UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Isolation, characterisation, and total synthesis of secondary metabolites from marine sponges Linington, Roger Gareth 2004

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

Item Metadata

Download

Media
831-ubc_2004-994333.pdf [ 10.63MB ]
Metadata
JSON: 831-1.0061160.json
JSON-LD: 831-1.0061160-ld.json
RDF/XML (Pretty): 831-1.0061160-rdf.xml
RDF/JSON: 831-1.0061160-rdf.json
Turtle: 831-1.0061160-turtle.txt
N-Triples: 831-1.0061160-rdf-ntriples.txt
Original Record: 831-1.0061160-source.json
Full Text
831-1.0061160-fulltext.txt
Citation
831-1.0061160.ris

Full Text

ISOLATION, C H A R A C T E R I S A T I O N , A N D T O T A L S Y N T H E S I S OF S E C O N D A R Y METABOLITES FROM MARINE S P O N G E S  by  R O G E R G A R E T H LININGTON  B. Sc., The University of Leeds, 1998  A T H E S I S S U B M I T T E D IN PARTIAL F U L F I L M E N T O F THE REQUIREMENTS FOR THE D E G R E E OF D O C T O R OF P H I L O S O P H Y  in THE FACULTY OF GRADUATE STUDIES Department of Chemistry  W e accept this thesis as conforming to the required standard  T H E UNIVERSITY O F BRITISH C O L U M B I A  June 2004  © Roger Gareth Linington, 2004  II  ABSTRACT  Investigations into the chemistry of the marine sponge Caminus  sphaeroconia  have led to the isolation of the glycolipids caminosides A - D (35, 42, 43, 44). Bioassay guided fractionation showed this class of compounds to be active in a screen designed to identify type III secretion inhibitors. The type III secretion system is the mechanism by which a number of pathogenic bacteria achieve invasion of target human host cells. Compounds capable of disrupting the type III secretion system have the potential to eliminate this pathogenic behaviour, and it has been proposed that small molecules exhibiting this activity could find future utility as antibacterial therapeutics. OH  Caminoside Caminoside Caminoside Caminoside  A B C D  (35): (42): (43) (44)  R=Ac, R=Bu, R=Ac, R=Bu,  R'=H R'=H R'=Bu R'=Bu  Previous investigations into the chemistry of the marine sponge Stylissa carteri led to the isolation of two novel alkaloids, latonduines A and B (66 and 67). The paucity of structural information provided by one- and two-dimensional NMR experiments and mass spectrometric analyses allowed only a tentative structural assignment to be made.  Ill  Completion of a total synthesis of 66 has confirmed the proposed structure and provided material for further biological testing.  Latonduine A (66)  Latonduine B (67)  Despite structural similarities with a number of previously reported biologically active marine alkaloids, 66 exhibited no biological activity against a range of protein kinases. In an attempt to generate a novel protein kinase inhibitor by rational drug design, synthesis of a regioisomer of 66, isolatonduine A (70), was attempted. Progress towards the synthesis of 70 will be presented.  Isolatonduine A (70)  iv T A B L E OF CONTENTS  Abstract  '.  ii  Table of Contents  iv  List of Tables  vii  List of Figures  viii  List of Schemes.....  xii  List of Abbreviations  xiii  Acknowledgements  xvi  Chapter 1: General Introduction  1  1.1. Introduction  1  1.2. Drugs from the Sea  2  1.3. The Rise in Antibiotic Resistance  4  1.4. Antibiotics from Marine Sponges  7  1.5. References  12  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia  13  2.1. Introduction  13  2.2. Methods of Bacterial Infection  13  2.3. Enterop_athogenic E. coli ( E P E C ) as a Therapeutic Target  16  2.4. The Type III Secretion Inhibitor Screen  19  2.5. Overview of Glycolipids from Marine Sponges  20  2.5.1. Glycoglycerolipids  20  2.5.2. Glycosphingolipids  23  2.5.3. Steroidal Glycosides  26  2.5.4. Other Glycolipids  26  2.6. Isolation of and Structural Determination of Caminoside A Peracetate (34)  28  2.7. Experimental  59  2.7.1. General Experimental Procedures  59  2.7.2. Isolation Procedure  60  2.7.3. Caminoside A Peracetate (34)-Physical Data  61  2.7.4. Methanolysis of Caminoside A Peracetate (34)  61  V  2.7.5. |3-1-Methyl-2,3,4,6-0-acetyl-L-glucose (36) Standard  62  2.7.6. a-1-Methyl-2,3,4-0-acetyl-D-quinovose (38) Standard  62  2.8. References Chapter 3: Isolation and Characterisation of Caminosides B-D  63 65  3.1. Introduction  65  3.2. Isolation and Characterisation of Caminoside B (42)  65  3.3. Isolation and Characterisation of Caminoside C (43)  80  3.4. Isolation and Characterisation of Caminoside D (44)  86  3.5. Biological Activities of Caminosides A - D  99  3.6. Chemical Ecology  100  3.7. Conclusions  104  3.8. Experimental  105  3.8.1. General Experimental Procedures  105  3.8.2. Isolation of Caminosides B (42), C (43) and D (44)  105  3.8.3. Isolation of Caminoside B Peracetate (45)  106  3.8.4. Isolation of Caminoside D Peracetate (46)  106  3.9. References Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa carteri  108 109  4.1. Introduction  109  4.2. Isolation and Characterisation of Latonduines A (66) and B (67)  111  4.3. Total Synthesis of Latonduine A (66)  120  4.3.1. Introduction  120  4.3.2. Overview of Oroidin Alkaloid Syntheses  122  4.3.3. Proposed Synthesis for Latonduine A (66)  124  4.3.4. Synthetic Results  128  4.4. Biological Activities of the Latonduines  138  4.5. Biogenetic Implications of Latonduines Isolation  138  4.5.1. Introduction 4.5.2. Biogenesis of Latonduines A (66) and B (67)  138 ...141  4.6. Conformational Analysis of Latonduines A (66) and B (67)  143  4.7. Experimental  150  4.8. References  159  vi Chapter 5: Progress Towards the Synthesis of Isolatonduine  161  5.1. Introduction  161  5.2. Overview of Oroidin Alkaloid Biological Activities  161  5.3. Pharmacophore Model of Functional Group Spatial Distribution  162  5.4. Retrosynthetic Analysis for Isolatonduine (70)  171  5.5. Synthetic Results  173  5.6. Future Directions  183  5.7. Experimental  184  5.8. References  192  Summary  196  vii LIST O F T A B L E S  Table 2.1.  Human pathogens that employ the type III secretion system  15  Table 2.2.  1D and 2D N M R data for caminoside A (35)  35  Table 2.3.  1D and 2D N M R data for caminoside A peracetate (34)  42  Table 3.1.  1D and 2D N M R data for caminoside B (42)  74  Table 3.2.  1D and 2D N M R data for caminoside B peracetate (45)  78  Table 3.3.  1D and 2D N M R data for caminoside C (43)  84  Table 3.4.  1D and 2D N M R data for caminoside D (44)  92  Table 3.5.  1D and 2D N M R data for caminoside D peracetate (46)  97  Table 3.6.  Biological activities of caminosides A - D (pg/disc)  99  Table 5.1.  Kinase enzyme inhibitory activity of selected oroidin alkaloids  168  viii LIST O F FIGURES  Figure 1.1.  Origin of all commercially available small molecule therapeutics, 1981-2002 (N = 877)  Figure 1.2.  1  Examples of marine natural products currently in late-stage clinical trials  2  Figure 1.3.  Annual number of publications pertaining to "antibiotic resistance"  5  Figure 1.4.  Anti-microbial compounds from marine sponges (part 1)  8  Figure 1.5.  Anti-microbial compounds from marine sponges (part 2)  9  Figure 2.1.  Secretion systems for bacterial attachment to host cells.;  Figure 2.2.  The roles of EspA, EspB and EspD in the insertion of bacterial  14  proteins into the host cell  17  Figure 2.3.  Glycoglycerolipids from marine sponges  21  Figure 2.4.  Crasserides (26) and isocrasserides (27)  22  Figure 2.5.  Structure of glycosphingolipid 28  24  Figure 2.6.  Structures of plakosides A (29) and B (30)  25  Figure 2.7.  Structures of the simplexides (31), erylusamines (32) and plaxyoside (33)  27  Figure 2.8.  Structure of caminoside A (35) and caminoside A peracetate (34)  28  Figure 2.9.  Photograph of Caminus sphaeroconia  30  Figure 2.10.  Expansion of the anomeric region of the H S Q C spectrum for 34  31  Figure 2.11.  Expansion of the carbohydrate region of the C O S Y spectrum for 34  32  H N M R of caminoside A (35) acquired at 500 M H z  33  Figure 2.12. Figure 2.13. Figure 2.14. Figure 2.15. Figure 2.16.  1  1 3  1  C N M R of caminoside A (35) acquired at 100 MHz  H N M R of caminoside A peracetate (34) acquired at 800 MHz  1 3  C N M R of caminoside A peracetate (34) acquired at 100 MHz  38  39  H S Q C spectrum of caminoside A peracetate (34) acquired at 800 MHz  Figure 2.18.  37  C O S Y spectrum of caminoside A peracetate (34) acquired at 800 MHz  Figure 2.17.  34  40  H M B C spectrum of caminoside A peracetate (34) acquired at 400 MHz  41  ix Figure 2.19.  Expansion of the carbohydrate region of the 2 D - J R E S spectrum for 34  Figure 2.20.  1 3  46  C chemical shifts for the sugar residues of caminoside A  peracetate (34) Figure 2.21.  48  H chemical shifts and coupling constants for the sugar residues of  1  caminoside A peracetate (34)  48  Figure 2.22.  H R E I M S fragmentation of aglycon 40  50  Figure 2.23.  1D-TOCSY experiments for 34, irradiating at 5 0.94  52  Figure 2.24.  1 D-TOCSY experiments for 34, irradiating at 5 2.03  53  Figure 2.25.  1 -Methyl-2,3,4,6-0-acetyl-glucose comparison  55  Figure 2.26.  1-Methyl-2,3,4-0-acetyl-quinovose comparison  56  Figure 2.27.  Selected JCH H M B C correlations for caminoside A (35)  58  Figure 2.28.  Expansion of the carbohydrate region of the H M B C for 34  58  Figure 3.1.  Caminoside B (42)  65  Figure 3.2.  Expansion of the anomeric region for the H M Q C for 42  67  Figure 3.3.  Expansion of the carbohydrate region of the H M B C for 42 (part 1)  68  Figure 3.4.  Expansion of the carbohydrate region of the H M B C for 42 (part 2)  68  Figure 3.5.  Caminoside B peracetate (45)  69  Figure 3.6.  Selected JHH coupling constant analyses for 45  70  Figure 3.7.  Aglycon structural fragments for 42  70  Figure 3.8.  Comparison of the H NMR spectra of caminosides A (35) and B  3  3  1  (42) Figure 3.9.  72  Comparison of the C NMR spectra of caminosides A (35) and B 1 3  (42) Figure 3.10. Figure 3.11.  1  73  H N M R of caminoside B peracetate (45) acquired at 500 MHz  1 3  C N M R of caminoside B peracetate (45) acquired at 100 MHz  Figure 3.12.  Caminoside C (43)  Figure 3.13.  H S Q C spectrum for caminoside C (43) acquired in DMSO-c/6 at 800  Figure 3.15.  77 80  MHz Figure 3.14.  76  81  Comparison of the H NMR spectra of caminosides A (35) and C 1  (43)  83  Caminoside D (44)  86  X  Figure 3.16. Expansions of the carbonyl regions of the C N M R spectra for 35 1 3  and  .87  44  Figure 3.17. Caminoside D peracetate (46)  87  Figure 3.18. Expansion of the carbonyl region of the H M B C spectrum for 44  88  Figure 3.19. Comparison of the H NMR spectra of caminosides A (35) and D 1  90  (44)  Figure 3.20. Comparison of the C NMR spectra of caminosides A (35) and D 1 3  (44)  91  Figure 3.21. H N M R of caminoside D peracetate (46) acquired at 500 MHz  95  Figure 3.22.  96  1  1 3  C N M R of caminoside D peracetate (46) acquired at 100 MHz  Figure 3.23. Photograph of Caminus sphaeroconia  100  Figure 3.24. ESI mass spectra for constituents of Caminus sphaeroconia  103  Figure 4.1.  Previously reported metabolites from Stylissa carteri  109  Figure 4.2.  Novel secondary metabolites from marine sponges of the genus Stylissa  Figure 4.3.  110  Previously reported secondary metabolites re-isolated from marine sponges of the genus Stylissa  110  Figure 4.4.  Photograph of Stylissa carteri  112  Figure 4.5.  Latonduines A (66) and B (67)  112  Figure 4.6.  H N M R spectrum of latonduine A (66) recorded in DMSO-d6 at 500  1  MHz Figure 4.7.  1 3  C N M R spectrum of latonduine A (66) recorded in DMSO-cf6 at 100  MHz Figure 4.8.  1  115  H N M R spectrum of latonduine B ethyl ester (69), DMSO-d6 at 500  MHz Figure 4.9.  114  1 3  116  C N M R spectrum of latonduine B ethyl ester (69), DMSO-d6 at 100  MHz  117  Figure 4.10. H N M R spectrum of latonduine B methyl ester (68), DMSO-d6, 500 1  MHz Figure 4.11. Latonduine A (66) and its reg/oisomer 70  118 121  Figure 4.12. Possible interpretations for selected H M B C correlations for latonduine A (66)  121  xi Figure 4.13. Chronology of reported oroidin alkaloid syntheses since 1971  123  Figure 4.14. Dibromoisophakellin (72) and dibromophakellstatin (73)  125  Figure 4.15. Steric interactions present in hymenin (49) and latonduine A (66)  126  Figure 4.16. Comparison of the H NMR spectra for synthetic and naturally 1  occurring latonduine A (66) Figure 4.17. Predicted structures for future oroidin alkaloids  137 143  Figure 4.18. Lowest energy conformation for 69 as predicted by MM2 calculations  144  Figure 4.19. D N M R results for latonduine B ethyl ester (69), DMSO-c/6 at 500 MHz  145  Figure 4.20. D N M R results for latonduine B methyl ester (68), DMSO-c/6 at 500 MHz  148  Figure 4.21. D N M R results for latonduine A (66) run in MeOD-c/4 at 500 MHz  149  Figure 5.1.  Oroidin alkaloids shown to exhibit anti-tumour activity (part 1)  163  Figure 5.2.  Oroidin alkaloids shown to exhibit anti-tumour activity (part 2)  164  Figure 5.3.  Oroidin alkaloids shown to be inactive against selected cancer cell lines  165  Figure 5.4.  Oroidin alkaloids tested as inhibitors of the MEK-1 protein  167  Figure 5.5.  Isolatonduine A (70) and debromoisolatonduine A (113)  169  Figure 5.6.  Calculated minimum energy conformations for 56, 70 and 113  170  Figure 5.7.  Steric constraints to anti addition at C-9/C-10 of 76  174  Figure 5.8.  2,3-Dihydroisolatonduine (122)  178  xii LIST O F S C H E M E S  Scheme 2.1.  Degradation of caminoside A peracetate (34)  49  Scheme 4.1.  Biomimetic synthesis of dibromophakellin (71)  124  Scheme 4.2.  Non-biomimetic synthesis of (±) hymenin (49)  125  Scheme 4.3.  Proposed transformations for the synthesis of latonduine A (66)  127  Scheme 4.4.  Proposed mechanism for the dehydration of 78 to generate 77  127  Scheme 4.5.  Synthetic route to 76  129  Scheme 4.6.  Proposed formation of 77 via epoxide 81  130  Scheme 4.7.  Revised approach to the formation of ketone 77  131  Scheme 4.8.  General reactivity of diborane with olefins  132  Scheme 4.9.  Reg/oselectivity of the hydroboration of 76  133  Scheme 4.10.  Conversion of olefin 76 to ketone 77  135  Scheme 4.11.  Conversion of ketone 77 into latonduine A (66)  136  Scheme 4.12.  Representative cyclisation products formed via oroidin (51)  139  Scheme 4.13.  Literature proposal for the biogenesis of 84 and 85  140  Scheme 4.14.  Proposed biogenesis for latonduines A (66) and B (67)  142  Scheme 5.1.  Synthetic route to latonduine A (66)  172  Scheme 5.2.  Proposed route for the construction of isolatonduine A (70)  172  Scheme 5.3.  Transition states for the hydroboration and oxymercuration of 76  174  Scheme 5.4.  Alternative methods for the conversion of 76 to 115  175  Scheme 5.5.  Conversion of 76 to 117  176  Scheme 5.6.  Revised route to the synthesis of 115  177  Scheme 5.7.  Acid catalysed addition of triethylorthoformate to 120  179  Scheme 5.8.  Mechanisms for the conversion of 77 to 80 and 120 to 125....  180  Scheme 5.9.  C-Silation of 120 with TIPSCI under strongly basic conditions  181  Scheme 5.10.  Conversion of 120 to 128 with DMF-DMA  182  LIST O F A B B R E V I A T I O N S 0 .  1D 2D  N  2 5 D  Ac AcOH Ac 0 AIBN APT Ar 9-BBN br Br bs c °C calcd 2  2  CDCI3  (CD ) CO C D CH CH CH CH CI cm" COSY-gr 5 d D DCC dd DDQ deoxytal DEPT DMAP DMF DMF-DMA DNMR D 0 dt DMSO-c/6 Av 3  6  2  6  2  3  2  2  1  2  £  EHEC EPEC Et N 3  -degree(s) -one-dimensional -two-dimensional -specific rotation at wavelength of sodium D line at 25°C -acetate -acetic acid -acetic anhydride -2,2'-azobisisobutyronitrile -attached proton test -argon -9-borabicyclo[3.3.1]nonane -broad -bromine -broad singlet -concentration -degrees celcius -calculated -deuterated chloroform -deuterated acetone -deuterated benzene -methine -methylene -methyl -dichloromethane -wavenumbers -gradient selected two-dimensional correlation spectroscopy -chemical shift in parts per million -doublet -c/ex/rorotatory -1,3-dicyclohexylcarbodiimide -doublet of doublets -2,3-dichloro-5,6-dicyano-1,4-benzoquinone -deoxytalose -distortionless enhancement by polarisation transfer spectroscopy -4-(dimethylamino)pyrdine -A/,A/-dimethylformamide -dimethylformamide dimethyl acetal -dynamic nuclear magnetic resonance -deuterium oxide -doublet of triplets -deuterated dimethyl sulphoxide -chemical shift difference at T -extinction coefficient -enterohemorraghic E. coli -enteropathogenic E. coli -triethyl amine c  xiv Et 0 EtOAc FTIR 2  9 AG* G glu h HCI HIV-1 H 0 H 0 HMBC HMQC +  2  2  2  2  -diethyl ether -ethyl acetate -Fourier transform infra-red spectrometry -gram(s) -Gibbs free energy of activation - G phase in cell division -glucose -Plank constant -hydrochloric acid -human immunodefficiency virus-1 -water -hydrogen peroxide -two-dimensional heteronuclear multiple bond correlation spectroscopy -two-dimensional heteronuclear multiple quantum coherence spectroscopy -high-performance liquid chromatography -high-resolution electron impact mass spectrometry -high-resolution electrospray ionisation mass spectrometry -two-dimensional heteronuclear single quantum coherence spectroscopy -hertz -two-dimensional J resolved spectroscopy -coupling constant in hertz -rate constant -degrees Kelvin -Boltzmann constant -kilocalorie(s) per mole -potassium carbonate -kilojoule(s) per mole -/evorotatory -mouse lymphoma cells -lithium diisopropylamine -lithium borohydride -wavelength at maximum intensity -human colon cancer cells -low-resolution chemical ionisation mass spectrometry -low-resolution electron impact mass spectrometry -low-resolution electrospray ionisation mass spectrometry -multiplet -molecular ion -m-chloroperbenzoic acid -acetonitrile -Mitogen-activated protein kinase kinase-1 -deuterated methanol -methanol -methylsulphonic acid -milligram(s) -magnesium sulphate -megahertz  2  HPLC HREIMS HRESIMS HSQC Hz 2D-JRES J k K k kcalmol" B  1  K2CO3  kJmol" L L5178y LDA LiBH 1  4  ^max  L0V0  LRCIMS LREIMS LRESIMS m M mCPBA MeCN MEK-1 MeOD-c/4 MeOH MeS0 H mg MgS0 MHz +  3  4  MIC mL mm mmol mp MRS A m/z N N Na NaOH NBS NH nm NMR NSCLC-N6 PCC PCI Pd on C PPA PrOH p.s.i. q qui R ROESY s SAR SCUBA 2  +  4  5  n  SP , sp t T* 2  3  2  To  TFA THF TIPSCI TLC TOCSY  uv  VRE W w/v w/w  -minimum inhibitory concentration -millilitre(s) -millimetre(s) -millimol(s) -melting point -methicillin-resistant Staphylococcus aureus -mass to charge ratio -normal -nitrogen -sodium -sodium hydroxide -/V-bromosuccinimide -ammonium ion -nanometre(s) -nuclear magnetic resonance -human non-small lung cancer -pyridinium chlorochromate -phosphorus pentachloride -activated palladium on charcoal -polyphosphoric acid -1-propanol -pounds per square inch -quartet -quinovose -universal gas constant -rotating frame Overhauser enhancement spectroscopy -singlet -structure-activity relationship -self-contained underwater breathing apparatus -sp hybrid orbital -sp hybrid orbital -triplet -transverse relaxation -coalescence temperature -trifluoroacetic acid -tetrahydrofuran -triisopropylsilyl chloride -thin-layer chromatography -total correlation spectroscopy -microgram(s) -microlitre(s) -microsecond(s) -ultraviolet -vancomycin-resistant Enterococci -watt(s) -weight to volume ratio -weight to weight ratio 2  3  xvi ACKNOWLEDGEMENTS  My sincere thanks go to my supervisor, Dr Andersen, for welcoming me into his group so openly, and for providing me with the opportunity to learn about, and become fascinated with, marine natural products chemistry. Without his leadership I am sure I would be treading a very different path. I am indebted to Dave Williams for discussions about both science and the mountains, and have enjoyed the conversations that we have had on both fronts. He has proved to be an excellent companion, both underwater and on skis, and I have enjoyed the many trips that we have taken together. Mike LeBlanc has provided much assistance in the lab and in the field, where he has been a memorable S C U B A mentor. Through him I have learnt a lot about the art of collecting, often in sub-optimal conditions, and for this I am grateful. The companionship of the members of the Andersen lab has helped to make my time at U B C a very enjoyable one, and Todd, Rob, Kelsey, and Urmila deserve special mention in this regard. The contributions of the B C Brew Crew members are also duly acknowledged. I thank the staff of the NMR and M S facilities at UBC, and the staff of N A N U C in Alberta, for assistance with the acquisition of both N M R and mass spectra, and the staff of the Biological Services facility at U B C and the members of the Finlay group for performing a number of biological assays.  Finally, I would like to thank Mum, Dad, and Ian for their continued interest, encouragement, and enthusiasm throughout the course of my studies.  1  Chapter 1: General Introduction.  Chapter 1: General Introduction. 1.1. Introduction. Investigation of the chemistry of marine flora and fauna has blossomed into a mature science since the first comprehensive review of marine natural products chemistry in 1974. In this time, the field has broadened and diversified to encompass 1  topics  including:  chemical  ecology;  biosynthesis;  total  synthesis;  enzymology;  spectroscopy and genetics. The challenges faced in pursuing investigations in these areas have forced researchers to develop new methods and techniques that range from the development of new N M R experiments to probe stereochemical issues to the 2  invention of methods for culturing previously uncultivatable marine bacteria and fungi.  3  Underpinning this whole field is the premise that marine natural products represent an under-utilised and valuable resource for the development of novel therapeutics. This 4  concept has driven much of the effort in this field as researchers employ a wide variety of methods to screen for compounds that display specific biological activities, and is central to the work presented herein.  Natural Product Derivatives 27% ^  Synthetic 39%  Natural Products 6% Synthetic Derivatives Developed from Natural Product Lead Compounds 28% Figure 1.1. Origin of all commercially available small molecule therapeutics, 1981-2002 (N = 877).  5  Chapter 1: General Introduction.  2  1.2. Drugs from the Sea. The proposal that marine natural products have potential utility as drug leads is borne out by consideration of the pipeline of compounds of marine origin that are currently in clinical trials. Several recent reviews highlight the growing number of marine natural products that have reached advanced stages of clinical testing and offer predictions of increases in both the number of marine natural products as drug leads, and the proportion of overall drug leads derived from natural product sources. " It is 5  clear, despite the  recent shift away from natural products  research  9  by many  pharmaceutical companies, that this remains an important avenue of investigation in the efforts to control treat numerous serious diseases.  Ecteinascidin 743 (1)  Dehydrodidemnin B (2)  Figure 1.2. Examples of marine natural products currently in late-stage clinical trials.  10,11  One might reasonably question why marine invertebrates have been highlighted as such an important resource in the search for new drug leads. The answer lies in the consideration of the factors required for survival for all competing organisms in the marine environment. Just as in the terrestrial environment where plants accumulate  Chapter 1: General Introduction.  