amine, 11, but in a complex mixture with a poor yield. Due to the difficulties in obtaining compound 11, the model compound 15 was prepared via butylation of the 2-chloroaldehyde 8 (Figure 3.16). Attempts to 169  substitute the chlorine at the 2 position of the imidazole ring with an azide on the model compound 15 were unsucessful. Additionally, explorations into this substitution using compounds 8, 9 and 10 resulted in decomposition of the starting materials. Thus, this route was abandoned.  Cl  O  OH O  N N BOM  CHO  1) 2, Bu2 BOTf, Et3N, CH2Cl2 -78o C to 20o C, 2h  N N Cl BOM  Cl  2) Add in 8  N  8  OH O NH4 OH, THF, O 20 oC, 30min  N Cl  9 n-BuLi, THF -78o C  NH2 N Cl BOM 10  OH N Cl  N BOM 15  Figure 3.16 Synthesis of the amide 10 and the model compound 15.  The coupled compound 10, as well as the model compound 15, were deprotected via two different methods, shown in Figure 3.17. Exposure to 1 atm of H2 with a Pd(0) catalyst removed both the protecting group and the chlorine at the 2-position. Aluminum chloride, however, removed just the BOM protecting group.126. The resulting compounds 16-19 were tested for biological activity. None of these compounds showed any activity.  170  AlCl3, DCM, 30 min, 20oC  OH O N N Cl BOM  Cl  10  OH N BOM 15  N NH  Cl  NH2 Cl 16  NH2 OH O  H2(1 atm), Pd(C), EtOH, 20oC, 24 hr  N NH  NH2 Cl 17 OH  AlCl3, DCM, 30 min, 20oC  N NH 18  Cl  N Cl  OH O  OH  H2(1 atm), Pd(C), EtOH, 20oC, 24 hr  N NH 19  Figure 3.17 The synthesis of analogues 16-19 of girolline(1).  Given the difficulties encountered in placing the amino group at the 2-position on the imidazole ring, an approach that featured the formation of the amino imidazole rings as one of the initial synthetic steps was pursued. In a method utilized by Horne and coworkers in the synthesis of oroidin127, a sodium-mercury amalgam was used to reduce the methyl ester of an amino acid (in this case, ornithine), to form an amino imidazole, with the side chain from the amino acid then forming the substituent at the 4(5)-position on the imidazole ring (Figure 3.18).  171  O  O  HO  NH2 NH2  SOCl2, MeOH, MeO 0oC, 4 hr NH2  1) Na(Hg), HCl, H2O 0oC, pH 1.5-2.5, 1 hr N NH2 2) NaOH (to pH 4.3) H2N HN 3) NH2-CN, 95oC, 2.5 hr  NH2 22  Figure 3.18 Synthesis of the 2-amino imidazole derivative 22.  The 2-amino imidazole ornithine derivative, 22, was then purified by recrystallization from ethanol. A small aliquot of the crude reaction product before recrystallization was submitted for testing, and showed moderate biological activity that was reproducible. However, the pure recrystallized 22 did not show any activity. A repeat of this synthesis and testing of both the crude and recrystallized ornithine derivative also showed no activity, raising questions about the validity of the one positive test result on the crude compound 22.  3.3 Isolation of Massacreones and 20-Hydroxy-Ecdysone In addition to identifying the sponge that was found to contain girolline, the TRL5 bioassay also identified a number of other hits from the Andersen lab marine extract library, including an extract called RJA06-08. The source of this extract was an unidentified cnidarian, tentatively identified as a zoanthid, harvested by hand using SCUBA on April 17, 2006, at a depth of -10 m off of a reef in Dominica at a dive site named “Rodney's Rock”, located along a straight stretch of coastline on Dominica's west coast at coordinates 15o 22.8 N, 61o 24.8 W. The cnidarian was a light greyish-white color with a rubbery texture, was approximately four centimeters across and three centimeters high and grew in clusters of approximately five to fifteen members per cluster. The principal compounds isolated from this extract were named after the nearby town of Massacre, Dominica. To properly identify the organism, recollection  172  was attempted two years later. However, new samples of the organism could not be located.  Whole, frozen RJA06-08 was completely submerged in methanol and extracted for twenty four hours. This was repeated two more times, and the combined extracts were concentrated in vacuo, then re-suspended in water. This aqueous suspension was partitioned against ethyl acetate (twice), and then against butanol (twice) to yield three fractions based on solubility. Bioassay guided fractionation indicated that the activity, which was somewhat diminished as compared to the crude extract, was in the aqueous and butanolic fractions.  The butanolic fraction from the solvent partitions was subjected to size exclusion chromatography using LH-20 preswollen in methanol. LH-20 column fractions were grouped together into two fractions based on color, thin layer chromatography profiles, and charring colors using a p-anisaldehyde stain. The first fraction showed activity in the bioassay, and was thus subjected to reversed phase HPLC using a step gradient starting at 10% aqueous acetonitrile for 10 minutes, then 30% aqueous acetonitrile for another 50 minutes. From this HPLC run, fourteen fractions were collected, including five that were major peaks. These five peaks, eluting at 24.5 minutes, 27.1 minutes, 34.3 minutes, 39.5 minutes, and 43.4 minutes showed similar 1H spectra and appeared to be terpenoid in origin. Unfortunately, none of these peaks, nor any of the other fractions from the HPLC run, showed any appreciable bioactivity.  The peak at 24.5 minutes was purified using reversed phase HPLC with 15% acetonitrile to yield massacreone D (28) as a clear glass (Rt = 15 min, 0.7 mg). The peak at 27.1 minutes was purified on reversed phase HPLC using 25% acetonitrile to yield the known 20-hydroxyecdysone (29) as a clear glass (Rt = 3.9 min, 1.1 mg). The peak at 34.3 minutes was purified on 173  reversed phase HPLC using a gradient from 25%-45% acetonitrile over 30 minutes to yield massacreone A (25) as a white powder (Rt = 15.6 minutes, 37.7 mg). The peak at 39.5 minutes was purified on reversed phase HPLC using a gradient from 25% - 45% acetonitrile over 30 minutes to yield massacreone B (26) as a clear glass (Rt = 18.4 min, 7.5 mg). The peak at 43.4 minutes was identified as massacreone C (27) (1.6 mg). O OH OH  OH  OH O  OH  OH H HO HO  H  OH H  HO  H  HO  HO  OH  H  OH HO  HO  H O  Massacreone A ( 25 )  HO  Massacreone B (26)  OH OH  OH  H  HO  HO  Massacreone C ( 27 ) OH  OH  OH OH  OH H  H HO  HO H  HO  OH  HO  H  OH  HO HO Massacreone D (28)  HO 20-Hydroxy-Ecdysone (29)  Figure 3.19. The massacreones (25-28) and 20-hydroxy-ecdysone (29), isolated from the Dominican cnidarian RJA06-08.  174  3.3.1 Structure Elucidation Of Massacreone A  HO HO  26 21 OH OH 27 22 18 OH 17 20 11 24 28 H 19 13 HO 9 15 1 OH 3 5 H 7 H O 25 Figure 3.20 Massacreone A (25)  Isolated as a white, fluffy powder, massacreone A (25) gave a [M+Na]+ pseudomolecular ion at m/z 533.3104, in its high resolution electrospray ionization time-of-flight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C28H46O8 (C28H46O8Na calculated for m/z 533.3090), indicating six degrees of unsaturation. The numbering scheme used in the structure elucidation of massacreone A and its NMR data is summarized in Table 3.2.  175  OH OH OH H HO HO  H  HO  OH  H O  Figure 3.21 600 MHz 1H NMR spectrum of massacreone A (25), recorded in Pyridine-d5.  176  OH OH OH H HO HO  H  HO  OH  H O  Figure 3.22 150 MHz 13C NMR spectrum of massacreone A (25), recorded in Pyridine-d5.  177  OH OH OH H HO HO  H  HO  OH  H O  Figure 3.23 600 MHz COSY spectrum of massacreone A (25), recorded in Pyridine-d5.  178  OH OH OH H HO HO  H  HO  OH  H O  Figure 3.24 600 MHz HSQC spectrum of massacreone A (25), recorded in Pyridine-d5.  179  OH OH OH H HO HO  H  HO  OH  H O  Figure 3.25 600 MHz HMBC spectrum of massacreone A (25), recorded in Pyridine-d5.  180  OH OH OH H HO HO  H  HO  OH  H O  Figure 3.26 600 MHz TROESY spectrum of massacreone A (25), recorded in Pyridine-d5.  181  Table 3.2 1D and 2D NMR data for massacreone A (25), recorded at 600 MHz (1H) and 150 MHz (13C) in Pyridine-d5.  HO HO  26 21 OH OH 27 22 18 OH 17 20 11 24 28 H 19 13 HO 9 15 1 OH 3 5 H 7 H O 25  C#  C δ (ppm)a  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  ROESYc (H→H)  1  38.4  H-1α 2.17 m  H-1β, H-2  C-2, C-3, C-5, C-8, C-9, C-10  H-1β, H-2, H-3, H-4β, H-4α  H-1β 1.96 m  H-1α, H-2, H-5  C-2, C-3, C-5, C-6, C-8, C-9, C-10  H-1α, H-2, H-3, H-5  1  2  68.5  4.20 m  H-3  -  H-1β, H-1α, H-9  3  68.4  4.25 br s (W1/2=8.4)  H-2, H-4  C-1, C-5  H-1β, H-1α, H-4β, H-4α  4  21.5  H-4α 1.88 m  H-5  -  H-1β, H-1α, H-3, H-4β, H-5, H-7, H-9  H-4β 1.76 m  H-3, H-5  C-2  H-3, H-4α, H-5  5  51.8  3.04 m  H-7, H-1β, H-4  C-6, C-7, C-9, C-10, C-11  H-1β, H-4β, H-4α, H-19  6  203.9  -  -  -  -  7  122  6.27 s  H-5, H-9  C-5, C-9, C-14  H-4α, H-15α  8  166.7  -  -  -  -  9  34.8  3.62 t (7)  H-7, H-11α, H-11β  C-4, C-6, C-8  H-2, H-4α, H-11α, H-11β, H-12β, H-12α  10  39  -  -  -  -  11  32.9  H-11α 2.07 m  H-9  C-12  H-9, H-11β  H-11β 1.82 m  H-9  C-9, C-10, C-12  H-9, H-11α, H-18, H-19  H-12α 2.62 m  H-11, H-12β  C-13, C-16, C-17, C-18  H-9, H-15α, H-21  H-12β 2.06 m  H-11, H-12α  C-9, C-13, C-14, C-16  H-9, H-17  12  a  13  32.4  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz. b  c  182  Table 3.2 (Continued) 1D and 2D NMR data for massacreone A (25), recorded at 600 MHz (1H) and 150 MHz (13C) in Pyridine-d5. C#  C δ (ppm)a  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  ROESYc (H→H)  13  48.5  -  -  -  -  14  84.5  -  -  -  -  15  32.1  H-15α 2.63 td (12.8, 4.2)  H-15β  C-16  H-7, H-12α, H-16β, H-17, H-23  H-15β 2.19 m  H-15α  C-13, C-14  H-17, H-18  H-16α 2.53 q (11)  H-16β, H-17  C-14, C-15, C-17, C-20  H-15β, H-16β, H-17, H-22  H-16β 2.34 m  H-16α  C-13  H-15β, H-16α, H-17, H-18, H-22, H-27  16  a  13  21.75  1  17  50.4  3.04 m  H-16α  C-13, C-15, C-16, C-18  H-12α, H-15β, H-15α, H-16β, H-16α, H-21, H-22  18  18.3  1.26 s, 3H  -  C-12, C-13, C-17  H-11β, H15β, H-16β  19  17.6  1.10 s, 3H  -  C-1, C-5, C-9, C-10  H-5, H-11β  20  77.3  -  -  -  -  21  21.8  1.65 s, 3H  -  C-17, C-20, C-22  H-12α, H-17  22  74.4  4.46 d (10.9)  H-23a, H-23b  C-20, C-21, C-23  H-16β, H-16α, H-17, H-23a, H-23b  23  35.8  H-23a 2.32m  H-22, H-23b  C-22, C-24, C-26, C-28  H-15α, H-22  H-23b 2.06 m  H-22, H-23a  C-24, C-28  H-22  24  76.8  -  -  -  -  25  34  2.33 m  H-26, H-27, H-28b  C-24, C-27  H-26, H-27, H-28a, H-28b  26  24.9  1.15 d (6.8), 3H  H-25  C-24, C-25  H-25, H-27  27  17.7  1.23 d (6.8), 3H  H-25  C-24, C-25, C-26  H-16β, H-25, H-26  28  66.7  H-28a 4.18 d (10.8)  H-28b  C-23, C-24, C-25  H-25, H-26, H-27  H-28b 4.09 d (10.8)  H-25, H-28a  C-23, C-24, C-25  H-25, H-26, H-27  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  183  Analysis of 1H and 13C NMR, as well as 2D COSY and HSQC spectra, identified several key substructures, shown in Figure 3.26 below. 11 9  12  26  OH 22  25 27  23 HO 2 HO  COSY 8  4H  O  6  17 15  Figure 3.27 Substructures of massacreone A (25) determined from analysis of COSY and HSQC data.  The C1-C5 substructure was pieced together as follows. A pair of protons, resonating at δH 2.17 ppm (H-1β) and at δH 1.96 ppm (H-1α) were attached to the same carbon, C-1 (δC 38.5 ppm) and showed a COSY correlation to a resonance at δH 4.19 ppm (H-2), which in turn showed a COSY correlation to a resonance at δH 4.26 ppm (H-3). This proton, H-3, displayed a COSY correlation to the resonance at δH 1.86 ppm (H-4α), and also a weak COSY correlation to a proton at δH 1.76 ppm (H-4β), both of which were attached to the same carbon, resonating at δC 21.6 ppm (C-4). These two protons H-4β and H-4α displayed a COSY correlation to a proton resonance at δH 3.04 ppm (H-5).  A proton resonance at δH 3.62 ppm (H-9) showed COSY correlations to two resonances at δH 2.07 ppm (H-11α) and δH 1.82 ppm (H-11β), which were both attached to the same carbon, resonating at δC 33.0 ppm (C-11). In turn, these protons both showed COSY correlations a resonance at δH 2.62 ppm (H-12α), which was attached to a carbon that resonates at δC 32.5 ppm (C-12) along with a second proton, resonating at δH 2.06 ppm (H-12β). This small spin system established the C9-C12 substructure. 184  A proton resonating at δH 3.04 ppm (H-17, isochronous with H-5) showed COSY correlations to two resonances at δH 2.34 ppm and δH 2.53 ppm (H-16β and H-16α, respectively), which were both attached to the same carbon resonating at δC 21.85 ppm (C-16). In turn, these protons showed COSY correlations to two more proton resonances at δH 1.93 ppm and δH 2.19 ppm (H-15β and H-15α, respectively), which were both attached to the same carbon at δC 32.2 ppm (C-15). This spin system established the C15-C17 substructure.  Two protons resonating at δH 2.06 ppm and δH 2.32 ppm (H-23a and H-23b), which were attached to the same carbon C-23 (δC 35.9 ppm), both showed a COSY correlation to the resonance at δH 4.46 ppm (H-22), establishing the C22-C23 substructure. Additionally, two proton resonances that appeared as doublets in the 1H NMR spectrum (δH 1.16 ppm (H-26) and δH 1.22 ppm (H-27)) and both integrated to three protons, showed a distinct COSY correlation to a proton resonance at δH 2.33 ppm (H-25), establishing the isopropyl moiety as an additional substructure (C25-C27).  Examination of the proton and carbon spectra, with these substructures in mind, revealed the C6-C8 substructure - a conjugated ketone δC 204.0 ppm (C-6), δC 166.8 ppm (C-8), and δC 122.1 ppm (C-7) that had an attached proton that resonates at δC 6.27 ppm (H-7). In addition to this, a terminal -CH2OH group was noted, with its characteristic carbon resonance at δC 66.8 ppm (C-28) showing two attached and coupled proton resonances at δH 4.09 ppm and δH 4.19 ppm (H28a and H-28b), as well as three methyl groups, seen as three singlet peaks in the 1H NMR spectrum integrating to three protons each at δH 1.65 ppm (H-21), δH 1.26 ppm (H-18), and δH 1.13 ppm (H-19). Lastly, two quaternary carbon resonances (δC 39.0 ppm (C-10) and δC 48.6 185  ppm (C-13)), as well as three oxygenated quaternary carbon resonances at δC 76.9 ppm (C-24), δC 77.4 (C-20) and δC 84.6 (C-14) were observed.  The substructures of massacreone A were pieced together using the 2D HMBC data. The protons from the methyl at position 19 (resonating at δH 1.19 ppm) displayed HMBC correlations into the quaternary carbon resonance at δC 39.0 ppm (C-10), as well as into carbons C-1 and C-5 (at δC 38.5 ppm and δC 51.9 ppm, respectively), which were already linked together via COSY correlations in the C1-C5 substructure. This established the presence of a six-membered ring (ring A) comprised of carbons C-1, C-2, C-3, C-4, C-5 and C-10, with a methyl substituent at C10 and oxygen attached at C-2 and C-3. In addition to the HMBC correlations used to establish the presence of this ring, the proton resonance at δH 1.13 ppm (H-19) also showed an HMBC correlation to a carbon resonance at δC 34.9 ppm (C-9) allowing this ring to be linked to the C9C12 substructure, as shown in Figure 3.27. 19  O O  12  1 10 9 5  COSY HMBC  Figure 3.28 structure elucidation of the A ring of massacreone A (25) – key HMBC correlations from the methyl group at position 19.  Proton H-5, resonating at δH 3.04 ppm, displayed an HMBC correlation to C-9, further confirming the placement of the C9-C12 substructure, as well as correlations into carbon resonances at δC 204.0 ppm (C-6) and δC 122.1 ppm (C-7), linking the C6-C8 substructure to ring A. Additionally, the proton at δH 6.27 ppm (H-7) showed a faint COSY correlation to H-5, as 186  well as HMBC correlations to the carbon resonances at δC 51.9 ppm (C-5) and δC 34.9 ppm (C9). In addition to this, the proton resonance at δH 3.62 ppm (H-9) showed an HMBC correlation to the carbon resonance at δC 166.8 ppm (C-8). Thus, a second six membered ring (ring B) was established, fused to the already elucidated ring A, as shown in Figure 3.28. 11 O O  5  12  9  8 7 6  COSY HMBC  O Figure 3.29 Structure elucidation of the B ring of massacreone A (25) – key HMBC correlations from the protons at positions 5, 9 and 11.  The protons at position 12 (δH 2.06 ppm (H-12β) and δH 2.62 ppm (H-12α)) showed HMBC correlations to the oxygenated quaternary carbon resonance at δC 84.6 ppm (C-14), as well as an HMBC correlation between H-12α and the quaternary carbon at δC 48.6 ppm (C-13). Moreover, the methyl singlet at δH 1.26 ppm (H-18) showed HMBC correlations into these same two quaternary carbons (C-13 and C-14), as well as into the carbon at δC 32.5 ppm (C-12) Additionally, the proton at δH 6.27 ppm (H-7), showed an HMBC correlation to the oxygenated quaternary carbon at δC 84.6 ppm (C-14). Thus, a third six membered ring (ring C) was found to be fused to ring B, as shown in Figure 3.30.  187  18 12 13  O O  7  COSY  14 O  HMBC  O Figure 3.30 Structure elucidation of the C ring of massacreone A (25) – key HMBC correlations from protons at positions 7, 12 and 18.  The protons from the methyl at position 18 (δH 1.26 ppm) showed an HMBC correlation to the carbon resonance at δC 50.5 ppm (C-17), establishing the linkage of the C15-C17 spin system to the growing steroidal core. The proton resonance at δH 3.04 ppm (H-17) showed an HMBC correlation to C-13. The protons at position 15 (δH 1.93 ppm (H-15β) and δH 2.19 ppm (H-15α)) were found to both have HMBC correlations to the carbon resonance at δC 84.6 ppm (C-14), as well as H-15β having an HMBC correlation into the carbon resonance at δC 48.6 ppm (C-13). One of the protons at position 16 – δH 2.34 ppm (H-16β) – showed an HMBC correlation to the C-13 resonance, while the other – δH 2.53 (H-16α) – showed an HMBC correlation to the C-14 resonance. These correlations clearly established the final ring (ring D) of the steroidal core, as shown in Figure 3.31. 