3  toxins in their leaves, stems and roots in order to defend themselves from both predation and attack by bacteria or fungi, so marine invertebrates must employ similar tactics for self-defense. Many marine invertebrates are insentient, immobile, and lack the physical protection of a hard outer shell. In these cases, the threat of predation by other organisms requires an alternative strategy for self-protection. One possibility is the sequestration or biosynthesis of chemical defenses at the surface of the organism, and there are numerous examples in the chemical ecology literature of compounds isolated from marine invertebrates that have subsequently been shown to exhibit anti-feedant properties against reef fish at biologically significant concentrations. " 12  15  An additional issue for invertebrates without locomotive ability is the threat of overgrowth by neighbouring organisms. Tropical reefs are often densely populated, to the point that the underlying substrate can be completely obscured by the high concentration of sponges, tunicates, soft corals and other organisms encrusting its surface. In these cases it becomes important for all organisms to defend their position on the reef, and therefore their access to both light and nutrients. This is another area 16  in which chemistry plays a defensive role, and again a number of compounds have been identified that exhibit anti-fouling activities. Finally, though marine invertebrates often contain large numbers of symbiotic and/or commensal bacteria, the potential exists for invasion by marine bacteria or fungi harmful to the host organism. The presence of selective anti-microbial compounds at the surface of the host safeguard it from infection, and while difficulties with culturing techniques for marine microbes have limited the opportunities for exploration of this hypothesis, a number of groups have alluded to this activity when exploring the ecological role of marine natural products.  17  Chapter 1: General Introduction.  4  If the range of biological roles required of marine natural products is combined with the number of different species currently identified in the marine environment (>300,000 in 1999),  18  it becomes obvious that the potential for chemical variability is  very high. Given the success that has been achieved by exploring terrestrial natural products as potential therapeutics, and given the current decrease in the number of 5  new chemotypes being identified annually from terrestrial plants and microbes, it seems clear that the future for marine natural products chemistry is bright indeed. 1.3. The Rise in Antibiotic Resistance.  In contrast to the optimism surrounding the future of marine natural products chemistry, the outlook for current anti-microbial treatments is anything but bright. Overconsumption and frequent mis-use of antibiotics since their discovery in the early 1940's has led to the development of numerous strains of multi-drug resistant bacteria. In the most serious cases, bacteria have been isolated that show resistance to all known classes of antibiotics. These include strains of Mycobacterium methicillin-resistant  Staphylococcus  aureus  (MRSA).  19  tuberculosis  Coupling  (TB) and  the increase in  immunity to current anti-microbial treatments with the recent closure of infectious disease research centres in a number of major pharmaceutical companies provides the potential for the emergence of large-scale infections by multi-drug-resistant bacteria amongst high-risk groups such as hospital patients, children, and the elderly. Doomsday predictions of this type may seem melodramatic, however both the scientific literature and the popular press have seen large increases in the number of articles that confront this issue over the last ten years. A search of the academic  5  Chapter 1: General Introduction.  literature' using the term "antibiotic resistance" provides 24724 responses, of which 7293 were published between 1941 (the year of the discovery of penicillin) and 1991, and the remaining 17431 were published between 1992 and 2004.  3000 w c g  o  2500 2000  0_ o n E  1500 1000  zs c c  <  500 0  i  i i M  1941  i M  i II  i  1951  i i  i i i i i  1961  1971  1981  1991  2001  Year Figure 1.3. Annual number of publications pertaining to "antibiotic resistance"  Performing the same search limited by year allows us to generate a plot showing the annual number of publications containing the term "antibiotic resistance" (figure 1.3). While these search criteria are somewhat crude, this graph does give some indication of the dramatic rise in interest in this topic over the last twelve years. 1992 can be considered the watershed year for this rise. In August of that year, Science published a special issue dedicated to the topic of antibiotic resistance. In an article entitled "The Crisis in Antibiotic Resistance" Harold C. Neu stated that:  Literature searches performed using the search engine SciFinder® Scholar.  6  Chapter 1: General Introduction.  "[t]he synthesis  of large numbers  of antibiotics  over the past three  decades has caused complacency about the threat of bacterial Bacteria have become resistant to antimicrobial chromosomal  resistance.  agents as a result of  changes or the exchange of genetic material via plasmids  and transposons  The extensive use of antibiotics in the community  and hospitals has fueled this crisis. Mechanisms such as antibiotic control programs,  better  antimicrobial  hygiene,  activity need  and  synthesis  to be adopted  of agents in order  with  improved  to limit bacterial  resistance".  20  A number of prominent scientists took up this message, and in the following years several books were written that aimed to bring the causes and effects of widespread antibiotic resistance to the attention of the general public. One of the most successful of these, "The Plague Makers" by Jeffrey A. Fisher M.D., presented the origins and mechanisms of bacterial resistance to anti-bacterial therapy and provided recommendations for policy initiatives to slow the rate of generation of resistant organisms. The book closed with the following words: 'This, then, with all its ramifications, is the problem facing us today. It is one of enormous magnitude. Bacterial resistance to antibiotics to develop inexorably,  every day and everywhere  continues  in the world, and the  number of patients with infections that can't be controlled mounts. Since this process is inherent in the molecular makeup of bacteria, we can't stop it entirely. But by beginning right now to use the antibiotics  appropriately,  we can slow it down. If we do, it will buy us some time to begin intense development  and implementation  of the alternative methods of thwarting  Chapter 1: General Introduction.  bacteria.  7  If we don't, the consequences  are almost too horrible to  imagine". ^ 2  The exponential growth in the number of scientific publications on this topic stands testament to the fact that both governments and scientists have indeed begun to take this subject seriously. With over 2100 publications in this field annually for the last three years the topic seems to have reached saturation point in terms of publication rate. This is not to suggest that further efforts are unnecessary. Sadly the substantial volume of work performed in the search for novel antibiotics in the last ten years has yielded few clinically  useful  results. Any explorations  into the identification of  compounds that act via novel modes of action are to be welcomed as these offer the potential for alternative methods of controling bacterial infection. It is within this area that a significant portion of the work presented in this thesis is focussed.  1.4. Antibiotics from Marine Sponges.  It is indisputable that the identification of strong biological activity for a given natural product increases its profile in the literature. Anti-microbial assays have long been a popular method for the identification of biological activity due to their simplicity and their robust nature. Because of the widespread use of assays of this type, a very large number of natural products have been reported as exhibiting anti-microbial activities.  Many  of these  compounds,  however,  are only  active  at very  high  concentrations (>100 u.g/ ml_), making their utility as therapeutics unlikely. If one reexamines the literature for compounds that show antibacterial activity at or below 20 u.g/ mL then a rather different picture emerges.  8  Chapter 1: General Introduction.  5-ep/-isospongiaquinone (4) S. aureus 20 ng/ disc  15a-methoxypuupehenol (3) S. aureus 1 ug/ disc  E. coli  Z-2,3-Dihydroneomanoalide (5) 1 ng (TLC autobiographic bioassay) OMe  H N O  W  NH  H N NH  N H  2  Bistellettadine A (6)  X  ©  2  NH  2  N  H  2  .4X  MeO  e  R =H  Bistellettadine B (7) R = E. coli 10 ng/ disc (6 + 7)  Cribrostatin 4 (8) S. pneumoniae 1 |ig/ mL  ,CH  0 CF C0 3  H  e.  '  I ^.NH NH  N  F R =R =H, R =Me, R =Et G Ri=R =R =Me, R =H H R ^ H , R =OH, R =R =Me 2  3  3  4  4  2  2  3  2  3  Aminozooanemonin (11) S. aureus 8.5 ng/ mL  O^NH  1  '^NH  CH  HN-  Dragmacidin D (10) £ coli 15.6 (ig/ mL  Cribrostatin 6 (9) S. aureus 16 ng/ mL S. pneumoniae 0.5 to 2 ng/ mL  2  3  O  4  Discodermin F-H (12, 13, 14) S. aureus 10 ng/ disc E. coli 10 ng/ disc  Figure 1.4. Anti-microbial compounds from marine sponges (part 1). 22  2  9  Chapter 1: General Introduction.  Phloeodictine A (15) S. aureus E. coli  1 ug/ mL 1 ug/ mL  Pseudoceratidine (17) S. aureus 4 ug/ mL E. co// 32 ug/ mL  Phloeodictine B S. aureus  (16)  3 ug/ mL  E. co//' 30 ug/ mL  Pyridinebetaine A S. aureus  (18)  5 ug/ mL OMe  Palau'amine (19) S. aureus 10 ug/ disc  Theoneberine (20) S. aureus 16 ug/ mL  Currently undergoing preclinical trials for anti-fungal, anti-tumour and immunosuppressive activities  Xestokerol A (21) S. aureus  16 ug/ mL  Wondosterols A (22) and C  (23)  E. coli 10 ug/ disc  Figure 1.5. Anti-microbial compounds from marine sponges (part 2 ) .  22  Chapter 1: General Introduction.  10  Figures 1.4 and 1.5 illustrate a cross-section of the sponge metabolites reported with anti-bacterial activities <20 u.g/ mL during the period 1990-1999. For comparison, typical in vitro MICs for fluoroquinolone antibiotics against non-resistant strains of S. pneumoniae  are in the 0.5 to 5 u,g/ mL range.  23  Of these structures, a number show  activity with comparable MICs to the fluoroquinolone antibiotics, while possessing simple skeletons that are seemingly amenable to large-scale synthesis. Despite this, as of 2003, there are no reports of marine natural products in clinical trials for their antibiotic activity.  4,7  The reasons for this lack of development are unclear. It is possible  that more detailed examinations of the biological effects of these compounds, either in vitro or in vivo, identified side effects that made them unsuitable as drug candidates. It is also possible that these compounds act via modes of action similar to those of currently available antibiotics. Were this the case, there would be little incentive to develop compounds that exhibit no therapeutic advantage over existing commercially available treatments. Due to the competitive nature of the drug discovery industry much of this information is unavailable to the public, making it impossible to do more than speculate upon the reasons why these compounds have not been subject to further pre-clinical development. Despite the apparent lack of progress in this area, the chemistry of marine sponges remains an important resource in the area of lead structure identification for antibiotic development. This claim is justified when one considers the number and diversity of compounds isolated over a ten year period that display antibiotic activity against one or more bacterial pathogens (figures 1.4 and 1.5). Rapid developments in molecular biology, coupled with continuous improvements in the understanding of the mechanisms by which bacterial achieve host cell invasion are providing novel and  Chapter 1: General Introduction.  11  selective targets for potential antibiotic therapeutics. Marine natural products chemistry is ideally suited to take advantage of the identification of these targets, and provides potential for the future evolution of clinically useful antibacterial therapeutics.  12  Chapter 1: General Introduction.  1.5. References. 1. Faulkner, D. J.; Andersen, R. J . In Sea, 1974; Vol. 5, pp 679-714. 2. Marquez, B. L; Gerwick, W. H.; Williamson, R. T. Magn. Reson. Chem. 2001, 39, 499-530. 3. Mincer, T. J.; Jensen, P. R.; Kauffman, C. A.; Fenical, W. Appl. Environ. Microbiol. 2002, 68, 5005-5011. 4. Haefner, B. Drug Discovery Today 2003, 8, 536-544. 5. Newman, D. J.; Cragg, G . M.; Snader, K. M. J. Nat. Prod. 2003, 66, 1022-1037. 6. Capon, R. J . Eur. J. Org. Chem. 2001, 633-645. 7. Proksch, P.; Edrada, R. A.; Ebel, R. Appl. Microbiol. Biotechnol. 2002, 59, 125-134. 8. Newman, D. J.; Cragg, G. M.; Snader, K. M. Nat. Prod. Rep. 2000, 17, 215-234. 9. Donia, M.; Hamann, M. T. Lancet Infect. Dis. 2003, 3, 338-348. 10. Erba, E.; Bergamaschi, D.; Bassano, L ; Damia, G.; Ronzoni, S.; Faircloth, G. T.; D'lncalci, M. Eur. J. Cancer 1990, 37, 97-105. 11. Urdiales, J . L ; Morata, P.; Nunez De Castro, I.; Sanchez-Jimenez, F. Cancer Lett. 1996, 102, 31-37. 12. Hay, M. E. J. Exp. Mar. Biol. Ecol. 1996, 200, 103-134. 13. Harborne, J . B. Nat. Prod. Rep. 1999, 16, 509-523. 14. Kats, L. B.; Dill, L. M. Ecoscience 1998, 5, 361-394. 15. McClintock, J . B.; Baker, B. J . Marine Chemical Ecology.; C R C , 2001. 16. Stachowicz, J . J.; Whitlatch, R. B.; Osman, R. W. Science 1999, 286, 1577-1579. 17. Kubanek, J . M.; Jensen, P. R.; Keifer, P. A.; Sullards, M. C ; Collins, D. O.; Fenical, W. P. Natl. Acad. Sci. U.S.A. 2003, 100, 6916-6921. 18. Pomponi, S. A. J. Biotechnol. 1999, 70, 5-13. 19. Amyes, S. G . B. In Magic Bullets, Lost Horizons; Taylor & Francis: London, 2001, pp 250-253. 20. Neu, H. C. Science 1992, 257, 1064-1073. 21. Fisher, J . A. The Plague Makers; Simon & Schuster: New York, 1994. 22. Kuniyoshi, M.; Higa, T. Recent Advances in Marine Biotechnology2001,  6, 21-83.  23. Madaras-Kelly, K. J.; Demasters, T. A. Diagn. Micr. Infec. Dis. 2000, 37, 253-260.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  13  sphaeroconia.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  2.1. Introduction. Enteropathogenic E. coli ( E P E C ) is a major cause of infantile diarrhea, causing significant mortality in children in developing countries due to dehydration, malnutrition and other complications of the disease. In an effort to identify compounds that inhibit the pathogenicity of bacteria such as E. coli, a library of sponge extracts were tested for their ability to prevent the attachment of bacteria to host cells. Caminus  sphaeroconia,  collected from Toucari Caves in Dominica, showed significant activity in this initial assay. Bioassay guided fractionation followed by N M R based structural elucidation led to the discovery of the glycolipids caminosides A - D.  2.2. Methods of Bacterial Infection. The invasive methods by which pathogenic bacteria achieve insertion of bacterial proteins into host cells has long been an area of interest in the field of microbiology. Recent studies " have shown there to be four principle pathways by which bacteria may 1  4  introduce effector proteins into the host cell. These pathways are known as secretion systems, and are illustrated in figure 2.1.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  14  Figure 2.1. Secretion systems for bacterial attachment to host cells (adapted from Donnenberg ). 5  In all cases, pathogenic bacteria secrete proteins into the bacterial supernatant that assist both in the attachment to the surface of the host cell and in the penetration of the host cell membrane. The four known secretion systems differ in that types II and IV require two separate processes to export these proteins from the cytoplasm while types I and III are able to achieve this export as a concerted process. In type II and IV secretions the required factors are first transported across the inner membrane into the periplasmic space before being secreted to the surface of the bacterial cell. By contrast types I and III secrete these proteins directly into the bacterial supernatant where they form an attachment point on the surface of the host cell and subsequently construct a  Chapter 2: Antimicrobial Metabolites from the Marine Sponge  Caminus sphaeroconia.  15  'needle' injector to deliver one or more effector proteins to the host which then modulate host cellular functions.  6  Particular emphasis has been placed upon understanding the type III secretion system because of the large number of human pathogens that employ this system to invade host cells (table 2.1). A substantial body of work performed over the last ten years has led to a deeper understanding of the functions of a number of proteins associated with type III secretion systems, although there are still many facets of the 7  process which remain poorly understood. One result of these investigations has been that for certain bacteria the functions of a number of proteins that are instrumental in the attachment phase of the secretion process (known as adhesin proteins) have been identified. A s will be seen in sections 2.3 and 2.4 this knowledge of adhesin protein function is critical to the development of selective antibacterial agents.  Table 2.1. Human pathogens that employ the type III secretion system. Consequence of Infection  Bacterial Species Enteropathogenic E. coli  Diarrhea in young children  Enterohemorraghic E. coli  Hemorrhagic colitis and hemoyitic-uremic syndrome  Shigella spp.  Dysentery  Yerisinia spp.  Plague and gastroenteritis  Chlamydia spp.  Sexually transmitted, respiratory, and ocular diseases  P. aeruginosa  Opportunistic infections  S. enterica  Food poisoning and typhoid fever  It is important to note at this point that only pathogenic bacteria utilise the type III secretion system for host cell attachment and that none of the bacteria that make up the commensal flora within the human system employ the type III secretion system for  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  16  interactions with the host cell. It is clear that such a marked difference between commensal and pathogenic bacteria in their methods of interaction with host cells makes this system an ideal candidate for developing selective therapeutic agents for treating and preventing harmful bacterial infections.  2.3. Enteropathogenic E. coli (EPEC) as a Therapeutic Target.  E.  coli represent  the predominant  anaerobic  species  present  iii human  commensal flora. Colonisation typically occurs within the first few hours of life, and the 9  resulting relationship is a symbiotic one, which continues for the lifetime of the human host. While the majority of E. coli strains are benign towards the human system, there are a number of strains which cause infective symptoms ranging from mild stomach upsets to chronic diarrhea, which in infants is responsible for thousands of deaths a year worldwide. The recent case of infection of the water supply in Walkerton, Ontario in which seven people were killed and a further thirteen hundred people infected with Enterohemorraghic E. coli (EHEC; a close relative of E P E C ) highlights the persistent threat to health of E. coli strains even in a developed nation in a population that was in good general health prior to the outbreak.  10  E P E C has been widely studied because of the significance of E P E C and E H E C infections in the healthcare industry and because of its amenability to current culturing techniques. One of the upshots of the use of E P E C as a model system for studying bacterial pathogenesis is that more is now known about the type III secretion system in E P E C than in any other bacterium, with the possible exception of Yersinia spp. (the bacterium responsible for the plague) which has also been extensively investigated because of it's use of the type III secretion system for host cell attachment.  11  Exploration of the type III secretion system in E P E C has led to the identification of a  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  17  suite of more than twenty proteins that have been implicated in host cell attack. Of these, three proteins in particular: EspA; EspB; and EspD (E. coli secreted protein) have garnered particular attention due to their respective functions in the formation of the needle injector. EspA has been identified as the protein used to construct the translocation tube,  12  whereas EspB and EspD together form a translocation pore in the  host cell membrane (figure 2.2).  13  CesT Chaperone  EPEC  P I  EPEC ner r. Membrane  Peptidoglycan EPEC Outer Membrane  iWififippiiin  m i  Esp A - Translocation Tube  Host Membrane Pore  Host Plasma Membrane  cytosol HOST  Figure 2.2. The roles of EspA, EspB and EspD in the insertion of bacterial proteins into the host cell (reproduced with permission from Gauthier ). 14  It has been shown that E P E C mutants that lack the gene responsible for the production of EspB show a marked decrease in their ability to cause diarrhea in adult  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  volunteers.  15  sphaeroconia.  18  In a remarkable experiment performed in the U S A in 2000, twenty  volunteers were fed solutions containing either wild-type E P E C or an isogenic AEspB mutant strain, which lacked the ability to secrete EspB. Over the following three days the subjects were monitored for signs of the development of diarrhea. Of the ten people fed wild-type E P E C , nine were shown to have contracted intestinal infection by E P E C in contrast to only one of the ten volunteers fed the mutant strain. This evidence of the importance of the presence of EspB for E P E C pathogenicity makes the production of the EspB protein an attractive target for chemical intervention. By screening for extracts that inhibit the secretion of EspB by E P E C , it was proposed that compounds could be isolated which would selectively inhibit pathogenic E. coli from attaching to host cells without disrupting the commensal flora, including the benign E. coli strains required in the gut for regular intestinal function. The attraction of developing inhibitors of EspB secretion is that the resulting therapeutics would be highly selective for bacteria employing the type III secretion system for host cell attachment. In contrast to current antibiotic treatments, which leave the patient susceptible to opportunistic infections by other species such as Staphylococcal bacteria,  9  enterocolitis due to the eradication of regulatory  it was hoped that type III secretion inhibitors would render pathogenic  bacteria harmless without invoking the side effects typical of these current treatments. In addition, because inhibition of EspB secretion has no selective pressure on viability for the target organism it was hoped that these theoretical therapeutics might be less susceptible to mutation driven resistance than current antibacterial treatments. Finally, by exploiting this novel mode of action it was hoped that inhibitors of the type III secretion system could show efficacy against strains of E. coli that show resistance to current antibiotic treatments.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  19  2.4. The Type III Secretion Inhibitor Screen.  In order to search for potential EspB secretion inhibitors, a novel screen was developed by Annick Gauthier and Marilyn Robertson from the laboratory of Professor B. Brett Finlay in the Department of Microbiology and Immunology at the University of British Columbia. In this new screen, E P E C cells were cultured for twenty four hours in the presence of a given extract and the resulting cellular suspension centrifuged to give a bacterial pellet and a supernatant containing all secreted proteins. The bacterial pellet was examined by eye and any extracts that caused a significant decrease in pellet size were discarded from the experiment on the grounds that these extracts contained compounds with traditional antibiotic or cytotoxic properties. The supernatants of the remaining samples were analysed for the presence of both EspB and EspC. E s p C is an E P E C secreted protein not involved in the type III secretion pathway. By analysing for the presence of both type III and non-type III derived proteins it is possible to isolate extracts which have a specific effect on the type III system from those extracts that disrupt general bacterial secretion. Extracts that showed decreased levels of EspB but similar levels of E s p C when compared with control cultures were therefore designated as genuine type III secretion inhibitors. From an initial screen of four hundred marine extracts the screening process identified eight positive hits, from which the marine sponge Caminus  sphaeroconia  was chosen for further investigation. Early N M R  evidence suggested the presence of one or more glycolipids in the crude extract. Glycolipids are well represented in marine natural products chemistry, and a summary of the general features of glycolipids from marine sponges is therefore presented below.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  20  2.5. Overview of Glycolipids from Marine Sponges. Glycolipid chemistry continues to draw attention in the field of marine natural products. The subject has warranted several reviews, " 16  18  and is regularly the focus of a  number of publications per annum. There exist a diverse array of structures that fall under the title of glycolipids. These compounds can be broadly divided into three subcategories: glycoglycerolipids; glycosphingolipids; and steroidal glycosides. In addition, there are also a smaller number of unusual glycolipids whose structures and likely biosynthetic origins mean that they do not belong to any of these categories.  2.5.1. Glycoglycerolipids. Glycoglycerolipids are defined as having a central glycerol unit with either a mono- or di-saccharide at one of the primary alcohol positions. The two other hydroxyl positions are also capped, either by fatty acid chains or by long chain alkanes. Glycoglycerolipids are ubiquitous in the cell membranes of organisms that are capable of photosynthesis, and as such are found extensively in algae and cyanobacteria. By contrast there are few examples of glycoglycerolipids isolated from sponges. 24, isolated from Phyllospongia  foliascens,™ and 25, isolated from Trikentrion  loeve,  20  represent two examples of this smaller subset (figure 2.3). The latter is unusual in that the  glycerol  moiety  is di-glycosylated,  in contrast  to the vast  glycoglycerolipids, which contain only one site of glycosylation.  majority of  Chapter 2: Antimicrobial Metabolites from the Marine Sponge  Caminus sphaeroconia.  21  S0 H 3  O  HO;  R  V  H O - ^ V ^ OH  0  6  0 25  24 Figure 2.3. Glycoglycerolipids from marine sponges.  Two other families of glyceroglycolipid-like molecules have also been identified from sponges. Although the crasserides (26)  20  and isocrasserides (27)  21  are not true  glyceroglycolipids due to the replacement of the hexose with a five-membered cyclic polyalcohol, they share many of the other characteristics of glyceroglycolipids and are therefore included in this overview (figure 2.4).  2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia.  Chapter  22  Figure 2.4. Crasserides (26) and isocrasserides (27).  There is also one recent paper describing the identification of glycoglycerolipids from a marine sponge-associated Microbacterium  species.  22  This report highlights the  ever present possibility that 24, 25, 26 and 27 could have originated from microbial symbionts within sponge samples. The  medicinal  chemistry  literature  contains  reports  of  a  number  of  glycoglycerolipids that exhibit biological activity. In recent years they have been shown to have a regulatory influence against mouse skin carcinoma c e l l s virus,  24  23  and the influenza A  and have been implicated in the replication pathway of the HIV-1 virus.  25  Although these bioactive glycoglycerolipids are all either synthetic or terrestrial in origin  Chapter  2 :  Antimicrobial Metabolites from the Marine Sponge  Caminus sphaeroconia.  23  these results show that there is a genuine interest in glycolipids as potential therapeutic agents.  2.5.2. Glycosphingolipids.  Glycosphingolipids differ from glycoglycerolipids in the replacement of the glycerol backbone with a long chain aminoalcohol commonly known as the sphingoid base. Similarly to glycoglycerolipids, glycosphingolipids contain an acyl fatty acid group, though attached to the nitrogen of the sphingoid  base via an amide linkage.  Glycosphingolipids also contain a carbohydrate portion, which exhibits a far higher degree of possible variation than is shown for glycoglycerolipids. The carbohydrate subunit for glycosphingolipids can contain anywhere from one to five sugars in either linear or branched chain arrangements. Glycosphingolipids are ubiquitous in both plant and animal cell membranes and are therefore much more widely distributed in sponges than glycoglycerolipids. While it is impractical to provide a comprehensive list of known glycosphingolipids from sponges, it is worth examining a couple of representative examples. A recent review  16  does present all marine glycolipids reported prior to 1997  and is recommended to readers who have an interest in this topic. Both examples have been chosen based upon their structural novelty and their interesting biological activities. The first, 28, was isolated from a sample of the sponge Aplysinella  rhax collected from the shores of New Caledonia as one of a number of  related glycosphingolipids isolated from the same sponge sample.  26  28 consists of a  trisaccharide head containing fucose, /V-acetylglucosamine and A/-acetylgalactosamine, a sphingoid base with an isopropyl group at its terminus, and an unbranched 2-hydroxy fatty acid. The presence of these unusual monosaccharides is not unique among glycosphingolipids, although they are not found in glycoglycerolipids. The sphingoid  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  24  base is representative of a type seen in many glycosphingolipids, as is the 2-hydroxy fatty acid.  Figure 2.5. Structure of glycosphingolipid 28.  28 has been tested for its activity as a nitric oxide release inhibitor. Nitric oxide has been identified as a signaling molecule in animal systems with implications for human health in areas that include heart disease, cancer, lung disease, impotence, and treatment of shock. Three prominent researchers in the field, Robert F. Furchgott, Louis J. Ignarro, and Ferid Murad, shared the 1998 Nobel Prize in Physiology or Medicine for their efforts in identifying the role of nitric oxide in intercellular signalling processes. In tests 28 exhibited good inhibitory activity on lipopolysaccharide (LPS)-induced NOV release by mammalian cells, with a MIC of 10 ^g/ mL. The second glycosphingolipid example was identified in 1997 from the marine sponge Plakortis  simplex.  27  In contrast to many glycolipid isolations, which report  compounds as mixtures of homologues that differ in the constitution of the acyl chains, plakosides A (29) and B (30) were isolated as chemically homogenous samples. In this case, the carbohydrate head consisted of a single galactose residue with an isoprene unit attached to the oxygen at C-2'. The alkyl chains of both the sphingoid base and the  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  25  acyl chain contained cyclopropyl rings whose positions were determined by chemical degradation. It was not possible to determine the absolute stereochemistry of these cyclopropyl rings; however it was possible to assign the ring substituents as c/'s to one another by considering J H coupling constant information. 3  H  Figure 2.6. Structures of plakosides A (29) and B (30).  Plakoside A has been the subject of two total syntheses since its discovery.  28,29  In both cases the absolute stereochemistry of each cyclopropyl moiety was randomly chosen, though the second synthetic effort was designed to give the opposite absolute stereochemistry to the first at the cyclopropyl centres. Interestingly, both products gave indistinguishable N M R data to that of the natural product. It required a separate investigation employing degradation and subsequent derivatisation of the natural product and comparison with known standards by H P L C to finally determine the absolute stereochemistry for 29.  30  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  26  Interest in the plakosides has been fuelled by the report that they act as immunosuppressive agents.  27  This is in contrast with many of the other biologically  active glycosphingolipids, which are reported as having immunostimulating activity. A recent review  17  covers the topic of the biological activities of marine glycolipids in depth,  and is currently the comprehensive resource on the subject. 2.5.3. Steroidal Glycosides.  The chemistry of steroidal glycosides has developed to the point where it is considered by many to be a separate field to that of marine glycolipids. Again, several reviews exist that present this chemistry in d e p t h .  18,31  The topic will not be covered in  this overview, other than to note that a large part of this chemistry derives from invertebrates of the phylum Echinodermata,  and is often known by the general moniker  of saponin chemistry.  2.5.4. Other Glycolipids.  This group of marine glycolipids, of which the caminosides are members, is the most difficult to generalise. To belong to this class, compounds must contain both a carbohydrate portion and a hydrophobic isoprenoid or lipid portion. A number of such compounds exist, of which the simplexides (31 ), (33)  34  32  erylusamines (32),  33  and plaxyoside  are all examples. These compounds are presented here as illustrations of the  types of chemistry one can encounter within this class of glycolipids.  Chapter 2 : Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  27  Simplexides (31)  R — C H 2 C H 2 C H 3 , R' — H  R R R R  = CH = CH = CH =CH  2  2  2  2  CH(CH CH(CH CH CH CH CH  3  3  ) , R' = H ) , R' = A c C H , R' = Ac C H C H , R' = Ac  2  2  2  2  2  2  3  2  3  Erylusamines  (32)  Plaxyoside (33) Figure 2.7. Structures of the simplexides (31), erylusamines (32) and plaxyoside (33).  There exist a wide variety of alkyl chains in this subclass of glycolipids, from the simple saturated chains of the simplexides (31) to the more highly functionalised chains of the erylusamines (32) which contain ketone, amide, amine and secondary alcohol  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  28  functional groups. In all cases, including plaxyoside (33), which contains a terpenoid side chain, the carbohydrate portion is coupled to the alkyl chain via a glycosidic acetal linkage. A s with glycosphingolipids, the variety and complexity of the carbohydrate portion is high, with linear and branched  polysaccharide  chains  of up to six  monosaccharide sub-units having been reported. Investigations into the biological activities for this subclass provide many reports of immunomodulating activity. While the dominance of a small number of research groups in the area of marine glycolipid chemistry has likely led to a bias in the variety of biological screens in which glycolipids have been examined, independent investigations into the mechanisms involved in mammalian immune responses have highlighted glycolipids as important messengers in intracellular signalling,  35  lending weight to the  idea that marine glycolipids could find future utility as therapeutics in the area of immune system control. 2.6. Isolation of and Structural Determination of Caminoside A Peracetate (34). 0  OR  K  •  11  12/  t  c  6-Deoxytalose |3  0  Quinovose  " f  ,OR  °0R 0,  ^^0 ?  R = H (35)  sOR Tr / T ^ 3  1  v  0*  ii  0  G  i  V/°  '  U  C  °  S  e  1  R = Ac(34)  R  17'  19'  Figure 2.8. Structure of caminoside A (35) and caminoside A peracetate (34). Caminus sphaeroconia  Sollas (468 g wet w t , figure 2.9) was collected by hand  using S C U B A at - 1 0 m from Toucari Caves in Dominica, and identified by Dr. R. van  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  29  Soest (University of Amsterdam). A voucher specimen has been deposited at the Zoologisch Museum, Amsterdam (ref. No. Z M A P O R 16775). The material was frozen immediately upon collection and kept at - 1 5 °C until workup. A portion of the frozen material (83 g wet wt.) was extracted three times with M e O H (3 x 500 mL) and the combined extracts concentrated to dryness in vacuo to give a yellow solid (5.5 g). A portion of this material (1.5 g) was purified by Sephadex™ LH-20 size exclusion chromatography eluting with 100% MeOH and the contiguous active fractions pooled and concentrated to dryness in vacuo to give a clear yellow glass (0.467 g). A portion of this material (0.150 g) was then recycled twice on Sephadex™ LH-20 size exclusion chromatography eluting with EtOAc/MeOH/H 0 20:5:2. The active fractions were again 2  identified by bioassay guided fractionation, combined, and concentrated to dryness in vacuo to give a clear colourless glass (0.015 g). This material was identified as an inseparable mixture of caminosides which differed only in the composition of their lipid aglycon components. Acetylation of this mixture and purification of the products by gradient flash silica gel column chromatography ( C H C I to CH CI /MeOH 9:1) and 2  2  2  2  R P H P L C (Ci8, H 0 / P r O H 6:4) afforded pure caminoside A peracetate (34, 3.4 mg) as n  2  an optically active ([a]  25 D  -27° (c = 0.024, MeOH)) clear colourless glass.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  Figure 2.9. Photograph of Caminus  sphaeroconia.  30  sphaeroconia.  Caminoside A peracetate (34) gave a [M + Na] ion at mlz 1427.6470 in the +  H R E S I M S appropriate for a molecular formula of  C 7Hi 4O iNa 6  0  3  (calcd 1427.6459) that  required 16 sites of unsaturation. The H NMR spectrum obtained for 34 at 800 MHz in 1  CeD (Figure 2.14)' showed a series of deshielded methine resonances between £ 3 . 4 6  and 5.9 characteristic of protons on s p carbons attached to oxygen atoms, a series of 3  methyl resonances between £ 1.55 and 2.1, assigned to acetate methyls, and a series of aliphatic methylene and methyl proton resonances between £0.8 and 2.4, suggestive of a linear hydrocarbon fragment. These proton N M R features indicated that caminoside A (35) was a glycolipid. 13  C / D E P T and H S Q C N M R data recorded for 34 identified four ketal methine  carbon resonances at £96.2 (Qui-C1: 1H £5.46), 100.1. (Glu2-C1: 1H £4.88), 100.2  ' All 800MHz N M R data run and processed by the staff of the Canadian National High Field N M R Centre (NANUC) at the University of Alberta.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  31  sphaeroconia.  (Deoxytal-C1: ,1H 8 4.77), and 100.9 (Glu1-C1: 1H S 4.64), which were assigned to sugar anomeric carbons (figure 2.10). Deoxytal-H1 Qui-H1  Glu1-H1  94.0  96.0  Qui-C1  98.0  Glu2-C1 MOO.O  Deoxytal-C1  Glu1-C1 102.0  5.60  5.40  5.20  5.00  4.80  4.60  Figure 2.10. Expansion of the anomeric region of the H S Q C spectrum for 34.  The anomeric proton resonances provided entry points for making complete  1 3  C  and H NMR assignments for each of the four monosaccharides in the molecule via the 1  1 3  C NMR, C O S Y , H S Q C , and H M B C data (Figures 2.15, 2.16, 2.17, 2.18 and Table  2.3). Because each monosaccharide represents an isolated spin system it was possible to identify all of the signals present in each case from information obtained from the C O S Y spectrum. Starting with each anomeric proton in turn, each subsequent proton resonance was sequentially identified as illustrated by the example for the assignment of Glu1 in figure 2.11.  Chapter  2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia.  32  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  33  Figure 2 . 1 3 . C N M R of caminoside A (35) acquired at 100 MHz 13  Table 2.2. 1D and 2D N M R data for caminoside A (35). 5  Pos.  «c  b  Multiplicity 1  H  (J, Hz)  COSY  3  (H H)  a  HMBC (H  a  C)  aqlvcon 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19  29.7 208.5 42.6 23.3 28.5 28.7 28.8 24.4° 34.3 80.7 33.7 24.0° 29.1 29.3 29.6° 29.7 31.3 22.1 13.9  2.05  s  -  -  -  -  2.38 1.43 1.20  t  7.0  1.25 1.26 0.83  19 18  101.5 76.5  4.14 3.32  Glu1-2 Glu1-1  10 Glu1-1, Glu2-1  70.0 76.5 60.7  3.16 3.03 3.49 3.61  Glu1-5 Glu1-4, Glu1-6, Glu1-6' Glu1-5 Glu1-5  Glu1-2, Glu1-6(w) Glu1-4(w), Glu1-6(w)  C  -  4 3, 5 4  C  C  d  3.47  d  c  C  C  Glu1 Glu1-1 Glu1-2 Glu1-3 Glu1-4 Glu1-5 Glu1-6 Glui-6'  -  dd ddd  9.4, 7.9 9.4, 4.9, 2.4  Glu2  Glu2-1 Glu2-2 Glu2-3 Glu2-4 Glu2-5 Glu2-6 Glu2-6' Glu2-7 Glu2-8  99.5 77.0 73.7 69.7 73.3 67.3  Glu2-8' Glu2-9 Glu2-10 Glu2-11 Glu2-12  -  172.0 35.4 17.7 13.3 169.2 20.4  Glu2 Glu2 Glu2 Glu2 Glu2  5.05 3.41 5.16 4.57 3.66 3.47 3.69  d  7.7  dd dd  9.4, 9.4 9.4, 9.6  -  -  -  2.17 2.29 1.47 0.83  dt dt  7.0, 16.0 7.4, 16.0  s  -  -  -  1.96  s  -  -2 -1, Glu2-3 -2, Glu2-4 •3, Glu2-5 •4  Glu1-2, Glu2-3 Glu2-2, Glu2-3, Glu2-4,  Glu2--3, Glu2- 5 Glu2 -4, Glu2- 7 Glu2 -5, Glu2- 6, Glu2-11 Glu2 -6  Deoxytal-1 Glu2 -8', Glu2-9 Glu2 -8, Glu2-9 Glu2 -8, Glu-2-8", Glu2-10 Glu2 -9  Glu2-7, Glu2 -9, Glu2- 10 Glu2-7, Glu2 -9, Glu2- 10 Glu2-7, Glu2 -8, Glu2 •10  Glu-2-11  Deoxytal Deoxytal-1 Deoxytal -2 Deoxytal -3 Deoxytal -4 Deoxytal -5  101.3 71.3 68.4 71.8 70.7  4.62 3.64 3.61 3.29 3.39  Deoxytal -6  16.5  1.13  100.1 72.2 72.7 75.8 67.3 17.9  4.64 3.13 3.34 2.75 3.93 1.09  Qui Qui-1 Qui-2 Qui-3 Qui-4 Qui-5 Qui-6  Deoxytal-2  6-3 d  5-3  6.4  Qui-2  dd m d  10.3, 8.7 6.4  Spectrum acquired at 500MHz in D M S O - d 6 Spectrum acquired at 100MHz in D M S O - d 6 A s s i g n m e n t s interchangeable. Chemical shifts are overlapped in H  Qui-4 Qui-3, Qui-5 Qui-4, Qui-6 Qui-5  a  b  c,d  1  .15-1.35 ppm  Glu2-6, Deoxytal-2, Deoxytal-3 Deoxytal-3, Deoxytal-4(w) Deoxytal-2(w), Deoxytal-3 Deoxytal-3(w), Deoxytal-4, Deoxytal-6 Deoxytal-4, Deoxytal-5  Qui-2, Qui-3 Qui-2, Qui-3, Qui-3, Qui-4,  Qui-3 Qui-4 Qui-5, Qui-6 Qui-4, Qui-6 Qui-5  Chapter  2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia.  37  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  39  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  Figure 2.17. H S Q C spectrum of caminoside A peracetate (34) acquired at 800 MHz.  40  Chapter  2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia.  Figure 2.18. H M B C spectrum of caminoside A peracetate (34) acquired at 400 MHz.  41  Table 2.3. 1D and 2D NMR data for caminoside A peracetate "c aqlvcon 1 2 3 4 5 6 7 8 9 9' 10 11 12 12' 13 14 15 16 17 18 19 d  Multiplicity  6  Pos.  (J, Hz)  c  29.3 206.0 43.3 24.1 29.6 30.0 30.6 25.7 35.3  3  (34).  COSY (H  0  H)  -  -  2, 3  -  -  -  -  2.03 1.57 1.28 1.40  t  7.2  4 3, 5 4, 6 5  2, 4, 5 2, 3, 5, 6 3,4, 6  4, 5, 6(w) 3, 5, 6 3, 4 , 6 4  9' 9' 8, 9, 10 9, 11 10  9, 10 8, 10, 11 8, 11 8, 9 10  9,11,12 8, 10 8, 10(w) 8, 9, 11, 12 12  18 17, 19 17, 18  16, 17, 18  10, Glu1-2(w), Glu1-5(w) Glu1-1, Glu1-3, Glu2-1 Glu1-2, Glu1-4  t  7.2  18 17, 19 18  Glu1-1  100.9  4.64  d  8.0  Glu1-2  Glu1-2  75.9  4.21  dd  8.0, 9.7  Glu1-1, Glu1-3  Glu1-3  75.1  5.62  dd  9.7, 9.4  Glu1-2, Glu1-4  0  C)  -  1.36 1.37 1.38 0.95  d  (H  s  -  d  C)  HSQC- TOCSY  1.69  29.9 30.3 30.4 29.8 32.3 23.1 14.3  d  81.7 34.7 25.5  (H  D  -  1.52 1.71 1.87 3.76 1.69 1.61 1.73  -  HMBC  ROESY (H  H)  -  Glu1 Glu1-2, Glu1-3, Glu1-4 Glu.1-3, Glu1-4, Glu1-5 Glui-2, Glu1-4, Glu1-5  Glu1-5  6  CD CD  3  3  o LO"  3  oo  CD C D  CO  oo  CNJ  CNI CNI  CNI  CNJ  O O  O  O O  O  o  LO  •4, Gl  •3, Gl  3  T—  T —  3  3  o  o  CNJ C D C O  3  3  3  sphaeroconia.  3  IT  LO  CD  CNJ  CNI  LO  CNI  O  O  o  o  o  o  o  3  3  3  in  LO  CO  'tf  •*)-"  LO  **—  T —  C N CNI  CNJ  CNJ  CNI  CNJ  CNI  CNJ  CNI  CNI  CNI  oo  O O  O  O  o  O  O  o  o  1" 1"  CNJ C O  CN  CO  CO  LO CO  O.  03  oo  O op  O oo  CNI C N  CNI  CNI  CNI  CNI CNJ  CNI  CN 13  CNI  CNI 13  CNI  O O  o  o  o  oo  , O  o  O  o  O  °? CNI  03  oo"  oo"  CNI  CNI  CNI  o  o-  O  3  1—  3  3  3  CD  3  3  1 -  Iu2-  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  3  T —  13  3  oo  3  3  3  3  3  3  3  3  3  3  3  3  3  3  3  3  O)  3  3  CD i  .  T -  CN|  LO  LO LO  LO CO  T—  T—  CNI CNJ  CNI  "? V -2 CNI CNJ O 3 3  o o  o  oo  CNI  1 -  CNI I - C O C D  LO L O  i i -  CNI  CNJ CNI CNI CNI CNI  CNI CNI  __2.2J5.2_2 O O O O O O O O  O O  I  i  _3  _3 _3  0 o  o CO C D •  1  i  T—  _3 _3  _3  o  I T —  _3 _3  O O LO  i t—  O O CD  i  _3  o  3  --1 3  3  O O  i  3  3  S  3  3  3  CD  CD CD  CN  CNJ CNI  CNI  O  o  O O  o  3  3  3  co"  CNI CNI  CNI  CNI  CNI CNI CNI CNI  O O  o  o  O OO O  3  3  3  3  CD  3  d  CNJ CNJ  T—  d  T3  d  CD" CO"  - -  LO  1^ 1^  JJ JJ  JD JD  JJ JD JJ JD  X3  o  CO  CNI CN CO  d  d  oo oo  CN  CO CNJ l~-  O) CN CD  d  LO  CD CD  T °^  d  i  J2  5 O CD  3  o  O  T—  CN  T3  -  JD  CO  T-  '<  3  d  «  O CD •<- h -  3  O O  CNI CN 3 3  O O  d  JJ JD  CNI C D  d  CN  od  T3 JD  JD JD JJ  JD JD JD JD  LO CD  CD  q  d  00 d  CN 3  o  CNI  °? CNI  o  o  o  o  00  3  - o  00  3  o  -  T—  „  CN  i  i  00  CO  co  T—  d  JD JD JD  T—  CO CNJ  O.  d  i  CN  d  CN 3  LO CNI 3  op CO d CN CN CN CN 3 3 3 3  CD  O  d  JD JJ JJ  03 d  o  3  d  r-d  o  00  CNI  CD  d  cp  op  „  CD  3  •3-  T—  o  3  ooooooooo d  d  3  _2_2_2-2_2.__..2_2.2  3  CNI  3  O OOO O  T-  T—  co  3 3  C N C N C N Coo" NCNCNCNCNCN  d r—T —  3  CNI  3  CO  - —  3  b> rora  LO" LO"  3  3  I  _3  3  o  T-"  3  CNI C N CNI CNI CNI C N CNJ  , oO O O  CN  CN  3  3  LO  O •"t"  CNI  CNI C N  CN  3  3  O O o o -o -o hp" T - C- T ~ r-~ - r - r- 1 - oo  CNJ  3  oo 3  _3  o  3  CO  CD C D  LO" LO"  co"  3  as CNI  CN CN 1  d  CO  o o o  00  LO •<-  E  _  CN CD  CN  d  ^  d  o  co  CNI _3  O  CO d  CO C D  1  CO CNJ 3  00  °?  ° T T  <_ij  CN CN CN  — O  _2 _2 .2 O O O  43  Chapter  2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia.  CNcom->-oocNLnm  i-o o  »_ i_ i_ i_ i_ i_ i_ i _ rocororororororo rococo  i  i  i  X X X X X X X X  X X X  o o o o o o o o  o o o  CDCDCDCDCDCDCDCD Q Q Q Q Q Q Q Q  CM ^ ', CN T - in  0) 0 0 O O O  ro  •5.  X 0 CD Q  X 0 CD O  erf  CN  ro %  ro  X 0 CD Q  X 0 CD O  X 0  '5 O LO CD  ro  O O 00 _l_  X X 0 CD 0 CD O O  in •<* in  '5 '5 '5 OOO  'zj  ro  % x  o  CD O  I  I  'zi 'zi  CN LO "ro  ro  •5. x  O  erf CN erf  o  ZJ ZJ ZJ  zi  CD  O  o  OOO  '5  co  •5, x o CP  O  CD CN  _ZJ (3  o  CO  X o CD Q  ro %  '5  oooo  ro  erf  1  CO*  Z! ZJ Z!  CD  ro  erf  ^  "ro  ro  CN erf i n  in  ZJ  ro ro  o  TJ-" co  CIO CID CIO C1 D ICO  " ^ " ^  XX 0 0 CD CD O O  X X X X X XX 0 0 0 0 0 0 0 CD CD 0 CD CD CD CD O O O O O O O  CN CN  o in" 1  • 'ZJ 'ZJ  OO  13  o  ZJ  O  erf  '5 O  o  ZJ ZJ Z)  OO0 'Z!  1  1  m  ZJ '5 O CO O 1  1  o  E o  •o  aerf  •5 a '5 ZJOZSOZ! O O iZJo O ^ t O ZJ O •  CN  ZJ  CN CD erf  in 00 CN erf  E Oo.  cz CO Q_  O  E o. o.  CD  X 0 CD O  in  X X X X X X 0 0 0 0 0 0 CD CD CD CD CD CD Q O O O OO  XX 0 0 C D 0CD 0  X 0 CD O  I  in  I  iri r~f  CO  CO  CN  '5 '5 '5 OOO  "co  CN 1  CO  T - " CN erf 1  1  Z!  Z!  1  OOO  Zi  m  Qui  CO 11 ro co  Qui  ro %  00 CN erf m *CD "C1 O CO CIO CIO 1CD T - "  Qui  CN  in1  cf  CN CD  T _  CM  'Z!  O  X3 CD Q. Q.  O  ro  X  o CD  "O CD O.  CD >  OO CO CO O 0 0 T—  O  • "~ <  CN CN c\f O)  CO CD  k_ ro co  CL  ro CD > O  ro o T3 T3  •0  •0 "O  T3 T3 T3 T3 T3 T3  CT  TJ  T3  X3  CO  CM  b o  "143 o  CD  Q  CN  iri  in  CO  co  v-  CO  T—  iri  CO CO  CN CO  CN CN 00 ocq^  m  CO "3-  ih iri in  CO  O O CC  o  CO  ro L_  O  CO  O 3JO  .S o -E  H- N N  • N -O X X u X  ro ^ ro  00  co  00  CO  CN  001  CO  CO  I  x  CN CM  ro  CD O J3  •<t 1^ iri  44  X 0 CD  0  %  X 0 CD Q  CO CO  CO  in ro X 0 CD  0  CO " ^  X 0 CD O  a>  •r-  00  m 1  O CO  N-  CO  m  CD  co ro  CD  ro %  X 0 CD O  ro 0 •o u-o CD Z3 C Zi o Z! O" " _co o °O" ' ^CO ° O  3 •-  00 I I 'ZJ 'Z! 