18 13  17 16  O 14 O  O  15  COSY HMBC  O Figure 3.31 Structure elucidation of the D ring of massacreone A (25) – key HMBC correlations from protons at positions 15, 16, 17 and 18.  188  The protons from the methyl at position 21, resonating at δH 1.65 ppm, showed an HMBC correlation into the carbon resonance at position 17 (δC 50.5 ppm) and also into an oxygenated quaternary carbon resonance at δC 77.4 ppm (C-20) and into the carbon resonance at δC 74.5 ppm (C-22), which was also oxygenated and was assigned previously to the two carbon spin system, C22-C23. In addition to this, the protons on position 16 (H-16β and H-16α) in ring D also showed an HMBC correlation to C-20 (δC 77.4 ppm), and additionally, the proton resonance at δH 4.46 ppm (H-22) showed an HMBC correlation into C-20 as well (Figure 3.32). 21 20 17  O O O  O  O  22 23 16 COSY HMBC  O Figure 3.32 Structure elucidation of the steroidal side chain of massacreone A (25) – key HMBC correlations from protons at positions 16, 21 and 22.  The protons at position 23 (δH 2.06 (H-23a) and δH 2.32 (H-23b)) both show HMBC correlations to an oxygenated quaternary carbon resonance at δC 76.9 ppm (C-24) and to the carbon resonance at δH 66.8 ppm (C-28). The pair of protons at position 28 (δH 4.09 ppm (H28a) and δH 4.19 ppm (H-28b)) both show HMBC correlations back into C-24, as well as into the carbon resonance at δC 34.0 ppm, C-25. The two resonances assigned as the methyl groups in the C25-C27 substructure (δH 1.16 ppm (H-26) and δH 1.22 ppm (H-27)) both show HMBC correlations into C-24, placing the final pieces of the side chain together (Figure 3.33). Given the molecular formula established by HRESIMS (C28H46O8), all of the oxygen atoms placed in 189  the structure (with the exception of the carbonyl) were established as alcohols. HO OH  26 25  23 HO OH HO  24 HO 28  27  OH  COSY HMBC  O Figure 3.33 Structure elucidation of the steroidal side chain of massacreone A (25) – key HMBC correlations from protons at positions 23, 26, 27 and 28.  With the planar structure of massacreone A established, attention was turned onto the relative configuration at the various centers. A clear and strong ROESY correlation was observed between the methyl protons resonating at δH 1.13 ppm (H-19) and the resonance at δH 3.04 ppm (H-5), establishing a cis ring fusion between the A and B rings. This was further emphasized by a ROESY correlation between the resonance at δH 6.27 ppm (H-7) and the resonance at δH 1.86 ppm (H-4α). Ecdysteroids typically have a cis ring fusion between rings A and B. The relative configuration at position 3 was determined by observing the W1/2 value of the proton resonance H-3, which appeared as a broad singlet at δH 4.25 ppm in the 1H NMR spectrum. In ecdysteroids, the resonance attributed to H-3 is typically a broad singlet, owing to the fact that there is a slow conformational change of ring A on the NMR time scale at 278K and 600 MHz. Ecdysteroids that have the 3-OH in the α position, typically have a W½ value that is in the 15-25 Hz range, while those with the 3-OH in the β position typically show a W½ for H-3 in the 5-15 Hz range.128. Thus, with a W½ value of 8.4 Hz (in pyridine), position 3 was assigned with its OH in the β position, and the proton H-3 in the α position as shown in Figure 3.34. The relative configuration of the 3, 5 and 10 positions were assigned as 3S*, 5R* and 10R*. These 190  orientations are very typical for edysteroids.129. 19  CH3  CH3  5  H  O HO  Hb  O H  4  HO  H  7  H  HO  Ha  H  3  H  HO  Figure 3.34 Relative configuration of massacreone A (25): key ROESY correlations between the methyl protons H-19 and H-5, as well as between H-4α and H-7, establishing a cis ring fusion. The A ring flips, leading to broadening of the H-3 peak in the 1H NMR spectrum.  To determine the relative configuration at positions 2 and 9, the ROESY correlations seen for those protons resonances (δH 4.20 ppm, H-2 and δH 3.62 ppm, H-9) were examined. First and foremost, these two resonances showed a strong ROESY correlation, which would only be possible if both were placed in an axial orientation in their respective rings. Moreover, there was no evidence of a ROESY correlation between H-19 and H-9, giving credence to H-9 being on the opposite face of ring B from the methyl group placed on position 10. The resonance at δH 1.13 ppm (H-19) did however show a ROESY correlation to one of the protons at position 11, resonating at δH 2.07 ppm (H-11β). The resonance at δH 4.20 ppm (H-2) and the resonance at δH 3.62 ppm (H-9) both showed a ROESY correlation to one of the protons at position 4 (δH 1.88 ppm, H-4α). This allowed the assignment of the relative configuration of positions 2 and 9 as being 2S* and 9R*, as shown in Figure 3.35.  191  Hb  19 CH3 Ha  2 HO  4  HO  9  11  O H  H H  Ha  Hb  Figure 3.35 Relative configuration of massacreone A (25): key ROESY correlations between H-19 and H-11β, as well as between H-2, H-4α and H-9.  The resonance at δH 3.62 ppm (H-9) showed an additional ROESY correlation to a resonance at δH 2.62 ppm, H-12α. Furthermore, the resonance at δH 2.07 ppm (H-11β) showed a ROESY correlation to the methyl protons at position 18 (δH 1.26 ppm). These correlations, along with the fact that C-8 is an sp2 hybridized carbon, allows ring C to be assigned as a boat, with the methyl group at position 18 to be placed axial off of position 13, establishing the relative configuration of position 13 as 13R* (Figure 3.36). 18  CH3  Hb  Hb  CH3 Ha  9  O HO  HO  11 12  H  Ha  OH  H  Figure 3.36 Relative configuration of massacreone A (25): key ROESY correlations between H-18 and H-11β and between H-9 and H-12α.  192  As shown in Figure 3.37, the fusion between rings C and D was established as trans due to the presence of ROESY correlations between the proton resonances at δH 1.26 ppm (H-18) and δH 1.93 ppm (H-15β), as well as between H-18 and the resonance at δH 2.53 ppm (H-16β). If the ring fusion were cis between rings C and D, the distance from the protons at position 18 to protons at positions 15 and 16 would be too great to allow observation of ROESY effects. Moreover, the standard ring fusion typically seen with ecdysteroids is trans between rings C and D. Thus, position 14 was established as having a relative configuration of 14S*. The final stereocenter in the steroid nucleus, position 17, was assigned as having a relative configuration of S based upon the presence of a ROESY correlation between its proton H-17 (δH 3.04 ppm) and the resonance at 2.62 ppm (δH H-12α). 18CH  3  Hb  Hb  CH3  12 O HO  HO  H  Ha  15 Ha OH H  17  Hb  16 Ha  Figure 3.37 Relative configuration of massacreone A (25): key ROESY correlations between H-18 and H-15β, H-18 and H-16β, H-12α and H-17, establishing the C/D ring fusion as trans.  The assignment of the relative configuration of the core steroid ABCD ring system for massacreone A (25) is thus 2S*, 3R*, 5R*, 9R*, 10R*, 13R*, 14S*, 17S*. Unfortunately, due to free rotation about the C-17/C-20 bond, as well as in the remainder of the steroid side chain, the relative configuration of positions 20, 22 and 24 could not be determined. Attempts to grow crystals of this compound were not successful. However, because 20-hydroxy-ecdysone (29) 193  was also isolated from this same organism, it is likely, due to common biosynthetic origins, that the relative configuration of positions 20 and 24 are 20R* and 22R* (Figure 3.38). This assignment is supported by comparing the very similar δc values for massacreone A (25) recorded in pyridine, and literature values for 20-hydroxy-ecdysone (29) recorded in pyridine129 at C-20 (77.3 ppm and 77.1 ppm), C-21 (21.8 ppm and 21.8 ppm) and C-22 (74.4 ppm and 77.7 ppm). It is exceedingly common in ecdysteroids129 for these positions to possess this geometry. There is no literature precedent for attachment of the oxygenated C-28 at an oxygenated C-24, thus, the configuration position 24 cannot be mused upon. OH  OH OH  OH OH OH  H  H HO  HO H HO  HO  OH  H  OH  HO HO 25  HO 29  Figure 3.38 Massacreone A (25), and 20-hydroxy-ecdysone (29). The relative configuration at C-24 of massacreone A is unknown.  194  3.3.2 Structure Elucidation of Massacreone B  29  O  26 OH O 21 OH 27 18 22 17 20 24 11 19 H 28 13 9 15 HO 1 OH 3 5 7 HO H O 26 Figure 3.39 Massacreone B (26).  Isolated as a clear glass, massacreone B (26) gave a [M+Na]+ pseudomolecular ion at m/z 557.3085, in its high resolution electrospray ionization time-of-flight mass spectrum (HRESITOFMS), corresponding to a molecular formula of C30H46O8 (C30H46O8Na calculated for m/z 557.3090), indicating eight degrees of unsaturation. Analysis of 1H and 13C NMR, as well as 2D COSY, HSQC, HMBC and ROESY spectra, and comparison to the data from massacreone A, allowed the quick identification of massacreone B as being an ecdysteroid, with a steroidal side chain being comprised of C10H19O4, and two degrees of unsaturation. The structure of massacreone B, plus the numbering scheme used in its structure elucidation, are shown in Figure 3.38, and its 1D and 2D NMR data is summarized in Table 3.3.  195  O OH O  OH  H HO HO  H  OH  H O  Figure 3.40 600 MHz 1H NMR spectrum of massacreone B (26), recorded in MeOD-d4.  196  O OH O  OH  H HO HO  H  OH  H O  Figure 3.41 150 MHz 13C NMR spectrum of massacreone B (26), recorded in MeOD-d4.  197  O OH O  OH  H HO HO  H  OH  H O  Figure 3.42 600 MHz COSY spectrum of massacreone B (26), recorded in MeOD-d4.  198  O OH O  OH  H HO HO  H  OH  H O  Figure 3.43 600 MHz HSQC spectrum of massacreone B (26), recorded in MeOD-d4.  199  O OH O  OH  H HO HO  H  OH  H O  Figure 3.44 600 MHz HMBC spectrum of massacreone B (26), recorded in MeOD-d4.  200  O OH O  OH  H HO HO  H  OH  H O  Figure 3.45 600 MHz TROESY spectrum of massacreone B (26), recorded in MeOD-d4.  201  Table 3.3 1D and 2D NMR data for massacreone B (26), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. 29  O  26 OH 21 OH O 27 18 22 17 20 24 11 19 H 28 13 9 15 1 OH 3 5 7  HO HO  H O 26  a  C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  ROESYc (H→H)  1  37.9  H-1α 1.8 m  H-1β, H-2  C-2, C-3, C-5, C-8, C-9, C-10  H-1β, H-2, H-3, H-4β, H-4α  H-1β 1.44 t (12.9)  H-1α, H-2  C-2, C-3, C-4, C-5, C-8, C-9, C-10, C-19  H-1α, H-2, H-3, H-5, H-10  2  68.6  3.83 d (11.7)  H-1β, H-1α, H-3  -  H-1β  3  68.9  3.94 br s (W ½ = 8.1)  H-2, H-4β, H-4α  -  H-4α  4  21.5  H-4α 1.78 m  H-3, H-5  -  H-1β, H-1α, H-3, H-4β, H-5, H-7, H-9  H-4β 1.68 m  H-3, H-5  C-2, C-5  H-3, H-4α, H-5  5  51.9  2.39 m  H-4β, H-4α  C-6, C-7, C-9, C-10, C-11, C-19  H-1β, H-4β, H-4α, H-19  6  206.5  -  -  -  -  7  122.4  5.8 s  H-9  C-5, C-9, C-14  H-4α, H-15β  8  167.8  -  -  -  -  9  35.2  3.15 t (8.7)  H-7, H-11α, H-1α  -  H-2, H-4α, H-11α, H-12β, H-12α  10  39.4  -  -  -  -  11  33  H-11α 1.69 m  H-9  C-12, C-13, C-17  H-9, H-11β  H-11β 1.74 m  H-9  C-9, C-10, C-12  H-9, H-11α  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  202  Table 3.3 (Continued) 1D and 2D NMR data for massacreone B (26), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. C#  C δ (ppm)a  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  ROESYc (H→H)  12  32.7  H-12α 2.11 m  H-11α, H-11β, H-12β  C-13, C-16, C-17, C-18  H-9, H-15α, H-21  H-12β 1.88 m  H-11α, H-11β, H-12α  C-9, C-13, C-16  H-9, H-16β, H-17  1  13  49.7  -  -  -  -  14  85.4  -  -  -  -  15  31.9  H-15α 1.98 m  H-15β  C-16  H-7, H-12α, H-16β, H-17, H-23  H-15β 1.63 m  H-15α  C-13, C-14  H-7  H-16α 1.88 m  H-16β, H-17  -  H-15β, H-16β, H-17, H-22  H-16β 1.82 m  H-16α  C-13  H-12α, H-16α, H-17, H-18, H-22, H-27  16  21.7  17  50.9  2.39 m  H-16α  C-13, C-15, C-16, C-18  H-12α,H-16β, H-16α, H-21, H-23  18  18.2  0.89 s, 3H  -  C-12, C-13, C-17  H-15β, H-15α, H-16α, H-12α, H-21  19  24.5  0.96 s, 3H  -  C-1, C-2, C-5, C-9, C-10  H-1β, H-4β, H-5, H-11α  20  77.4  -  -  -  -  21  21.7  1.3 s, 3H  -  C-17, C-20, C-22  H-12α, H-17, H-18  22  78.6  5.05 d (10.4)  H-23a, H-23b  C-20, C-21, C-23, C-24, C-29  H-16α, H-17, H-23a, H-23b, H-25  23  37.5  H-23a 2.1 m  H-22, H-23b  C-22, C-24, C-25, C-28  H-22  H-23b 2.42 m  H-22, H-23a  C-24, C-25, C-28  H-22  24  150.1  -  -  -  -  25  43  2.25 q (6.7)  H-26, H-27a, H-27b  C-23, C-24, C-26, C-27, C-28  H-26  26  17.8  1.07 d (6.9), 3H  H-25  C-24, C-25, C-27  H-25, H-28a  27  67.2  H-27a 3.37 dd (10.6, 5.4)  H-26, H-27b  C-24, C-25, C-26  -  H-27b 3.59 dd (10.6, 8.0)  H-26, H-27a, H-27b  C-24, C-25, C-26  -  H-28a 4.77 s  H-28b  C-23, C-24, C-25  H-26  H-28b 4.84 s  H-28a  C-23, C-24, C-25  -  -  -  -  -  2.02 s, 3H  -  C-29  -  28  a  13  112.3  29  173.1  30  21.4 b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  203  Analysis of the 1D carbon spectrum revealed three new peaks in the sp2 region of the spectrum as compared to the 13C NMR spectrum of massacreone A (25): δC 112.3 ppm (C-28), δC 150.2 ppm (C-24) and δC 173.1 ppm (C-29). These values suggested that the two unsaturations in massacreone B's side chain are due to a double bond (between δC 112.3 ppm, C-28 and δC 150.2 ppm, C-24) as well as an ester (δC 173.1 ppm, C-29).  A proton resonance, which appeared as a doublet in the 1H NMR spectrum of massacreone B at δH 5.05 ppm (H-22), displays COSY correlations to two proton resonances assigned to position 23 (δH 2.13 ppm (H-23a) and δH 2.43 ppm (H-23b)). Additionally, H-22 displays HMBC correlations to quaternary carbon resonances at δC 77.4 ppm (C-20) and δC 173.1 ppm (C-29) (Figure 3.46). The only other correlation into the carbon resonance at δC 173.1 ppm was from a proton resonance that appears as a singlet, integrating to three protons, that has a chemical shift characteristic of a methyl group of an acetyl moiety (δH 2.03 ppm (H-30)). O 30 29 O O 20 22  COSY 23  HMBC  Figure 3.46 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations from H-22 and H-30, allowing the placement of the C-29/C-30 acetate at position 21.  The proton resonance at δH 2.13 ppm (H-23a) also displayed an HMBC correlation to the oxygenated carbon resonance at δC 77.4 ppm (C-20). The proton resonance at δH 1.31 ppm (H21) also showed an HMBC correlation to C-20, as well as to a carbon resonance at δC 50.5 ppm (C-17). The resonance at δH 1.31 ppm also showed an HMBC correlation to the carbon  204  resonance at δC 78.6 ppm (C-22). This information is summarized in Figure 3.47.  O OAc 21 20 22  19  23 17  COSY HMBC  OH Figure 3.47 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations from H-21 and H-23.  The resonances at δH 4.85 ppm (H-28a) and δH 4.78 ppm (H-28b) both show HMBC correlations to the carbon resonances at δC 37.9 ppm (C-23), δC 150.2 ppm (C-24), and δC 43.0 ppm (C-25). Moreover, the two protons at position 23 (δH 2.13 ppm (H-23a) and δH 2.43 ppm (H-23b)) showed HMBC correlations to the carbon resonances at δC 150.2 ppm (C-24) and δC 112.3 ppm (C-28). This established the presence of a gem-di substituted double bond off of the main steroidal side chain (Figure 3.48). 28  AcO HO  COSY 23  24 25  HMBC  R Figure 3.48 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations used to position a gem-di substituted double bond in the steroidal side chain.  The resonance at δH 2.26 ppm (H-25) showed a COSY correlation to a resonance that appeared as a singlet at δH 1.08 ppm (H-26). H-25 also showed COSY correlations to two protons, resonating at δH 3.37 ppm (H-27a) and δH 3.60 ppm (H-27b), which were attached to 205  the same carbon. All of these proton resonances (H-26, H27a and H-27b) showed HMBC correlations into the carbon at δC 150.2 ppm (C-24), establishing the end of the steroidal side chain as shown in Figure 3.49. All of the oxygen atoms not assigned as part of the acetyl group on C-22 were determined to be alcohols in order to account for the remainder of the mass as indicated by HRESIMS. 28  AcO HO  24 23 R  25  27 OH  COSY HMBC  26  Figure 3.49 Structure elucidation of the steroidal side chain of massacreone B (26): key HMBC correlations used to place the C25-C27 group.  The relative configuration of massacreone B was determined using the same arguments invoked in the analysis of massacreone A. Just as with massacreone A, the relative configuration at positions 20, 22 and 25 in the steroidal side chain could not be determined, but biosynthetic precedent from the volume of literature on the ecdysteroids found thus far in marine invertebrates, as well as comparison between the carbon chemical shifts for C-20, C-21 and C-22 reported for various ecdysteroids – including 20-hydroxy-ecdysone (29) – suggested that the orientation at these positions is (20R*, 22R*). As with massacreone A, this assignment is supported by comparing the very similar δc values for massacreone B (25) recorded in methanol, and literature values for 20-hydroxy-ecdysone (29) recorded in methanol129 at C-20 (77.4 ppm and 77.9 ppm), C-21 (21.7 ppm and 21.0 ppm) and C-22 (78.6 ppm and 78.4 ppm). The configuration at position 25 remains ambiguous (Figure 3.50).  206  O OH O  OH  H HO H HO  OH  HO 26  Figure 3.50 Massacreone B (26). The relative configuration at C-25 is unknown.  3.3.3 Structure Elucidation Of Massacreone C Massacreone C (27) was isolated as a clear glass that gave a [M+Na]+ pseudomolecular ion at m/z 517.3151, in its high resolution electrospray ionization time-of-flight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C28H46O7 (C28H46O7Na calculated for m/z 517.3141), indicating six degrees of unsaturation. Analysis of 1H and 13C NMR, as well as 2D COSY, HSQC, HMBC and ROESY spectra, and comparison to the data from massacreone A, allowed the quick identification of massacreone C as being an ecdysteroid, with a strong degree of similarity to massacreone A. Massacreone C differed from massacreone A (25) by having one less oxygen atom, suggesting that perhaps it simply had one less alcohol group in its side chain. The structure of massacreone C (27), plus the numbering scheme used in its structure elucidation, are shown in Figure 3.51 and its 1D and 2D NMR data is summarized in Table 3.3.  