'ZJ  m j_  CD  OOO  a  o  CN  I  'ZJ  'zj  co cz ro  E §E E roZ! ^ c Z! o .9? o CZ CD CO CD c Q . CO Q . o CO < CO  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  45  Two of the sugar spin systems, originating with the anomeric protons at S 4.64 (Glu1-H1) and 4.88 (Glu2-H1), could be assigned to hexoses. Weak H M B C correlations observed between Glu1-H1 (S 4.64) and Glu1-C5 {S 72.3), and between Glu2-H5 (5 3.67) and Glu2-C1 (S 100.1), indicated that the hexoses existed in the pyranose forms. Analysis of the vicinal coupling constants for the H1 to H5 regions in both systems showed only large coupling constants consistent with all axial/axial coupling (Table 2.3) demonstrating that both monosaccharides were glucose residues with p anomeric configurations. This coupling constant information was derived from interpretation of the 2D J-resolved spectrum. In this simple experiment the chemical shift and spin-spin coupling information is presented on different axes of the 2D matrix. The chemical shift information is decoupled on the F axis, providing each proton resonance as a sharp 2  singlet and therefore improving the resolution of the information in this dimension, and potentially resolving signals that are overlapped in the coupled H spectrum. The F-i axis 1  contains only the spin-spin coupling information and is symmetric about the central 0 Hz line. The presentation of the coupling constant information in this fashion allows greater accuracy in determining coupling constant values than is achievable by the analysis of signals from the H spectrum, and is sometimes helpful in simplifying the deconvolution 1  of complex multiplet patterns. An example of the interpretation of the 2 D - J R E S experiment for 34 is presented in figure 2.19.  Figure 2.19. Expansion of the carbohydrate region of the 2 D - J R E S spectrum for 34.  Chapter  2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia.  47  The remaining two sugar spin systems, originating with the anomeric protons at £ 4.77 (Deoxytal-H1) and 5.46 (Qui-H1), could be assigned to 6-deoxy hexoses. Once again, H M B C correlations observed between Deoxytal-H5 (£3.43) and Deoxytal-C1 (8 100.2), and between Qui-H5 (£4.57) and Qui-C1 (£96.2), showed that both of the deoxyhexoses were in the pyranose forms. Vicinal coupling constant analysis (Figure 2.21) showed that the deoxyhexose with the anomeric resonance at £ 5.46 was a quinovose with an a anomeric linkage. R O E S Y correlations observed between the H-5 resonance at £ 3.43 and both the H-3 (£ 5.22) and H-1 (£ 4.77) resonances in the remaining deoxyhexose spin system showed that H-1, H-3, and H-5 were all axial. The H-1/H-2, H-2/H-3, H-3/H-4, and H-4/H-5 vicinal coupling constants in this spin system were all small (1.4 to 5.7 Hz), consistent with a 6-deoxytalose residue having a p anomeric configuration.  Chapter  2:  Antimicrobial Metabolites from the Marine Sponge  Caminus sphaeroconia.  H  H  Glucose 2  Glucose 1  o 6-Deoxytalose Figure 2.20.  1 3  48  H  Quinovose  C chemical shifts for the sugar residues of caminoside A peracetate (34).  4.95  5.39  dd (9.0, H  dd (9.4, H  10.1  Hz) '  4.21 Hdd(8.0,9.7 Hz)  5.62 H  dd (9.7, 9.4  H 4.64 d (8 Hz)  Hz)  Glucose 1  O | d (6.4  10.3  Hz)  4.01 3.79  Hz)  Hz)  o  o  H 3.67 O ddd (2.6, 5.47 H ' ' > H 4.88 dd (9.7, 9.4 Hz) d (7.7 8  2  1 0 3 Hz  Hz)  Glucose 2  5.02 dd(4.2, 10.3 Hz)  1 1 1  3.97  H dd (7.7, 9.1  1-  43  d (6.3 H z )  5.13 Hdd(9.2, I  1  0  3  H  z  )  5.24  dd (3.7, H 2.0  o  Hz)  H  3.43 H 5.22 H 1 < - ' J ' , , , H 4.77 dd(3.7, 6.4 Hz) dd(1.4, .4Hz) 5.7 Hz) 5.7 Hz) d£  2  0  4  7  5.82  H dd (9.2, 10.3  d ( 1  6-Deoxytalose Figure 2.21.  1  Hz)  Quinovose  H chemical shifts and coupling constants for the sugar residues of  caminoside A peracetate (34).  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  49  sphaeroconia.  Methanolysis of peracetate 34 with HCI and M e O H yielded, after purification, the aglycon 40 and a mixture of the methyl glycosides of glucose, quinovose, and 6deoxytalose (Scheme 2.1). OAc  c.HCI, M e O H , reflux, 18hrs  A c 0 , pyridine, D M A P , 2  |rt, 18hrs  36  37  38  Scheme 2.1. Degradation of caminoside A peracetate (34).  39  Chapter  2:  Antimicrobial Metabolites from the Marine Sponge  50  Caminus sphaeroconia.  The aglycon 40 gave a [M + N H ] ion at mlz 316 in the LRCIMS and a [M +  4  H 0]  +  2  The  13  ion at mlz 280 in the LREIMS consistent with a molecular formula of C i H 0 2 . 9  38  C / A P T and H N M R data for 40 identified a methyl ketone ( C D : £ H 1.64, s, 1  1  6  6  3H; C 207), a carbinol methine ( J H 3.76; C 81.7), an aliphatic methyl ( £ H 0.90, t, 1 3  1  1 3  1  J = 6.0 Hz), and 15 aliphatic methylene carbons. These data were only consistent with a C-19 linear aliphatic chain having a methyl ketone at one terminus, the aliphatic methyl at the other terminus, and a secondary alcohol at some point in the interior of the chain. The base peak in the HREIMS spectrum of 40,  which appeared at mlz  171.1386  corresponding to a molecular formula of C10H19O2 (calcd 171.1385), was assigned to an a cleavage of the bond adjacent to the secondary alcohol on the side remote from the methyl ketone terminus  (Figure 2.22). This fragmentation  located the  hydroxyl  functionality at C-10 in the aglycon 40.  CioH-|g0  2  Figure 2.22. HREIMS fragmentation of aglycon 40.  Analysis of the N M R data obtained for the peracetate 34 (Table 2.3) confirmed the presence of the aglycon 40 (methyl ketone: £ H 1.69, s, 3H and C 206.0; carbinol 1  methine: H 3.76, m, 1H and 1  1 3  C  14.3). H M B C  1 3  1 3  C 81.7; terminal methyl: H 0.95, t, J = 7.2 Hz, 3H and  correlations observed  1  between the aglycon carbinol  methine  resonance at < _ > 3.76 (H-10) and the anomeric carbon resonance of one of the glucose residues {8 100.9, Glu1-C-1), and between the Glu1 anomeric proton resonance (S  Chapter  2: Antimicrobial Metabolites from the Marine Sponge Caminus sphaeroconia.  51  4.64) and the carbinol methine resonance at 8 81.7 (C-10) demonstrated their connection via an acetal linkage. A series of selective one-dimensional T O C S Y experiments provided added confirmation of the connectivity of the aglycon, and gave further information about the chemical shifts of protons on the aglycon chain. In the 1D-TOCSY  experiment  magnetisation is introduced to the molecule at a specified position by the irradiation of a selected proton resonance. Through-bond magnetisation transfer then occurs during the subsequent mixing time, during which the bulk magnetisation is rapidly subjected to an extended sequence of composite 180° pulses that reduce the loss of signal due to transverse relaxation (T* ). The degree to which this relay of coherence proceeds 2  through a given spin system is directly correlated to the length of the allowed 'spin-lock' mixing time. In short, the longer the spin-lock, the greater the degree of propagation of the magnetisation through the spin system.  36  For caminoside A peracetate (34) a series of experiments were run with a variety of mixing times in order to observe the propagation of the magnetisation along the aliphatic chain. While it was not possible to make an accurate count of the number of methylenes between each functional group on the chain due to the degree of overlap of the methylene signals  in the  1  H  NMR, the  1D T O C S Y  did provide additional  confirmation of the connectivity of these functional groups within the aglycon. Irradiation of the terminal methyl group at 8 0.94 (H-19) followed by mixing times ranging from 50 u.s to 400 |as gave a series of spectra that showed signals from H-18 (§ 1.37) as far as H-11 (5 1.69) in the case where the mixing time was 400u.s (figure 2.23). More interestingly, irradiation of the signal at 5 2.03 (H-3) with mixing times ranging from 50 \xs to 1500 u.s gave a series of spectra that showed signals from H-4 (1.56) as far as H-  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  52  sphaeroconia.  10 (5 3.76) (figure 2.24). The C O S Y correlation between H-10 and H-11 had already shown these two protons to be on adjacent carbons, and this last piece of information allowed the connection of the partial spin systems identified from the 1D-TOCSY experiments and confirmed that the aglycon consisted of one contiguous spin system from H-3 to H-19.  Mixing time = 50 \xs  Mixing time = 100 |is  J Mixing time = 200 (is  Mixing time = 400 (is  *U Hunt i | OT IITHTp'f 11IP 1 IfT'Ttl11 TT 1111 M 1111 11 111 M l I ITp I I M I M I | l 1 l l l l l l l | l l l f l I'TTTp'tl I? 111 I [ M11 M I M 11111'H HTf I'M III I M | I H M 1 I M | I H I TITTT fl111 n 11 I [ III11 l'TfTf**f UTI M11 M II ? II111111 It I f t^TTTTlTrn  1.9  1.8  1.7  1.6  1.5  1.4  1.3  1.2  1.1  1.0 (ppm)  0.9  0.8  0.7  0.6  0.5  Figure 2.23. 1 D - T O C S Y experiments for 34, irradiating at 8 0.94.  0.4  0.3  0.2  0.1  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  53  sphaeroconia.  Mixing time = 25 (is (NS = 32 unless otherwise stated)  Mixing time = 50 |is  Mixing time = 100 (is  Mixing time = 500 (is  Mixing time = 1000 (is  Mixing time = 1500 |is (NS = 128)  —i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—>—i—i—i—<—i— —i— —i— —i— —i— 1  3.B  3.6  3.4  3.2  3.0  2.8  2.6  2.4  2.2  2.0  1.8  1.6  1.4  (ppm)  Figure 2.24. 1 D-TOCSY experiments for 34, irradiating at 5 2.03.  1  1.2  1  1.0  1  0.8  0.6  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  The  mixture  of  monosaccharide  methyl  sphaeroconia.  glycosides  obtained  54  from  the  methanolysis reaction was peracetylated with Ac 0/pyridine and the products were 2  purified via H P L C to give a-1-methyl-2,3,4,6-0-acetyl-glucose (36) and p-1-methyl2,3,4-0-acetyl-6-quinovose  (38) (Scheme 2.1). Comparison of the NMR data and  specific rotations for 36 and 38 with authentic material prepared from L-glucose and Dquinovose, using the reaction conditions described above for transformation of the natural product, confirmed the presence of these monosaccharides in caminoside A (35) and showed that the glucose residues had the D configuration and the quinovose residue had the L configuration (Figure 2.25 and 2.26). The N M R data obtained for 1methyl-2,3,4-0-acetyl-6-deoxytalose  (39) were  in  complete  agreement  with  the  assigned structure. Authentic 6-deoxytalose was not available for comparison purposes, so 39 was hydrolyzed to the native sugar (41) and its specific rotation ([CC]D = +14°) 25  and  1 3  C N M R data were compared with literature values for 6-deoxy4_-talose ([ct]  25 D  =  -20°), confirming its identity and showing that it had the D configuration. 37  H M B C data provided evidence for the nature of the linkages between the four monosaccharides. Correlations between the 6-deoxytalose anomeric proton resonance at 5 4.77 and the Glu2 C-6 resonance at 5 69.4 demonstrated that there was a 1,6glycosidic linkage between the 6-deoxytalose and Glu2 residues. A strong H M B C correlation between the Glu2 H-2 resonance at 5 3.97 and the Qui anomeric carbon resonance at 596.2 established a 1,2 glycosidic bond between Qui and Glu2. Finally, a correlation between the Glu1 H-2 resonance at 54.21 and an anomeric resonance at 5 100.1 (Glu2-C-1) showed there was a 1,2- glycosidic link between Glu2 and Glu1 (figure 2.27).  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  55  sphaeroconia.  d oo  d  1  o c\i  CM  1 co CM  CM  E a  co  o  O Z5 "D O  CL c/)  co  CM LO  o  CD  d  o d  "D  c 05  •*-»  00  Figure 2.25. 1-Methyl-2,3,4,6-0-acetyl-glucose comparison by H N M R at 500 MHz. 1  CQ C —.  CD  Hydrolysis Product  ro 05  CD i— ZT »<_ i  ro  CO 6  i o 5.6  5.2  4.8  4.4  4.0  3.6  3.2  X.  2.8  2.4  2.0  1.6  1.2  0.8  0.4  2.8  2.4  2.0  1.6  1.2  0.8  0.4.  (ppm)  CD  Standard  c ZJ" o < o C/3  CD  O O 3  03  w'  O CT 73 i—t-  cn o o N  56  5.2  4.8  4.4  4.0  3.6  3.2  (ppm)  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  57  As described above, coupling constant data showed that glucose-1 and glucose2 had p anomeric configurations and that quinovose had an a anomeric configuration. R O E S Y data showed that the 6-deoxytalose residue had the p anomeric configuration. Comparison of the L R E S I M S of the underivatised natural product mixture of caminosides, differing only in the aglycon fragment, injected in M e O H ([M + Na] m/z +  1049) and MeOD ([M + Na] m/z 1058) showed that there were only nine exchangeable +  protons, even though the identified tetrasaccharide fragment had eleven available hydroxyls. In addition, the N M R data for the underivatised natural product mixture contained resonances that could be assigned to acetyl ( £ H 1.96, s, 3H and 1  1 3  C 20.4,  169.2) and butyryl ( £ H 0.83, t, 3H; 1.47, m, 2H; 2.29, dt, 1H, 2.17, dt, 1H and C 13.3, 1  17.7, 35.4, 172.0) residues, respectively. Therefore, caminoside A (35)  1 3  had to already  contain an acetate and a butyrate ester before derivatisation with acetic anhydride to give the peracetate 34. With the structure of the tetrasaccharide fragment of caminoside A in hand from analysis of the N M R data for the peracetate 34, it was possible to go back and assign 1  H and  1 3  C N M R resonances to the sugar portion of caminoside A (35) using the data  for the mixture of natural products that differed only in the aglycon (table 2.2). Using these assignments for the underivatised compound, H M B C correlations were observed between the Glu2 H-4 resonance at £4.57 and the acetyl carbonyl resonance at £169.2 (Glu2-C-11), and between the Glu2 H-3 resonance at £ 5 . 1 6 and the butyryl carbonyl resonance at £ 172.0 (Glu2-C-7) (figures 2.27 and 2.28). This placed the butyrate ester at C-3 of Glu2 and the acetate ester at C-4 of Glu2, completing the structure of caminoside A and its peracetate as shown in 35 and 34, respectively (Figure 2.8).  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  58  sphaeroconia.  R = Aglycon H(X ^  O^tJL^ / O H  ~  Reciprocal 3 Bond H M B C correlations = 3 Bond H M B C correlations from proton to carbon  =  '  OH  Figure 2.27. Selected J H H M B C correlations for caminoside A (35). 3  C  Glu2-3  Glu2-4  • (C  -  166  -  168  Glu2-11  d •  -  170  -  172  -  174  Hi ,  5.2  5.0  4.8  4.6  Glu2-7  4.4  Figure 2.28. Expansion of the carbohydrate region of the H M B C for 34.  The lack of dispersion in the methylene proton resonances near the C-10 region of the aglycon made it impossible to use Mosher-type methodology to determine the  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  59  absolute configuration at C-10, leaving the sfereochemistry of the last remaining stereogenic carbon centre undefined.  2.7. Experimental.  2.7.1. General Experimental  Procedures.  Infrared spectrum was recorded on a Perkin Elmer 1710 FTIR spectrometer equipped with a helium-neon laser (633nm) and controlled by Perkin Elmer Spectrum v.2 software. Optical rotations were determined with a J A S C O J-1010 polarimeter equipped with a halogen lamp (589 nm) and a 10 mm micro cell. UV spectrum was recorded on a Waters 2487 spectrophotometer. H , COSY-gr, HSQC-gr and HMBC-gr 1  spectra (optimised for J H = 8 HZ) for 35 were recorded on a Varian I NOVA 500MHz 2,3  C  NMR spectrometer. H , COSY-gr, HSQC-gr, HMBC-gr (optimised for ' J H=8 HZ), 1  2  3  C  T O C S Y , and H S Q C - T O C S Y for 34 were recorded on a Varian I N O V A 800 MHz N M R spectrometer at the N A N U C facility at the University of Alberta. R O E S Y spectra were recorded on a Bruker AMX500 N M R spectrometer.  1 3  C spectra and some H spectra 1  were recorded on a Bruker AM400 N M R spectrometer. Chemical shifts were referenced to solvent peaks (5 7.15 ppm, 8 H  C  128 ppm for C D ; 5 2.49 ppm, 8 39.5 ppm for 6  6  H  C  DMSO-c/6). Low resolution ESI mass spectra were recorded on a Bruker Esquire L C mass spectrometer. High resolution ESI mass spectra were recorded on a Macromass LCT mass spectrometer. Both low and high resolution El mass spectra were recorded on an AEI MS-50 mass spectrometer. Both low and high resolution CI mass spectra were recorded on a Kratos Concept II HQ mass spectrometer. Flash silica gel column chromatography was performed using 230-400 mesh silica gel 60 (Silicycle). H P L C separations were achieved using a Waters 600 pump and a Waters 486 tuneable  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  60  absorbance detector. Solvents were all H P L C grade (Fisher) and those used for H P L C were filtered prior to use and sparged with helium. Pyridine, acetic anhydride and 4(dimethlyamino)pyridine  were  reagent  grade  (Aldrich)  and  used without  further  purification.  2.7.2. Isolation Procedure. Caminus sphaeroconia  Sollas (468 g wet wt.) was collected by hand using  S C U B A at - 1 0 m from Toucari Caves in Dominica, and identified by Dr. R. van Soest (University of Amsterdam). A voucher specimen has been deposited at the Zoologisch Museum, Amsterdam (ref. No. Z M A P O R 16775). The material was frozen immediately upon collection and kept at - 1 5 °C until workup. A portion of the frozen material (83 g wet wt.) was extracted with MeOH (3 x 200 mL) and the combined  extracts  concentrated to dryness in vacuo to give a yellow solid (5.5 g). A portion of this material (1.5 g) was purified by Sephadex™ LH-20 size exclusion chromatography eluting with 100% M e O H to give a clear yellow glass (0.467 g). A portion of this material (0.150 g) was then recycled twice on Sephadex™ LH-20 size exclusion chromatography eluting with EtOAc/MeOH/H 0 20:5:2 to give a clear colourless glass (0.015 g). This material 2  was acetylated by stirring in pyridine (9 mL, 111 mmol) and acetic anhydride (3 mL, 31 mmol) with 4-(dimethylamino)pyridine (1 mg, 0.008 mmol) under N at 25 °C for 18 hrs. 2  The reaction mixture was concentrated to dryness in vacuo and partitioned between H 0 and EtOAc. The organic phase was washed with water and concentrated to 2  dryness to give a yellow solid which was subjected to flash silica gel column chromatography (gradient elution from C H C I to C H C I / M e O H 9:1) to give a clear 2  2  2  2  colourless glass (0.020 g). Finally this material was purified by R P H P L C (Inertsil C i  8  Chapter  2:  Antimicrobial Metabolites from the Marine Sponge  61  Caminus sphaeroconia.  9.4x250 mm, eluting with H 0 / P r O H 6:4, UV detection at 210 nm) to give pure n  2  Caminoside A peracetate (34, 3.4 mg, 0.002 mmol).  2.7.3. Caminoside A Peracetate (34)-Physical Data.  Clear colourless glass. [a]  25 D  -27° (c=0.019, MeOH); IR (thin film): v  max  1748,  1712 cm" ; UV (c=0.001, MeOH): \ * 206 (s 473); L R E S I M S m/z 1428; HRESIMS: m/z 1  m a  1427.6470 (calcd for C Hio403iNa 1427.6459); H and C N M R (800 and 100 M H z 1  1 3  67  respectively) see Table 2.3. 2.7.4. Methanolysis of Caminoside A Peracetate (34).  Caminoside  A  peracetate  (34, 3.4 mg, 0.002  mmol) was subjected to  methanolysis by refluxing in M e O H (10 mL) and cone. HCI (1 drop) for 18 hours. Upon cooling, the reaction mixture was concentrated to dryness in vacuo and partitioned between H 0 (100 mL) and E t 0 (100 mL). The phases were separated and the 2  2  aqueous phase extracted with E t 0 2  (2x100 mL). The combined organics were  concentrated to dryness in vacuo to give the crude aglycon (40) as a white solid (0.8 mg). Purification by flash silica gel column chromatography (16x115 mm, hexanes 7:3 hexanes/EtOAc) gave pure 40 as a white amorphous solid (0.7 mg, 0.002 mmol). The  aqueous  portion  of the partition  was concentrated  to dryness by  lyophilisation to give a crude mixture of methyl glycosides. Purification by flash silica gel column chromatography (20x235 mm, C H C I to 9:1 CH CI /MeOH) gave the individual 2  2  2  2  methyl glycosides, which were acetylated by stirring in pyridine (9 mL, 11Tmmol) and acetic anhydride (3 mL, 31 mmol) with D M A P (1 mg, 0.008 mmol) under N at 25 °C for 2  18 hrs. Further purification by R P H P L C (Whatman Partisil 10 ODS-3, eluting with 6:4 H 0 / M e O H , detection at 210 nm) gave B-1-methyl-2,3,4,6-0-acetyl-D-glucose (36, clear 2  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  colourless glass, 3.2 mg, [ a ]  24 D  sphaeroconia.  62  +119.6°), a-1-methyl-2,3,4-0-acetyl-L-quinovose (38,  clear colourless glass, 3.2 mg, [ a ]  24 D  +19.3°) and a-1-methyl-2,3,4-0-acetyl-6-deoxy-D-  talose (39, clear colourless glass). a-1-Methyl-2,3,4-0-acetyl-6-deoxy-D-talose (39) was hydrolysed by refluxing in H 0 (10 mL) and cone. HCI (1 drop) for 18 hours to give 62  deoxy-D-talose (41, clear colourless glass, 1.4 mg, [ a ]  23 D  +14.0°).  2.7.5. p-1-Methyl-2,3,4,6-0-acetyl-L-glucose (36) Standard.  Commercially available L-glucose (Aldrich, 35mg, 0.19 mmol) was refluxed with MeOH (10 mL) and cone. HCI (1 drop) for 18 hrs. Upon cooling the reaction mixture was concentrated to dryness in vacuo to give a mixture of a and (3-1-methyl-L-glucose which was acetylated by stirring in pyridine (9 mL, 111 mmol) and acetic anhydride (3 mL, 31 mmol) with D M A P (1 mg, 0.008 mmol) under N at 25 °C for 18 hrs. The reaction 2  mixture was concentrated to dryness in vacuo and partitioned between H 0 (100 mL) 2  and EtOAc (100 mL). The aqueous layer was washed (EtOAc, 2x100 mL) and the combined organics were concentrated to dryness in vacuo to give crude a and 8-1methyl-2,3,4,6-0-acetyl-L-glucose (36) as an orange solid. Purification by flash silica gel column chromatography (16x115 mm, hexanes to 7:3 hexanes/EtOAc) followed by R P H P L C (Whatman Partisil 10 ODS-3, eluting with 6:4 H 0 / M e O H , detection at 210 2  nm) gave 3-1-methyl-2,3,4,6-0-acetyl-L-glucose (36, clear colourless glass, 6.7 mg, [a]  21 D  -85.0°).  2.7.6. a-1-Methyl-2,3,4-0-acetyl-D-quinovose (38) Standard.  Commercially available D-quinovose (Sigma, 22 mg, 0.134 mmol) was treated in an identical fashion to L-glucose above to give a-1-methyl-2,3,4-0-acetyl-D-quinovose (38, clear colourless glass, 6.4 mg, [a]o -18°). 24  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  63  2.8. References.  1. Finlay, B. B.; Falkow, S. Microbiol. Mol. Biol. R. 1997, 61, 136-169. 2. Fath, M. J.; Kolter, R. Microbiol. Rev. 1993, 57, 995-1017. 3. Salmond, G . P.; Reeves, P. J . Trends Biochem. Sci. 1993, 18, 7-12. 4. Van Gijsegem, F.; Genin, S.; Boucher, C. Trends Microbiol. 1993, 1, 175-180. 5. Donnenberg, M. S. Nature 2000, 406, 768-774. 6. Hueck, C. J . Microbiol. Mol. Biol. R. 1998, 62, 379-433. 7. Lee, V. T.; Schneewind, O. Gene: Dev. 2001, 15, 1725-1752. 8. Galan, J . E.; Collmer, A. Science 1999, 284, 1322-1328. 9. Drasar, B. S.; Hill, M. J . Human Intestinal Flora; 1 ed.; Academic Press Inc.: London, 1974. 10. Paulson, E. "Canada Communicable Disease Report," Government of Canada, 2000. 11. Ramamurthi, K. S.; Schneewind, O. Annu. Rev. Cell Dev. Bi. 2002, 18, 107-133. 12. Knutton, S.; Rosenshine, I.; Pallen, M. J.; Nisan, I.; Neves, B. C ; Bain, C ; Wolff, C ; Dougan, G.; Frankel, G . EMBOJ.  1998, 17, 2166-2176.  13. Ide, T.; Laarmann, S.; Greune, L.; Schillers, H.; Oberleithner, H.; Schmidt, M. A. Cell. Microbiol. 2001, 3, 669-679. 14. Gauthier, A.; Finlay, B. B. In ASM News, 2002; Vol. 68, pp 383-387. 15. Tacket, C. O.; Sztein, M. B.; Losonsky, G.; Abe, A.; Finlay, B. B.; McNamara, B. P.; Fantry, G. T.; James, S. P.; Nataro, J . P.; Levine, M. M.; Donnenberg, M. S. Infect. Immun. 2000, 68, 3689-3695. 16. Fattorusso, E.; Mangoni, A. Prog. Ch. Org. Nat. Prod. 1997, 72, 215-301. 17. Costantino, V.; Fattorusso, E.; Mangoni, A. In Bioactive Compounds from Natural Sources.; Tringali, C , Ed., 2001, pp 556-575. 18. Minale, L.; Riccio, R.; Zollo, F. Prog. Ch. Org. Nat. Prod. 1993, 62, 75-308. 19. Kikuchi, H.; Tsukitani, Y.; Manda, T.; Fujii, T.; Nakanishi, H.; Kobayashi, M.; Kitagawa, I. Chem. Pharm. Bull. 1982, 30, 3544-3547. 20. Costantino, V.; Fattorusso, E.; Mangoni, A.; Aknin, M.; Fall, A.; Samb, A.; Miralles, J. Tetrahedron 1993, 49, 2711-2716.  Chapter 2: Antimicrobial Metabolites from the Marine Sponge Caminus  sphaeroconia.  64  21. Costantino, V.; Fattorusso, E.; Imperatore, C ; Mangoni, A. J. Nat. Prod. 2002, 65, 883-886. 22. Wicke, C ; Hueners, M.; Wray, V.; Nimtz, M.; Bilitewski, U.; Lang, S. J. Nat. Prod. 2000, 63, 621-626. 23. Colombo, D.; Compostella, F.; Ronchetti, F.; Scala, A.; Toma, L.; Kuchide, M.; Tokuda, H.; Nishino, H. Cancer Lett. 2000,161,  201-205.  24. Nakata, K.; Guo, C.-T.; Matsufuji, M.; Yoshimoto, A.; Inagaki, M.; Higuchi, R.; Suzuki, Y. J. Biochem. 2000, 727, 191-198. 25. Fujimoto, N. Kurume Igakkai Zasshi 2002, 65, 186-194. 26. Borbone, N.; De Marino, S.; lorizzi, M.; Zollo, F.; Debitus, C ; lanaro, A.; Pisano, B. Eur.J. Org. Chem. 2001, 4651-4656. 27. Costantino, V.; Fattorusso, E.; Mangoni, A.; Di Rosa, M.; lanaro, A. J. Am. Chem. Soc. 1997, 119, 12465. 28. Nicolaou, K. C ; Li, J.; Zenke, G. Helv. Chim. Acta 2000, 83, 1977-2006. 29. Seki, M.; Kayo, A.; Mori, K. Tet. Lett. 2001, 42, 2357-2360. 30. Mori, K.; Tashiro, T.; Akasaka, K.; Ohrui, H.; Fattorusso, E. Tet. Lett. 2002, 43, 3719-3722. 31. Minale, L.; lorizzi, M.; Palagiano, E.; Riccio, R. Adv. Exp. Med. Biol. 1996, 404, 335356. 32. Costantino, V.; Fattorusso, E.; Mangoni, A.; Di Rosa, M.; lanaro, A. Bioorg. Med. Chem. Lett. 1999, 9, 271-276. 33. Sata, N.; Asai, N.; Matsunaga, S.; Fusetani, N. Tetrahedron 1994, 50, 1105-1110. 34. Costantino, V.; Fattorusso, E.; Imperatore, C ; Mangoni, A. Eur. J. Org. Chem. 2001, 4457-4462. 35. Cox, T. M. Philos. T. Roy. Soc. B 2003, 358, 967-973. 36. Claridge, T. D. W. High-resolution  NMR Techniques in Organic Chemistry;  Pergamon: Oxford, 1999; Vol. 19. 37. Defaye, J.; Gadelle, A.; Angyal, S. J . Carbohyd. Res. 1984, 126, 165-169.  65  Chapter 3: Isolation and Characterisation of Caminosides B-D.  Chapter 3: Isolation and Characterisation of Caminosides B-D. 3.1. Introduction.  In addition to the marine metabolite caminoside A (35), whose isolation and 1  structural identification has been extensively discussed in chapter two, the marine sponge Caminus sphaeroconia  also contained a number of related glycolipids. Isolation  and purification of three of these compounds led to the discovery of caminosides B (42), C (43), and D (44) as novel antibacterial metabolites.  3.2. Isolation and Characterisation of Caminoside B. O  OH  Figure 3.1. Caminoside B (42).  Caminoside B (42) was isolated in an analogous fashion to caminoside A (35), with the crude methanolic sponge extract being first chromatographed by Sephadex™ LH-20 size exclusion column chromatography eluting with M e O H to give a complex mixture of caminosides. This material was recycled twice on Sephadex™ LH-20 size exclusion chromatography eluting with E t O A c / M e O H / H 0 20:5:2. Caminoside B (42) 2  was isolated as a clear colourless glass (0.028g), as a mixture of compounds that were identical in the polysaccharide portion, but differed in the aglycon region.  Chapter 3: Isolation and Characterisation of Caminosides B-D.  66  Initial consideration of the H NMR spectrum for 42 (figure 3.8) indicated a high 1  degree of similarity between the structures of caminosides A (35) and B (42). There were, however, several differences between the upfield regions of the two spectra. The spectrum for 42 lacked the sharp singlet at 5 1.96 present in the H NMR spectrum of 35 1  and assigned in that spectrum as the methyl group of the acetyl functionality at Glu2-4. The H spectrum for 42 contained an additional methyl resonance within the methyl 1  envelope at S 0.81-0.89. The signals at 5 2.17 and 2.29, which were assigned as the diastereotopic methylene protons a to the butyrate ketone at Glu2-3 in caminoside A (35), now integrated for two protons each rather than one, as was the case with 35. Comparing the  1 3  C N M R spectra for the two compounds indicated the loss of the  acetate methyl signal at 5 20.4 and the inclusion of three new carbon signals at 5 13.4, 17.5, and 35.1. In addition, the carbonyl signal at 5 169.2 had shifted downfield to 5 171.5. H R E S I M S analysis of 42 gave an [M + Na] ion at 1077.5815 consistent with a +  formula of C i H O 2 2 N a (calc. 1077.5821). This represented an increase in atomic 5  90  composition of two carbons and four protons which, when coupled with the changes in both the H and C spectra for 42 suggested the replacement of the acetate attached to 1  1 3  Glu2-4 in caminoside A (35) with a butyrate unit in caminoside B (42). As with 35, interpretation of the carbohydrate region of the H M Q C spectrum for 42 allowed the assignment of the signals due to the presence of four anomeric protons and carbons (figure 3.2). With this information in hand, it was possible to identify the proton resonances belonging to each monosaccharide by consideration of the C O S Y spectrum (as with 34, figure 2.10).  67  Chapter 3: Isolation and Characterisation of Caminosides B-D.  Glu2-H1  Qui-H1  Deoxytal-H1  Glu1-H1  96  98  J •U®—  Glu2-C1 Qui-C1 Deoxytal-C1  102  Glu1-C1  104 5.0  4.8  4.6  4.4  4.2  4.0  Figure 3.2. Expansion of the anomeric region for the H M Q C for 42.  Interpretation of the H M Q C spectrum provided the chemical shifts of the carbons at each position in the carbohydrate portion of the molecule. These assignments were confirmed with information from the H M B C spectrum that once again also allowed for assignment of the connectivity of the monosaccharide sub-units in the natural product due to  3  J H C  correlations  observed  across  the glycosidic  linkages  (figure 3.3).  Consideration of the A P T , COSY-gr, HMQC-gr and HMBC-gr spectra for 42 confirmed the presence of two butyrate subunits, and defined their positions of attachment to the carbohydrate as being at Glu2-C3 and Glu2-C4 (figure 3.4).  68  Chapter 3: Isolation and Characterisation of Caminosides B-D.  Glu2-H1  Deoxytal-H1 Glui-H1  Glu2-C6 Glu1-C1 Glu2-C2 —Aglycon-C10  Figure 3.3. Expansion of the carbohydrate region of the H M B C for 42 (part 1).  Glu2-H3  Glu2-H4  Glu2-C7 Glu2-C11  Figure 3.4. Expansion of the carbohydrate region of the H M B C for 42 (part 2).  Chapter 3: Isolation and Characterisation of Caminosides B-D.  69  Due to the poor dispersion of the proton signals in the H N M R spectrum of 42 in 1  DMSO-c/6 it was not possible to complete a total structural assignment of the natural product. As stated above, 42 was isolated as a mixture of natural products that varied only in the composition of the aglycons. Conversion of 42 to the corresponding nonaacetate (45) provided greater dispersion in the H N M R spectrum, and allowed for 1  the isolation of 45 as a single pure compound. Glucose 2  o  OAc  Figure 3.5. Caminoside B peracetate (45).  Treatment of 42 with acetic anhydride and 4-(dimethylamino) pyridine (DMAP) in pyridine overnight followed by reversed phase H P L C (43% PrOH/57% H 0 ) provided n  2  caminoside B peracetate (45) as a clear colourless glass (0.002 g). Detailed NMR analysis allowed for the complete assignment of the planar structure of the glycosidic portion of 45, and coupled with the consideration of the JHH coupling constant 3  information obtained from the 2D-J Resolved (2D-JRES) experiment showed that the connectivity and relative stereochemistry of this portion of 45 were identical to that of 34 (figure 3.6). It was assumed that the absolute stereochemistry for each individual sugar did not differ between caminosides A (35) and B (42).  70  Chapter 3: Isolation and Characterisation of Caminosides B-D.  J= 10.1 Hz  O  (axial) \  J = 1.9 Hz J = 7.9 Hz •>^(axial)  O  (equatorial) O  J = 9.3 Hz (axial)  J = 3.1 H z ' (equatorial) J = 4.1 Hz (equatorial)  J = 9.8 Hz (axial)  J= 1-7 Hz (equatorial)  6-Deoxytalose  Glucose 1 Figure 3.6. Selected JHH coupling constant analyses for 45. 3  The remaining signals in the H and 1  1 3  C N M R spectra of 45 were due to the  presence of the aglycon portion of the molecule. Consideration of the A P T and H M Q C spectra for 45 showed the aglycon to contain: 2 x C H ; 15 x C H 2 ; 1 x CH; and 1 x 3  ketone. It was possible to identify the presence of a methyl ketone fragment, a terminal saturated hydrocarbon and an oxygenated methine flanked by methylenes on each side (figure 3.7). O  Figure 3.7. Aglycon structural fragments for 42.  Knowledge about the constitution of the aglycon, coupled with the fragments identified from the N M R spectra meant that the aglycon had to be a saturated linear chain with a methyl ketone at one terminus and an acetal linkage to the polysaccharide portion at an unknown position along the length of the chain. Consideration of the fragments presented in figure 3.7 ruled out a number of positions, leaving only eight of a  Chapter 3: Isolation and Characterisation of Caminosides B-D.  71  possible sixteen sites for the location of the acetal bridge. While it was not possible to complete an assignment of all aglycon NMR resonances due to the overlapping nature of both the H and 1  1 3  C the signals, all the available information suggested that the  aglycon had an identical composition to that of the aglycon in 3 5 , completing the assignment of 42 as that depicted in figure 3.1.  Figure 3.8. Comparison of the H NMR spectra of caminosides A (35) and B (42). 1  Chapter 3: Isolation and Characterisation of Caminosides B-D.  Figure 3.9. Comparison of the C NMR spectra of caminosides A (35) and B (42). 1 3  73  Table 3.1. 1D and 2D NMR data for caminoside B (42). Pos.  Multiplicity  "c aalycon 1 2 3 4 5 6 7 8 9 9' 10 11 12 12' 13 14 15 16 17 18 19  b  1  H  (J, Hz)  a  COSY"  HMBC  ROESY"  (H  (H  (H  H)  C)  29.5 208.0 42.7 23.0 29.3  2.04  s  -  -  2, 3  -  -  -  -  -  4 3, 5 4  2, 4, 5 2, 3 3,4  80.8  3.44  13.9  0.83  101.2 76.5 68.4 69.9 76.8 60.7  4.13 3.33  2.37 1.42 1.19  H)  Glu1-1  Glu1 Glu1-1 Glu1-2 Glu1-3 Glu1-4 Glu1-5 Glu1-6 Glu1-6'  d  8.0  Glu1-2 Glu1-1  10  Glu1-5  Glu1-4  Glu1-1  C  -  3.16 3.03 3.48 3.62  Glu1-5 Glu1-4, Glu1-6, Glu1-6' Glu1-5, Glu1-6' Glu1-5, Glu1-6  Glu2 Glu2-1 Glu2-2 Glu2-3 Glu2-4 Glu2-5 Glu2-6 Glu2-6' Glu2-7 Glu2-8 Glu2-8' Glu2-9 Glu2-10 Glu2-11 Glu2-12 Glu2-12' Glu2-13 Glu2-14  99.2 77.1 73.7 69.5 73.4 67.0 171.0 35.1 17.5 13.3 171.5 35.4 17.6 13.4 d  e  f  d  e  f  5.04 3.40 5.17 4.57 3.66 3.42 3.63  d dd dd  8.0 9.4,9.4 9.4,9.4  Glu2-2 Glu2-1, Glu2-2, Glu2-3, Glu2-4, Glu2-5, Glu2-5,  Glu2-3 Glu2-4 Glu2-5 Glu2-6, Glu2-6' Glu2-6' Glu2-6  Glu1-2  Glu2-5  Glu2-2, Glu2-4, Glu2-7 Glu2-6, Glu2-11  Glu2-5 Glu2-1, Glu2-3  2.14 2.27 1.47 0.84  dt dt  7.2, 16.0 7.2, 16.0  Glu2-8\ Glu2-9 Glu2-8, Glu2-9 Glu2-8, Glu2-8', Glu2-10 Glu2-9  Glu2-7, Glu2-7, Glu2-7, Glu2-8,  Glu2-9, Glu2-10 Glu2-9, Glu2-10 Glu2-8, Glu2-10 Glu2-9  2.14 2.27 1.47 0.84  dt dt  7.2, 16.0 7.2, 16.0  Glu2-12', Glu2-13 Glu2-12, Glu2-13 Glu2-12, Glu2-12', Glu2-14 Glu2-13  Glu2-11, Glu2-11, Glu2-11, Glu2-12,  4.64 3.64  Deoxytal-2 Deoxytal-1  Glu2-6  3.39 1.13  Deoxytal-6 Deoxytal-5  Glu2-13, Glu2-14 Glu2-13, Glu2-14 Glu2-12, Glu2-14 Glu2-13  Deoxytal  Deoxytal-1 Deoxytal-2 Deoxytal-3 Deoxytal-4 Deoxytal-5 Deoxytal-6  101.0 71.3° 72.6 71.7 70.9 16.4 C  Deoxytal-4, Deoxytal-5  Qui Qui-1 99.7 4.65 Qui-2 72.1 3.13 dd 3.34 Qui-3 77.0° Qui-4 75.6 2.75 Qui-5 67.0 3.92 Qui-6 1.08 17.9 Spectrum acquired at 500MHz in DMSO-c/6 "Spectrum acquired at 100MHz in DMSO-tVe 'Assignments interchangeable. a  9.6,11.2  Qui-2 Qui-1, Qui-2, Qui-3, Qui-4, Qui-5  Glu2-2 Qui-3 Qui-4 Qui-5 Qui-6  Qui-4 Qui-4 Qui-4, Qui-5  Qui-2 Qui-1 Qui-5 Qui-6 Qui-3, Qui-6 Qui-4, Qui-6  Figure 3.10. H N M R of caminoside B peracetate (45) acquired at 500 MHz. 1  Table 3.2. 1D and 2D NMR data for caminoside B peracetate Pos.  Multiplicity  S "C  c  1  H  (J, Hz)  (45)  a  COSY (H  D  0  H)  HMBC (H  D  C)  aglycon  1 2 3 4 5 6 7 8 9 10 11 12 12' 13 14 15 16 17 18 19  28.8 206.1 42.8 23.7 29.3 29.5 29.8 25.5 34.7 81.5 35.4 25.8 9  1.70  s  2.03 1.56 1.27  -  2  -  t  4 3, 5 4  2,4 2, 3, 5  9  s  f  d  e  d  e  1.71 3.77 1.86  f  30.0 30.4 30.5 30.7 32.4 23.1 14.1  9  9  9  9  1.37 0.95  t  4.65 4.23  d dd  5.63 5.42 3.74 4.32 4.42  dd dd ddd dd dd  18 17, 19 18  17 17, 18  7.9 7.9,9.8  Glu1-2 Glu1-1, Glu1-3  10 Glu1-1, Glu1-3 Glu2-1  9.3,9.8 9.3,10.1 2.7,5.5,10.1 2.7,12.1 5.5,12.1  Glu1-2, Glu1-4 Glu1-3, Glu1-5 Glu1-4, Glu1-6 Glu1-6' Glu 1-5, Glu 1-6' Glu1-5, Glu1-6  Glu1-2 Glu1-4' Glu1-3, Glu1-5 Glu1-6  Glu1 Glu1-1 Glu1-2 Glu1-4 Glu1-5 Glu1-6 - ' G  l  u  G l u 1  1  3  6  100.7 75.4 7  4  7  69.8 72.0 62.4 "  Glu1-5  -si CO  Glu2  Glu2-1 Glu2-2 Glu2-3 Glu2-4 Glu2-5 Glu2-6 Glu2-6' Glu2-7 Glu2-8 Glu2-8" Glu2-9 Glu2-10 Glu2-11 Glu2-12 Glu2-13 Glu2-14  Deoxytal Deoxytal-1 Deoxytal-2 Deoxytal-3 Deoxytal-4 Deoxytal-5 Deoxytal-6 Qui Qui-1 Qui-2 Qui-3 Qui-4 Qui-5 Qui-6 a b c d,  e, f,  g  99.8 75.9 75.4 70.0 73.7 69.0  172.6 36.2  18.3 13.4 172.1 35.8 18.2 13.3  99.8 66.9 68.6 67.9 69.3 15.8  4.89 4.00 5.52 4.98 3.69 3.82 4.03  d dd dd dd ddd dd dd  7.8 7.8, 9.0, 9.0, 2.5, 8.6, 2.5,  -  -  -  2.19 2.34 1.64 0.84  dt dt m t  -  -  1.98 1.42 0.70  t m t  4.82 5.75 5.22 5.25 3.46 1.11  d dd dd dd dq d  9.1 9.1 10.3 8.6, 10.3 11.4 11.4  Glu2-2 Glu2-1, Glu2-2, Glu2-3, Glu2-4, Glu2-5, Glu2-5,  Glu2-3 Glu2-4 Glu2-5 Glu2-6, Glu2-6' Glu2-6' Glu2-6  Deoxytal-2 Deoxytal-1, Deoxytal-2, Deoxytal-3, Deoxytal-4, Deoxytal-5  95.8 5.45 d 4.1 Qui-2 71.5 4.1, 10.4 5.03 dd Qui-1, 70.1 5.83 dd 9.2, 10.4 Qui-2, 73.4 5.13 dd 9.2, 10.3 Qui-3, 65.4 4.59 dq 5.4, 10.3 Qui-4, 17.0 1.44 d 5.4 Qui-5 Non-natural acetates omitted from table. Chemical shifts are overlapped in C Spectrum acquired at 500MHz in C D Spectrum acquired at 100MHz in C D Assignments interchangeable. 1 3  6  6  6  6  Deoxytal-1  Glu2-9, Glu2-9, Glu2-8, Glu2-9,  Glu2-10 Glu2-10 Glu2-10 Glu2-10  Glu2-13 Glu2-12, Glu2-14 Glu2-13  1.7 1.7 3.1, 4.1 1.9 1.9, 6.3 6.3  Glu2-3, Qui-1 Glu2-4, Glu2-7 Glu2-5, Glu2-11  Glu2-8', Glu2-9 Glu2-8, Glu2-9 Glu2-8, Glu2-8', Glu2-10 Glu2-9  -  Glu2-5 Glu2-1, Glu2-2, Glu2-3, Glu2-6 Glu2-5,  Deoxytal-3 Deoxytal-4 Deoxytal-5 Deoxytal-6  Glu2-13, Glu2-14 Glu2-12, Glu2-14 Glu2-12, Glu2-13  Deoxytal-4 Deoxytal-2 (w), DeoxytalDeoxytal-4, Deoxytal-6 Deoxytal-4, Deoxytal-5  Qui-3, Qui-5 Qui-3 Qui-4 Qui-5 Qui-6  Qui-2, Qui-4 Qui-3, Qui-5, Qui-6  Qui-4, Qui-5 from 169.3-170.4 ppm and in H from 1.59-2.02 ppm 1  80  Chapter 3: Isolation and Characterisation of Caminosides B-D.  3.3. Isolation and Characterisation of Caminoside C . Glucose 2  O  OH  Figure 3.12. Caminoside C (43).  Caminoside C (43) was isolated using a modified protocol to that employed for the isolation of caminoside B (42). The crude methanolic sponge extract was first chromatographed by Sephadex™ LH-20 size exclusion column chromatography, then eluted with M e O H to give a complex mixture of caminosides. This material was recycled twice  on  Sephadex™  LH-20  size  exclusion  chromatography  eluting  with  EtOAc/MeOH/H 0 20:5:2. The resulting yellow glass was purified by R P H P L C (50% 2  n  PrOH/50% H 0 ) to give caminoside C (43) as a clear colourless glass (0.001 g). 2  NMR analysis provided evidence for the presence of both acetate and butyrate subunits as for 35. Several signals were observed that were consistent with the presence of two butyrate units. The H N M R contained an additional methyl group at 5 1  0.88, and the multiplets at 5 2.15 and 2.27 integrated for two protons each, rather than one as was the case for 35. The small quantity of material isolated (0.001 g) precluded the acquisition of a complete  1 3  C N M R spectrum, but assignment of carbon chemical  shift information was possible by interpretation of the H S Q C spectrum (figure 3.13).  Chapter 3: Isolation and Characterisation of Caminosides B - D .  81  O  m  o  ^  ?  G  CM  O  o  o CO  9  -  8  • •  0  o  o  J  •  o  to  1 O  O  O  O  O  O  Q  O  Q  Figure 3.13. H S Q C spectrum for caminoside C (43) acquired in D M S 0 - d 6 at 800 MHz.  Chapter 3: Isolation and Characterisation of Caminosides B-D.  82  Consideration of the H M B C spectrum provided the points of attachment for the acetate and butyrate esters, and defined the positions of the glycosidic linkages between the monosaccharide sub-units as for 42, providing the flat structure of 43. The overlapping nature of the H N M R spectrum made it impossible to determine complete coupling 1  constant information for 43. The partial coupling constant information that was available was, however, fully consistent with 43 containing the same monosaccharide sub-units as observed for both 35 and 42. This information provided the structure of caminoside C (43) as that presented in figure 3.12.  Figure 3.14. Comparison of the H NMR spectra of caminosides A (35) and C (43). 1  T a b l e 3.3.  1 D a n d 2D N M R d a t a f o r c a m i n o s i d e C  Pos.  Multiplicity  5 1 J  C  D  1  H  (43).  (J, Hz)  COSY (H  a  HMBC  H)  (H  C)  aqlvcon 1 2 3 4 5 6 7 8 9 9' 10 11 12 12' 13 14 15 16 17 18 19  28.0 208.2 42.2 24.1  2.04  80.4  3.43  13.7  0.82  101.1  4.12 3.34 3.33 3.17 3.04 3.47 3.62  s  2.39  -  2  -  m  4 3, 5  1,2, 4  C  c  c  c  Glu1 Glu1-1 Glu1-2 Glu1-3 Glu1-4 Glu1-5 Glu1-6 Glu1-6'  Glu1-2 GluM, Glu1-2, Glu1-3, Glu1-4, Glu1-5, Glu1-5,  Glu1-3 Glu1-4 Glu1-5 Glu1-6, Glu1-6' Glu1-6  a  Glu2 Glu2-1 Glu2-2 Glu2-3 Glu2-4 Glu2-5 Glu2-6 Glu2-6' Glu2-7 Glu2-8 Glu2-8' Glu2-9 Glu2-10 Glu2-11 Glu2-12  98.9 73.6 69.4 67.1  171.3 35.0  17.7 13.3 169.1 29.5 C  5.08 3.41 5.16 4.55 3.65 3.49 3.69  d dd dd dd d  -  -  2.17 2.28 1.46 0.86  m m m t  d  -  -  1.94  s  100.7  4.60  d dd  15.6  3.40 1.14  dt d  4.4 4.4, 9.7 9.7, 9.7 9.7, 9.7 9.7  Glu2-2 Glu2-1, Glu2-3 Glu2-2, Glu2-4 Glu2-3, Glu2-5 Glu2-4  Glu2-8\ Glu2-9 Glu2-8, Glu2-9 Glu2-8, Glu2-8', Glu2-10 Glu2-9  Glu2-7 (w) Glu2-11  Glu2-7, Glu2-7, Glu2-7, Glu2-8,  Glu2-9, Glu2-10 Glu2-9, Glu2-10 Glu2-8, Glu2-10 Glu2-9  Glu2-11  Deoxytal Deoxytal-1 Deoxytal-2 Deoxytal-3 Deoxytal-4 Deoxytal-5 Deoxytal-6  Deoxytal-2 Deoxytal-1  6.3 6.3  Deoxytal-6 Deoxytal-5  5.9 5.9  Qu Qu Qu Qu Qu Qu  Qui Qui-1 99.1 4.64 Qui-2 69.9 3.24 Qui-3 75.8 4.41 Qui-4 76.4 2.92 Qui-5 64.2 4.22 dt Qui-6 16.8 0.97 d Qui-7 172.0 Qui-8 35.0 2.17 m Qui-8' m 2.28 Qui-9 17.7 1.46 m Qui-10 13.4 0.87 Spectrum acquired at 500MHz in D M S O - d 6 Spectrum acquired at 100MHz in D M S O - d 6 Assignments interchangeable. C  b  , d  d  -2 -1, Qui-3 (w) -2 (w), Qui-4 -3 -6 -5  Qui-7  Qui-8', Qui-9 Qui-8, Qui-9 Qui-8, Qui-8', Qui-10 Qui-9  Qui-7, Qui-7, Qui-7, Qui-8,  Qui-9, Qui-10 Qui-9, Qui-10 Qui-8, Qui-10 Qui-9  co cn  86  Chapter 3: Isolation and Characterisation of Caminosides B-D.  3.4. Isolation and Characterisation of Caminoside D. OH  O  OH  Figure 3.15. Caminoside D (44).  Caminoside D (44) was isolated using a similar protocol to that employed for the isolation of caminoside B (42). The crude methanolic sponge extract was first chromatographed by Sephadex™ LH-20 size exclusion column chromatography, then eluted with MeOH to give a complex mixture of caminosides. This material was recycled twice  on  Sephadex™  LH-20  size  exclusion  chromatography  eluting  with  EtOAc/MeOH/H 0 20:5:2 to give caminoside D (44) as a mixture of natural products 2  that varied only in the aglycon portion of the molecule. In this case, there was an absence of signals for the acetate group present at Glu2-4 in 35 and 43. Instead, signals consistent with the presence of three butyrate units were observed in both the 1  H and  of the  1 3  1 3  C spectra. As an illustrative example, the differences in the carbonyl regions  C spectra for 35 and 44 are presented in figure 3.16. Carbon chemical shifts for  acetate carbonyls are typically observed further upfield than those for butyrate carbonyls. In figure 3.16 the acetate carbonyl of 35, present at 5 169.2 in the upper trace, is absent in the trace for 44. Instead, two more deshielded signals are observed at 5 171.8 and 172.3, implying the presence of three butyrate units.  87  Chapter 3: Isolation and Characterisation of Caminosides B-D.  Caminoside A  Caminoside D  ~T~  182  '  1  1  1  1  1  '  178  1  '  1  174  1  — — i — i — i — i — i —  1  170  166  Figure 3.16. Expansions of the carbonyl regions of the  1 3  162  C N M R spectra for 35 and 44  Once again, detailed N M R analysis allowed for the partial assignment of the structure and the positioning of the three butyrate units by interpretation of JCH cross3  peaks in the H M B C spectrum (figure 3.18). In order to obtain a single compound from which to perform a more thorough structural assignment, however, it was necessary to acetylate 44 to the corresponding octaacetate 46. Glucose 2  o  Figure 3.17. Caminoside D peracetate (46).  OAc  88  Chapter 3: Isolation and Characterisation of Caminosides B-D.  Glu2-H3  Qui-H3  Glu2-H4  160  - 164  168  (Uf  w  /  h. <m*•  »  \  Glu2-C7 Glu2-C11 Qui-C7  - 176  180 184 5.2  5.0  4.8  4.6  Figure 3.18. Expansion of the carbonyl region of the H M B C spectrum for 44.  Treatment of 44 with acetic anhydride and 4-(dimethylamino) pyridine (DMAP) in pyridine overnight followed by reversed phase H P L C (50% PrOH/50% H 0 ) provided n  2  caminoside D peracetate (46) as a clear colourless glass (0.002 g). HRESIMS of 44 gave an [M + Na] ion at 1483.7057, consistent with a formula of C i H n 0 3 i N a (calc. +  7  2  1483.7085) and acceptable for an analogue of 43 with the acetate group at Glu2-4 replaced by a butyrate unit. In order to confirm that this reaction had generated the peracetate of 44, separate LRESI mass spectra of 44 were run using both MeOH and MeOD-d4 as solvents. In the case where deuterated solvent was employed, a quantitative deuterium shift of eight mass units, from 1147 to 1155, was observed, confirming the presence in 44 of eight exchangeable protons.  Chapter 3: Isolation and Characterisation of Caminosides B-D.  89  Detailed N M R analysis provided the complete flat structure for 44 and confirmed the positions of the three butyrate units. Consideration of the available JHH coupling 3  constant information for 46 showed once again that the relative stereochemistry for the polysaccharide portion was identical to that of 34. The absolute stereochemistry was again presumed to be unchanged from the stereochemistry of 34. Signals for the aglycon were identical to those observed for 34 and 45, and conservation of the aglycon sub-structure as a linear hydrocarbon, with a methyl ketone at C2 and an acetal linkage to the carbohydrate segment at C10, was determined. This information completed the structural elucidation for caminoside D (44) presented in figure 3.15.  and provided the final structure as  Figure 3.19. Comparison of the H NMR spectra of caminosides A (35) and D (44). 1  Table 3.4. 1D and 2D N M R data for caminoside D (44).  i a  aqlvcon 1 .2 3 4 5 6 7 8 9 9' 10 11 12 12' 13 14 15 16 17 18 19  Multiplicity  5  Pos.  c  b  1  H  (J, Hz)  COSY (H  a  HMBC  3  (H  H)  2.05  s  -  -  80.6  3.45  101.2  4.14 3.33  Glu1-2  69.6 76.3 60.4  3.16 3.03 3.51 3.63  Glu1-3, Glu1-4, Glu1-5, Glu1-5,  - • 4, 5 3, 5 3, 4, 6  2.38 1.42 1.20  Glu1 Glu1-1 Glu1-2 Glu.1-3 Glu1-4 Glui-5 Glu1-6 Glu1-6'  C)  2, 3  -  29.4 208.0 42.6 23.0 28.9  3  10, Glu1-5(w)  Glu1-5 Glu1-6, Glu1-6' Glu1-6' Glu1-6  Glu1-5  Chapter 3: Isolation and Characterisation of Caminosides B-D.  •<* t  o o o o T— T—  CD CD CD CD  CD CD_ CD CD _  CO  CO" CO" CN CN  op  CN CN CNI CN 3 3 3  CNI 3  CD  CD CD CD CD  1 1  CD  <c—  CNI CNI CNI CN g 3 3 CD  O CD O O  Q  CO  g  O  ™  I CN  CNI  CD  0  co  co"  J2  3  O0  1  CNI  3  3  3  CN  3  „ co C D co  CN  CN  i2-9,  CN  10  1 11  CN CN  _3 _3 _3 _3  T—  CN  CO  ^J"  LO LO  CNI C N  CN  CN  CN  CN  3  3  3  3  3  3  CN  CN  3  3  T-  „  - - " 00 • • 00  CN  3  CN  3  CNI  CN  3  3  CN  3  CN  3  '5 O  O O O O  CN  3  v V  CNI ^ < N C M  CD O CD CD co" CO" CNI CN| C O T— T— T— T  a  CN  3  CNJ CNJ CNJ  3  3  3  CO .-L .-L 3 3  CD CD CD CD CD  3  -  CM CN" C N " CN" CO" T - T— T— T— CN  CN CN  3  CN  3  CN I C •N  3  3  CN  3  CN CN  3  3  CN  CNI  3  3  a CN a CO CO 3  3  LO  10  d  d  h-  3  d  TJ  1^ CO 00 O CD CN LO CD 0 LO CO LO CO CO CO  CN  CN CN CN  ,_:  CN d  LO  3  •t"  T3 TJ  d CO  3  O O O O O O  -tf" oS  1^  i  CN CO" ^_L _L _L J . '3 '3 '3 '3  cp  CD CDCD CD CD CD CD , O CD CDCD CD , CD CD CD CD CD  10  LO  3  3  CD CD CD CD CD 00 C O 00 00 a>  O CD CD CD CD CD CN  a.  CN  . C O co 1—  CD CD CD CD  CD CO  O.  a>" 1  i  CD CD CD CD  CD CD CN  ' 3 ' 3 .—.  0 3  co  o  CD CD CD CD CN CN CN CN _3 _3 _3 _3  CD CD CD CD  Q  3  CN CN CNI CN _3 _3 _3 _3  0 0 5  CN CN CN CN 3 3 3 3  CD  ^  CN CN CN CN _3 _3 _3 _3  CNI CNI CN CNJ 3 3 3 3 T— T-  93  d CD  1^ CD  00 d CO  1  LO  "siLO 00  T—  T—  1^ 1^  d  CN  CN CNCN  T—  1  OO  LO LO LO OO T—  T—  00  x—  d  |--  d CO  CO  T—  1 •<—  CN CN CO  T - ( N I O C ) T f L O ! D c o N C O c o C n o ) J - ^ , - ^ , C N C N C N C N C N C N C N C N C N C N C N C N ^ C M C N C Nv ]  J 2 J 5 J 2 J 5 J - . J 2 j 3 j 5 i 2 ^ J 3 _ 3  3  3  3 -fj  3  -5  CD CO  00 Oi  "-f  d  ^  CM  CD  00  C N CNII  d  -tf- d  I -; s  d  CO  O  1 d , _  I CD S  d  CN 1  CO CD  CD CO  CO  ,  3  OOOOOOOCDOOCDOOCDOOCDOCD  o  0)1  Q  T-  CN CO  -5J-  (0  CO  CO  CO CO  X  X  X  X  X X  Q  O O O O O CD CD CD CD CD Q  Q  Q  ^  1^ N(D  d 1-  LO CD  CO  O (D  S  CNI C N CD T -  Q  Q  , T -  CN CO -ST LO CD  3 — — .