207  26  21 OH 18 19 HO HO  11 9  1  3 5 H  17 20  13 7  15  22  27  OH 24 28 H HO  OH  HO 27 Figure 3.51 Massacreone C (27).  Due to difficulties in getting high quality data on massacreone C when dissolved in pyridine (due to having a small amount of sample), massacreone C was subjected to NMR in methanol. Thus, to compare its spectra to those for massacreone A, a data set that was recorded in methanol on massacreone A before it was purified via HPLC was used as a point of comparison. Figure 3.52 below shows the 13C spectrum of massacreone A (top) and of massacreone C (bottom). As compared to massacreone C, massacreone A had an extra carbon resonance at around δC 75 ppm and another around δC 35 ppm. Conversely, massacreone C shows two new carbon resonances at δC 38 ppm and δC 26.3 ppm as compared to massacreone A.  208  Figure 3.52 A comparison of the 13C NMR spectra of massacreone A (top) and massacreone C (bottom). Relevant differences in the spectra are marked with *.  209  OH OH H HO H  HO  OH  HO HO  Figure 3.53 600 MHz 1H NMR spectrum of massacreone C (27), recorded in MeOD-d4.  210  OH OH H HO H  HO  OH  HO HO  Figure 3.54 150 MHz 13C NMR spectrum of massacreone C (27), recorded in MeOD-d4.  211  OH OH H HO H  HO  OH  HO HO  Figure 3.55 600 MHz COSY spectrum of massacreone C (27), recorded in MeOD-d4.  212  OH OH H HO H  HO  OH  HO HO  Figure 3.56 600 MHz HSQC spectrum of massacreone C (27), recorded in MeOD-d4.  213  OH OH H HO H  HO  OH  HO HO  Figure 3.57 600 MHz HMBC spectrum of massacreone C (27), recorded in MeOD-d4.  214  Table 3.4 1D and 2D NMR data for massacreone C (27), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. 26  21 OH 18 19 HO HO  11 9  1  3 5 H  17 20  13 7  15  22  27  OH 24 28 H HO  OH  HO 27  C#  13  C δ (ppm)a  1  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  1  37.5  H-1α 1.78 m  H-1β, H-2  C-2, C-3, C-5, C-6, C-8, C-9, C-10,  H-1β 1.43 t (12.4)  H-1α, H-2  C-2, C-3, C-9, C-10, C-19  2  68.8  3.84 m  H-3, H-1β, H-1α  -  3  68.6  3.95 br s (W ½=8.7)  H-2, H-4β, H-4α  C-1, C-2, C-5  4  21.7  H-4α 1.79 m  H-3, H-4β, H-5  C-2, C-3, C-5, C-10  H-4β 1.70 m  H-3, H-4α, H-5  C-2, C-3, C-5, C-7, C-10  5  53.9  2.32 t (9.1)  H-4β, H-4α  C-3, C-6, C-7, C-9, C-10, C-11, C-19  6  206.6  -  -  -  7  122.2  5.81 d (2.4)  H-9  C-5, C-9, C-14  8  168.3  -  -  -  9  35.2  3.14 t (7.1)  H-7, H-11α, H-11β  C-4, C-7, C-8, C-10, C-19  10  39  -  -  -  11  33  H-11α 1.68 m  H-9, H-11β, H-12α  -  H-11β 1.75 m  H-9, H-11α  -  H-12α 2.11 td (13.1, 4.9)  H-11α, H-12β  C-13, C-16, C-18  H-12α  C-9, C-13, C-14  12  32.6  H-12β 1.86 m a  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  215  Table 3.4 (Continued) 1D and 2D NMR data for massacreone C (27), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. C#  C δ (ppm)a  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  13  48.2  -  -  -  14  85.7  -  -  -  15  31.7  H-15α 1.98 m  H-15β  C-16  H-15β 1.62 m  H-15α  C-13, C-14  H-16α 1.96 m  H-16β  -  H-16β 1.89 m  H-16α, H-17  -  16  13  22.2  1  17  52  2.38 dd (12.8, 4.4)  H-16β  C-13, C-15, C-16, C-18, C-20  18  18.1  0.86 s, 3H  -  C-12, C-13, C-14, C-17  19  24.5  0.97 s, 3H  -  C-1, C-2, C-5, C-9, C-10  20  77  -  -  -  21  26.3  1.27 s, 3H  -  C-17, C-20, C-22  22  38  H-22a 1.50 s  H-22b, H-23  C-17, C-20, C-21, C-23  38  H-22b 1.62 s  H-22a, H-23  C-17, C-20, C-23, C-28  23  29.2  1.52 s, 2H  H-22a, H-22b  C-20, C-21, C-22  24  76.2  -  -  -  25  34  1.88 s  H-26, H-27  C-22, C-24, C-28  26  17.6  0.94 d (4.6), 3H  H-25  C-24, C-25, C-27  27  17.4  0.93 d (4.6), 3H  H-25  C-24, C-25, C-26  28  66.1  H-28a 3.49 d (4)  H-28b  C-23, C-24, C-25  H-28a  C-23, C-24, C-25  H-28b 3.49 d (4) a  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  216  Examination of the COSY, HSQC and HMBC spectra quickly revealed that massacreone C was not oxygenated at position 22, and that the remainder of the molecule was the same as massacreone A. This was easily seen by examining the HMBC correlations from the methyl protons at position 21 (δH 1.28 ppm), which showed HMBC correlations onto the carbon at position 20 (δC 77.0 ppm), into the steroid core at position 17 (δC 52.0 ppm) and finally the carbon at position 22 at δC 38.0 ppm. The HSQC spectrum revealed that the protons attached to position 22 both resonated at δH 1.50 ppm, and showed a COSY correlation to another proton resonance at δH 1.62 ppm (H-23b). This proton at δH 1.62 ppm was also found to be attached to a carbon at δC 29.2 ppm (C-23) along with another proton at δH 1.52 ppm (H-23a). HMBC correlations to C-23 were seen from a proton resonance at δH 1.88 ppm (H-25) and from the proton resonance at δH 3.50 ppm (H-28a and H-28b), as shown in Figure 3.58. OH 21  26 20  22  17 HO  25 OH 28  27  COSY HMBC  Figure 3.58 Structure elucidation of massacreone C (27): key HMBC and COSY correlations in the side chain.  Thus, the structure of massacreone C was established as shown in Figure 3.59, with the relative configuration at position 24 unresolved.  217  OH OH H HO  HO H HO  OH  HO 27  Figure 3.59 Massacreone C (27). The relative configuration at C-24 is unknown.  3.3.4 Structure Elucidation of Massacreone D  OH 26 OH OH 27 22 18 OH 17 20 24 11 28 19 13 9 15 HO 1 OH 3 5 7 H O 28 21  HO HO  Figure 3.60 Massacreone D (28).  Massacreone D (28) was isolated as a clear glass that gave a [M+Na]+ pseudomolecular ion at m/z 549.3025, in its high resolution electrospray ionization time-of-flight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C28H46O9 ( C28H46O9Na calculated for m/z 549.3040), indicating six degrees of unsaturation. Analysis of 1H and 13C NMR, as well as 2D COSY, HSQC, HMBC and ROESY spectra, and comparison to the data from massacreone A (25), allowed the quick identification of massacreone D as being an ecdysteroid, with a strong  218  degree of similarity to massacreone A. Massacreone D differs from massacreone A by having one more oxygen atom, suggesting that perhaps it was oxygenated at one additional position in its side chain.  Taking the same approach as was used to solve the structure of massacreone C, the carbon spectrum of massacreone D was compared alongside that of massacreone A (Figure 3.61). Due to limited amounts of massacreone D, it was not possible to see any new oxygenated carbon resonances in the 13C NMR spectrum of massacreone D (as compared to that of massacreone A). It was possible to note that the carbon assigned as the 25 position of the isopropyl group in the steroid side chain of massacreone A (δC 34.0 ppm) was conspicuously absent in the 13C NMR spectrum of massacreone D. Also noteworthy is the fact that the methyl carbons at around δC 17 ppm in the spectrum of massacreone A seem to have shifted downfield to around δC 25 ppm in the spectrum of massacreone D. A quick glance at the COSY spectrum of massacreone D revealed that there were no correlations between the methyl groups of the isopropyl group and any other proton resonances. This suggested that the additional oxygenation of massacreone D was at position 25.  219  Figure 3.61 A comparison of the 150 MHz 13C NMR spectra of massacreone A (top) and massacreone C (bottom), recorded in MeOD-d4. Relevant differences in the spectra are marked with *.  220  OH OH  OH OH  H HO  HO H  OH  HO HO  Figure 3.62 600 MHz 1H NMR spectrum of massacreone D (28), recorded in MeOD-d4.  221  OH OH  OH OH  H HO  HO H  OH  HO HO  Figure 3.63 150 MHz 13C NMR spectrum of massacreone D (28), recorded in MeOD-d4.  222  OH OH  OH OH  H HO  HO H  OH  HO HO  Figure 3.64 600 MHz COSY spectrum of massacreone D (28), recorded in MeOD-d4.  223  OH OH  OH OH  H HO  HO H  OH  HO HO  Figure 3.65 600 MHz HSQC spectrum of massacreone D (28)), recorded in MeOD-d4  224  OH OH  OH OH  H HO  HO H  OH  HO HO  Figure 3.66 600 MHz HMBC spectrum of massacreone D (28), recorded in MeOD-d4.  225  OH OH  OH OH  H HO  HO H  OH  HO HO  Figure 3.67 600 MHz TROESY spectrum of massacreone D (28), recorded in MeOD-d4.  226  Table 3.5 1D and 2D NMR data for massacreone D (28), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. OH 26 OH OH 27 22 18 OH 20 17 11 24 28 19 13 9 15 HO HO 1 OH 3 5 7 HO H O 28 21  C#  C δ (ppm)a  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  ROESYc (H→H)  1  37.5  H-1β 1.43 m  H-1α, H-2  C-2, C-3, C-9, C-10, C-19  H-1α, H-19  H-1α 1.80 m  H-1β, H-2  C-2, C-3, C-5, C-10  H-1β, H-5  1  2  68.8  3.84 m  H-3, H-1β, H-1α  -  -  3  68.7  3.95 br s (W ½=8.3)  H-2, H-4β  C-1, C-2, C-5  -  4  21.1  H-4β 1.70 m  H-3, H-4α, H-5  -  H-19  H-4α 1.79 m  H-4β, H-5  -  H-19  5  51.9  2.38 m  H-4β, H-4α  C-6  H-1α, H-11β, H-19  6  206.9  -  -  -  -  7  122.2  5.81 s  -  C-5, C-9, C-14  -  8  168.3  -  -  -  -  9  35.2  3.15 m  H-11α, H-11β  -  -  10  39  -  -  -  -  11  33  H-11α 1.68 m  H-9, H-11β, H-12β, H-12α  -  H-18  H-11β 1.75 m  H-9, H-11α, H-12α, H-12β  -  H-5, H-18, H-19  H-12β 1.86 m  H-11α, H-11β, H-12α  -  H-12α  H-12α 2.12 m  H-11α, H-11β, H-12β  -  H-12β  12  a  13  32.6  Recorded at 150 MHz. b Assigned according to HSQC recorded at 600 MHz. c Recorded at 600 MHz.  227  Table 3.5 (Continued) 1D and 2D NMR data for massacreone D (28), recorded at 600 MHz (1H) and 150 MHz (13C) in MeOD-d4. C#  C δ (ppm)a  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  ROESYc (H→H)  13  49  -  -  -  -  14  85.7  -  -  -  -  15  31.7  H-15β 1.62 m  H-15α, H-16β, H-16α  -  H-15α  H-15α 1.98 m  H-15β, H-16β, H-16α  -  H-18  H-16β 1.89 m  H-15β, H-15α, H-16α, H-17  -  -  H-16α 1.96 m  H-15β, H-15α, H-16β, H-17  -  H-18  16  13  1  21.5  17  50.6  2.32 m  H-16β, H-16α  -  -  18  18.2  0.88 d (15.9), 3H  -  C-12, C-13, C-14, C-17  H-11α, H-15α, H-16α, H-21  19  21.7  0.97 s, 3H  -  C-1, C-2, C-5, C-9, C-10  H-1β, H-4β, H-4α, H-5  20  78  -  -  -  -  21  24.6  1.19 s, 3H  -  C-17, C-20, C-22  H-18  22  74.8  3.9 d  H-23a, H-23b  -  -  23  31.9  H-23a 1.62 m  H-22, H-23b  C-25  H-23b  31.9  H-23b 1.97m  H-22, H-23a  C-25  H-23a  24  76.5  -  -  -  -  25  77  -  -  -  -  26  25.1  1.22 s, 3H  -  C-24, C-25, C-27  H-27  27  25.5  1.24 s, 3H  -  C-24, C-25, C-26  H-26  28  66.2  H-28a 3.70 d (10.6)  H-28b  C-25  -  H-28b 3.76 d (10.6)  H-28a  C-25  -  66.2 a  b  c  Recorded at 150 MHz. Assigned according to HSQC recorded at 600 MHz. Recorded at 600 MHz.  228  Examination of the COSY, HSQC and HMBC spectra quickly revealed that massacreone D was indeed oxygenated at position 25, and that the remainder of the molecule was the same as massacreone A.  H-21 ( δH 1.24 ppm) displayed HMBC correlations to C-20 (δC 78.0 ppm), C-17 (δC 50.6 ppm) and C-22 (δC 75.0 ppm). A proton resonance at δH 3.89 ppm (H-22), was attached to the carbon resonating at δC 75.0 ppm (C-22) and showed COSY correlations to the two proton resonances at δH 1.43 ppm (H-23a) and δH 1.89 ppm (H-23b). These two proton resonances (H23a and H-23b) showed HMBC correlations to a carbon resonance at δC 68.6 ppm (C-28) that had two attached protons resonating at δH 3.68 ppm (H-28a) and δH 3.76 ppm (H-28b). H-28a and H-28b both showed HMBC correlations to two carbons at very similar chemical shifts – δC 76.5 ppm and δC 77.0 ppm (C-24 and C-25). Proton resonances at δH 1.23 ppm (H-26) and δH 1.25 ppm (H-27) were assigned as being two methyl groups attached to carbons at δC 24.6 ppm (C-26) and δC 25.5 ppm (C-27). These two proton resonances (H-26 and H-27) showed HMBC correlations to each others' carbons, as well as to the two similar oxygenated carbon resonances at δC 76.5 ppm (C-24 or C-25) and δC 77.0 ppm (C-24 or C-25). This data is summarized in Figure 3.68. 21 20  OH OH 26 OH  22 23 24 17  25 27 OH  COSY HMBC  HO 28 Figure 3.68 Structure elucidation of massacreone D (28): key HMBC correlations in the side chain.  The relative configuration of the side chain was assigned as 20R* and 22R*, in keeping 229  with the rest of the ecdysteroids found in this organism, with the configuration at C-24 unresolved (Figure 3.69). OH  OHOH  OH H HO  HO H HO  OH  HO 28  Figure 3.69 Massacreone D (28). The relative configuration at C-24 is unknown.  3.3.5 Structure Elucidation Of 20-Hydroxy-Ecdysone OHOH  OH  H HO H HO  OH  HO 29  Figure 3.70 20-hydroxy-ecdysone (29).  20-Hydroxy-ecdysone (29) was isolated as a clear glass that gave a [M+Na]+ pseudomolecular ion at m/z 503.2997, in its high resolution electrospray ionization time-of-flight mass spectrum (HRESI-TOFMS), corresponding to a molecular formula of C27H44O7  230  ( C27H44O7Na calculated for m/z 503.2985), indicating six degrees of unsaturation. Due to the fact that this compound was isolated along side other ecdysteroids, a quick literature search revealed a common ecdysteroid with the formula C27H44O7 whose data seemed to match that of the isolated compound. A side by side analysis of the 13C and 1H resonances in MeOD- d4 of the isolated substance with the values for 20-hydroxy-ecdysone as reported in the “Ecdysone Handbook”129 showed that it was an exact match for 20-hydroxy-ecdysone, and was thus identified as such.  3.4 Proposed Biosynthesis of the Massacreones Ecdysteroids are well known as sex and moulting hormones in both insects and crustaceans. As of November 2009, the ecdysone handbook (the Ecdybase) – the de facto catalog of all known ecdysteroids – contained over 400 different examples of ecdysteroids from plants, microorganisms, and marine invertebrates.129 Numerous reports of ecdysteroids isolated from zooanthids130-135 and other cniderians136,137 have been reported with highly oxygenated side chains (Figure 3.71).138  231  OH OH HO  H HO  H  HO HO  OH OH  OH OH  H  OH HO  H  H  OH  HO OH  HO  H  HO  HO  HO  30  31  32 OH OH  OH OH  OH  OH HO  HO HO  H  H  HO  H  OH  H  HO  O 33  OH  HO 34  OH OH  OH  H HO HO  H HO  OH  OH  OH HO HO  OH  H  OH H  OH  HO 36  35  Figure 3.71 Ecdysteroids from cnidarians: ajugasterone C (30)134, gerardiasterone (31)132, 2deoxyecdysterone (32)133, 4-dehydroecdysterone (33)130, palythoalones A (34) and (35)135, zoanthusterone (36)131.  Numerous studies have been performed on the biosynthesis of ecdysteroids – namely, ecdysone (37) and 20-hydroxy ecdysone (29) – in insects139, plants140, and crustaceans141. However, in coelenterates and sponges, attempts to deduce the biosynthetic pathways leading to ecdysteroids have so far been unsuccessful or have not been attempted142. Currently, the most reasonable explanation for the presence of ecdysteroids in such marine invertebrates is dietary 232  origin, with the possibility of a few endogenous metabolic transformations. In insects, as well as in some crustaceans, ecdysteroids are formed from cholesterol (38). In plants, including algae, they can alternatively be formed from lathosterol (39). In either case, ecdysone (37) is a commonly seen product from these biosynthetic pathways, as well as 20-hydroxy-ecdysone (29) (Figure 3.72).  H  H  H  HO  H  H  H  HO  38  H Plants  OH  Insects Crustaceans  39 OH OH OH  OH  H  H HO  HO H HO HO 37  OH HO  H  OH  HO 29  Figure 3.72 Biosynthesis of ecdysone (37) in insects and crustaceans (via cholesterol, (38)), or in plants (via lathosterol (39)). An additional oxidation step at C-20 yields 20-hydroxy-ecdysone (29).  Thus, it is reasonable to imagine either of these ecdysteroids being sequestered from a food source, or formed from an available precursor, in the Dominican cnidarian, RJA06-08. With 20-hydroxy-ecdysone available, a series of endogenous biosynthetic transformations could then lead to the different side chain oxygenation and methylation patterns seen in the  233  massacreones. However, without any evidence for or against this theory, it is also equally plausible to imagine that the massacreones are sequestered in their reported form from some available food source, or from a symbiotic microorganism. Possible biosynthetic transformations from 20-hydroxy-ecdysone to massacreones A-D are outlined in Figure 3.73. OH  OHOH  OHOH  H H HO OH  H  HO  Dehydration C-25 Methylation at C-24 by SAM Rearrangement  HO  HO HO  OH  H HO  29  4 C2 28 t na C tio n at a id io Ox xidat O  OHOH  Oxidation at C26 Acetylation at C-22 O  OH H HO HO  H  OHO  HO H  OH  HO  H O 25 Ox ida tio na tC 25  Dehydration and Reduction at C22  HO  H O  OHOH  H HO H O 27  HO  26 OH OH  OH H  OH  H  OH  HO  OH  H HO  HO  OH  OH  H HO HO  28  Figure 3.73 Proposed biosynthetic transformations from 20-hydroxy-ecdysone (29) to the massacreones.  234  Interestingly, in a number of species of sponges and cnidarians, as well as one notable marine arachnid, very high concentrations of ecdysteroids have been found, which is not to be expected for compounds having a hormonal role.142 In the case of the marine arachnid Pycnogonum littorale143, it was clearly demonstrated that ecdysteroids are stored within the epidermis and released when the animal is attacked by a crab, resulting in an efficient deterrent/antifeedant effect.144 This is alluringly analogous to the role of ecdysteroids seen in plants (protection against insects)145 and also in some insects against predators such as spiders146, and it is tempting to postulate that the ecdysteroids in the Dominican cnidarian RJA06-08 thus perform a defensive role against some predator – however, without any evidence for or against this theory, it is impossible to say whether this is true or not.  3.5 Biological Activity While the fraction that was identified as predominantly girolline showed promising inhibition of the TRL5 activation in the presence of flagellin, the synthesized analogues based on the girolline carbon skeleton (16, 17, 18, 19, 22) also did not show any activity. Although the exact requirements for activity are not completely understood, it appears that the alcohol and chlorine groups are required, because compound 22, which differed from girolline by lacking these two groups, was inactive.  