—  U  •— •—  '3 '3 '3 '3 '3  r>\  |  '3  o o o o a a  Qui-7 Qui-8 Qui-8' Qui-9 Qui-9' Qui-10  172.3 35.8  17.7  13.3°  2.12 2.24 1.45 1.55 0.84  Qui-8', Qui-9, Qui-9' Qui-8, Qui-9, Qui-9' Qui-8, Qui-8', Qui-9', Qui-10 Qui-8, Qui-8', Qui-9, Qui-10 Qui-9, Qui-9'  Qui-7, Qui-7, Qui-7, Qui-7,  Qui-9, Qui-9, Qui-8, Qui-8,  Qui-10 Qui-10 Qui-10 Qui-10  Spectrum acquired at 500MHz in DMSO-cf6. "Spectrum acquired at 100MHz in DMSO-c/6. Assignments interchangeable.  a  d  CD  Figure 3.21. H N M R of caminoside D peracetate (46) acquired at 500 MHz. 1  Chapter 3: Isolation and Characterisation of Caminosides B-D.  97  CD i T -  Ti CN  2  _zj  O  O  co"  m"  2 2 2  01  a> co CO  ^  CO  CN  CN C N  oo r-~" N-"  O  O (D (D (3 O  CD  2 O co  m  co" c o  CD  2 2 2 2 2 2 CM  >  oooooo T-"  o  CO  oo N." oo  O  O  CO  aj  |2  co"  m"  O O O O O  CO  "O  o  CD ID  CO  CM  1-'  co  in  a> CO q * o o * c o (J) m N CD iri b o CO ^i- b b b CM CM CM CM CM 0 0 CM CM CO  05  oo 00  CO  CD" L O "  iri  — ro  q  4-  ^t n %  i r i o i d d b CO CO CM CM CO CO CO  00  b CM CO CO  c  "5  CD  T3  O T - N c o ^ r m c o N c o o ) ^ - ( M n ^ - L O C D N C D 0 5  T5  T3 "O  o o T—  iri  T3 "O  T3 XI  uo  00  d  CN CN CO  CO o  iri iri h-  CM  CM T— CO  CM 0 0 CO CM c o  q  •<- ^ in ^ ^2 ^ ^ 2 CM  O  oi .  CO b  -a -a  q  CM CM T— •«-  ai  d  •o 0> CO N  i r i  O  q  CO  T— T— co" CD" 00  CO  CM"  J3 ^ J3 J3 J2 ^  CD CO  ooooooo  J2  Glu2 Glu2-1 Glu2-2 Glu2-3 Glu2-4 Glu2-5 Glu2-6 Glu2-6' Glu2-7 Glu2-8 Glu2-8' Glu2-9 Glu2-10 Glu2-11 Glu2-12 Glu2-13 Glu2-14  100.1 76.3 75.8 70.4 74.0 69.3  172.6 36.3  18.3 13.7 172.0 35.8 18.3 13.6  4.88 4.00 5.52 4.99 3.68 4.02 3.80  d dd dd dd ddd dd dd  7.9 7.9, 8.9 8.9,8.9 8.9, 10.3 2.4, 9.0, 10.3 2.4, 11.3 9.0, 11.3  -  -  -  2.19 2.34 1.62 0.83  m m t  7.7  -  -  -  1.92 1.42 0.69  m t  7.6  4.81 5.75 5.22 5.25 3.45 1.09  d dd dd dd dq d  1.6 1.6, 3.7, 1.0, 1.0, 6.4  Glu2-2 Glu2-1, Glu2-2, Glu2-3, Glu2-4, Glu2-5, Glu2-5,  Glu2-3 Glu2-4 Glu2-5 Glu2-6, Glu2-6' Glu2-6' Glu2-6  Glu1-2, Glu2-5 Glu2-1, Glu2-3, Qui-1 Glu2-2, Glu2-4, Glu2-7 Glu2-3, Glu2-5, Glu2-6, Glu2-11 Glu2-1, Glu2-6 Deoxytal-1 Deoxytal-1, Glu2-5  Glu2-8', Glu2-9 Glu2-8, Glu2-9 Glu2-8, Glu2-8', Glu2-10 Glu2-9  Glu2-7, Glu2-7, Glu2-7, Glu2-8,  Glu2-9, Glu2-10 Glu2-9, Glu2-10 Glu2-8, Glu2-10 Glu2-9  Glu2-13 Glu2-12, Glu2-14 Glu2-13  Glu2-11, Glu2-13, Glu2-14 Glu2-11, Glu2-13, Glu2-14 Glu2-12, Glu2-13  Deoxytal-2 Deoxytal-1, Deoxytal-3 Deoxytal-2, Deoxytal-4 Deoxytal-3 Deoxytal-6 Deoxytal-5  Glu2-6, Deoxytal-2 Deoxytal-4  Deoxytal Deoxytal-1 Deoxytal-2 Deoxytal-3 Deoxytal-4 Deoxytal-5 Deoxytal-6  100.1 67.3 69.0 68.4 69.7 16.2  3.7 3.7 3.7 6.4  Deoxytal-2, Deoxytal-3 Deoxytal-1, Deoxytal-4, Deoxytal Deoxytal-4, Deoxytal-5  Qui Qui-2 Qui-1 96.2 5.45 d 4.3 Qui-2 5.04 Qui-1, Qui-3 72.0 dd 4.3, 10.2 Qui-3 70.2 5.86 10.2, 10.2 Qui-2, Qui-4 dd Qui-4 73.7 10.2, 10.3 5.15 dd Qui-3, Qui-5 6.2, 10.3 Qui-4, Qui-6 Qui-5 65.9 4.59 dq Qui-6 17.3 1.45 d 6.2 Qui-5 Qui-7 172.1 Qui-8 36.1 2.02 m Qui-9 18.7 Qui-9 1.48 Qui-8, Qui-10 Qui-10 13.5 0.71 t 7.6 Qui-9 Non-natural acetates omitted from table. Chemical shifts are overlapped in C from 169.3-170.4 ppm and Spectrum acquired at 500MHz in C D °Spectrum acquired at 100MHz in C D ' ' 'Assignments interchangeable. a  1 3  b  d  e  6  6  6  6  Qui-3, Qui-5, Glu2-2 Qui-3 Qui-2, Qui-4, Qui-7 Qui-3, Qui-5, Qui-6 Qui-4(w) Qui-4, Qui-5  Qui-7, Qui-9, Qui-10 Qui-7, Qui-8, Qui-10 Qui-8, Qui-9 in H from 1.59-2.02 ppm 1  99  Chapter 3: Isolation and Characterisation of Caminosides B-D.  3.5. Biological Activities of Caminosides A - D.  In addition to the activity of the caminoside family in the initial type III secretion inhibition screen, we were interested in exploring the activities of the caminosides against a wider range of human pathogens. Agar disc diffusion assays were performed on a number of bacteria and fungi, with the positive results presented in table 3.6.' In this assay, activity was defined as being shown for any disc with a clear zone of exclusion beyond its perimeter. In addition to the activities presented in table 3.6, the same samples were tested and shown to be inactive as anti-microbial agents against the following microbes: E. coli; Psaab; Peep; Calbi; and botrytis. Pure samples of caminosides A (35), B (42) and D (44) were also tested as anti-mitotic agents and G2 checkpoint inhibitors but showed no activity in either case. Table 3.6. Biological activities of caminosides A - D (ux)/disc). MRS A  VRE  Xm  Pythium  Mixture of caminosides A - D  12.5  6.3  >100  25  Type III secretion inhibition active  Caminoside A (35)  12.5  12.5  25  50  active  Caminoside A peracetate (34)  >100  >100  >25  >100  inactive  Caminoside B (42)  6.3  3.1  >25  25  active  Caminoside B peracetate (45)  >25  >25  >25  >25  -  Caminoside D (44)  6.3  6.3  >25  >100  active  Caminoside D peracetate (46)  >25  >25  >25  >25  -  Compound  ' Anti-microbial biological assays  performed by Helen Wright, Bio-services  Chemistry, University of British Columbia.  Section, Department of  100  Chapter 3: Isolation and Characterisation of Caminosides B-D.  It is not surprising that the caminosides proved inactive against E. coli, given that the initial screen was designed to identify compounds that could selectively disable the attachment mechanism of pathogenic bacteria such as E. coli to host cells without exhibiting cytotoxicity towards these same bacteria. That the caminosides inhibited the growth of some but not all of the other pathogens tested should also be viewed as a positive result since a high degree of selectivity is a primary requirement for drug candidates. The structures of the caminosides lack the simplicity of structure normally required for drug candidates that do not possess a viable, renewable natural source, however this selectivity of action does make the caminosides potentially useful as molecular probes to further explore mechanisms of bacterial pathogenesis.  3.6.  Chemical Ecology.  The morphology of the sponge Caminus sphaeroconia  is extremely unusual in  that it has a hard 'armour-like' outer shell that encompasses a softer inner core. The vast majority of marine sponges show little to no variation in texture throughout their cross section, frequently maintaining their shape by the use of a silicate lattice-like internal skeleton.  Internal Flesh Figure 3.23. Photograph of Caminus  External Shell  sphaeroconia.  Chapter 3: Isolation and Characterisation of Caminosides B-D.  The inner flesh of Caminus sphaeroconia  101  also exhibits this silicate skeleton but  the outer shell provides the sponge with the unusual advantage of a physical defensive barrier with which to augment it's chemical armoury. Because we possessed a quantity of frozen sponge that had yet to be extracted with methanol we thought it an interesting experiment to segregate the shell and the flesh and to extract them separately in order to explore the possibility that the two components might exhibit different chemical constitutions. A portion of frozen sponge (50 g) was carefully dissected and both samples examined to ensure no cross contamination. Each sample was extracted in MeOH (50 mL) and a portion of the solution concentrated to dryness in vacuo. Comparison by mass spectrometry showed an unexpectedly large variation in the metabolic constitution of the shell in comparison to that of the flesh, as exhibited in figure 3.24. The internal flesh contained significant quantities of all four known caminosides ([M + Na] = 1049 (caminoside A (35)); 1077 (caminoside B (42)); 1119 (caminoside C +  (43)); 1147 (caminoside D (44))). By contrast, the external shell contained almost exclusively caminoside A (35), with only trace amounts of caminosides B to D present. One can hypothesise that 35 may play a more significant ecological role than 42, 43 or 44, perhaps as an anti-feedant or as an anti-fouling agent. In line with this hypothesis, pure samples of 35, 42, 44, 34, 45 and 46 were tested for antibacterial activity against ten species of bacteria isolated from marine sources. A s with the assays performed to investigate the activities of the caminosides versus human pathogens, agar disc diffusion assays were performed and activity defined as a clear zone of exclusion beyond the perimeter of the disc. 35 proved to be more active than either 42 or 44, showing anti-bacterial activity at 25 u.g/disc against four species, compared with 42 and 44, which showed activity at the same concentration against only two of the bacterial  Chapter 3 : Isolation and Characterisation of Caminosides B-D.  102  species. 34, 45 and 46 were completely inactive in all cases. While the observed activities are not strong, these results do lend some credence to the idea that 35 may play a heightened role as an anti-microbial agent in defending the sponge from pathogenic bacteria within the water column. In particular, this defence may take the form of preventing surface fouling by bacteria that would otherwise cut off both light and nutrient flow to the sponge.  Chapter 3: Isolation and Characterisation of Caminosides  103  B-D.  Flesh  Figure 3.24. ESI mass spectra for constituents of Caminus  sphaeroconia.  Chapter 3: Isolation and Characterisation of Caminosides B-D.  104  The question of the true origin of secondary metabolites isolated from marine sponges has been the subject of a great deal of speculation in the marine natural products literature over the years. With a few exceptions, however, little effort has been made experimentally to differentiate between genuine sponge metabolites and those produced by the symbiotic bacteria present in sponge samples. While the difference in 2  metabolite concentrations observed between the shell and flesh samples is not a de facto guarantee that the caminosides are genuine sponge metabolites it makes a strong case in favour of this viewpoint. There are numerous reports of marine invertebrates transporting metabolites to strategic sites at the exterior of the organism. example is that of the marine mollusc Cadlina  luteomarginata,  3,4  A classic  which transports  metabolites to glands on the perimeter of the dorsum, presumably as a method of defence against predation. It is plausible that Caminus sphaeroconia  could selectively  transport compounds required for its own self-defence to the outer shell in a similar fashion. It is less easy to accept the idea that bacteria in different regions of the sponge are capable of the same level of control over metabolite concentration. In the end however, only further experimentation such as the cell sorting experiments performed by Faulkner " and by Kerr would definitively answer the question of the origin of the 5  8  9  caminosides.  3.7. Conclusions. The caminosides possess a number of structural features not found in sponge glycolipids to date. They possess a fully functionalised glucose residue (Glu2), as well as a 6-deoxytalose functional group, which is rarely seen in nature. The aglycon also contains several unusual features, including the C19 linear chain, and the methyl ketone terminus, both unprecedented in sponge natural products chemistry. Finally, the  Chapter 3: Isolation and Characterisation of Caminosides B-D.  105  caminosides are the first compounds to show activity (IC  50  = 20uJv1) in a screen  designed to identify small-molecule inhibitors of the type III secretion pathway of bacterial pathogenesis.  3.8. Experimental.  3.8.1. General Experimental Procedures.  See section 2.7.1.  3.8.2. Isolation of Caminosides B (42), C (43) and D (44).  A portion of the frozen material (83 g wet wt.) was extracted with MeOH (3 x 200 mL) and the combined extracts concentrated to dryness in vacuo to give a yellow solid (5.5 g). A portion of this material (1.5 g) was purified by Sephadex™ LH-20 size exclusion chromatography eluting with 100% MeOH to give a clear yellow glass (0.467 g). A portion of this material (0.150 g) was then recycled twice on Sephadex™ LH-20 size  exclusion  chromatography  eluting  with  EtOAc/MeOH/H 0 2  20:5:2  to  give  caminosides B (42), C (43) and D (44). Caminoside C (43) was further purified by R P H P L C (50% PrOH/50% H 0 ) . n  2  Caminoside B (42): Clear colourless glass (0.028 g). [ a ]  25 D  - 2 2 ° (c=0.175, MeOH). For  1D and 2D N M R data see table 3.1. HRESIMS: [M + Na] = 1077.5815 ( C 5 i H o 0 N a ; +  9  22  calcd 1077.5821). Caminoside C (43): Clear colourless glass (0.003 g). [ a ]  25 D  - 9 ° (c=0.019, MeOH). For  1D and 2D N M R data see table 3.3. HRESIMS: [M + Na] = 1119.5913 (CsaH^O^Na; +  calcd 1119.5927).  Chapter 3: Isolation and Characterisation of Caminosides B-D.  106  Caminoside D (44): Clear colourless glass (0.031 g). [ct]  25 D  - 5 4 ° (c=0.194, MeOH). For  1D and 2D N M R data see table 3.4. HRESIMS: [M + Na] = 1147.6262 (CssHgeOzsNa; +  calcd 1147.6240).  3.8.3. Isolation of Caminoside B Peracetate (45).  Caminoside B (42, 0.012 g) was acetylated by stirring in pyridine (9 mL, 111 mmol) and acetic anhydride (3 mL, 31 mmol) with 4-(dimethylamino)pyridine (1 mg, 0.008 mmol) under N at 25 °C for 18 hrs. The reaction mixture was concentrated to 2  dryness in vacuo and partitioned between 1N HCI (100 mL) and EtOAc (100 mL). The organic phase was washed with water (2 x 100 mL) and concentrated to dryness to give a yellow solid. This was subjected to flash silica gel column chromatography (gradient elution from C H C I to 2% M e O H in C H C I ) to give a clear colourless glass (0.008 g). 2  2  2  2  Finally this material was purified by R P H P L C (Inertsil C n  1 8  9.4x250 mm, eluting with 43%  PrOH/57% H 0 , UV detection at 210 nm) to give pure Caminoside B peracetate (45, 2  0.003 g) as a clear colourless glass. [a]  25 D  -30° (c=0.019, MeOH). For 1D and 2D N M R  data see table 3.2. H R E S I M S : [M + Na] = 1455.6814 (CegH^OsiNa; calcd 1455.6772). +  3.8.4. Isolation of Caminoside D Peracetate (46).  Caminoside D (44, 0.018 g) was acetylated under identical conditions to those utilised in the acetylation of 42 {vide supra), and the reaction mixture worked up in an similar fashion. The crude reaction mixture was subjected to flash silica gel column chromatography (gradient elution from C H C I to 2% M e O H in C H C I ) to give a clear 2  2  2  2  colourless glass (0.010 g). This material was purified by R P H P L C (Inertsil C i 9.4x250 8  mm, eluting with 50% PrOH/50% H 0 , UV detection at 210 nm) to give pure n  2  Caminoside D peracetate (46, 0.002 g) as a clear colourless glass. [a]  25 D  -41° (c=0.013,  Chapter 3: Isolation and Characterisation of Caminosides B-D.  107  MeOH). For 1D and 2D N M R data see table 3.5. H R E S I M S : [M + Na] = 1483.7057 +  (C iH 7  1 1 2  0 i N a ; calcd 1483.7085). 3  Chapter 3: Isolation and Characterisation of Caminosides B-D.  108  3.9. References.  1. Linington, R. G.; Robertson, M.; Gauthier, A.; Finlay, B. B.; van Soest, R.; Andersen, R. J . Org. Lett. 2002, 4, 4089-4092. 2. Kobayashi, J.; Ishibashi, M. Chem. Rev. 1993, 93, 1753-1769. 3. Barsby, T.; Linington, R. G.; Andersen, ,R. J . Chemoecology 2002, 72, 199-202. 4. Fontana, A.; Tramice, A.; Cutignano, A.; d'lppolito, G.; Gavagnin, M.; Cimino, G. J. Org. Chem. 2003, 68, 2405-2409. 5. Haygood, M. G.; Schmidt, E. W.; Davidson, S. K.; Faulkner, D. J . J. Mol. Microb. Biotech. 1999, 7, 33-43. 6. Faulkner, D. J.; Unson, M. D.; Bewley, C. A. Pure Appl. Chem. 1994, 66, 1983-1990. 7. Faulkner, D. J.; He, H. Y.; Unson, M. D.; Bewley, C. A.; Garson, M. J . Gazz. Chim. Ital. 1993, 723, 301-307. 8. Bewley, C. A.; Holland, N. D.; Faulkner, D. J . Experientia 1996, 52, 716-722. 9. Mydlarz, L. D.; Jacobs, R. S.; Boehnlein, J.; Kerr, R. G. Chem. Biol. 2003, 10, 10511056.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  109  carteri.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa carteri. 4.1. Introduction. A review of the natural products literature reveals only one publication that has probed the chemistry of the tropical marine sponge Stylissa carteri.^ This investigation led to the isolation of a number of pyrrole containing secondary metabolites as illustrated in figure 4.1. Of these, two (47 and 48) were novel compounds, with nine having previously been reported from other sponge species.  50  49  48  47  Metabolites unique to Stylissa carteri  NH  Br,  Br  N' H  Ii  O  51  52 R = R = H 55 R = R = H 53 R = Br, R = H 56 = Br, R = H 54 R = R = Br 57 R = R = Br 1  2  1  1  1  2  2  2  2  1  2  Figure 4.1. Previously reported metabolites from Stylissa carteri.  In addition, a review of the literature pertaining to the chemistry of sponges of the genus Stylissa  reveals only four further articles. Of these, two present novel carbon  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  110  Figure 4.3. Previously reported secondary metabolites re-isolated from marine sponges of the genus Stylissa.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  skeletons as illustrated in figure 4.2,  2,3  carteri.  111  while the other two provide discussions of the  biological activities of previously reported secondary metabolites as illustrated in figure 4.3.  4,5  Curman  et  a/  5  reported  G  2  checkpoint  inhibition  activity  for 10-Z-  debromohymenialdisine (55) and 10-Z-hymenialdisine (56) but no activity for the structurally similar debromoaxinohydantoin (63). In addition Tasdemir et at  have  reported 10-Z-hymenialdisine (56) and 10-E-hymenialdisine (53) as MEK-1 inhibitors, but reported little to no inhibitory activity against the same protein by: aldisine (60); 2bromoaldisine (61); 10-Z-debromohymenialdisine (55); hymenin (49); oroidin (51); or 4,5-dibromopyrrole-2-carbonamide (62). W e will return to the topic of biological activity for this broad class of marine alkaloids in chapter five, where a full overview and discussion of the reported biological activities to date will be presented. 4.2. Isolation and Characterisation of Latonduines A (66) and B (67).'  As part of an ongoing effort in the Andersen lab to discover novel selective cytotoxic agents from marine  sponges, a sample  of Stylissa  carteri  (Dendy)  (Demospongiae, order Halichondrida, family Dictyonellidae) was subjected to chemical investigation due to its initial in vitro cytotoxic activity against several human cancer cell lines. A methanolic extract of the sponge sample (figure 4.4), collected off Latondu Island in the Taka Bonerate region of Indonesia, was partitioned between H 0 and 2  EtOAc and the EtOAc fraction was then further purified by Sephadex™ LH-20 size exclusion column chromatography eluting with 8:2 MeOH/ C H C I . The initial cytotoxic 2  2  activity was located in fractions that remain under investigation. However, N M R and  Isolation and characterisation work for latonduines A and B performed by Dr D. E. Williams, Andersen research group, University of British Columbia.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  112  T L C analysis of a separate inactive fraction indicated the presence of two related alkaloids. Consequently, this fraction was subjected to reversed phase H P L C (eluent: 45% MeOH/55% H 0 ) which afforded pure samples of latonduine A (66) (2.9 mg), 2  latonduine B methyl ester (68) (0.5 mg) and latonduine B ethyl ester (69) (2.8 mg) (figure 4.5).  6  Figure 4.4. Photograph of Stylissa carteri. NH  R R R R  =H = C0 Me = C0 Et =C0 H 2  2  2  2  Latonduine Latonduine Latonduine Latonduine  Figure 4.5. Latonduines A (66) and B (67).  A B B B  (66) methyl ester (68) ethyl ester (69) (67)  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  113  Latonduine A (66) gave a 1:2:1 M ion cluster at m/z 371, 373, and 375 in the +  LREIMS, indicating that the molecule contained two bromine atoms. A analysis of 66 showed that the mass of the M  +  HREIMS  cluster peak at m/z 372.9002 was  appropriate for a molecular formula of C i o H N 0 B r B r (calcd 372.8997) requiring 7 9  7  seven sites of unsaturation. The  1 3  8 1  5  C NMR spectrum (DMSO-c/6) of 66 (figure 4.7)  contained ten well-resolved resonances consistent with the H R E I M S data, and the H M Q C spectrum demonstrated that only three of the protons in the molecule were attached to carbon (1 x C H ( £ H 3.90 (d, J = 5.2 Hz, 2H), C 46.4); 1 x C H ( £ H 8.76 1  1 3  1  2  s,  1 3  C 155.9); 8 x C (£  1 3  C 96.0, 107.9, 113.4, 120.0, 125.1, 161.8, 162.1, 163.7)).  Exchangeable resonances were observed at £ 6 . 8 8 (s, 2H), 8.14 (t, J = 5.2 Hz, 1H), and 13.10 (s, 1H) in the H N M R spectrum (DMSO-d6) of 66 (figure 4.6), accounting for 1  the remaining hydrogen atoms. The C O S Y spectrum contained a single cross-peak confirming the scalar coupling between the aliphatic methylene protons (£3.90) and the exchangeable proton at £8.14. As already mentioned, a previous chemical investigation of S. carteri resulted in the isolation of several oroidin alkaloids, including Z-hymenialdisine (56) and Z-3bromohymenialdisine (57).  1  The H M B C spectrum of latonduine A (66)  contained  correlations that were consistent with the presence of the 2,3-dibomo-4-alkyl-5-amido fragment found in 57. Thus, the exchangeable resonance at £ 13.10, assigned to the pyrrole NH, showed H M B C correlations to carbon resonances at £ 9 6 . 0 and 120.0, assigned to C-3 and C-4, respectively, and the exchangeable resonance at £ 8.14, assigned to the N-7 amide NH, showed correlations to carbon resonances at £ 125.1, assigned to C-5, and 46.4, assigned to C-8. By analogy with Z-3-bromohymenialdisine (57), the carbon resonance at £ 107.9 in the spectrum of 66 was assigned to C-2.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  114  Figure 4 . 6 . H N M R spectrum of latonduine A (66) recorded in DMSO-d6 at 500 MHz. 1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  115  Figure 4.7. C N M R spectrum of latonduine A (66) recorded in DMSO-d6 at 100 MHz. 1 3  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  116  Figure 4.8. H N M R spectrum of latonduine B ethyl ester (69), DMSO-af6 at 500 MHz. 1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  117  Figure 4.9. C N M R spectrum of latonduine B ethyl ester (69), DMSO-c/6 at 100 MHz. 1 3  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  118  Figure 4.10. H N M R spectrum of latonduine B methyl ester (68), DMSO-d6, 500 MHz. 1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  HQ  The remaining fragment of latonduine A (66)  had to account for four sp 2  hybridized carbons (1 x C H , 3 x C), an isolated aromatic proton (£8.76), three nitrogen atoms, two equivalent exchangeable protons (£ 6.88), and five sites of unsaturation. These structural requirements could be satisfied by linking the C-4 position of the pyrrole and the C-8 methylene to adjacent carbons of a 2-aminopyrimidine fragment as shown in 66. A n H M B C correlation observed between the aromatic methine resonance at 8 8.76 (H-11) and the carbon resonance at 8 120.0, assigned to C-4, tentatively suggested the orientation of the 2-aminopyrimidine ring as shown. Additional H M B C correlations observed between the C-9 resonance at £163.7 and the H-11 (£8.76), H8/H8' (£3.90), and NH -15 (£6.88) resonances, between the C-10 resonance at £113.4 2  and the H-11 (£ 8.76) and H-8/H-8'(£ 3.90) resonances and between the C-11 resonance at £ 155.9 and the NH -14 (£6.88) resonance, were consistent with the 2  assigned structure 66. The four-bond NH -14 to C-11 and C-9 correlations seemed to 2  be reasonable since "W coupling" pathways existed for both long-range correlations. Latonduine B ethyl ester (69)  gave a M ion at m/z 444.9193 in the HREIMS +  appropriate for a molecular formula of C H n N 0 3 B r B r 7 9  1 3  5  8 1  (calcd 444.9195) that  required eight sites of unsaturation. The NMR data obtained for 69 showed strong similarities to the data for latonduine A (66),  indicating that the two compounds were  closely related. The major differences in the N M R data for 69, compared to the data for 66, were the absence of an aromatic resonance in the region of £8.76, assigned to H11 in 66, and the presence of a complex two-proton multiplet at £ 4 . 