The massacreones and 20-hydroxy-ecdysone were isolated by following the TRL5 bioassay, but the activity of the purified compounds was not very potent and was only seen at concentrations where cytotoxicity was also observed. Also, only massacreones A and B demonstrated any biological activity – however, the smaller amounts of massacreone C, 235  massacreone D and 20-hydroxy-ecdysone did not allow for any extended testing in the TLR5 bioassay. As shown in Figure 3.74 below, massacreone A did not show any suppression of TLR5 activation in the presence of flagellin (as compared to the control, which is +DMSO and +DMSO+flag), except at concentrations high enough to also cause cell death (at 0.4 ng/μL), while for massacreone B, some activity was present without causing much cell death (at 0.08 ng/μL). However, at 0.4 ng/μL, massacreone B is toxic to the cells, and at 0.016 ng/μL, it does not show any activity.  Figure 3.74 Results of the TLR5 bioassay for massacreone A (left) and massacreone B (right).  Thus, the potential therapeutic window for these compounds in suppressing the activation of TLR5 in the presence of flagellin is far too narrow, and these compounds are not potential leads for pharmaceutical development. Massacreone C, massacreone D and 20-hydroxyecdysone did not show any activity.  236  3.6 Discussion and Conclusions The cycle of blockage-infection-inflammation in cystic fibrosis patients presents a complicated challenge in the search for potential pharmaceuticals to help manage the disease. Recent work done in the Turvey lab has shown that bacterial flagellin from infections in the cystic fibrosis lung activates TLR5, initiating an inflammation response. Thus, inhibition of TLR5's activation in the presence of bacterial flagellin presents an alluring route for new pharmaceuticals to aid in the management of the disease.  Girolline, an alkaloid from the marine sponge 47643, was identified in a fraction that showed potent activity in the TLR5 bioassay. Synthetic analogues of girolline showed no activity. The source of the potent activity seen in the active fraction is still under investigation.  Massacreones A-D were isolated along with 20-hydroxy-ecdysone. These new, polyoxygenated ecdysteroids add to a growing body of ecdysteroids isolated from zoanthids. Ecdysteroids are well known as insect moulting hormones; acting as moulting hormones to some crustaceans as well. They were mildly active in the TLR5 bioassay, with a very small window between observed activity and cytotoxicity.  237  3.7 Experimental 3.7.1 General Experimental Procedures 1  H, 13C, COSY, HSQC, HMBC, TOCSY and ROESY spectra were recorded on a Bruker  AV600 NMR spectrophotometer equipped with a cryoprobe. 1H chemical shifts were referenced to the residual MeOD-d4 signal (δH 3.31 ppm), and 13C chemical shifts were referenced to the MeOD-d4 solvent peak (δC 49.15 ppm). All NMR solvents were obtained from Cambridge Isotope Laboratories. All NMR data was processed using Bruker XWINNMR® software.  All chromatography was performed using HPLC grade solvents from Fisher Scientific with no additional purification. Water was purified by use of a Millipore MQ filter system. HPLC separations were performed using a Waters 600 pump and a Waters PDA900 detector, using an Inertsil C18 column, 9.4 x 250 mm, flow 2 mL/min. All solvents were filtered prior to use, then sparged with helium. Optical rotations were determined by using a JASCO J-1010 polarimeter equipped with a halogen lamp (589 nm) and a 10 mm micro cell. UV spectra were recorded on a Waters 2487 spectrophotometer. ESI mass spectra were recorded using a Micromass LCT mass spectrometer. Thin layer chromatography was performed using Merck KggA 60F254 plates with TLC visualization performed by UV at 254 nm, or by spraying a solution of 4% p-anisaldehyde, 2.5% acetic acid and 2.5% H2SO4 in ethanol and heating to yield colored spots.  All non-aqueous reactions were carried out in flame dried glassware under a nitrogen atmosphere unless otherwise noted. Air and moisture sensitive liquid reagents were manipulated  238  via a dry syringe. Anhydrous tetrahydrofuran was obtained via distillation over sodium. All other solvents and reagents were obtained from commercial sources without further purification. Flash chromatography was performed on Silicycle Ultra Pure silica gel (230-400 mesh) using solvent conditions indicated in each procedure.  3.7.2 Isolation of Girolline The sponge #47643 was an orange rope sponge with branching fingers that was collected by hand using SCUBA on August 14, 1993, at Mont Pass in Pohnpei, at coordinates 6o 52 N, 158o 18 E, from a depth of -12 m, and frozen at -20oC until extracted. It was was immersed in methanol (300 mL) and extracted for twenty four hours. This was repeated two more times, and then a further two times with dichloromethane. These combined extracts were concentrated in vacuo (yielding 4.99 g crude extract), then resuspended in water. This aqueous suspension was partitioned against ethyl acetate (twice), and than against butanol (twice) to yield three fractions based on solubility. Bioassay guided fractionation indicated that the activity was in the aqueous fraction (1.98 g).  This active aqueous fraction was then subjected to size exclusion chromatography in a column containing Sephadex LH-20 preswollen in methanol (400 mg applied each column, five columns done in total with fractions grouped based on TLC profiles), resulting in three subfractions. The first of these LH-20 fractions was found to display the biological activity. By TLC this fraction contained mainly very polar material that did not move off of the TLC baseline (normal phase silica, developed in 4:2:1 butanol : water : acetic acid).  This active fraction (379 mg) was then suspended in 10 mL methanol and adsorbed onto 239  2 g C18 silica. This adsorbed material was then applied to a Waters C18 Sep-pak (10 g) eluted with 500 mL water, then 500 mL 50% aqueous methanol, and finally 500 mL 1:1 methanol:dichloromethane. The fraction eluting with water (82 mg) was then subjected to reversed phase HPLC using a gradient of water to 40% aqueous acetonitrile over 40 minutes (2 mL/min). The majority of the material eluted in the first ten minutes. This HPLC run yielded five fractions, with fractions one to four representing a peak or peaks, the most massive of which was peak cluster number two (fraction name: 47643.3.1.1.2, 31.5 mg). This fraction also turned out to display the biological activity. The active fraction was subjected to a second round of HPLC on C18, run with water as an eluent, injecting 1 mg per run. This HPLC run was monitored at 210 nm and 254 nm, with a major peak eluting at 6.5 minutes with a strong absorbance at 210 nm, and negligible absorbance at 254 nm. A number of peaks with more substantial absorbance at 254 nm were collected as well for a total of seven fractions. Bioassay results showed that the first fraction – the substantial peak at 210 nm, was the active fraction. This fraction, when developed on normal phase TLC using 4:2:1 butanol : water : acetic acid, showed a spot at Rf = 0.1 that charred yellow with a p-anisaldehyde stain.  3.7.3 Girolline Physical Data Yellow/Brown oil, [a]D20 = 9.1º (c=0.02, MeOH), UV λmax MeOH nm: 214.1. HRESIMS:[M+H+] m/z 191.0611 (calculated 191.0700 for C6H12ON4Cl). For 1D and 2D NMR data please refer to Table 3.1.  240  3.7.4 Preparation of Acylated Oxazolidinone 2 O  O N  Cl  O  2  R-(+)-4-Isopropyl-2-oxazolidinone (200 mg, 1.55 mmol) was dissolved in 5 mL of dry THF, and the solution was cooled to -78oC. n-BuLi (1 mL, 1.6 M, 1.6 mmol) was added over a period of 5 minutes, followed by the addition of chloroacetyl chloride (0.12 mL, 1.71 mmol). The solution was stirred for 30 minutes, then allowed to warm to room temperature over an additional 30 minutes. The solution was then quenched with approximately 5 mL of saturated NH4Cl(aq) solution. The crude reaction mixture was extracted into dichloromethane and washed with 3x H2O. The organic phase was dried over Na2SO4, filtered and concentrated. The crude product was purified by column chromatography (20% thyl acetate in hexanes) to yield 2 (232 mg, 1.13 mmol, 73%) as a pale yellow oil. 1H NMR (400 MHz, CDCl3) δ 0.89 (d, J=7.0 Hz, 3H), 0.93 (d, J=7.0 Hz, 3H), 2.40 (doublet of septets, J1=3.8 Hz, J2=7.0 Hz, 1H), 2.54 (s, 2H), 4.23 (td, J1=3.2 Hz, J2=8.7 Hz, 1H), 4.27 (t, J=8.7 Hz, 1H), 4.43 (dt, J1=3.6 Hz, J2=8.7 Hz, 1H); 13CNMR (100 MHz, CDCl3) δ 14.8, 18.1, 23.9, 28.5, 58.5, 63.4, 154.4, 170.4; ESIMS [M+Na]+ calcd for C8H12ClNO3Na 228.0403, found 228.0401.  241  3.7.5 Preparation of Protected Imidazole 3 N N O  3  Imidazole (2.21g, 32.4 mmol) and K2CO3 (4.48g, 32.4 mmol) were combined. Dry DMF (60 mL) was added and the solution stirred untill homogeneous as a fine suspension. BOMCl (4.50 mL, 32.35 mmol) was added slowly over 5 minutes. The solution was stirred for 22 hours, and then quenched with 30 mL H2O to destroy any residual BOMCl. Both the organic and aqueous phases were concentrated, with the resulting residue then resuspended into ethyl acetate and then washed with 3x H2O. The organic phase was dried over Na2SO4, filtered and concentrated. The crude product was purified by column chromatography (10% methanol in ethyl acetate) to yield 3 (4.83g, 25.6 mmol, 79%) as a colorless oil. 1H NMR (400 MHz, CD2Cl2) δ 4.36 (s, 2H), 5.25 (s, 2H), 6.99 (s, 1H), 7.01 (s, 1H), 7.26 (m, 5H), 7.55 (s, 1H); 13CNMR (100 MHz, CD2Cl2) δ 69.6, 74.5, 118.5, 127.5, 127.8, 128.2, 129.7, 136.0, 137.1; ESIMS [M+Na]+ calcd for C11H12N2ONa 211.0847, found 211.0841.  242  3.7.6 Preparation of Aldehyde 8 N Cl  H N  O  O  8  BOM protected imidazole 3 (435 mg, 2.31 mmol) was dissolved in 10 mL of dry THF, and the solution was cooled to -78oC. n-BuLi (1.44 mL, 1.6M, 2.31 mmol) was added over a period of 10 minutes, and the mixture was stirred for 10 minutes. In another flask, hexachloroethane (546.7 mg, 2.31 mmol) was dissolved into 2 mL dry THF, with the resulting solution then added, drop-wise, to the original flask. The flask that contained the hexachloroethane was rinsed with an additional 2 mL dry THF, with the rinse then added to the reaction mixture, which was then allowed to stir for 1hr. n-BuLi (1.44 mL, 1.6 M, 2.31 mmol) was then added over a period of 10 minutes, followed by the addition of DMF (0.72 mL, 9.34 mmol). The solution was stirred for 30 minutes, then allowed to warm to room temperature over an additional 30 minutes. The solution was then quenched with approximately 10 mL of saturated NH4Cl(aq) solution. The crude reaction mixture was extracted into dichloromethane and washed with 3x H2O. The organic phase was dried over Na2SO4, filtered and concentrated. The crude product was purified by column chromatography (40% ethyl acetate in hexanes) to yield 8 (317.8 mg, 1.27 mmol, 55%) as a colorless solid. 1H NMR (400 MHz, CD2Cl2) δ 4.61 (s, 2H), 5.83 (s, 2H), 7.31 (m, 5H), 7.67 (s, 1H), 9.67 (s, 1H); 13CNMR (100 MHz, CD2Cl2) δ 70.9, 73.7, 127.1, 127.5, 127.9, 132.0, 136.5, 140.5, 141.8, 178.0; ESIMS [M+Na]+ calcd for C12H11ClN2O2Na 273.0407, found 273.0417. 243  3.7.7 Preparation of Amide 9 O  OH O N Cl  N  O  N Cl BOM 9  Acylated oxazolidinone 2 (751 mg, 3.70 mmol) was suspended under Ar in 30 mL of dry dicholoromethane, and put on ice to cool to 2oC. Dibutyl boron triflate (Bu2BOTf, 4.44 mL, 1.0 M, 4.44 mmol) was added slowly over 5 minutes, followed by the slow addition (over 5 minutes) of dry triethylamine (0.68 mL, 4.88 mmol). The solution was then allowed to stir for 10 minutes. In a second flask, the aldehyde 8 (1.017g, 4.07 mmol) was suspended under Ar in 5 mL of dry dichloromethane, and the resulting solution was then added drop-wise over 10 minutes into the reaction mixture. The flask that contained the aldehyde was rinsed with 2 mL dry dichloromethane, with this rinse then added drop-wise over 4 minutes into the reaction mixture. The chilled flask was then closed up using a small piece of duct tape over the needle holes in the septum and sealed using parafilm, then placed in a 0oC freezer for 18 hours. The reaction mixture was then placed in an ice bath, and quenched with 1 mL pH 7 phosphate buffer, 1 mL MeOH, and 2 mL of a 30%H2O2 solution in methanol, then left to stir for one hour on ice. The crude reaction mixture was extracted into dichloromethane and washed with 3x H2O. The organic phase was dried over Na2SO4, filtered and concentrated. The reaction mixture was a mixture of unreacted starting material 8 and product 9. The crude product was purified by column chromatography (40% ethyl acetate in hexanes) to yield 9 as a white solid (852.5 mg, 1.87 mmol, 51%). Approximately 30% (345 mg, 1.36 mmol) of the Aldehyde 8 starting material was unreacted and was thus recovered. 1H NMR (400 MHz, CD3COCD3) δ 0.86 (d, J=7.1 Hz,  244  6H), 2.30 (d septet, J1=3.9 Hz, J2=7.0 Hz, 1H), 4.20 (m, 2H), 4.29 (m, 1H), 4.53 (d, J=4.5 Hz, 2H), 5.24 (d, J=8.3 Hz, 1H), 5.47 (d, J=4.5 Hz, 2H), 6.02 (d, J=8.3 Hz, 1H), 6.91 (s, 1H), 7.30 (m, 5H); 13CNMR (100 MHz, CD3COCD3) δ 16.8, 19.7, 30.7, 59.7, 61.0, 66.5, 67.6, 73.0, 76.0, 129.2, 130.6, 131.1, 135.9, 140.0, 156.1, 170.0; ESIMS [M+Na]+ calcd for C20H23Cl2N3O5Na 478.0912, found 478.0909.  3.7.8 Preparation of Amide 10 OH O N Cl  N Cl BOM  NH2  10 The amide 9 (412 mg, 0.91 mmol) was dissolved in 4 mL of 1:1 THF:H2O and cooled to 2oC. Aqueous ammonia (120.5µL, 14.8M, 1.80 mmol) was added, and the reaction was left stirring for 3 hours. Na2SO3 (450 mg, 3.6 mmol) was added to quench the reaction and then the reaction mixture was concentrated. After resuspending in ethyl acetate, the crude reaction mixture was washed with 3x H2O and concentrated, then purified by column chromatography (5% methanol in ethyl acetate) to yield 10 as a colorless glass (45 mg, 0.13 mmol, 14%).  1  H NMR (400 MHz,  CD3OD) δ 4.59 (d, J=4.7 Hz, 2H), 4.70 (d, J=8.1 Hz, 1H), 5.29 (d, J=8.1 Hz, 1H), 5.56 (d, J=4.7 Hz, 2H), 7.03 (s, 1H), 7.34 (m, 5H); 13CNMR (100 MHz, CD2Cl2) 60.8, 65.2, 70.1,72.7, 125.8, 127.2, 127.3, 127.7, 132.9, 136.3. 170.2 ; ESIMS [M+Na]+ calcd for C14H15Cl2N3O3Na 366.0388, found 366.0395.  245  3.7.9 Preparation of Alcohol 15 OH N Cl  N BOM 15  The aldehyde 8 (100 mg, 0.40 mmol) was dissolved in 4 mL of dry THF, and the solution was cooled to -78oC. n-BuLi (0.3 mL, 1.6M, 0.48 mmol) was added over a period of 5 minutes. The solution was stirred for 30 minutes, then allowed to warm to room temperature over an additional 30 minutes. The solution was then quenched with approximately 0.3 mL of saturated NH4Cl(aq) solution. The crude reaction mixture was extracted into dichloromethane and washed with 3x H2O. The organic phase was dried over Na2SO4, filtered and concentrated. The crude product was purified by column chromatography (40% ethyl acetate in hexanes) to yield 15 (43.2 mg, 0.14 mmol, 35%) as a white solid. Aldehyde 8 was also recovered (27.5 mg, 0.11 mmol).  1  H  NMR (400 MHz, CD2Cl2) δ 0.91 (t, J=6.8 Hz, 3H), 1.36 (m, 3H), 1.49 (m, 1H), 1.89 (q, J=7.5 Hz, 2H), 4.55 (s, 2H), 4.69 (t, J=6.8 Hz, 1H), 5.44 (d, J=11.0 Hz, 1H), 5.57 (d, J=11.0 Hz, 1H), 6.87 (s, 1H), 7.31 (m, 5H); 13CNMR (100 MHz, CD2Cl2) 14.3, 23.0, 28.6, 35.4, 65.1, 71.1, 74.2, 125.7, 128.2, 128.5, 128.9, 133.3, 137.3, 137.5; ESIMS [M+Na]+ calcd for C16H21ClN2O2Na 331.1189, found 331.1179.  246  3.7.10 Preparation of Amide 16 OH O N Cl  NH2  NH  Cl 16  Amide 10 (65.2 mg, 0.19 mmol) was dissolved in 2 mL dichloromethane. AlCl3 (255 mg, 1.91 mmol) was added and the reaction stirred at room temperature for 60 minutes. The reaction mixture was quenched with 0.5 mL H2O, and the organic layer separated off and washed with 3 x H2O, then filtered through a cake of celite and concentrated. The crude product was purified by column chromatography (ethyl acetate, then 50% methanol in ethyl acetate) to yield the yellow solid 16 (38.5 mg, 0.17 mmol, 89%) as a mixture of two diastereomers. 1H NMR (400 MHz, D2O) δ 4.47 (s, 1H), 4.56 (s, 1H), 4.67 (s, 1H), 4.77 (s, 1H), 7.42 (s, 1H), 7.51 (s, 1H); 13CNMR (100 MHz, D2O) 49.2, 49.9, 70.0, 70.1, 118.2, 130.5, 130.7, 131.8, 173.2, 173.3; ESIMS [M+Na]+ calcd for C6H7Cl2N3O2Na 245.9813, found 245.9820.  3.7.11 Preparation of Amide 17 OH O N NH  NH2 Cl 17  Amide 10 (55.2 mg, 0.16 mmol) and 10% Pd/C (4 mg, 0.1g per mmol starting material) were placed in a flask that was flushed with, then maintained at, 1atm H2. 2 mL Ethanol was added, and the reaction mixture was left to stir at room temperature for 24 hours. The reaction mixture was concentrated, then purified by column chromatography (ethyl acetate, then 20% methanol in  247  ethyl acetate) to yield 17 (28.8 mg, 0.15 mmol, 94%) as a colorless glass. 1H NMR (400 MHz, D2O) δ 4.87 (m, 1H), 5.42 (d, J=4.2 Hz, 1H), 7.41 (s, 1H), 8.21(s, 1H); 13CNMR (100 MHz, D2O) 60.4, 65.8, 113.4, 115.4, 133.0, 169.8; ESIMS [M+Na]+ calcd for C6H8ClN3O2Na 212.0203, found 212.0195.  3.7.12 Preparation of Alcohol 18 OH N Cl  NH 18  Alcohol 15 (40 mg, 0.13 mmol) was dissolved in 2 mL dichloromethane. AlCl3 (174 mg, 1.30 mmol) was added and the reaction mixture stirred at room temperature for 60 minutes. The reaction mixture was quenched with 0.5 mL H2O, and the whole mixture concentrated. The crude product was purified by column chromatography (25% methanol in ethyl acetate) to yield 18 (24.4 mg, 0.13 mmol, 100%) as a white solid.  1  H NMR (400 MHz, CD3OD) δ 0.93 (t, J=7.0  Hz, 3H), 1.37 (m, 5H), 1.79 (q, J=6.7 Hz, 2H), 4.72 (t, J=6.5 Hz, 1H), 7.42 (s, 1H); 13CNMR (100 MHz, CD3OD) 13.9, 23.0, 28.5, 34.6, 75.5, 119.0, 131.7, 137.0; ESIMS [M+Na]+ calcd for C8H13ClN2NaO 211.0614, found 211.0611.  248  3.7.13 Preparation of Alcohol 19 OH N NH  19  Alcohol 15 (12.3 mg, 0.04 mmol) and 10% Pd/C (4 mg, 0.1g per mmol starting material) were placed in a flask that was flushed with, then maintained at, 1atm H2. 2 mL ethanol was added, and the reaction mixture was left to stir at room temperature for 24 hours. The reaction mixture was concentrated, then purified by column chromatography (ethyl acetate, then 20% methanol in ethyl acetate) to yield 19 (4.7 mg, 0.03 mmol, 75%) as a colorless glass.  1  H NMR (400 MHz,  CD3OD) δ 0.91 (t, J=7.0 Hz, 3H), 1.21-1.41 (m, 5H), 1.81 (m, 2H), 4.65 (t, J=6.9 Hz, 1H), 6.94 (s, 1H), 7.42 (s, 1H); 13CNMR (100 MHz, CD3OD) 14.2, 22.9, 28.3, 36.8, 68.2, 117.0, 136.3, 141.3; ESIMS [M+Na]+ calcd for C8H14N2ONa 177.1004, found 177.1011.  3.7.14 Preparation of Amino Imidazole 22 N H 2N  NH  NH2 22  D/L-Ornithine methyl ester 21 (947.8 mg, 4.35 mmol) was dissolved into 100 mL water and cooled down to 2oC. The pH was adjusted to 1.5-2.0 by addition of 15% HCl. Over the course of 1 h, 10% Na/Hg (1.0 g, 43.5 mmol) was added to the solution while the temperature and pH within were maintained in the given range. When the pH remained constant and the evolution of gas had ceased, the solution was decanted to remove Hg. The pH was then adjusted to 4.3 by the addition of 1 N NaOH. Cyanamide (0.9130 mg, 21.7 mmol) was added and the solution heated at 95 °C for 2.5 hr. Removal of the solvent in vacuo afforded a thick, light yellow residue that 249  was washed with ether 3 x 40 mL. Methanol was added to the residue and NaCl removed by filtration. The filtrate, after evaporation, gave a pale yellow oil that was purified by recrystallization from ethanol to give 22 as colorless needles (347.4 mg, 2.48 mmol, 57%). 1H NMR (400 MHz, CD3OD) δ 1.98 (p, J=7.5 Hz, 2H), 2.63 (t, J=7.6 Hz, 2H), 2.96 (t, J=7.9 Hz, 2H), 6.61 (s, 1H); 13CNMR (100 MHz, CD3OD) δ 21.2, 25.9, 38.6, 109.1, 125.9, 147.5; ESIMS [M+Na]+ calcd for C6H12N4Na 163.0960, found 163.0969  3.7.15 Isolation of Massacreones A-D For a general isolation scheme see section 3.3. An unidentified cnidarian, was harvested by hand using SCUBA off of a reef in Dominica. Upon collection, the organism was drained of sea water and packaged in a plastic bag, and then frozen. Whole, frozen 06-08 was completely submerged in methanol and extracted for twenty four hours. This was repeated two more times, and the combined extracts were concentrated in vacuo, then resuspended in water. This aqueous suspension (250 mL) was partitioned against ethyl acetate (2x150 mL), with the first ethyl acetate layer back extracted with 100 mL additional water. The aqueous layer (now 350 mL) was then partitioned against butanol (2x250 mL) to yield three fractions.  The butanolic fraction (614.6 mg) from the solvent partitions was subjected to size exclusion chromatography using LH-20 preswollen in methanol (200 mg at a time, three columns, column size 200 cm x 2 cm). LH-20 column fractions were grouped together into two fractions based on color, thin layer chromatography profiles, and charring colors. The first fraction (148.1 mg) showed activity in the bioassay, and was thus subjected to reversed phase HPLC using a step gradient starting at 10% aqueous acetonitrile for 10 minutes, then 30% aqueous acetonitrile for another 50 minutes. From this HPLC run, fourteen fractions were 250  collected, including five that were major peaks. These five peaks, which eluted at 24.5 minutes, 27.1 minutes, 34.3 minutes, 39.5 minutes, and 43.4 minutes, showed similar 1H spectra and appeared to be terpenoid in origin.  The peak at 24.5 minutes was purified using reversed phase HPLC with 15% acetonitrile to yield massacreone D as a clear glass (4 mL/min Rt = 15 min, 0.7 mg). The peak at 27.1 minutes was purified on reversed phase HPLC using 25% acetonitrile to yield the known 20hydroxy-ecdysone as a clear glass (Rt = 3.9 min, 1.1 mg). The peak at 34.3 minutes was purified on reversed phase HPLC using a gradient from 25%-45% acetonitrile over 30 minutes to yield massacreone A as a white powder (Rt = 15.6 minutes, 37.7 mg). The peak at 39.5 minutes was purified on reversed phase HPLC using a gradient from 25%-45% acetonitrile over 30 minutes to yield massacreone B as a clear glass (Rt = 18.4 min, 7.5 mg). The peak at 43.4 minutes was identified as massacreone C (1.6 mg).  3.7.16 Massacreone A (25) Physical Data White Powder, [a]D20 = 42.8º (c=0.5, MeOH), UV λmax MeOH nm: 242.2. HRESIMS: [M+Na]+ m/z 533.3104 (calculated 533.3090 for C28H46O8Na). For 1D and 2D NMR data please refer to Table 3.2.  3.7.17 Massacreone B (26) Physical Data Clear Glass, [a]D20 = 61.1º (c=0.37, MeOH), UV λmax MeOH nm: 242.0. HRESIMS: [M+Na]+ m/z 557.3085 (calculated 557.3090 for C30H46O8Na). For 1D and 2D NMR data please refer to Table 3.3.  251  3.7.18 Massacreone C (27) Physical Data Clear glass, [a]D20 = 60.0º (c=0.16, MeOH), UV λmax MeOH nm: 245.3. HRESIMS: [M+Na]+ m/z 517.3151 (calculated 517.3141 for C28H46O7Na). For 1D and 2D NMR data please refer to Table 3.4  3.7.19 Massacreone D (28) Physical Data Clear Glass, [a]D20 = 68.6º (c=0.07, MeOH), UV λmax MeOH nm: 243.5. HRESIMS: [M+Na]+ m/z 549.3025 (calculated 549.3040 for C28H46O9Na). For 1D and 2D NMR data please refer to Table  3.7.20 20-Hydroxy-Ecdysone (29) Physical Data Clear Glass, [a]D20 = 66.4º (c=0.11, MeOH), UV λmax MeOH nm: 247.3. HRESIMS: [M+Na]+ m/z 503.2997 (calculated 503.2985 for C27H44O7Na). Spectral data was in accord with that previously reported for the natural product.  252  4. Isolation of a Potential Inhibitor of IDO from Caulerpa sp. 4.1 Brief Introduction to Algal Metabolites Over 3000 secondary metabolites have been described from marine red, brown and green algae, representing approximately 20% of all reported marine natural products. In the 1960s, a significant amount of attention was cast onto metabolites from macro algae – they comprised over 50% of reported marine natural products during that time.147 Now the number hovers closer to 10% annually.27 In contrast, since 2000, natural products from microalgae and cyanobacteria have dominated the literature, comprising 50% of the reported marine natural products. Despite this shift away from macroalgal metabolites, there is still high interest in several macroalgal natural products for pharmaceutical development (Figure 4.1).  Tropical macroalgae make a greater diversity of structurally unique and biologically active natural products than do their temperate counterparts.147,27 This may be due to a greater variety of hrbavores and competing organisms. Like all plants, the majority of macroalgal compounds are terpenoids (especially sesqui- and di-terpenoids), acetogenins (acetate-derived compounds), amino acid derivatives (including alkaloids), and polyphenols. Algal secondary metabolites tend to contain nitrogen less often, and halogen atoms more often, as compared to terrestrial plant natural products . This is likely due to the relative availabilities of these elements in the marine and terrestrial environments. Many defense functions, such as antifoulant activity, have been reported for algal metabolites.147  253  NH2 O O NH N  O  O N HN  O  O  N H HN O N H  O  R1  O  Br HN NH O NH HN O  NH  O  O  O  R2 O  R3 R1= H, OH, OAc R2 = H, Cl, Br or I R3 = H, Cl, Br or I  NH O  1  2: R1 = R3 = H, R2 = Br  Figure 4.1 Natural products from macro algae: kahalide F (1), a potent antitumor compound, and frimbrolide I (2), one of 17 known frimbrolides, which are antifouling compounds.  4.2 Brief Introduction to Indole Alkaloids The indole alkaloids are a large and complex group of natural products derived from the amino acid tryptophan (3), containing an indole or dihydroindole nucleus. Tryptophan is derived from the shikimate pathway, and diverges from the biosynthesis of L-tyrosine and Lphenylalanine at the common precursor, chorismic acid (4). As shown in Figure 4.2, chorismic acid (4) is converted to anthranilic acid (5), which then couples with phosphoribosyl phosphate and undergoes a series of rearrangements to form the indole nucleus in indole-3-glycerol phosphate (6), which is then transformed, via release of glyceraldehyde-3-phosphate followed by  254  the incorporation of L-serine, into L-tryptophan (3).1,51,52 CO2H  CO2H  L-Gln  CH2OPi  OH  O  O-  L-Glu +  -PiOH  OH  O  OPi  O PiO OH OH  O  O  OH 4  NH2  CO2H HO H N  5  6  O  O OH  COOH COOH OH HO  OH  O  OPi OH  N H  HO OH H  OPi  OH  N H  N H+  OPi  OH  -CO2 -H2O OH OPi N H+ H  O  L-Ser  NAD  OH  OH NADH/H+  N H  H2O  N H  NH2  HO 3 PiO  O  Figure 4.2 Biosynthetic pathway from chorismic acid (4) to tryptophan (3).  This class of alkaloids includes a number of physiologically active substances: the psychedelic compounds lysergic acid diethylamide (7),, a semi-synthetic drug based on the naturally occurring alkaloid ergine (8)148; bufotenin (9), which is secreted from the parotoid glands of Bufo toads; psilocybin (10), from mushrooms of the genus Psilocybe149; and dimethyltryptamine (11), which is found in psychedelic plants such as Diplopterys cabrerana150.  255  O  NEt2 H  O H  N  HN  NH2  N HO  N  HO  O  7  P  8  HN  OH O  O  O  N H  HN  N  N  N H  N H  9  10  N H 12  11  N HN  NH2 H  N HO  H N H  N H O  13  14  OH CH2CH3  H3COOC H O H  N N  H3CO H N  CH2CH3  OCOCH3 H HO COOCH3 15  H  N H H H H3COOC  OH  16  Figure 4.3 Various indole alkaloids with biological activity: lysergic acid diethylamide (7), ergine (8), bufotenin (9), psilocybin (10), dimethyltryptamine (11), melatonin (12), serotonin (13), strychnine (14) , vinblastine (15), and yohimbine (16).  Other indole alkaloids with physiological effects are the human hormones melatonin (12)151 and serotonin (13)149, the convulsant poison strychnine (14) (from the seeds of the Strychnos vomica tree)152, the vinca alkaloids such as vinblastine (15) (anti-mitotic/anti-tubule cancer chemotherapeutics from Catharanthus roseus)153, and yohimbine (16), a stimulant drug used in the treatment of erectile dysfunction found in Pausinystalia yohimbe154. This wide range of indole alkaloids, found in a variety of plants, with significant physiological activities in humans demonstrates the fundamental nature of this structural motif gnerated in various  256  biochemical pathways.  In plants, the indole alkaloids are also found to have important physiological functions. Auxins, a general name given to plant hormones that stimulate growth, are an important class of plant hormones that contain a number of examples of indole alkaloids.155 Indole-3-acetic acid (17) is the most potent of all plant auxins, but indole-3-butyric acid (18) and 4-chloroindole-3acetic acid (19) are also known to be strong plant growth promoters. Indole-3-propanoic acid (20) and indole-pyruvic acid (21) also show plant growth inducing activity, but are much weaker than the indole auxins with an even number of carbons in their side chain. Indole-3-acrylic acid (22) is believed to be the principal auxin in certain members of the pea family Fabaceae.155 O  O  OH N H  Cl  OH  OH N H 18  17  O  19  N H O  O OH  OH  O N H 20  OH  O N H 21  N H 22  Figure 4.4 Auxins, plant hormones: Indole-3-acetic acid (17), indole-3-butyric acid (18), 4-chloroindole3-acetic acid (19), indole-3-propanoic acid (20), indole-pyruvic acid (21), indole-3-acrylic acid (22).  257  4.3 General Introduction to Algae Algae are a diverse group of simple, plant-like organisms. Algae, like most plants, use sunlight as a source of energy. Algae are not, however, vascular plants – they lack roots, leaves, stems and other structures commonly associated with 'true' plants. Algae are the most important photosynthesizing organisms on the planet, capturing more of the sun's energy and producing more oxygen than all other plants combined.156  Among all algae, there is a huge variety in size and habitat. Microalgae, or phytoplankton, float and/or swim in lakes and oceans; thousands to tens of thousands are easily contained within a milliliter of water. Macroalgage, including the seaweeds, can stretch as long as 100 m from the bottom of the ocean to the surface. Algae can also grow in soil, on trees and animals, or even inside porous rocks like sandstone. Algae can tolerate a wide range of temperatures, and are known to grow in hot springs and even deep within polar ice.157  4.3.1 Introduction to Caulerpa sp. Members of the phylum Chlorophyta, or green algae, contain chloroplasts, resulting in a characteristic green color. More than 500 genera and 8000 species have been identified. Green algae range in size and shape from unicellular plankton, to colonial filaments of pond scum, to leafy-like seaweeds in rocky and sandy intertidal areas.157 Caulerpa is a genus of seaweeds, under the Chlorophyta division. These macro green algae are prolific worldwide, living in mostly temperate and tropical waters at depths down to -50 m. Growing in a variety of bright green shades, Caulerpa sp. can take on many different forms and shapes, from tall fronds to thick mats; from razor-like blades to feathers to small clusters of 'grapes'.157,158  258  Caulerpa sp. are unusual because they are siphonous green algae and are coenocytic, meaning they consist of only one cell with many nuclei. In external appearance, these algae resemble a vascular plant, having stolons (horizontal runners) that bear root-like rhizoids for anchorage upon substrate, and frond-like assimilatory shoots. Unlike a vascular plant, which has roots, Caulerpa sp. is a single cell – its protoplast consists of a central vacuole surrounded by cytoplasm which contains chloroplasts, amyloplasts and many small nuclei.157 Caulerpa sp. obtains mechanical support via turgor and also an extensive internal network of projections from the cell wall called trabeculae, along with the thickness of the cell wall. When wounded, Caulerpa sp. has a wound response, quickly forming a wound plug from proteins and sulphonated polysaccharides. As a result, when attacked by a herbivore, the resultant pieces of Caulerpa sp. are able to attach onto substrate and start growing independently. This, in addition to defensive secondary metabolites, forms Caulerpa sp.'s main mode of defense.159  4.3.2 Overview of Known Metabolites From Caulerpa sp. There are currently 194 taxonomically distinct species of Caulerpa, with over a hundred more proposed.158 The current body of literature contains a rich number of compounds from the genus Caulerpa, most of which are terpenoid in origin.27 Caulerpa sp. were some of the first green algae to be investigated by marine natural product chemists, so much is understood about the ecological purpose and origin of secondary metabolites from Caulerpa sp. The earliest studies identified a myriad of diterpenes and sesquiterpenes, including caulerpol (23), flexilin (24) and trifarin (25).160  259  OH 23 OAc 24  AcO OAc  25  AcO  Figure 4.5 Caulerpol (23), flexilin (24) and trifarin (25).  Flexilin and trifarin are the first natural products reported with the bis-enol acetate functional group – a common feature among compounds found in green algae from the genera Caulerpa, Udotea and Halimeda, among others. Another metabolite well known in Caulerpa sp. is caulerpenyne (26), a unique acetylenic diterpenoid that is believed to be an antifeedant, with activity generally associated with the bis-enol-acetate functional group.161 Caulerpenyne is the major metabolite found in Mediterranean populations of Caulerpa taxifola, an invasive species that covers extensive areas of the southern Mediterranean coast as well as areas along the California coastline. The Mediterranean species of Caulerpa taxifola transforms caulerpenyne to oxytoxins 1 (27) and 2 (28) via deacetylation as a result of a wound-activated process. Because of the aldehyde functional groups, oxytoxins 1 and 2 are probably more potent chemical defenses than caulerpenyne.162,163  260  AcO  OAc 26  OAc  AcO  OAc  O  O  O 28  27  Figure 4.6 Caulerpenyne (26) is converted to oxytoxins 1 (27) and 2 (28) in a process that results from wounding, in Caulerpa taxifola.  In addition to these bis-enol acetate terpenes, numerous aromatic and nitrogenous compounds have also been reported in Caulerpa sp. including caulerpin (29)164, caulerpinic acid (30)164, and caulersin (31).165 MeO  N H O  OMe 29  HO  O H N N H O  O H N  H N  O  N H OH 30  O  OMe 31  Figure 4.7 Caulerpin (29), caulerpinic Acid (30) and caulersin (31).  First isolated in 1966, caulerpin (29) is a dimer of indole-3-acrylic acid (22), a known plant hormone. Its structure was initially incorrectly reported, but has been confirmed by 261  extensive chemical degradation, X-ray crystallography, and modern NMR techniques.166 Caulerpin, and its related mono- and diacids, have been reported in a number Caulerpa sp.167,164,165 Originally labeled as a toxin, caulerpin does not show appreciable levels of toxicity to vertebrates and microorganisms, and is now commonly referred to as a pigment molecule. Due to its structural similarity to the auxin indole-3-acrylic acid (22), plus the isolation of other auxins in some Caulerpa sp.168, has led to speculation that caulerpin serves a growth hormone purpose in Caulerpa sp..169  4.4 Indoleamine-2,3-dioxygenase The essential amino acid tryptophan (3) is metabolized in mammals by two enzymes: tryptophan-2,3-dioxygenase (TDO) and indoleamine-2,3-dioxygenase (IDO). TDO is found in the liver, whereas IDO is found in the placenta, lung, small intestine, large intestine, colon, spleen, liver, kidney, stomach and brain. IDO is a 45 kDa monomeric protein containing a heme as its sole prosthetic group. It is encoded by a gene on human and mouse chromosome 8, that is over 15 kb in length; human IDO cDNA encodes a protein that is 403 amino acids in length. IDO is ubiquitous amongst all mammals but is not found in bacteria. IDO is an enzyme involved in the first and rate limiting step of the kynurenine pathway (Figure 4.8) for the degradation of tryptophan, and towards the biosynthesis of nicotinate-adenine dinucleotide. Active ferric-IDO catalyzes the oxidative cleavage of the pyrrole ring of tryptophan to generate N-formylkynurenine (32). IDO has a high affinity for tryptophan (Km = 0.02 mM), and thus can rapidly deplete a local tissue micro-environment of the essential amino acid.170  262  O  COOH  COOH O  NH2  NH2  COOH NH2  NH O 3  OH  N  NH2  H  32  33  COOH  34  O  COOH NH2  NH2 OH 35 O COOH N  COOH  O OH  O  HO  OH  NH2  NH2  O  OH  37  36  NAD Figure 4.8 The Kynurenine pathway. Tryptophan is degraded into quinolinic acid (37), a precursor to NAD.  The tertiary structure of human IDO was recently determined by x-ray crystallography.171 IDO is folded into two distinct alpha helical domains (one smaller, one larger) with the heme group positioned between them. The heme is coordinated to the active site by a the imidazole  263  group of a histidine.172 Heme contains a four coordinate iron atom, thus the histidine imidazole group coordinates as the fifth ligand, leaving the distal heme-iron position open for ligand binding.  IDO is purified in an inactive state. Activation requires the single electron reduction of the ferric iron to ferrous iron, which allows the binding of tryptophan and O2 to the active site (Figure 4.9). Formation of the intermediate 38 may occur via the ionic mechanism that is shown, but may also proceed via radical or pericyclic mechanisms. A rearrangement yields the cyclic hemiacetal 39, from which N-formyl-kynurenine is formed upon release from the enzyme.173  264  His  His  Fe+2  Fe+2  O  O O  +  :B  BH  O  R  R N H  N 38  His  His  Fe+2  Fe+2  O  O R  B: H N  R  H  O N  B H 39  O  Figure 4.9 Possible mechanism for formation of N-formyl-kynurenine from tryptophan, as catalyzed by IDO.  4.4.1 IDO as a Drug Target IDO depletes tryptophan from its local tissue micro environment, and consequently promotes the formation of kynurenine pathway metabolites. IDO is modulated by the adaptive 265  immune system - interferon-γ, one of the most potent inducers of IDO, is produced by T-cells in response to microbial or viral infection.174 In the absence of sufficient tryptophan, T-lymphocytes arrest in G1 and fail to proliferate. Moreover, tryptophan metabolites from the kynurenine pathway have a strong T-cell inhibitory action. Thus, IDO plays an immunosuppressive role 'hiding' its source tissues from T cells175,176. IDO is highly expressed in the placenta, where it plays a crucial role in preventing allogenic fetal rejection due to maternal T-cell immunity, as well as in many tumors.  Many tumors in humans will express IDO even though their tissue of origin does not.177 IDO is also expressed by antigen-presenting cells at the periphery of tumors, and in surrounding lymph nodes that drain the tumors, where T lymphocytes would otherwise be activated. The survival benefit of this mechanism is balanced by the need for the essential amino acid tryptophan.175,176 The relationship between cancer and elevated tryptophan catabolism was first recognized in the 1950s in the urine of bladder cancer patients.178 Several studies have suggested that IDO over-expression is associated with a poor prognosis in cancer patients.179-181  Some of the intermediate metabolites of the kynurenine pathway of tryptophan metabolism are neuroactive – specifically, quinolinic acid (37). Direct, intracerebral injection of quinolinic acid causes neural death, and in vivo application kills neurons.182 Furthermore, cerebrospinal fluid levels of quinolinic acid correlate directly to the severity of neurological deficits.183 Additionally, there is evidence that quinolinic acid levels and upregulation of IDO are increased in the brain of those suffering from Alzheimer's disease.184 It is thus suggested that the production of quinolinic acid plays a direct role in the pathogenesis of neurological disorders. Since the production of quinolinic acid closely reflects the local induction of IDO, it is theorized 266  that upregulation of IDO in the brain directly results in the accumulation of quinolinic acid and associated neural degeneration.185 IDO is also induced in various types of brain inflammation. The multiple and complex brain inflammations associated with Alzheimer's disease, plus the observed upregulation of IDO and accumulation of quinolinic acid, suggest that quinolinic acid may play an important role in the neurodegeneration associated with Alzheimer's disease.186  IDO also plays a role in cataract formation. The primary UV filter components in the human lens are kynurenine, 3-hydroxy-kynurenine, 3-hydroxy-kynurenine glucoside, and 4-(2amino-3-hydroxyphenyl)-4-oxobutanoic acid glucoside. All of these filter components are synthesized locally from tryptophan in lens epithelial cells, and they are present in high concentrations in the lens itself.187 With age, the human lens tends to progressively become more yellow, especially in age-related nuclear cataracts that irreversibly result in a loss of vision.188 When 3-hydroxy-kynurenine glucoside is deaminated, the product binds covalently to lens proteins. Since there is little or no turnover of lens proteins, these yellow 3-hydroxykynurenine glucoside adducts accumulate increasingly with age (3-hydroxy-kynurenine and kynurenine also form similar adducts).189 Additionally, 3-hydroxy-kynurenine is easily oxidized by molecular oxygen to generate hydrogen peroxide, a strong oxidant. Oxidation is a hallmark of an age related cataract, but hydrogen peroxide levels are low or undetectable in the vitreous humor. 3-hydroxy-kynurenine might be a potential internal source of H2O2 in the lens190, contributing to the oxidation of lens proteins. Thus, “kynurenilation” (adduct formation) and the oxidation of lens proteins by H2O2 from 3-hydroxy-kynurenine appear to play a critical role in the formation of age related cataract. If the levels of these UV filters in the lens are reduced by inhibiting IDO, it may be possible to slow the formation of cataracts.191  267  4.4.2 Inhibitors of IDO IDO is a promising target for therapeutic intervention in cancer therapy, neurological disorders and cataract formation, since IDO is the rate limiting enzyme in the kynurenine pathway. To that end, significant effort has been made to find specific inhibitors of IDO. The first reported IDO inhibitor was the noncompetitive norharman (40), a β-carboline. Among the derivatives of norharman, 3-butyl-β-carboline (41) was found to be the most active, with a Ki of 3.3 μM.192 1-Methyl tryptophan (1-MT, 42) is currently accepted as the optimal IDO inhibitor, showing competitive inhibition and a Ki of 13.0μM. 1-Methyl tryptophan has been used to demonstrate that inhibition of IDO can exert anti-tumor effects – a number of studies have shown that 1-MT will slow the growth of tumors.193 In all cases, 1-MT retarded tumor growth but did not elicit tumor regression when used alone. Unfortunately, 1-MT is poorly soluble, and also has been shown to be a very poor inhibitor for IDO in vitro, requiring more than 200 μM to block 50% of IDO-mediated tryptophan degradation due to low membrane permeability in tumor cells.194 Necorostatin (43), or methyl-thiohydantonin-tryptophan, has a slightly more potent Ki (11.3 μM) and is much more permeable through cell membranes, but unfortunately it is not specific to IDO and blocks a critical step in a nonapoptotic/necroptosis pathway.195 Clearly there is a need for more potent, soluble and specific lead compounds in order to further probe the efficacy of IDO inhibition.194  268  N  N  N H  N H 41  40  O COOH  N N H  NH2 N  S  N  42  43  Figure 4.10 IDO inhibitors norharman (40), 3-butyl-β-carboline (41), 1-methyl tryptophan (42) and necorostatin (43).  In order to observe IDO inhibition, Vottero and Mauk developed a high-throughput in vitro IDO bioassay (Figure 4.11).196 A solution of extract (or pure compound) was added to a mixture of phosphate buffer (pH 6.5), recombinant IDO, tryptophan, ascorbic acid and methylene blue, then left for 30 minutes at 37oC. If IDO remains active, N-formyl-kynurenine (32) will be formed. The reaction is stopped by the addition of trichloroacteic acid and heating to 65oC, which converts N-formyl-kynurenine to kynurenine. Addition of pdimethylaminobenzaldehyde (44) generates the yellow adduct, p-(Nmethylaminobenzylidene)kynurenine (45), which has a strong absorbance at 480 nm. The amount of this product formed is a measure of IDO activity.197  269  O  O  COOH NH 2  COOH NH2 N H 3  O2, IDO 30oC O 30min H  NH  O  TFA, 60oC 15min  COOH H2N 4  N H2N  O  N 44  COOH N  H2N 33  45 Yellow  Figure 4.11 Basic schematic of bioassay used to identify numerous inhibitors of IDO.  Using this bioassay, the Andersen lab has found some interesting and novel inhibitors of IDO. Polyketides from the marine hydroid Garveia annulata (annulins A (46), B (47) and C (48), graveatins A (49), C (50) and E (51), 2-hydroxygraveatin A (52) and 2-hydroxygarvin E (53)198) were shown to possess good inhibitory activity for IDO, with the most potent inhibitor, annulin C, having a Ki of 0.14 μM. The novel alkaloids exiguamine A (54)199 and B (55)200 from the sponge Neopetrosia exigua were also identified by this bioassay, and were found to have excellent nanomolar Ki's of 41 nM and 80 nM, respectively. Finally, the novel indole alkaloids, plectosphaeroic acid A (56), B (57) and C (58), isolated from the marine sediment fungus Plectosphaerella cucumerina, inhibit IDO with Ki values of approximately 2 μM.201 The plectosphaeroic acids, the exiguamines and the annulins represent important lead structures for the development of more potent IDO inhibitors.  270  O  O  O  O  O  O  O  O HO  O O  O  46 O  O  47  OH O  O  O  49  48  OH O  O  OH  O  HO HO  OH OH O  O  O  O  O  O  HO  O O  HO  O  OH O  HO HO  O 50  O  51  52  53  HN  H2N  N  O N  OO N H  OH  HN  HN  HN  N  N  O  HO S O  HO S O  HO S O  S  OH  54  N H O  S N H O  S N H O  N  COOH  COOH  N  N  N OH S  COOH  HN  H2N  N  O N  O  OO N H  COOH  OH O HO  N  O  N COOH NH2  NH2  O  COOH NH2  O  O  O  56  57  58  55  Figure 4.12 IDO inhibitors identified from marine organisms: annulin A (46), B (47) and C (48), graveatin A (49), C (50) and E (51), 2-hydroxygraveatin A (52), 2-hydroxygarvin E (53), exiguamine A (54) and B (55), and plectosphaeroic acid A (56), B (57) and C (58).  271  4.4.3 Yeast Based Screen for IDO Inhibition These newer IDO inhibitors, particularly the annulins, still suffer from issues of solubility and presumably with the permeability of the cell membrane. In an effort to address this specific issue, Roberge and colleagues set out to develop a yeast cell-based bioassay that has human IDO expressed heterologously in a yeast strain.202 Because yeast requires tryptophan for normal growth, expression of IDO and its subsequent degradation of tryptophan will reduce the yeast's ability to grow and proliferate. This quantifiable phenotypic response can then be used to measure the efficect of inhibitors of IDO, and will select for compounds that are able to enter the yeast cell.203 The yeast strain EIS-20-2B was selected for use in this bioassay because it is missing the major drug efflux pumps PDR5 and SNQ2, so it is generally more sensitive to drugs204. The human IDO gene was integrated as a single copy, with a GAL1 promoter for control of that gene. A control strain of yeast was prepared in the same manner, with an empty expression cassette used in place of the IDO gene. During testing, washed yeast cells with and without the IDO gene were diluted to a final A600nm of 0.02 with a synthetic medium containing 2% galactose (for activation of promoter) and 25 μM of tryptophan. To these cells, test compounds (or extracts) were added, and the solution incubated at 30oC for 40-45 hours. After incubation, the solutions were shaken to resuspend the cells, then their A600nm taken. Test wells containing compounds that were able to enter the cell and inhibit IDO would show growth, whereas those that were inactive or unable to enter the cell would not. Compounds that were toxic or growth promoting to the the yeast could be easily identified by observing the effect on the control strain.202  The known inhibitors, necorostatin and 1-methyl tryptophan were tested using this  272  bioassay. Necorostatin showed 25% growth restoration at 100 μM, but was toxic to yeast at concentrations above 30 μM. 1-MT showed no growth restoration at any concentration, but was also toxic to yeast above 10 μM. These results indicate that the bioassay has strict requirements for showing activity, and that growth restoration must occur at concentrations lower than those toxic to yeast to indicate a positive 'hit'. The yeast based IDO bioassay was used to screen 2500 pure compounds from the 'NCI Biodiversity Set', and 875 methanolic extracts of marine organisms from the Andersen lab. From the NCI compounds, nine were shown to restore growth with EC50 values ranging from 0.05 μM to 6 μM. One in particular, NCI 401366, was also shown to be an in vivo inhibitor of IDO with a Ki of 1.5 μM, validating the efficacy of the bioassay. Additionally, the extract 03-93, from a Caulerpa sp. was shown to restore growth, and was selected for further bioassay guided fractionation.202  4.5 Isolation of Caulerpin from Caulerpa sp. A sample of Caulerpa sp. was extracted repeatedly at room temperature with methanol, and the combined methanolic extracts evaporated to a powdery dry green solid that tested positive in the yeast growth restoration assay. The crude extract was resuspended in water and then partitioned with ethyl acetate. The ethyl acetate layer was evaporated to afford a green, oily solid that was subjected to size exclusion chromatography on Sephadex LH-20 swollen in a solution of 20% dichloromethane in methanol. The eluent fractions from the column were pooled into six fractions on the basis of their TLC profile. Growth restoration activity was exhibited by the fifth fraction, which showed only one spot by TLC as visualized with UV light at 254 nm. The 1H NMR spectrum of this fraction revealed that it was a nearly pure compound with a number of well resolved signals in the aromatic region, as well as an intense singlet at δH 3.89 ppm. This fraction was subsequently applied to a SiO2 Sep-pak, with the compound eluting 273  with 25% hexanes in ethyl acetate. Crystals of the compound were grown via vapor diffusion between ethyl acetate and hexanes. A small amount of the compound (2 mg) was dissolved, in a small vial, into 1 mL of ethyl acetate. This vial was then placed into a sealed glass jar containing an excess of hexanes, and left at room temperature for two days, during which small red crystals formed. NMR studies and X-ray crystallography allowed this compound to be identified as the known indole alkaloid, caulerpin.  4.5.2 Structure Elucidation of Caulerpin Caulerpin was isolated as bright red crystals, and identified by analysis of its 1H-NMR, 13  C-NMR, COSY, HMBC and HSQC spectra, and by HRESIMS. 1D and 2D NMR data for  caulerpin are summarized in Table 4.1. O  6 5  10  4  7 8 9  2  3  O 11 H N  N H O 12 O 13  Figure 4.13 Caulerpin (29), with numbering scheme.  274  O  6 5  10  4  7 8 9  2  3  O 11 H N  N H O 12 O 13  1  Figure 4.14 400 MHz H spectrum of caulerpin (29), recorded in CD2Cl2-d4.  275  O  6 5  10  4  7 8 9  2  3  O 11 H N  N H O 12 O 13  Figure 4.15 100 MHz 13C spectrum of caulerpin (29), recorded in DMSO-d6.  276  O  6 5  10  4  7 8 9  2  3  O 11 H N  N H O 12 O 13  Figure 4.16 400 MHz COSY spectrum of caulerpin (29), recorded in CD2Cl2-d4.  277  O  6 5  10  4  7 8 9  2  3  O 11 H N  N H O 12 O 13  Figure 4.17 400 MHz HSQC spectrum of caulerpin (29), recorded in CD2Cl2-d4, with inset  expansion of H-13/C-13 (methoxy) data.  278  O  6 5  10  4  7 8 9  2  3  O 11 H N  N H O 12 O 13  Figure 4.18 400 MHz HMBC spectrum of caulerpin (29), recorded in CD2Cl2-d4, with inset expansion  of aromatic region.  279  Table 4.1 1D and 2D NMR data for caulerpin (29), recorded at 400 MHz (1H) and 100 MHz (13C) in CD2Cl2-d2.  O  6 5 4  7 8 9  a  13  1  10 2  3  O 11 H N  N H O 12 O 13  C#  C δδ (ppm)a  Hb δ (ppm) mult. (J (Hz))c  COSYc (H→H)  HMBCc (H→C)  1  -  9.22 br s  -  -  2  137.3  -  -  -  3  111  7.34 d(6.9)  H-3  C-3, C-4, C-6  4  122.6  7.18 t (6.9)  H-2, H-4  C-1, C-5  5  119.9  7.11 t (7.1)  H-3, H-5  C-2, C-5, C-6  6  117.7  7.44 d (7.1)  H-4  C-1, C-3  7  128.4  -  -  -  8  111.6  -  -  -  9  132.7  -  -  -  10  126.8  -  -  -  11  141.5  8.07 s  -  C-2, C-3, C-4, C-11, C-12  12  165.6  -  -  -  13  52.1  3.89 s, 2H  -  C-12  Recorded at 100 MHz. b Assigned according to HSQC recorded at 400 MHz. c Recorded at 400 MHz.  The 1H NMR spectrum of caulerpin shows only seven distinct resonances: the amide proton (H-1) at δH 9.22 ppm, the cyclooctatetraene (C-10) proton at δH 8.07ppm, aromatic (C-5 to C-8) protons at δH 7.44 ppm (doublet, J = 7.5 Hz), δH 7.34 ppm (doublet, J = 8.0 Hz), δH 7.18 ppm (triplet, J = 7.5 Hz), δH 7.11 ppm (triplet, J = 8.0 Hz), and the methoxy (C-13) protons at δH 3.89 ppm. 280  From the splitting patterns, it was evident that the aromatic protons formed an isolated spin system characteristic of a 1,2-disubstituted phenyl ring. This was confirmed by examination of the COSY spectrum, which showed that there were only correlations amongst the adjacent aromatic protons, and nothing to the amide, cyclooctatetraene and methoxy protons.  Upon examination of the HSQC and HMBC spectra, it was possible to put together three substructures, as shown in Figure 4.19.  6  5  7 8  4  3  9  2  11  O 13  10 12 H O  Figure 4.19 Substructures of caulerpin (29).  These structures did not account for the exchangeable peak of the amide proton in the 1HNMR spectrum. Additionally, HRESIMS showed a [M+H]+ peak at m/z 399.1339, corresponding to a molecular formula of C24H18N2O4 (calculated mass for C24H18N2O4 m/z 399.1345). Given that there were only 12 peaks in the 13C-NMR spectrum, it was deduced that this molecule must contain an element of symmetry.  Examination of the HMBC spectrum reveled a weak correlation between the proton at δH 8.07 ppm (H-10) to the carbon resonance at δC 128 ppm (C-4), and another to the carbon resonance at δC 126 ppm (C-2). Based on these correlations, the substructures from Figure 4.19, and the molecular formula, the structure shown in Figure 4.20 was proposed.  281  O  COSY HMBC  N H O  O H N  O  Figure 4.20 Key COSY and HMBC correlations in caulerpin (29).  The structure of caulerpin (29) was compared with 1H and 13C NMR values found in the literature. The reported 13C chemical shifts in the literature (particularly C-2 and C-11) differed somewhat from the experimental data. Since the isolated caulerpin readily grew crystals, it was submitted for X-ray structure analysis (performed by Dr. Brian Patrick at UBC) in order to confirm the structure assignment. The resulting ORTEP diagram is shown in Figure 4.21. The crystal structure confirms the proposed structure assignment.  282  Figure 4.21 ORTEP diagram of caulerpin (29).  4.6 Discussion and Conclusions Caulerpin exhibited excellent activity in the yeast cell based assay; however, this activity did not translate into any observable inhibition of IDO in vitro using the free enzyme assay. The most likely explanation of this is that caulerpin is a prodrug that is metabolized once in the cellular environment into the actual active compound. Tryptophan, the substrate for IDO, structurally resembles half of the caulerpin molecule, with one of the main differences being the  283  presence of the methyl ester in caulerpin (Figure 4.22).  HO  O  O NH2 N H O  N H  3  O H N  O 29  Figure 4.22 A comparison of the structures of tryptophan (3) and caulerpin (29).  It was theorized202 that the methyl ester might aid in transport of caulerpin across the cell membrane, but may prevent caulerpin itself from effectively inhibiting IDO. Once inside the cell, caulerpin could be metabolized to its mono or diacid (caulerpinic acid) via hydrolysis of the methylester groups – giving a resultant structure that closely resembles tryptophan. Unfortunately, however, crude caulerpinic acid from a simple hydrolysis reaction of lithium hydroxide on caulerpin showed absolutely no activity in vitro on free IDO.  Another explanation for the lack of activity in vitro is simply that the in vitro bioassay has a bias towards identifying uncompetitive inhibitors of IDO, because it requires a large amount of tryptophan to be present in the experimental well. Presumably, a competitive inhibitor of IDO could be 'swamped out' by the excess of tryptophan, and would not show activity in the bioassay. Caulerpin is likely a competitive inhibitor of IDO due to its structural similarity to tryptophan. Recently, a method for quantifying IDO activity in an IDO-expressing tumor model has been used to observe the tumor evading properties of IDO inhibitors205.  In wild type mice, and IDO  284  deficient (control) mice, blood plasma levels of N-formyl-kynurenine and tryptophan are monitored via HPLC after an inhibitor is introduced. Inhibitory activity will then be a function of the amount N-formyl-kynurenine formed in the blood plasma of the wild type mice. Caulerpin will be subjected to this bioassay in an effort to determine if it is active against IDO.  4.7 Experimental 4.7.1 General Experimental Procedures 1  H, 13C, COSY, HSQC, HMBC, TOCSY and ROESY spectra were recorded on a Bruker  AV400 or Bruker AV600 NMR spectrophotometer equipped with a cryoprobe. 1H chemical shifts were referenced to the residual CD2Cl2-d2 signal (δH 5.31ppm), and 13C chemical shifts were referenced to the CD2Cl2-d2 solvent peak (δC 53.42 ppm). All NMR solvents were obtained from Cambridge Isotope Laboratories. All NMR data was processed using Bruker XWINNMR® software.  All chromatography was performed using HPLC grade solvents from Fisher Scientific with no additional purification. Water was purified by use of a Millipore MQ filter system. HPLC separations were performed using a Waters 600 pump and a Waters PDA900 detector, using an Inertsil C18 column, 9.4 x 250 mm, flow 2 mL/min. All solvents were filtered prior to use, then sparged with helium. Optical rotations were determined by using a JASCO J-1010 polarimeter equipped with a halogen lamp (589 nm) and a 10 mm micro cell. UV spectra were recorded on a Waters 2487 spectrophotometer. ESI mass spectra were recorded using a Micromass LCT mass spectrometer.  285  4.7.2 Isolation of Caulerpin For isolation scheme see section 4.5.  4.7.3 Caulerpin Physical Data Red prisms. HRESIMS: [M+Na]+ m/z 399.1339 (calculated for C24H18N2O4Na). For 1D and 2D NMR data please refer to Table 4.1.  286  5. Conclusions The study of bioactive natural products – secondary metabolites with interesting biological activity – occurs in a variety of ways, with outcomes and courses of future study equally varied. Once a bioactive natural product is isolated and characterized, understanding the requirements for the biological activity often becomes the focus of a project, with related family members or synthetic analogues providing information about the importance of various structural motifs in a molecule. This may be done by studying how the molecules perturb the biological activity, or through methods that allow study of protein-ligand interactions (such as STD NMR). Ultimately, this information may lead to an understanding of a simplified pharmacophore, and the identification of stronger lead compounds.  The montbretins, a family of flavonoids, were isolated as the active compounds from an extract of the corms of Crocosmia sp., a hardy perennial with invasive qualities. Montbretins AE are competitive inhibitors of human pancreatic α-amylase, an enzyme involved in the breakdown of complex carbohydrates and whose inhibition may help diabetics control their blood sugar levels after a meal. Ki's for the montbretins range from 1.3 nM (for montbretin A) to 12 μM (for montbretin E). Montbretins A and B were previously known, but the data reported for them in the literature was sparse and contained some uncertainly, and thus the full assignment is reported here. Montbretins C, D and E are new members of the family that provide important and significant structure-activity information. The presence of caffeic acid instead of p-coumaric acid or ferulic acid in the cinnamic acid moiety of the montbretins seems to be important for activity. A 1D STD NMR performed on montbretin B in the presence of human pancreatic α-  287  amylase confirms that the p-coumaric acid region comes into close proximity to the protein when the two interact. Montbretin A has shown promising results in a mouse 'starch tolerance test', suggesting it may be as effective – if not better – than the currently prescribed human pancreatic α-amylase inhibitor, acarbose.  When activated by bacterial flagellin, TLR5, which is expressed in the lungs of cystic fibrosis patients, initiates a pro-inflammatory response. In an effort to help control and understand the cycle of obstruction-inflammation-infection, this activation has been identified as an interesting target. Girolline, a known alkaloid, was isolated from an extract of a Phonpeian sponge following very potent activity. Girolline proved to be difficult to isolate, so its synthesis was undertaken. Girolline was not synthesized, and instead, a small portfolio of analogues was created – none of which displayed any activity. This suggests that girolline has strict structural requirements for activity. Efforts to obtain pure girolline in order to validate it as a hit are currently underway.  The massacreones, a new family of ecdysteroids from a Dominican cnidarian, were also isolated after pursuing promising activity in the TLR5 assay. Ecdysteroids are a large family of well known terrestrial insect moulting hormones (and plant defense chemicals). Some ecdysteroids act as moulting hormones for crustaceans. The massacreones, while somewhat active in the TLR5 assay, have a small window between toxicity and activity, and thus are not good candidates for drug development. They do however provide a potential tool for use in probing the TLR5+flagellin activation process. Additionally, the massacreones show interesting structural features – notably, the addition of the oxidized carbon C-28 at position 25 – not sen in 288  any reported ecdysteroids in the literature.  Caulerpin, an algal pigment molecule from Caulerpa sp., was found to inhibit indoleamine-2,3-dioxygenase, a key enzyme involved in tryptophan degradation that is used by tumors to hide from the body's natural defenses. Inhibition of IDO could also be of benefit in the treatment of neurodegenerative disorders and cataracts. The bioassay used to identify caulerpin was a yeast growth restoration assay, and caulerpin showed excellent activity. However, when tested in vitro on cell free IDO, caulerpin did not register as a hit. Clearly, one of the bioassays has provided incorrect, or misleading, results. It is thought that the cell free bioassay may be unable to detect caulerpin if it is a competitive inhibitor for IDO, because the cell free bioassay uses a large excess of tryptophan, IDO's substrate, effectively swamping out any competitive inhibitors present. A third bioassay will be used to investigate the validity of caulerpin's IDO activity.  289  References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.  P. M. Dewick, Medicinal Natural Products: A Biosynthetic Approach, Anthony Rowe Ltd., Chippenham, Wiltshire, 3rd edn., 2009. M. Heinrich, S. Gibbons, J. Barnes, and E. M. Williamson, Fundamentals of Pharmacognosy and Phytotherapy, Elsevier Science Limited, Livingston, London, 2004. D. Newman and G. Cragg, J. Nat. Prod., 2007, 70, 461-477. Y. Chin, M. Balanas, H. Chai, and A. Kinghorn, AAPS J., 2006, 8, 239-253. D. Newman, J. Med. Chem., 2008, 51, 2589-2599. T. Eisner and J. Meinwald, Eds., Chemical Ecology: The Chemistry of Biotic Interaction, National Academy Press, Washington, DC, 1995. J. B. McClintock and B. J. Baker, Eds., Marine Chemical Ecology, CRC Press LLC, Boca Raton, Flordia, 2001. P. E. Minnis, Ed., Ethnobotany: a reader, University of Oklahoma Press, Norman, Oklahoma, 2000. H. D. Neuwinger, African Ethnobotany: Poisons and Drugs: Chemistry, Pharmacology, Toxicology, Wissenschaftliche Verlagsgesellschaft mbH, Stuttgart, Germany, 1994. J. Kennedy, Nat. Prod. Rep., 2008, 25, 25-34. D. A. Hopwood, Complex Enzymes in Microbial Natural Product Biosynthesis, Academic Press, San Diego, CA, USA, 2009, vol. 459. M. Misick and D. Hoffmeister, Planta Med., 2007, 73, 103-115. H. Bode and R. Muller, Angew. Chem. Int. Ed., 2005, 44, 6828-6846. S. Caveney, D. A. Charlet, H. Freitag, M. Maier-Stolte, and A. N. Starratt, Am. J. Bot., 2001, 88, 1199-1208. D. Youssef and A. Frahm, Planta Med., 1994, 60, 267-271. The Ebers Papyrus: A New English Translation, Commentaries and Glossaries, Academy of Scientific Research and Technology, Cairo. W. Jones, Hippocrates and the Corpus Hippocraticum, Oxford University Press, London, 1945. S. Yang, The Divine Farmer's Materia Medica: A Translation of the Shen Nong Ben Cao Jing, Blue Poppy Press, Boulder, CO, 1998. F. Sertuerner, Annalen der Physik, 1817, 55, 56-89. A. Gilman, L. Goodman, and F. Murad, The Pharmacological Basis of Therapeutics, MacMillan Publishing Co., New York, 1985. S. Scheindlin, Mol. Interventions , 2007, 7, 236-242. Ladenburg, in American Druggist Volume 17, 1888, p. 168. W. Sneader, Drug Discovery: A History, John Wiley and Sons, West Sussex, England, 2005. A. Flemming, Br. J. Exp. Pathol., 1929, 10, 226-36. R. J. Andersen and D. E. Williams, in Chemistry of the Marine Environment, Royal Society Of Chemistry, Cambridge, 2000, vol. 17. W. Fenical, Oceanography, 2006, 19, 110-119. J. W. Blunt, B. R. Copp, W. Hu, M. H. Munro, and P. T. Northcote, Nat. Prod. Rep., 2009, 25, 170-244. M. H. Munro, J. W. Blunt, E. Dumdei, S. Hickford, R. Lill, C. Battershill, and A. Duckworth, J. Biotechnol. , 1999, 70, 15-25. 290  29. W. Bergmann and R. J. Feeney, J. Org. Chem., 1951, 16, 981-987. 30. W. Bergmann and D. Burke, J. Org. Chem., 1955, 20, 1501-1507. 31. B. M. Olivera, L. J. Cruz, V. D. Santos , G. LeCheminant, D. Griffin, R. Zeikus, J. M. McIntosh, R. Galyean, and J. Varga, Biochemistry, 1987, 28, 2086-2090. 32. T. F. Molinski, D. S. Dalisay, S. L. Lievens, and J. P. Saludes, Nat. Rev. Drug Discov., 2009, 8, 69-85. 33. A. E. Wright, D. A. Forleo, G. P. Gunawardana, S. P. Gunasekera, F. E. Koehn, and O. J. McConnell, J. Org. Chem., 1990, 55, 4508-4512. 34. K. L. Rinehart, T. G. Holt, N. L. Fregeau, J. G. Stroh, P. A. Kiefer, F. Sun, L. H. Li, and D. G. Martin, J. Org. Chem., 1990, 55, 4512-4515. 35. MarinLit database., Department of Chemistry, University of Canterbury. 36. R. A. Edrada, M. Heubes, G. Brauers, V. Wray, A. Berg, U. Grafe, M. Wohlfarth, J. Muhlbacher, K. Schaumann, Sudarsono, G. Bringmann, and P. Proksch, J. Nat. Prod., 2002, 65, 1598-1604. 37. C. A. Bewley and D. J. Faulkner, Angew. Chem. Int. Ed., 1998, 37, 2162-2178. 38. T. Yasumoto, Y. Oshima, W. Sugawara, T. Fukuyo, H. Oguri, T. Igarashi, and N. Fujita, Nippon Suisan Gakkaishi, 1980, 46, 1405-1411. 39. G. S. Jayatilake, M. P. Thornton, A. C. Leonard, J. E. Grimwade, and B. J. Baker, J. Nat. Prod., 1996, 59, 293-296. 40. M. Hildebrand, L. E. Waggoner, G. E. Lim, K. H. Sharp, C. P. Ridley, and M. G. Haygood, Nat. Prod. Rep., 2004, 21, 122-142. 41. L. Di and E. H. Kerns, Drug Discov. Today, 2006, 11, 446-451. 42. F. E. Koehn and G. T. Carter, Nat. Rev. Drug. Discov., 2005, 4, 206-220. 43. A. Rahman, M. Choudhary, and W. Thomsen, Bioassay Techniques for Drug Development, Academic Publishers, Amsterdam, 2001. 44. W. P. Janzen, High Throughput Screening: Methods and Protocols, Human Press, Dordrecht, New York, 2nd edn., 2009. 45. J. B. Lambert, H. F. Shurvell, D. A. Lightner, and R. G. Cooks, Organic Structural Spectroscopy, Prentice Hall Inc., Upper Saddle River, New Jersey, 1998. 46. R. Freeman, Chem. Heterocycl. Compd., 1995, 31, 1004-1005. 47. W. Voelter, Pure Appl. Chem. , 1976, 48, 105-126. 48. W. Reynolds and R. Enriquez, J. Org. Chem., 2002, 65, 221-244. 49. G. Lawrence, The Taxonomy of Vascular Plants, The Macmillan Company, New York, 1951. 50. J. D. McChesney, S. K. Venkataraman, and J. T. Henri, Phytochemistry, 2007, 68, 20152022. 51. P. B. Kaufman, L. J. Cseke, S. Warber, J. A. Duke, and H. L. Brielmann, Natural Products From Plants, CRC Press LLC, Boca Raton, Flordia, 1999. 52. A. R. Knaggs, Nat. Prod. Rep., 2003, 20, 119-136. 53. B. Winkel-Shirley, Plant Physiol., 2001, 126, 485-493. 54. J. Manning and P. Goldblatt, The Iris Family: Natural History & Classification, Timber Press, Portland, Oregon, 2008. 55. P. Kostelijk, The Plantsman, 1984, 5, 246-253. 56. Warringah Council, Montbretia. http://www.warringah.nsw.gov.au/environment/documents/factsheetmontbretia_001.pdf. (accessed January 30, 2010)., . 57. C. Stone and J. Scott, Hawaii's terrestrial ecosystems: preservation and management, University of Hawaii Press, Honolulu, Hawaii, 1985. 291  58. N. Nagamoto, H. Hoguchi, A. Itokawa, K. Nakata, K. Namba, H. Nishimura, M. Matsui, and M. Mizuno, Planta Med., 1988, 54, 305-307. 59. K. Masuda, S. Funayama, and I. Umezawa, J Nat Prod, 1987, 50, 418-421. 60. K. Masuda, S. Funayama, K. Komiyama, and I. Umezawa, J. Nat. Prod., 1987, 50, 958-960. 61. Y. Asada, T. Ueoka, and T. Furuya, Chem. Pharm. Bull., 1990, 38, 142-149. 62. Y. Asada, T. Ueoka, and T. Furuya, Chem. Pharm. Bull., 1989, 37, 2139-2146. 63. T. Furuya, T. Ueoka, and Y. Asada, Chem. Pharm. Bull., 1988, 36, 444-447. 64. Y. Asada, M. Ikeno, and T. Furuya, Phytochemistry, 1994, 35, 757-764. 65. Y. Asada, Y. Hirayama, and T. Furuya, Phytochemistry, 1988, 27, 1497-1501. 66. Y. Asada, M. Ikeno, T. Ueoka, and T. Furuya, Chem. Pharm. Bull., 1989, 37, 2747-2752. 67. K. Koji, N. Hideo, G. Igi, and K. Masako, Fermented beverage derived from Crocosmia crocosmiiflora bulb extracts and its production method. JP10084923 (A). April 7, 1998. , . 68. Y. Michio, Skin cleansers containing collagenase inhibitors derived from Crocosmia crocosmiiflora/Tritonia crocosmaeflora bulb. JP2000191512 (A). July 11, 2000., . 69. G. D. Brayer, Y. Luo , and S. G. Withers, Protein Sci., 1995, 4, 1730-1742. 70. E. H. Rydberg, C. Li, R. Maurus , C. M. Overall, G. D. Brayer, and S. G. Withers, Biochemistry, 2002, 41, 4492-4502. 71. E. H. Rydberg, G. Sidhu, H. Vo , J. Hewitt , H. Cote, Y. Wang, S. Numao, R. MacGillivray, C. M. Overall, G. D. Brayer, and S. G. Withers, Protein Sci., 1999, 8, 635-643. 72. E. E. Calle, C. Rodriguez, K. Walker-Thurmond, and M. J. Thun, N. Engl. J. Med., 2003, 38, 1625-1638. 73. K. Campbell, J. Am. Pharm. Assoc., 2009, 49, S3-S9. 74. A. J. Scheen, Drugs, 2003, 63, 1165-1184. 75. The World Heatlh Organization. http://www.who.int. (accessed January 30, 2010)., . 76. P. Zimmet, K. G. M. M. Alberti, and J. Shaw, Nature, 2001, 414, 782-787. 77. M. Janbom, J. Chaptal, A. Vedel, and J. Schaap, Montpellier Med., 1942, 441, 21-22. 78. J. A. Greene, Prescribing by numbers: drugs and the definition of disease, Johns Hopkins University Press, Baltimore, Maryland, 2007. 79. A. Harrower, Clin Pharmacokinet. , 1996, 31, 111-119. 80. M. Patlak, FASEB J., 2002, 1853-1870. 81. C. Bailey and C. Day, Practical Diabetes International, 2004, 21, 115-117. 82. S. Kahn, S. Haffner, M. Heise, W. Herman, R. Holman, N. Jones, B. Kravitz, J. Lachin, M. O'Neill, B. Zinman, and G. Viberti, N. Engl. J. Med., 2006, 355, 2427-2443. 83. International Diabetes Federation, in Global Guideline for Type 2 Diabetes, International Diabetes Federation, Brussels, France, 2005. 84. C. A. Tarling, K. Woods, R. Zhang, H. C. Brastianos, G. D. Brayer, R. J. Andersen, and S. G. Withers, ChemBioChem, 2008, 9, 433-438. 85. J. Chiasson, R. Josse, R. Gomis, M. Hanefeld, A. Karasik, and M. Laakso, Lancet, 2002, 359, 2072-2077. 86. R. J. Andersen, K. Woods, S. G. Withers, G. D. Brayer, and A. C. Tarling, Alpha-Amylase Inhibitors: The Montbretins and Uses Thereof. WO2009049428 (A1). April 23, 2009., . 87. P. Lincez and C. Hii, α-amylase inhibitors for Type-2 diabetes: Isolation, Purification, and Kinetic Analysis of Montbretins from Crocosima sp., Presented at the CDRD, University of British Columbia, 2008. 88. J. Wicki, University of British Columbia, Vancouver, BC. Personal Communication, 2009, . 89. M. Mayer and B. Meyer, Angew. Chem. Int. Ed., 1999, 39, 1784-1788. 90. M. Mayer and B. Meyer, J. Am. Chem. Soc., 2001, 123, 6108-6117. 292  91. J. Vavra, C. Deboer, A. Dietz, L. Hanka, and W. Sokolski, Antibiot. Annu., 1959, 7, 230-235. 92. F. Ratjen and G. Doring, Lancet, 2003, 361, 681-689. 93. Canadian Cystic Fibrosis Foundation. http://www.cysticfibrosis.ca/, accessed January 30, 2010., . 94. P. Buchanan, R. Ernst, J. Elborn, and B. Schock, Biochem. Soc. Trans., 2009, 37, 863-867. 95. C.A. Janeway Jr. and R. Medzhitov, Annu. Rev. Immunol., 2002, 20, 197-216. 96. T. Machen, Am. J. Physiol., 2006, 291, C218-C230. 97. T. Kahn, J. Wagener, T. Bost, J. Martinez, F. Accurso, and D. Riches, Am. J. Respir. Crit. Care. Med., 1995, 151, 1075-1082. 98. F. Ratjen, Paediatr. Respir. Rev., 2006, 7, S70-S72. 99. J. Morton and A. Glanville, Semin. Respir. Crit. Care. Med., 2009, 30, 559-568. 100. S. Kirkby, K. Novak, and K. McCoy, Expert. Rev. Anti. Infect. Ther., 2009, 7, 967-980. 101. J. Chmiel and M. Konstan, Clin. Chest Med., 2007, 28, 331-346. 102. S. Epelman, D. Stack, C. Bell, E. Wong, G. Neely, S. Krutzik, and K. Miyake, J. Immunol., 2004, 173, 2031-2040. 103. T. Noah, H. Black, P. Cheng, R. Woods, and M. Leigh, J. Infect. Dis., 1997, 175, 638-647. 104. R. Adamo, S. Sokol, G. Soong, M. Gomez, and A. Prince, Am. J. Respir. Cell. Mol. Biol., 2004, 30, 627-634. 105. H. Auerbach, M. Williams, J. Kirkpatrick, and H. Colten, Lancet, 1985, 2, 686-688. 106. M. Konstan, P. Byard, C. Hoppel, and P. Davis, N. Engl. J. Med., 1995, 332, 848-854. 107. A. Muir, G. Soong, S. Sokol, B. Reddy, M. Gomez, A. Van Heeckeren, and A. Prince, Am. J. Respir. Cell. Mol. Biol., 2003, 30, 777-783. 108. C. Greene, T. Carroll, S. Smith, C. Taggart, J. Devaney, S. Griffin, S. O'Neill, and N. McElvaney, J. Immunol., 2005, 174, 1638-1646. 109. C. J. Blohmke, R. E. Victor, A. F. Hirschfeld, I. M. Elias , D. G. Hancock, C. R. Lane, G. F. Davidson, P. G. Wilcox, K. D. Smith, J. Overhage, R. E. Hancock, and S. E. Turvey, J. Immunol., 2008, 180, 7764-7773. 110. D. N. Cook, D. S. Pisetsky, and D. A. Schwartz, Nat. Immunol., 2004, 5, 975-979. 111. K. Takeda, T. Kaisho, and S. Akira, Annu. Rev. Immunol., 2003, 21, 335-376. 112. S. E. Turvey and T. R. Hawn, Clin. Immunol., 2006, 120, 1-9. 113. A. F. Hirschfeld, J. A. Bettinger, R. E. Victor, D. J. Davidson, A. J. Currie, J. M. Ansermino, D. W. Scheifele, J. S. Orange, and S. E. Turvey, Clin. Immunol., 2007, 122, 271-278. 114. R. E. Victor, A. F. Hirschfeld, and S. E. Turvey, . 115. A. Ahond, , M. B. Zurita, M. Colin, C. Fizames, P. Laboute, F. Lavelle, D. Laurent, C. Poupat, J. Pusset , O. Thoison, and P. Potier, C. R. Séances Acad. Sci. Paris, 1988, 307, 145148. 116. A. Commerçon and J. Paris, Tetrahedron Lett. , 1991, 32, 4905-4906. 117. S. Marchais, A. Al Mourabit, A. Ahond, C. Poupat, and P. Potier, Tetrahedron Lett. , 1998, 39, 8085-8088. 118. D. A. Evans, J. Nelson, E. Vogel, and T. Taber, J. Am. Chem. Soc., 1981, 103, 3099-3111. 119. K. M. Cergol and M. J. Coster, Nat. Protoc., 2007, 2, 2568-2573. 120. D. A. Evans, Aldrichimica Acta , 1982, 15, 24. 121. J. R. Gage and D. A. Evans, Organic Syntheses, 1990, 68, 83. 122. B. L. Eriksen, P. Vedso, and M. Begtrup, J. Org. Chem., 2001, 66, 8344-8348. 123. B. L. Eriksen, P. Vedso, S. Morel, and M. Begtrup, J. Org. Chem., 1998, 63, 12-16. 124. E. F. Scriven and K. Turnbull, Chem. Rev., 1988, 88, 297-368. 125. W. Lin, X. Zhang, Z. He, Y. Jin, L. Gong, and A. Mi, Synth. Commun., 2002, 32, 3279-3284. 293  126. J. K. Groves, N. E. Cundasawmy, and H. J. Anderson, Can. J. Chem., 1973, 51, 1089-1098. 127. A. Olofson, K. Yakushijin, and D. E. Horne, J. Org. Chem., 1998, 63, 1248-1253. 128. S. Kumpun, B. Yingyongnarongkul, R. Lafont, J. Giraultc, and A. Suksamrar, Tetrahedron Lett. , 2007, 63, 1093-1099. 129. R. Lafont, J. Harmatha, F. Marion-Poll, L. Dinan, and I. Wilson, The Ecdysone Handbook, 3rd edition. http://ecdybase.org/, accessed January 30, 2010., . 130. P. A. Searle and T. F. Molinski, J. Nat. Prod., 1995, 58, 264-268. 131. A. Suksamrarn, A. Jankam, B. Tarnchompoo, and S. Putchakarn, J. Nat. Prod., 2002, 65, 1194-1197. 132. A. Guerriero, P. Traldi, and Pi, J. Chem. Soc. Chem. Commun., 1986, 40-41. 133. P. Parameswaran, C. Nair, C. Consalves, and C. Achuthankutty, J. Indian Inst. Sci., 2001, 81, 169-173. 134. A. Guerriero and F. Pietra, Comp. Biochem. Physiol. , 1985, 80B, 277-278. 135. H. Shigemori, Y. Sato, T. Kagata, and J. Kobayashi, J. Nat. Prod., 1999, 62, 372-374. 136. S. Jingyu, P. Tangshend, L. Kanghou, and Z. Longmei, Steroids, 1986, 48, 233-238. 137. B. Tursch, C. Hootele, D. Losman, M. Kaisin, and R. Karlsson, Steroids, 1976, 27, 137-142. 138. M. V. D'Auria, L. Minale, and R. Riccio, Chem. Rev., 1993, 93, 1839-1895. 139. R. Niwa, T. Matsuda, T. Yoshiyama, T. Namiki, K. Mita, Y. Fujimoto, and H. Kataoka, J. Biol. Chem., 2004, 279, 35942-35949. 140. J. H. Adler and R. J. Grebenok, Crit. Rev. Biochem. Mol., 1999, 34, 253-264. 141. C. Blais, C. Dauphin-Villemant, N. Kovganko, J. Girault, C. Descoins, and R. Lafont, Biochem. J., 1996, 320, 413-419. 142. R. Lafont and M. Mathieu, Ecotoxicology, 2007, 16, 109-130. 143. W. Behrens and D. Bückmann, Gen. Comp. Endocr., 1983, 51, 8-14. 144. K. Tomaschko, J. Chem. Ecol., 1994, 20, 1445-1455. 145. L. Dinan and R. Lafont, J. Endocrinol., 2006, 191, 1-8. 146. R. Lafont and J. Koolman, in Ecdysone: Structures and Functions, Springer, New York, 2009, pp. 47-70. 147. A. J. Smit, J. Appl. Phycol., 2004, 16, 245-262. 148. A. Hoffman, LSD - My Problem Child, McGraw-Hill, New York, 1980. 149. W. Chilton, J. Bigwood, and R. Jensen, J. Psychedelic. Drugs, 1979, 11, 61-69. 150. S. Baker, J. Monti, and S. Christian, in International Review of Neurobiology, Academic Press, New York, vol. 22, pp. 83-110. 151. S. Paredes, A. Korkmaz, L. Manchester, D. Tan, and R. Reiter, J. Exp. Bot., 2009, 60, 57-69. 152. A. Brossi, The Alkaloids: Chemistry and Pharmacology, Academic Press, New York, 1988. 153. V. Malikov and S. Y. Yunusov, in Chemistry of Natural Compounds, Springer, New York, 1977, pp. 497-512. 154. A. Morales, Int. J. Impot. Res., 2000, 12, S70-S74. 155. C. Delker, A. Raschke, and M. Quint, Planta, 2008, 227, 929-941. 156. C. D. Amsler, Algal chemical ecology, Springer, New York, 2009. 157. C. Hoek, D. Mann, and H. M. Jahns, Algae: an introduction to phycology, Cambridge University Press, Cambridge, 1995. 158. M. Guiry and G. Guiry, Algaebase. http://www.algaebase.org, accessed January 30, 2010., World-wide electronic publication, National University of Ireland, Galway. 159. V. Paul and W. Fenical, Bioorg. Mar. Chem., 1987, 1, 1-29. 160. A. Blackman and R. Wells, Tetrahedron Lett. , 1978, 19, 3063-3064. 161. V. Paul and W. Fenical, Mar. Ecol. Prog. Ser., 1986, 34, 157-169. 294  162. V. Jung and G. Pohnert, Tetrahedron Lett. , 2001, 57, 7169-7172. 163. V. Jung, T. Thibaut, A. Meinesz, and G. Pohnert, J. Chem. Ecol., 2002, 28, 2091-2105. 164. G. Aguilar-Santos, J. Chem. Soc. Perkin Trans. 1, 1970, 6, 842-843. 165. J. Su, Y. Zhu, L. Zeng, and X. Xu, J. Nat. Prod., 1997, 60, 1043-1044. 166. B. Maiti, R. Thomas, and M. Mahendran, J. Chem. Res. Synop., 1978, 126-127. 167. A. Anjaneyulu, C. Prakash, and U. Mallavadhani, Phytochemistry, 1991, 30, 3041-3042. 168. W. P. Jacobs, K. Falkenstein, and R. H. Hamilton, Plant Physiol., 1985, 78, 844-848. 169. M. Raob, J. H. Cardellina, and J. G. Schwede, Phytochemistry, 1987, 26, 619-620. 170. N. King and S. Thomas, Int. J. Biochem Cell. Biol., 2007, 39, 2167-2172. 171. H. Sugimoto, S. Oda, T. Otsuki, T. Hino, T. Yoshida, and Y. Shiro, Proc. Natl. Acad. Sci. USA, 2006, 103, 2611-2616. 172. T. Liitlejohn, O. Takikawa, R. Truscott, and M. Walker, J. Biol. Chem., 2003, 278, 2952529531. 173. W. P. Malachowski, R. Metz, G. C. Prendergast, and A. J. Muller, Drugs of the Future, 2005, 30, 897-909. 174. A. Mellor and D. Munn, Nat. Rev. Immunol, 2004, 4, 762-774. 175. D. Munn and A. Mellor, Trends Mol. Med., 2004, 10, 15-18. 176. D. Munn and A. Mellor, Immunol. Rev., 2006, 213, 146-158. 177. X. Zheng, J. Koropatnick, M. Li, X. Zhang, F. Ling, X. Ren, X. Hao, H. Sun, C. Vladau, J. Franek, B. Feng, B. Urqhart, R. Zhong, D. Freeman, B. Garcia, and W. Min, J. Immunol., 2006, 177, 5639-5646. 178. D. Hou, A. J. Muller, M. Sharma, J. DuHadaway, T. Banerjee, M. Johnsom, A. Mellor, G. Prendergast, and D. Munn, Cancer Res., 2007, 67, 792-801. 179. G. Brandacher, A. Perathoner, R. Ladurner, S. Schneeberger, P. Obrist, C. Winkler, E. Werner, G. Werner-Felmayer, H. Weiss, G. Gobel, R. Margreiter, A. Konigsrainer, D. Fuchs, and A. Amberger, Clin. Cancer Res., 2006, 12, 1144-1151. 180. A. Okamoto, T. Nikaido, K. Ochiai, S. Takakura, M. Saito, N. Ishii, N. Yanaihara , K. Yamada, O. Takikawa, R. Kawaguchi, S. Isonishi, T. Tanaka, and M. Urashima, Clin. Cancer Res., 2005, 11, 6030-6039. 181. N. Yoshida, K. Ino, Y. Ishida, H. Kajiyama, E. Yamoto, K. Shibata, M. Terauchi, A. Nawa, H. Akimoto, O. Takikawa, K. Isobe, and F. Kikkawa, CLin. Cancer Res., 2008, 14, 72517259. 182. J. Kim and D. Choi, Neuroscience, 1987, 23, 423-432. 183. M. Heyes , B. Brew, A. Martin, R. Price, A. Salazar, J. Sidtis, J. Yergey, M. Mouradian, A. Sadler, and J. Keilp, Ann. Neurol., 1991, 29, 202-209. 184. G. Guillemin and B. Brew, Redox Rep., 2002, 7, 199-206. 185. M. Heyes , M. Saito, J. Crowley, L. Davis, M. Demitrack, M. Der, L. Dilling, J. Elia, M. Kruesi, A. Lackner, S. Larsen, K. Lee, H. Leonard, S. Markey, A. Martin, S. Milstein, M. Mouradian, M. Pranzatelli, B. Quearry, A. Salazar, M. smith, S. Strauss, T. Sunderland, S. Swedo, and W. Tourtellotte, Brain, 1992, 1115, 1249-1273. 186. G. Guillemin, B. Brew, C. Noonan, O. Takikawa, and K. Cullen, Neuropathol. Appl. Neurobiol., 2005, 31, 395-404. 187. O. Takikawa, T. Littlejohn, and R. Truscott, Exp. Eye Res., 2001, 72, 271-277. 188. N. Serbecic and S. Beutelspacher, Exp. Eye Res., 2006, 82, 416-426. 189. H. Malina and X. Martin, Eur. J. Ophthalmol., 1996, 6, 250-256. 190. L. Goldstein, M. Leopold, X. Huang, C. Atwood, A. Saunders, M. Hartshorn, J. Lim, K. Faget, J. Muffat, R. Scarpa, L. Chylack, E. Bowden, R. Tanzi, and A. Bush, Biochemistry, 295  2000, 39, 7266-7275. 191. O. Takikawa, Int. Cong. Ser., 2007, 1304, 290-297. 192. N. Eguchi, Y. Watanabe, K. Kawanishi, Y. Hashimoto, and O. Hayaishi, Arch. Biochem. Biophys., 1984, 232, 602-609. 193. D. Hou, A. J. Muller, M. Sharma, J. DuHadaway, T. Banerjee, M. Johnsom, A. Mellor, G. Prendergast, and D. Munn, Cancer Res., 2007, 67, 792-801. 194. L. Stefan, A. Konigsrainer, H. Rammensee, O. Gerhard, and P. Terrness, Nat. Rev. Cancer, 2008, 9, 445-452. 195. A. J. Muller, J. DuHadaway, P. Donover, E. Sutanto-Ward, and G. C. Prendergast, Nat. Med., 2005, 11, 312-319. 196. E. Vottero, Doctorate, The University of British Columbia, 2006. 197. O. Takikawa, T. Kuroiwa, F. Yamazaki, and R. Kido, J. Biol. Chem., 1988, 263, 2041-2048. 198. A. Pereira, E. Vottero, M. Roberge, A. G. Mauk, and R. J. Andersen, J. Nat. Prod., 2006, 69, 1496-1499. 199. H. C. Brastianos, E. Vottero, M. Roberge, A. G. Mauk, and R. J. Andersen, J. Am. Chem. Soc., 2006, 128, 16046-16047. 200. M. Volgraf, J. Lumb, H. Brastianos, G. Carr, M. Chung, M. Munzel, A. Mauk, and R. Andersen, Nat. Chem. Biol., 2008, 4, 535-537. 201. G. Carr, W. Tay, H. Bottriell, S. Andersen, A. Mauk, and R. Andersen, Org. Lett., 2009, 11, 2996-2999. 202. E. Vottero, A. Balgi, K. Woods, S. Turgendreich, T. Melese, and R. J. Andersen, Biotechnol. J., 2006, 1, 282-288. 203. J. Simon and A. Bedalov, Nat. Rev. Cancer, 2004, 4, 481-492. 204. W. Mager and J. Winderickx, Trends Pharmacol. Sci., 2005, 26, 265-273. 205. H. K. Koblish, M. J. Hansbury, K. J. Bowman, G. Yang, C. L. Neilan, P. J. Haley, T. C. Burn, P. Waeltz, R. B. Sparks, E. W. Yue, A. P. Combs, P. A. Scherle, K. Vaddi, and J. S. Fridman, Mol Cancer Ther, 2010, 9, 489-498.  296  Appendix A NMR Spectra Of Selected Compounds For Chapter 3  297  O  O Cl  O  2 0.93ppm d (7) 3H  2 3 .9 p p m 1 5 4 .4 p p m  1 7 0 .4 p p m  0.89ppm d (7) 3H  1 8 .1 p p m 1 4 .8 p p m  2.40ppm d sep (7.0, 3.8) 1H  2 8 .5 p p m  2.54ppm s 2H  6 3 .4 p p m  4.23ppm td (8.7, 3.2) 1H  5 8 .5 p p m  4.27ppm t (8.7) 1H 4.43ppm dt (8.7, 3.6) 1H  N  Figure A.1 1H and 13C NMR of 2, recorded in CDCl3 -d4 at 400MHz and 100MHz, respectively.  298  N Cl  H N  O  O  5.83 ppm s 2H  4.61 ppm s 2H  s  1 3 6 .5 p p m  1 3 2 .0 p p m  1 4 1 .8 p p m  1 7 8 .0 p p m  7.31 ppm m 5H  7 3 .7 p p m 7 0 .9 p p m  7.67 ppm s 1H  8  1 2 7 .9 p p m 1 2 7 .5 p p m 1 2 7 .1 p p m  9.67 ppm s 1H  Figure A.2 1H and 13C NMR of 8 recorded in CD2Cl2 -d4 at 400MHz and 100MHz, respectively.  299  O  OH O N Cl  N  O  N Cl BOM 9  1 9 .7 p p m 1 6 .8 p p m  0.86 d (7.1) 6H  2.30 d sep (7, 3.9) 1H  1 4 0 .0 p p m 1 3 5 .9 p p m  1 5 6 .1 p p m  1 7 0 .0 p p m  1 3 1 .1 p p m 1 3 0 .6 p p m 1 2 9 .2 p p m  5.47 d (4.5) 2H  3 0 .7 p p m  6.02 ppm d (8.3) 1H  7.30 ppm m 5H  4.29 ppm 4.52 ppm m d (4.5) 1H 4.20 ppm 5.24 ppm 2H m d (8.3) 2H 1H  7 6 .o p p m 7 3 .0 p p m 6 7 .6 p p m 6 6 .5 p p m 6 1 .0 p p m 5 9 .7 p p m  6.91 ppm s 1H  Figure A.3 1H and 13C NMR of 9 recorded in (CD3)2CO-d6 at 400MHz and 100MHz, respectively.  300  OH O N  1 2 5 .8 p p m  4.70 ppm 4.59 ppm d (8.1) d (4.7) 1H 2H  1 3 6 .3 p p m 1 3 2 .9 p p m  1 7 0 .2 p p m  5.29 ppm d (8.1) 1H  6 5 .2 p p m  5.56 ppm d (4.7) 2H  10  1 2 7 .7 p p m 1 2 7 .3 p p m 1 2 7 .2 p p m  7.03 ppm s 1H  6 0 .8 p p m  7.34 ppm m 5H  7 2 .7 p p m 7 0 .1 p p m  Cl  NH2 N Cl BOM  Figure A.4 1H and 13C NMR of 10 recorded in CD2Cl2 at 400MHz and 100MHz, respectively.  301  5.57 ppm 5.44 ppm d (11) d (11) 1H 1H  1 3 7 .5 p p m 1 3 7 .3 p p m 1 3 3 .3 p p m  15  1 4 .3 p p m  N BOM  2 3 .0 p p m  Cl  2 8 .6 p p m  N  0.91 ppm t (6.8) 3H  3 5 .4 p p m  OH  1.89 ppm q (7.5) 2H  4.55 ppm s 2H  6 5 .1 p p m  4.69 ppm t (6.8) 1H  6.87 ppm s 1H  7 4 .2 p p m 7 1 .1 p p m  1 2 8 .9 p p m 1 2 8 .5 p p m 1 2 8 .2 p p m 1 2 5 .7 p p m  7.31 ppm m 5H  1.49 ppm 1.36 ppm m m 1H 3H  Figure A.5 1H and 13C NMR of 15 recorded in CD2Cl2- d4 at 400MHz and 100MHz, respectively.  302  OH O N Cl  NH2 Cl 16  4.67 ppm 1H 4.56 ppm 1H 4.47 ppm 1H  1 1 8 .2 p p m  4 9 .9 p p m 4 9 .2 p p m  7.42 ppm 1H  7 0 .1 p p m 7 0 .0 p p m  4.77 ppm 1H  1 3 1 .8 p p m 1 3 0 .7 p p m  1 7 3 .3 p p m 1 7 3 .2 p p m  7.51 ppm 1H  NH  Figure A.6 1H and 13C NMR of 16 recorded in CD3OD-d4 at 400MHz and 100MHz, respectively.  303  OH O N  5.42 ppm d (4.2) 1H  7.41 ppm s 1H  4.87 ppm 1H  1 1 5 .4 p p m 1 1 3 .4 p p m  i  1 3 3 .0 p p m  1 6 9 .8 p p m  6 5 .8 p p m  8.21 ppm s 1H  Cl 17  6 0 .4 p p m  NH  NH2  Figure A.7 1H and 13C NMR of 17 recorded in CD3OD-d4 at 400MHz and 100MHz, respectively.  304  OH N NH  Cl  18 1.79 1.37 q (6.7) m 5H 2H  i i  1 3 .9 p p m  3 4 .6 p p m  7 5 .5 p p m 1 1 9 .0 p p m  1 3 1 .7 p p m  1 3 7 .0 p p m  0.93 t (7) 3H  2 3 .0 p p m  4.72 t (6.5) 1H  2 8 .5 p p m  7.42 s 1H  Figure A.8 1H and 13C NMR of 18 recorded in D2O at 400MHz and 100MHz, respectively.  305  OH N  0.91 ppm t (7) 3H  1 4 .2 p p m  2 2 .9 p p m  2 8 .3 p p m  4.65 ppm t (6.9) 1H  6 8 .0 p p m  6.94 ppm s 1H  1.21-1.41 ppm m 1.81 ppm 5H m 2H  1 1 7 .0 p p m  1 3 6 .3 p p m  1 4 1 .3 p p m  7.61 ppm s 1H  19  3 6 .8 p p m  NH  Figure A.9 1H and 13C NMR of 19 recorded in D2O-d2 at 400MHz and 100MHz, respectively.  306  N 22  2.96 ppm 2.63 ppm t (7.6) t (7.9) 2H 2H  1 2 5 .9 p p m  1 4 7 .5 p p m  1 0 9 .1 p p m  6.61 ppm s 1H  1.98 ppm p (7.5) 2H  2 5 .9 p p m 2 1 .2 p p m  NH  3 8 .6 p p m  H2N  NH2  Figure A.10 1H and 13C NMR of 22 recorded in MeOD-d4 at 400MHz and 100MHz, respectively.  307  Appendix B Biological Data For Compounds In Chapter 3  308  Figure B.11: Bioassay data for the natural product fraction from which giroline(1) was isolated.  309  Figure B.12: Bioassay data, compound (16)  310  Figure B.13: Bioassay data, compound (17)  311  Figure B.14: Bioassay data, compound (18)  312  Figure B.15: Bioassay data, compound (19)  313  Figure B.16: Bioassay data, crude reaction mixture, compound (22) prior to recrystallization.  314  Figure B.17: Bioassay data, pure, recrystallized compound (22).  315  Figure B.18: Bioassay data, massacreone C (27)  316  Figure B.19: Bioassay data, massacreone D (28)  317  Figure B.20: Bioassay data, 20-hydroxy-ecdysone (29)  318  

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.24.1-0059410/manifest

Comment

Related Items