1 9 and a threeproton triplet at £ 1.19, both assigned to an ethoxy fragment. An H M B C correlation from the ethoxy methylene at £4.19 to a carbon resonance at £ 165.0 demonstrated that the ethoxy fragment was part of an ethyl ester. The ethyl ester had to be attached at C-11 in  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  120  69 to explain the absence of the H-11 methine resonance. All of the additional twodimensional N M R data obtained for latonduine B ethyl ester was consistent with the proposed structure 69. Latonduine B methyl ester (68) gave a 1:2:1 M ion cluster at m/z 429, 431, 433 +  in the LREIMS. H R E I M S analysis gave an M  +  ion at m/z 430.9050, providing a  molecular formula of C H 9 N 0 3 B r B r (calcd 430.9053). Analogous to 69, 68 also 79  1 2  81  5  lacked the aromatic singlet resonance at 58.76 present in the proton spectrum of 66. In place of the signals assigned to the ethyl ester in 69, 68 contained a sharp singlet (5 3.75, s, 3H) in the H N M R spectrum and new signals at 5 52.6 and 165.5 in the C 1  1 3  NMR. These variations allowed for the assignment of a methyl ester in place of the ethyl ester present in 69 and the absence of the resonance found at 58.76 in 66 once again allowed for the positioning of this new methyl ester at the C-11 position of the latonduine skeleton, thus completing the assignments for all three isolated compounds. Isolation of both the methyl and ethyl esters of latonduine B (67) from a sponge that had been exposed to both ethanol and methanol during workup suggested that latonduine B occurs naturally as the free acid 67 and that the methyl (68) and ethyl (69) esters are isolation artifacts.  4.3. Total Synthesis of Latonduine A (66).  4.3.1. Introduction.  While the data acquired for 66, 68 and 69 accurately fit the connectivity of the proposed structures, it was not possible to rule out by spectroscopic means an alternative carbon skeleton whereby the connectivity of the pyrimidine ring was reversed as exhibited by structure 70 in figure 4.11. Though a number of J H and J H H M B C 2  3  C  C  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  121  carteri.  correlations were present in the data acquired for latonduine A (66), a number of key correlations were not observed. In principle, consideration of the data provided by a complete H M B C spectrum should be sufficient to rule out one of the two possible /•eg/oisomers, however with the limited quantity of sample available many of the theoretically possible signals were either weak or entirely absent. Analysis of those signals present determined that insufficient data was available to differentiate between the two possible structures (figure 4.12). Mindful of this ambiguity, a total synthesis of 66 was undertaken, with the dual aims of confirming the tentative structural assignment by synthetic methods and supplying more material for further biological testing.  Figure 4.11. Latonduine A (66) and its reg/oisomer 70.  H,N  or  H  66  6 70  Figure 4.12. Possible interpretations for selected H M B C correlations for latonduine A (66).  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  Latonduine A (66)  carteri.  122  represents only the second oroidin alkaloid with a C10  framework, the majority of the others containing eleven carbons or more. In addition, latonduine A (66)  is the first reported compound to contain a 6 membered nitrogenous  heterocycle (in this case an amino-pyrimidine) and the first compound in the oroidin alkaloid class to exhibit a 5,7,6 fused ring system. These unique structural features made the total synthesis a desirable goal in its own right, and strengthened our enthusiasm for selecting 66 as a synthetic target.  4.3.2. Overview of Oroidin Alkaloid Syntheses. Marine alkaloids of the oroidin family have been synthetic targets for many researchers since the original isolation of oroidin (51) in 1971. To date over 30 papers 7  discussing the total syntheses of compounds from this class have been reported, with a further 9 papers presenting progress towards the construction of various members of the oroidin family. The wide range of biological activities attributed to members of this 8  class, as well as the relatively frequent discovery of novel carbon skeletons has helped to drive these synthetic investigations. As figure 4.13 shows, the number of reported syntheses is increasing yearly as researchers seek to exploit the potential therapeutic value of these small molecules.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  123  carteri.  Total Number of Reported Syntheses of Oroidin Alkaloids Since 1971 25 n  co  CD CO CD  20  J  c >^ CO 1 5 H CD  n E  10  co o  0  •••••••  1970  1975  • • • • • • • • 1980  1985  1990  1995  2000  Year  Figure 4.13. Chronology of reported oroidin alkaloid syntheses since 1971.  The continued interest in the oroidin family as targets for synthesis is in fact driven by four principle factors: the opportunity to construct novel carbon skeletons and to use these targets as a justification for the development of new synthetic methods; interest in the biomimetic cyclisations and dimerisations of non-cyclised imidazole alkaloids; investigations  pyrrole  into the reactivity and functionalisation of 2-  aminoimidazoles; and the generation of larger amounts of material for biological and structure-activity  relationship  (SAR) studies.  Of these  possible  motivations for  completing a total synthesis of latonduine A (66), only the third can be said to be irrelevant to our investigation, as the latonduine target lacks the 2-aminoimidazole  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  124  carteri.  functional group. The question of whether or not to pursue a biomimetic approach to the synthesis of 66 was to be the defining factor in deciding the style of the synthetic route.  4.3.3. Proposed Synthesis for Latonduine A (66). Historically the syntheses of monomeric oroidin alkaloids has fallen into one of two camps. One approach has been to construct the complete open carbon chain before cyclisation to form the final product, an example of which is shown in scheme 4.1. This approach has found greatest utility in the formation of dibromophakellin (71), dibromoisophakellin (72) and dibromophakellistatin (73) skeletons (figure 4.14), though there are also selected examples in the literature of the formation of small quantities of compounds containing fused 5,7 ring systems. The other approach has been to construct the rings sequentially, typically forming a CgN2 framework before the introduction of the second nitrogenous heterocycle (scheme 4.2). 0  Br,  H  ©  1.5 eq tBuOK,  H N 2  Br  71  Scheme 4.1. Biomimetic synthesis of dibromophakellin (71).  9  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  125  carteri.  H,N  72  73  Figure 4.14. Dibromoisophakellin (72) and dibromophakellstatin (73).  (  N  >  ^  H  C  O  C  ,  3  i  )  B  r  2  '  A  C  °  H  ii) M e C N , rt,  74  H z N ' ^ O ^  M e S 0 H , 40°C 3  H  2  7days  N ^ N  NH N  2  ^ v = T , M e S 0 H , rt  Br  3  7 days  Scheme 4.2. Non-biomimetic synthesis of (±) hymenin (49)  10  In the case of 66, several factors combined to favour a non-biomimetic approach. Literature precedent suggested that we would face difficulties finding favourable conditions for the formation of the fused 5,7 pyrroloazapinone ring system. As stated above, the majority of conditions utilised for the cyclisation of linear C n N  5  precursors  have led to the formation of fused or spiro 5,6 systems, and those conditions that result in the formation of pyrroloazapinone ring systems do so only as side products in modest to low yields. In the examples presented to date where such fused 5,7 ring systems have been observed, the final product has contained the pyrroloazapinone core with an  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  126  carteri.  additional five membered nitrogenous heterocycle attached a to the pyrrole. By contrast, the formation of 66 requires the formation of the aforementioned 5,7,6 ring system. The increased proximity of the third ring to the functionalised pyrrole leads to a significant increase in the steric interaction between the bromine attached at C-3 of the pyrrole, and the proton at C-11 on the pyrimidine ring (figure 4.15). This unfavourable steric interaction further decreases the likelihood of the successful formation of the latonduine skeleton by biomimetic methods. NH  2  66  49  Figure 4.15. Steric interactions present in hymenin (49) and latonduine A (66).  In contrast to the limitations inherent to the biomimetic approach, the reports of oroidin alkaloid syntheses by traditional synthetic methods offered the basis for a number of potential synthetic routes. Of principle interest was the report of Home and coworkers of the synthesis of hymenin (49) via the intermediate 76 (scheme 4.2).  11  A number of  transformations were envisioned that could be used to convert the olefinic group present in 76 into a ketone " 12  16  to give intermediate 77 with the newly introduced ketone  functionality (3 to the 4-position of the pyrrole (scheme 4.3). Of these possibilities, the selected  route  corresponding  proceeded ketone  77  via  1,2-diol 78  which  could  be converted  via a reg/oselective dehydration  across  into the  C-9/C-10.  mechanistic rationale for the predicted reg/oselectivity is proposed in scheme 4.4.  A  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  Scheme 4.3. Proposed transformations for the synthesis of latonduine A (66).  77 Scheme 4.4. Proposed mechanism for the dehydration of 78 to generate 77.  127  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  128  Introduction of a formyl group at C-10, a to the 4-position of the pyrrole, would generate compound 80 containing an a-formyl ketone moiety. The most widely used procedure in the literature for the formation of aminopyrimidines is via the condensation of p-diketone systems with guanidine under mild basic conditions. The introduction of the formyl group in 80 provides such a p-diketone system and allows for the condensation with guanidine as the final step in the synthesis. With this information in hand an initial synthesis was proposed as outlined in scheme 4.3.  4.3.4. Synthetic Results.  As presented in the retrosynthetic analysis of latonduine A (66) {vide supra), the initial key intermediate was the pyrroloazapinene 76. A s has already been discussed, this intermediate had been presented in the literature by Home and coworkers as a precursor in the synthesis of the marine natural product hymenin (49),  11  providing  conditions for the synthesis of this desired material. Following an established literature procedure  17  the commercially available pyrrole trichloromethyl ketone 74 was dissolved  in glacial acetic acid, and a solution of bromine in acetic acid added dropwise over fifteen minutes. After stirring at room temperature for twentyfour hours, standard workup conditions followed by flash silica gel column chromatography afforded 79 in excellent yield. The subsequent transformation required the formation of an amide linkage between the trichloromethyl ketone portion of 74 and the amine portion of 1,3dioxolane-2-ethylamine. This was accomplished by solublising 79 and 1,3-dioxolane-2ethylamine in dry acetonitrile and stirring at room temperature for eighteen hours. A s described by Home and coworkers the desired amide 75 is insoluble in acetonitrile, and consequently a large quantity of white precipitate was observed in the reaction vessel at the end of the reaction period. Filtration of the reaction mixture followed by washing of  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  129  the precipitate with a small volume of acetonitrile provided pure 75 as a fine white solid in good yield. The final step to 79 as described by Home required stirring of 75 in neat methylsulfonic acid at room temperature for seven days to give the cyclised olefin 76. In our experience, this procedure was unsuccessful at room temperature, with no reaction observed. Elevating the temperature led to generation of the required product; however, the choice of temperature proved critical to the yield of the final material. Performing the reaction at 35°C gave reproducibly good yields of 76;  however increasing the  temperature of the reaction mixture to even 40°C led to the production of large amounts of strongly coloured polymeric by-products which were uncharacterisable by NMR methods and led to a dramatic drop in the yield of 76. Utilising these three steps as described above, we were able to generate large quantities of the desired olefin 76 in a concise fashion and good yield as shown in scheme 4.5.  76  75  Scheme 4.5. Synthetic route to 76. At this point in the synthesis our proposed route diverged from that of Home with the attempt to oxidise the olefin of 76 to the corresponding diol 78. Treatment of 76 with t-butyl hydroperoxide and tetraethylammonium acetate in the presence of a catalytic quantity of osmium tetroxide  12  led to an inseparable mixture of oxidation products that  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  130  retained the characteristic H NMR signals for the olefinic protons at C-9 and C-10 of 1  76.  Use of alternative oxidising agents such as N-methylmorpholine-N-oxide  18  did  nothing to improve the reg/'oselectivity of the reaction and so this route to the desired ketone 77 was discarded. An alternative approach to the introduction of the ketone at C9 of 76 would be to convert the olefin in 76 to its corresponding epoxide (81), followed by treatment with aqueous acid to give ketone 77 (scheme 4.6). O  76  81  O  77  Scheme 4.6. Proposed formation of 77 via epoxide 81.  Treatment of 76 with dimethyldioxirane (DMDO, prepared via Murray protocol ) 19  in acetone at -78°C for two hours followed by warming to room temperature gave numerous products observable by TLC. This mixture of closely related products proved inseparable by chromatographic methods however N M R analysis of the crude sample again showed characteristic signals for the olefinic protons at C-9 and C-10 of 76. While it was not possible to perform a complete structural assignment on any of the products from this reaction due to their inseparable nature, there was some evidence from the H 1  and  1 3  C N M R spectra that competing oxidative reactions were occurring at various  positions on the pyrrole portion of the molecule. Attempts to perform the epoxidation using traditional methods with m-chloroperbenzoic acid (mCPBA) as the oxidising agent  20  met with a similar lack of selectivity for the double bond, and therefore attempts  to access the desired ketone 77 via direct oxidative treatment of the olefin 76 were  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  131  halted in the hope of altering our proposed synthetic scheme to utilise methods which did not require the direct oxidation of the pyrrole/olefin combination. An attractive possibility was the introduction of a functional group at the C-9 position of 76, p to the pyrrole, which could be subjected to a subsequent functional group interconversion to a secondary alcohol. There exist a large number of conditions in the literature for the selective oxidation of secondary alcohols to their corresponding ketones,, many of which are mild and applicable to situations where other potentially oxidisable functionalities are present. The revised general approach for the generation of 77 was therefore as presented in scheme 4.7.  77  82  Scheme 4.7. Revised approach to the formation of ketone 77.  One area of organic chemistry that is often employed in the functionalisation of olefins is the hydroboration chemistry developed extensively by Dr. Herbert C. Brown. Professor Brown's contributions to this field earned him the Nobel Prize in Chemistry jointly with Dr. Georg Wittig in 1979 for "their development of the use of boron- and  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  132  phosphorus-containing compounds, respectively, into important reagents in organic synthesis".  21  Brown's versatile methodology is applicable to olefins with all levels of  substitution, from the simple case of ethylene to highly substituted olefins such as 1,2dimethylcyclopentene. In the case of tetrasubstituted olefins, reactions tend to proceed slowly and often require more forcing conditions than with mono-, di- or tri-substituted olefins. However, our system contained only a single disubstituted olefin and was therefore an ideal candidate for the use of hydroboration chemistry.  Scheme 4.8. General reactivity of diborane with olefins.^  The predicted regioselectivity of the reaction of 76 with an alkyl borane species can be rationalised by the consideration of two complementary factors. In general the steric bulk of the alkyl portion of an activated borane species will orientate the alkylborane so as to minimise steric interactions with other portions of the molecule. Electronic effects may also play a role in determining the position of the formation of the new carbonboron bond if the potential exists for resonance stabilisation of the transition state intermediate. The addition of borane to double bonds is generally considered to occur via a four-centre transition state, with both the C-B and C-H bonds being formed simultaneously. One can also present the addition of borane species to double bonds as an electrophilic attack on the olefin by the borane. In our case, one reg/oisomer is favoured over the other on electronic grounds due to the resonance stabilisation of the carbocation intermediate (scheme 4.9). Conveniently in our case, this is also the  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  133  carteri.  regioisomer that is favoured on steric grounds, due to the steric clash between the borane and the pyrrole ring.  H  6  Scheme 4.9. Reg/'oselectivity of the hydroboration of 76.  Modern methods exist that allow for a one-pot conversion of olefins to the corresponding  alcohols  via  an  organoborane  intermediate.  23  Formation  of this  organoborane intermediate, as discussed above, is followed by aqueous alkaline oxidative workup to give the corresponding alcohol, most often by quenching the reaction mixture with a dilute solution of sodium hydroxide and treatment with aqueous hydrogen peroxide. Methods also exist for the direct conversion of organoboranes to aldehydes and ketones, though these techniques require stronger oxidising agents and more forcing conditions than those required for the formation of alcohols. " 24  26  In an attempt to achieve maximum reg/'oselectivity for the hydroboration reaction 76 was treated with a solution of 9-borabicyclo[3.3.1]nonane (9-BBN) in THF followed by oxidative workup employing aqueous chromic acid in order to convert 76 directly into the desired ketone 77. 9-BBN was chosen as the hydroborating reagent due to the large steric bulk of the alkyl portion, which has been reported to dramatically improve the  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  134  regioselectivity of the hydroboration of a number of cyclic olefins in comparison to the use of borane in the formation of the organoborane intermediate.  27  In our case, no  reaction was observed upon treatment of 76 with 9-BBN, presumably because the steric constraints were too great. By contrast, repeating the reaction using boron trifluoride etherate and sodium borohydride followed by oxidative aqueous chromic acid workup as for the previous reaction afforded not the ketone 77 as expected but compound 82 with a secondary alcohol at position C-9. These conditions provided only modest yields of 82 and numerous by-products. However, modifying the conditions by the use of catecholborane as the hydroborating agent,  28  coupled with an alteration to the workup  procedure to employ the milder conditions utilising hydrogen peroxide, as more commonly used for alcohol formation, provided the alcohol 82 in high yield. 29  Several reliable methods exist for the oxidation of secondary alcohols to their corresponding ketones under mild conditions in the presence of other potentially oxidisable functionalities. In this case, treatment of 82 with Dess-Martin periodinane at 30  room temperature in T H F for one hour provided an almost quantitative conversion of 82 to the ketone 77 with a combined yield for the two steps of 82% (scheme 4.10).  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  Dess Martin periodinane  LiBH , THF 4  ii) N a O H , H 0 2  135  carteri.  THF 2  84%  98%  Scheme 4.10. Conversion of olefin 76 to ketone 77.  The next step in the proposed synthesis of latonduine A (66) called for the introduction of a formyl group at C-10 of 77 in order to introduce the p-diketone required for the construction of the aminopyrimidine ring. Our initial approach was to employ strongly basic conditions in order to deprotonate 77 at C-10, and then to treat the resulting carbanion with ethyl formate, generating a new carbon-carbon bond at C-10/C11. Despite repeated attempts under a number of different conditions and with a variety of bases we were never able to observe any evidence for the introduction of the desired formyl group at C-10. Instead we chose to treat 77 with triethylorthoformate under acidic conditions in an attempt to promote an acid catalysed enolisation at C-9/C-10, followed by alkylation at C-10.  31  Refluxing 77 in neat triethylorthoformate in the presence of a  catalytic quantity of trifluoroacetic acid (TFA) proceeded as predicted to generate the hemiketal 83, albeit in only moderate yield. With this material in hand, the last remaining step was the construction of the aminopyrimidine ring by condensation of 83 with guanidine. This was achieved following a standard literature procedure by refluxing 83 in a mixture of T H F and water in the presence of guanidine hydrochloride and potassium carbonate to afford 66 as a white solid (scheme 4.11).  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  77  carteri.  136  83  66  Scheme 4.11. Conversion of ketone 77 into latonduine A (66).  Comparison of the N M R data for the synthetic sample of latonduine A (66) with the  data  collected  on  the  natural  product  showed  the  two  samples  to  be  spectroscopically identical (figure 4.16), thus providing definitive confirmation for the proposed structure of the natural product and verifying the completion of the total synthesis of Latonduine A (66).  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  137  Figure 4.16. Comparison of the H NMR spectra for synthetic and naturally occurring 1  latonduine A (66).  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  138  4.4. Biological Activities of the Latonduines.  Initial work on Stylissa carteri was driven by the activity of the crude extract of the sponge as a cytotoxic agent against human cancer cell lines. This activity was subsequently shown to be due to the presence of a complex mixture of metabolites that is still under investigation. None of 66, 68 or 69 proved active in this original screen, and subsequent re-screening of all three compounds failed to show any evidence of interaction of these metabolites with a small panel of protein kinases. This result means that the latonduines are without known biological activity, in itself a remarkable result given the wide range of biological activities reported for many of the other oroidin alkaloids. " 2  4  4.5. Biogenetic Implications of Latonduines Isolation.  4.5.1. Introduction.  Many of the publications in the literature reporting the identification of novel oroidin alkaloids have also contained suggestions about their possible biogenetic origin.  9,32  '  35  Indeed, this area of inquiry has become large enough to warrant a recent  review on the subject by Potier and Mourabit.  36  This review presented the common  structural motifs of this class of compounds and suggested possible biosynthetic routes in each case. The vast majority of these examples commenced with condensation of a pyrrole-2-carboxylic acid subunit with a primary alkyl amine containing an N-heterocycle to give the linear C n N  x  presented in scheme 4.12.  (x=4,5) precursor to further /rrtramolecular cyclisations as  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  NH  carteri.  139  2  Scheme 4.12. Representative cyclisation products formed via oroidin (51) (adapted from Mourabit ). 36  This generalised scheme has been strengthened by the natural co-occurrence in sponges of both the 4,5-dibromopyrrole-2-carboxylic aminoimidazolyl)-prop-1-ene (85) subunits,  37,38  acid (84)  and 3-amino-1-(2-  and further elaborated by the suggestion  of possible biosynthetic origins for both 84 and 85. Despite the inherent difficulties with performing feeding studies on marine sponges due to the problems of organism viability in laboratory environments, Kerr and coworkers performed the first investigation into the biogenetic  origin of an oroidin  alkaloid when they were  able to  demonstrate  incorporation of the radio-labelled amino acids [U- C]proline, [C - C]omithine and [U14  14  5  14  C]histidine into the natural product stevensine (50).  39  Separate cell cultures of the  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  140  marine sponge Teichaxinella morchella were incubated in the presence of one of each of four radio4abelled amino acids for forty eight hours, followed by isolation of 50 from the cell culture in each case and measurement of the radioactivity count for each sample. Interestingly, performance of the experiment with [U- C]arginine led to the 14  isolation of a sample of 50 which exhibited no radioactivity, and this combination of results enabled Kerr to postulate biogenetic pathways for the construction of both 84 and 85 (scheme 4.13).  Ornithine  Proline  Histidine  Scheme 4.13. Literature proposal for the biogenesis of 84 and 85.  Arginine  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  141  In their proposal, it was suggested that 84 could derive from either ornithine or proline by one of two complementary pathways, though no further speculation was offered as to the precise mechanisms by which these two conversions might occur. The authors went on to suggest, based on their experimental results, that 85 is derived solely from histidine via oxidative de-amination, followed by reduction of the carbonyl with subsequent amination at the terminal position, and that no evidence exists for the construction of 85 via a theoretical route employing arginine as a precursor. This work represents the only study to employ cell cultures of a marine invertebrate to probe the origin of a marine metabolite, and is the only study to date that has sought to investigate the biogenetic origin of an oroidin alkaloid. 4.5.2. Biogenesis of Latonduines A (66) and B (67).  If one attempts to postulate a mechanism for the construction of latonduine A (66) or B (67) using the general scheme presented in schemes 4.12 and 4.13, one immediately encounters a number of insurmountable problems. Most significant of these is that in order to introduce the 6-membered aminopyrimidine ring one is forced to consider a ring expansion reaction that would introduce a carbon between C-11/C-12 or C-12/N-13 of the existing aminoimidazole functionality. A survey of the literature shows that  there  are no reported  synthetic  methods  for the direct  conversion  of  aminoimidazoles to the corresponding aminopyrimidines. In addition, we have not been able to find any examples, either proven or postulated, for this conversion in Nature. Fortunately an alternative possible pathway exists that circumvents this difficulty with the aminopyrimidine formation while providing an additional possible route to the construction of many of the existing oroidin alkaloids. Scheme 4.14 shows our proposed biogenesis for latonduine B (67) that suggests that the building blocks of the molecule  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  142  carteri.  are 4,5-dibromopyrrole-2-carboxylic acid (84), ornithine, and guanidine. Latonduine A (66) would arise from decarboxylation of lantonduine B (67). The biogenetic scheme also suggests that the hymenialdisines might arise from an alternate cyclization of an intermediate like 86 that involves the ornithine carboxyl functionality and a guanidine nitrogen  to  give  the  five-membered  oxoaminoimidazole  ring  bromohymenialdisine (54).  NH  H  2  6  Latonduine A (66)  Br^-  H  o  Latonduine B (67)  E-3-Bromohymenialdisine (54)  Scheme 4.14. Proposed biogenesis for latonduines A (66) and B (67).  found  in E-3-  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  143  This scheme is attractive as an alternative biosynthetic pathway because of the ease with which one can rationalise the generation of both the latonduine and traditional oroidin alkaloid frameworks, and the simplicity with which oxidation and functionalisation patterns can be explained. If this proposal proves to be correct, then we can expect to see 87 and 88 as reported marine metabolites in the future, along with other possible cyclisation products such as 89. COOH  89  Figure 4.17. Predicted structures for future oroidin alkaloids.  4.6. Conformational Analysis of Latonduines A (66) and B (67).  The H N M R spectra for latonduine A (66) and the ethyl ester of latonduine B 1  (69) revealed an interesting variation between the two compounds. Whereas for 66 the signals for the enantiotopic C-8 methylene protons H-8/H-8' occurred as a sharp doublet at 5 3.89, the signals for the protons in the same position in 69 appeared as two distinct multiplets at 5 3.77 and 4.08. This variation can be attributed to the steric bulk of the  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  144  ester group at C-11 of 67 interacting with the bromine atom at C-3 on the pyrrole to restrict rotation about the C-4/C-10 bond on the NMR timescale at 500 MHz.  Line of sight of i MM2 projection '  \ \f  Figure 4.18. Lowest energy conformation for 69 as predicted by MM2 calculations.  MM2 free energy calculations" predict 66 and 69 to adopt similar lowest energy conformations, both with a pronounced twist to the seven membered ring which orientates one of the protons on C-8 towards the shielding cone of the pyrrole while orientating the other into free space, away from the rest of the molecule (present as the two atoms furthest to the left of the page in the three dimensional projection of 69, figure 4.18). While one must always remain cautious about the accuracy of predictions made using  computational  methods,  particularly  in  unusual  systems  such  as  pyrroloazapinones where the number of database examples is likely to be low, the results in this case were compatible with results obtained experimentally. Using information gleaned from the computational calculations we hypothesised that the steric interaction between the bromine on C-3 and the ester at C-11 in 69 prevents free  MM2 minimum energy calculations performed using CambridgeSoft C h e m 3D Ultra v.7.0.0.  Chapter 4: Novel Alkaloids from the Marine Sponge  Stylissa  145  carteri.  i n t e r c o n v e r s i o n b e t w e e n t h e t w o r e s u l t i n g e n a n t i o m e r s , t h u s m a k i n g t h e t w o p r o t o n s at C-8 NMR  m a g n e t i c a l l y i n e q u i v a l e n t w h e r e a s f o r 66 t h i s i n t e r c o n v e r s i o n o c c u r s r a p i d l y o n t h e t i m e s c a l e , c o n s e q u e n t l y m a k i n g H-8  thought that a s e r i e s of d y n a m i c  a n d H-8'  m a g n e t i c a l l y e q u i v a l e n t . It w a s  NMR e x p e r i m e n t s (DNMR) c o u l d p r o v i d e e s t i m a t e s f o r  the e n e r g y b a r r i e r s to t h e i n t e r c o n v e r s i o n of t h e s e t w o c o n f o r m a t i o n a l s t a t e s . A s e r i e s of one dimensional H 1  NMR  e x p e r i m e n t s w e r e r u n o n 69  at e l e v a t e d t e m p e r a t u r e s f r o m  307 K to 427 K w i t h t e m p e r a t u r e i n t e r v a l s o f 20 K g a v e t h e r e s u l t s in f i g u r e 387 K  Figure 4.19. D N M R results for latonduine B ethyl ester (69),  4.19.  ||  DMSO-c/6 at 500 MHz.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  146  As can be seen with only a cursory inspection of the data presented, the two signals for H-8 and H-8', while not initially magnetically equivalent, coalesce with increasing temperature, passing the coalescence point (7~) at 347 K until at 427 K the c  signals are observed as a sharp doublet. One can utilise the data provided by the determination of 7~ to provide an estimate for the value of the free energy of activation c  (AG*) for the interconversion between the two enantiomeric conformations by using an approximation derived from the Eyring equation.  40  h In this form of the Eyring equation, k = rate constant; k = Boltzmann constant; T B  = absolute temperature; h = Planck constant; and R = universal gas constant. In simple systems where the transition state can freely transfer energy to its surroundings K « 1 (commonly the case with small organic molecules). In addition, Dr. H. S. Gutowsky showed in 1959 that at T , the rate of rotation (k ) is given by the equation below. c  41  c  k =nvV2y c  Substituting this expression into the Eyring equation gives f  AG* = -R7"ln|  xAvh ^  which in turn can be converted to log™ and, inserting numerical vales for k , h, and R, B  be usefully converted into AG* =19.127" (10.32 + Iog7~ -\ogk ) C  c  c  Analysis of the spectra in figure 4.19 gives a Av value of 100 Hz that in turn gives a AG* value of 72 kJmol" (or 17.2 kcalmol" ). 1  1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  Performing the same DNMR analysis on latonduine B methyl ester (68)  147  gave  identical results for 7~ (347 K) and Av (100Hz) (figure 4.20). This gives a A G * value of c  72 kJmol" and suggests that the difference in steric bulk between the methyl and ethyl 1  esters is not large enough to significantly affect the energy barrier to interconversion. Success with the D N M R for 69 prompted us to investigate the energy barrier to interconversion of the two potential enantiomeric conformations for 66. In this case the resonances for the methylene protons on C-8 were magnetically equivalent, appearing as a sharp doublet at room temperature that intergrated for two protons. It was thought that by cooling the sample it might be possible to slow down the rate of interconversion of the two enantiomeric conformations, allowing us to estimate a value for AG*. A series of low temperature one dimensional DNMR experiments were performed on a sample of latonduine A (66) in MeOD-d4 at temperatures ranging from 298 K to 218 K. As can be seen in figure 4.21 the signals for H-8/H-8' gradually disperse with decreasing temperature from their initial appearance as a sharp singlet to a discrete pair of mutually coupled doublets at 218 K (J = 15 Hz). Analysis of this data in an analogous fashion to the analysis performed on 68 and 69 gave a Av value of 63 Hz, which in turn provided a value of 52 kJmol" (or 12.4 kcalmol" ) for AG*. 1  1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  148  carteri.  300 K H8  H8'  \  4 . 5 0  4 . 2 5  4 . 0 0  3 . 5 0  3 . 7 5  307 K  A  I 4 . 5 0  4 . 2 5  4 . 0 0  3 . 7 5  .  /J^. I 3 . 5 0  327 K  —I 4 . 5 0  4 . 0 0  3 . 7 5  4 . 2 5  4 . 0 0  3 . 7 5  4 . 2 5  4 . 0 0  3 . 7 5  4 . 2 5  1 3 . 5 0  347 K  4 . 5 0  3 . 5 0  367 K  4 . 5 0  3 . 5 0  Figure 4.20. D N M R results for latonduine B methyl ester (68), DMSO-d6 at 500 MHz.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  149  carteri.  298K  i  i  ppm  1  n  i  i  1  1  4.50  4.25  4.00  3.75  3.50  3.25  4.75  4.50  4.25  4.00  3.75  3.50  3.25  4.75  4.50  4.25  4.00  nr  3.75  3.50  3.25  4.75  4.50  4.25  4.00  3.75  3.50  T 3.25  4.75  4.50  4.75  :  278K  ppm  258K  ppm  248K  ppm  "T"  23 8 K  ppm  —r—— 4.25  1  1 4.00  1 3.75  :  1 3.50  1  :—i— 3.25  218K I  ppm  1  4.75  4.50  —i  4.25  1 4.00  1 3.75  1 3.50  1 — 3.25  Figure 4.21. D N M R results for latonduine A (66) run in MeOD-d4 at 500 MHz.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  150  4.7. Experimental. Stylissa  carteri  (Dendy)  (Demospongiae,  order  Dictyonellidae) was harvested by hand using S C U B A  Halichondrida,  family  on Latondu Island, Taka  Bonerate, 80 miles south of southern Sulawesi, Indonesia. Freshly collected sponge (50 g) was initially preserved in EtOH for two days at room temperature after which the EtOH was discarded and the sample frozen. The frozen sponge was subsequently repeatedly extracted with MeOH. The combined extracts were concentrated in vacuo, and the resulting aqueous suspension was partitioned between H 0 and EtOAc. 2  Fractionation of the EtOAc soluble material, by sequential application of Sephadex LH20  chromatography  (80% MeOH/CH CI ) 2  2  and  reversed  phase  H P L C (45%  MeOH/H 0), yielded latonduine A (66) (2.8 mg, pale yellow solid), latonduine B ethyl 2  ester B (69) (2.9 mg, pale yellow crystalline solid) and latonduine B methyl ester (68) (0.4 mg, pale yellow crystalline solid).  Latonduine A (66): mp decomposes at ~290°C; UV (MeOH) A 14222) nm; H R E I M S [ M ] m/z 372.9002 ( C H 7 O N B r +  10  5  m a x  245 (e 19830), 284 (e  calcd 372.8997); H NMR (500 1  2i  MHz, DMSO-d6): 5 13.10 (s, 1H), 8.76 (s, 1H), 8.14 (t, J = 5.2 Hz, 1H), 6.88 (bs, 2H), 3.90 (bd, J = 5.2 Hz, 2H); C N M R (100 MHz, DMSO-d6): 5 163.7, 162.1, 161.8, 155.9, 1 3  125.1,120.0,113.4,107.9,9.6.0,46.4.  Latonduine B ethyl ester (69): mp 228-231 °C; UV (MeOH) A  m a x  203 (e 24131), 252 (e  25922), 284 (shoulder), 348 (e 5887) nm; HREIMS [ M ] m/z 444.9193 (C^HnOaNsBrz, +  calcd 444.9195); H N M R (500 MHz, DMSO-d6): § 13.10 (s, 1H), 8.26 (m, 1H), 7.24 (bs, 1  2H), 4.19 (m, 2H), 4.10 (dd, J = 14.5, 3.1 Hz, 1H), 3.77 (dd, J - 14.5, 7.1 Hz, 1H), 1.19  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  151  (t, J = 7.1 Hz, 3H); C N M R (100 MHz, DMSO-d6): 5 168.8, 165.1, 162.0, 161.4, 155.4, 1 3  125.4, 120.2, 111.3, 106.4, 99.1, 61.5, 46.8, 13.5.  Latonduine B methyl ester (68): UV (MeOH) A  m a x  209 (e 11181), 252 (e 13507), 345 (e  3032) nm; HREIMS [M] m/z 430.9050 (C^HgOgNsBrz, calcd 430.9053); H NMR (500 +  1  MHz, DMSO-d6): 5 13.12 (bs, 1H), 8.26 (m, 1H), 7.25 (bs, 2H), 4.09 (bd, J = 14.4, 1H), 3.77 (dd, J = 14.4, 6.9 Hz, 1H), 3.75 (s, 3H); partial  1 3  1 3  C N M R (100 MHz, DMSO-d6) (only  C N M R assignments possible from H M Q C and H M B C data): 5 168.8, 165.5,  162.1, 111.3, 52.6,46.8.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  152  carteri.  Preparation of 4,5-dibromopyrrol-2-yl trichloromethyl ketone (79). Br,  Br  IT  N  H  CCI  3  O  To a stirred solution of pyrrol-2-yl trichloromethyl ketone (11.5 g, 54.1 mmol) in acetic acid (25 mL), was added a solution of bromine (5.86 mL, 113.7 mmol) in acetic acid (225 mL). After the solution had been stirred for 24 hours, an additional quantity of bromine (0.5 mL, 9.7 mmol) was added and the reaction mixture stirred for a further 3 hours. The reaction mixture was concentrated to dryness in vacuo and partitioned between aqueous K C 0 2  (10% w/w, 250 mL) and E t 0 (200 mL). The phases were  3  2  separated and the aqueous phase was extracted with E t 0 (2 x 200 mL). The combined 2  organic phases were dried ( M g S 0 ) , filtered through Celite™ and concentrated to 4  dryness in vacuo. Purification of the crude product by flash chromatography (40 x 150 mm, hexanes/EtOAc 19:1) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 18.65 g (93%) of the bromopyrrole 79 as a pale pink solid.  1  H NMR (400 MHz, CDCI ) 6 7.35 (s, 1H), 10.27 (bs, 1H). 3  1 3  C NMR (100 MHz, CDCI ) 5 93.8, 102.4, 113.5, 123.2, 123.8, 172.1. 3  Exact mass calculated for C H O N B r 79  6  2  35 2  CI  37 2  C I : 368.7539; found: 368.7540.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  153  Preparation of amide 75.  To a solution of the bromopyrrole 79 (7.52 g, 20.3 mmol) in M e C N (40 mL) was added, dropwise, 2-(2-ethylamino)-1,3-dioxolane  (2.24 mL, 20.4 mmol). After the  reaction mixture had been stirred for 17 hours at rt it was filtered and the precipitate washed with M e C N (30 mL) to yield 6.16 g (81%) of the amide 75 as a white solid which required no further purification.  1  H NMR (400 MHz, ( C D ) C O ) 5 1.90 (m, 2H), 3.48 (q, 2H), 3.80 (m, 2H), 3.92 (m, 2H), 3  2  4.89 (t, 1H), 6.86 (s, 1H), 7.58 (bs, 1H), 11.97 (bs, 1H).  1 3  C NMR (100 MHz, ( C D ) C O ) 8 34.3, 35.5, 65.5, 99.4, 103.5, 105.3, 113.0, 129.2, 3  2  160.0. Exact mass calculated for C H O N 1 0  1 2  3  7 9 2  B r B r : 367.9194; found: 367.9194. 8 1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  154  carteri.  Preparation of olefin 76. Br, NH  Br H  O  A stirred solution of the amide 75 (2.82 g, 7.68 mmol) in M e S 0 H (15 mL) was 3  heated to 35 °C. After the solution had been stirred for 20 hours an additional quantity of M e S 0 H (10 mL) was added. After the solution had been stirred for a further 22 hours 3  an additional quantity of MeSOsH (10 mL) was added and the solution was stirred for 5 days. The solution was cooled to room temperature and poured into cold (0 °C) water (400 mL). The resultant suspension was filtered to give a grey solid. Purification of the crude product by flash chromatography (40 x 270 mm, EtOAc/hexanes 7:3) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 1.602 g (69%) of the olefin 76 as a pale yellow solid.  1  H NMR (400 MHz, C D O D ) 5 3.56 (d, 2H, J = 6.4 Hz), 6.00 (dt, 1H, J = 6.4, 10.3 Hz), 3  6.64 (d, 1H, J = 10.3 Hz).  1 3  C NMR (100 MHz, C D O D ) 8 39.7, 100.2, 108.4, 126.5, 126.7, 126.8, 127.1, 164.7. 3  Exact mass calculated for C H O N 8  6  79 2  B r ! B r : 305.8826; found: 305.8830. 8  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  155  carteri.  Preparation of alcohol 82.  H  O  To a stirred solution of the olefin 76 (338 mg, 1.105 mmol) in dry THF (10 mL) was added L i B H  4  (3 mg, 0.136 mmol). The reaction was placed under an inert  atmosphere (Ar ) and 1,3,2-benzodioxaborole (141 u.L, 1.325 mmol) was added as a (g)  solution in dry T H F (5 mL). Following stirring for 1 hour the reaction mixture was quenched with 1N NaOH (1 mL) and treated with 30% H 0 (1.3 mL, 1.325 mmol). The 2  2  reaction mixture was concentrated to dryness in vacuo and partitioned between water (100 mL) and EtOAc (100 mL). The phases were separated and the aqueous phase was extracted with EtOAc (2 x 100 mL). The combined organic phases were concentrated  to dryness  in  vacuo.  Purification  of the  crude  product  by flash  chromatography (45 x 170 mm, EtOAc then EtOAc/MeOH 9:1) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 254 mg (71%) of the alcohol 82 as a white solid and 44 mg (13%) of the starting material 76.  1  H NMR (500 MHz, ( C D ) S O ) S 2.53 (dd, 1H, J = 8.8, 16.8 Hz), 2.84 (dd, 1H, J = 4.9, 3  2  16.8 Hz), 3 . 1 0 - 3 . 1 9 (m, 2H), 3.88 (m, 1H), 5.17 (d, 1H, J = 4.1 Hz) 7.71 (t, 1H, J = 5.4 Hz) 12.44 (s, 1H).  1 3  C NMR (100 MHz, ( C D ) S O ) 5 36.4, 47.9, 67.4, 100.2, 105.5, 121.9, 123.7, 161.5. 3  2  Exact mass calculated for C H 0 N 8  8  2  7 9 2  B r B r : 323.8932; found: 323.8932. 8 1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  156  carteri.  Preparation of ketone 77.  H  o  To a solution of the alcohol 82 (96 mg, 0.296 mmol) in T H F (5 mL) was added Dess-Martin periodinane (452 mg, 1.066 mmol) and the resulting mixture was stirred for 15 minutes. The reaction mixture was concentrated to dryness  in vacuo and  partitioned  between water (100 mL) and EtOAc (100 mL). The phases were separated and the organic phase was washed with water (2 x 100 mL). The organic phase was concentrated  in vacuo.  to dryness  Purification  of the crude  product  by flash  chromatography (30 x 270 mm, EtOAc/hexanes 9:1) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 79 mg (83%) of the ketone 77 as a white solid and 14 mg (15%) of the starting material 82.  1  H NMR (400 MHz, ( C D ) C O ) 5 3.70 (s, 2H), 3.91 (d, 2H, J = 4.9 Hz), 7.15 (bs, 1H), 3  2  11.85 (bs, 1H).  1 3  C NMR (100 MHz, ( C D ) C O ) 8 41.5, 52.1, 100.7, 106.9, 121.1, 125.7, 164.1, 205.9. 3  2  Exact mass calculated for C H 0 N 8  6  2  2  Br : 323.8755; found: 323.8757. 2  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  157  Preparation of ethyl ether 83. Et  O  To a stirred solution of the ketone 77 (25 mg, 0.078 mmol) in triethylorthoformate (10 mL, 60.12 mmol) was added trifluoroacetic acid (200 uL, 0.003 mmol). The reaction was placed under an inert atmosphere (Ar ) and refluxed for 18 hours, after which the (g)  reaction mixture was concentrated to dryness in vacuo. Purification of the crude product by flash chromatography (16 x 250 mm, EtOAc/hexanes 3:7) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 10 mg (35%) of the ethyl ether 83 as a white solid.  1  H NMR (500 MHz, CDCI ) 5 1.40 (t, J = 7.1 Hz, 3H), 3.86 (bs, 2H), 4.24 (q, J = 7.1 Hz, 3  2H), 6.40 (bs, 1H), 7.63 (s, 1H), 11.00 (bs, 1H).  Exact mass calculated for C H O N 1 1  1 0  3  7 9 2  B r B r : 377.9038; found: 377.9038. 8 1  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  158  Preparation of latonduine A (66). NH  2  NH  Br H  O  To a stirred solution of the ethyl ether 83 (3.5 mg, 0.009 mmol) in THF (5 mL) and water (1 mL) were added K C 0 (3.8 mg, 0.028 mmol) and guanidine hydrochloride 2  3  (2.3 mg, 0.024 mmol) and the solution stirred at reflux for 19 hours. The reaction mixture was concentrated to dryness in vacuo and partitioned between water (20 mL) and EtOAc (20 mL). The phases were separated and the aqueous phase was extracted with EtOAc (2 x 20 mL). The combined organic phases were concentrated to dryness in vacuo. Purification of the crude product by reversed phase C™ high performacy liquid chromatography (46% M e O H / H 0 ) and removal of trace amounts of solvent (vacuum 2  pump) from the resulting solid provided 2.8 mg (82%) of latonduine A (66) as a white solid.  1  H NMR (500 MHz, ( C D ) S O ) 5 3.90 (bd, J = 5.2 Hz, 2H), 6.88 (bs, 2H), 8.14 (t, J = 5.2 3  2  Hz, 1H), 8.76 (s, 1H), 13.10 (s, 1H).  Exact mass calculated for C H O N 1 0  7  8 1 5  B r : 374.8976; found: 374.8978. 2  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  carteri.  159  4.8. References  1. Eder, C ; Proksch, P.; Wray, V.; Steube, K.; Bringmann, G.; V a n Soest, R. W. M.; Sudarsono; Ferdinandus, E.; Pattisina, L. A.; Wiryowidagdo, S.; Moka, W. J. Nat. Prod. 1999, 62, 184-187. 2. Nishimura, S.; Matsunaga, S.; Shibazaki, M.; Suzuki, K.; Furihata, K.; Van Soest, R. W. M.; Fusetani, N. Org. Lett. 2003, 5, 2255-2257. 3. Assmann, M.; van Soest, R. W. M.; Koeck, M. J. Nat. Prod. 2001, 64, 1345-1347. 4. Tasdemir, D.; Mallon, R.; Greenstein, M.; Feldberg, L. R.; Kim, S. C ; Collins, K.; Wojciechowicz, D.; Mangalindan, G. C ; Concepcion, G. P.; Harper, M. K.; Ireland, C. M. J. Med. Chem. 2002, 45, 529-532. 5. Curman, D.; Cinel, B.; Williams, D. E.; Rundle, N.; Block, W. D.; Goodarzi, A. A.; Hutchins, J . R.; Clarke, P. R.; Zhou, B.-B.; Lees-Miller, S. P.; Andersen, R. J.; Roberge, M. J. Biol. Chem. 2001, 276, 17914-17919. 6. Linington, R. G.; Williams, D. E.; Tahir, A.; Van Soest, R.; Andersen, R. J . Org. Lett. 2003, 5, 2735-2738. 7. Forenza, S.; Minale, L.; Riccio, R.; Fattorusso, E. J. Chem. Soc. Chem. Comm. 1971, 18, 1129-1130. 8. Hoffmann, H.; Lindel, T. Synthesis 2003, 12, 1753-1783. 9. Foley, L. H.; Buechi, G. J. Am. Chem. Soc. 1982, 704, 1776-1777. 10. Xu, Y.-z.; Phan, G.; Yakushijin, K.; Home, D. A. Tet. Lett. 1994, 35, 351-354. 11. Xu, Y.-z.; Yakushijin, K.; Home, D. A. J. Org. Chem. 1997, 62, 456-464. 12. Akashi, K.; Palermo, R. E.; Sharpless, K. B. Journal of Organic Chemistry 1978, 43, 2063-2066. 13. Monson, R.; Deggary, P. J. Org. Chem. 1971, 36, 3826-3828. 14. Roussel, M.; Mimoun, H. J. Org. Chem. 1980, 45, 5387-5390. 15. Lusinchi, X.; Hanquet, G. Tetrahedron 1997, 53, 13727-13738. 16. Kishi, Y.; Aratani, M.; Tanino, H.; Fukuyama, T.; Goto, T.; Inoue, S.; Sugiura, S.; Kakoi, H. J. Chem. Soc. Chem. Comm. 1972, 7, 64-65. 17. Bailey, D. M.; Johnson, R. E. J. Med. Chem. 1973, 76, 1300-1302.. 18. Van Rheenen, V.; Kelly, R. C ; Cha, D. Y. Tet. Lett. 1.976, 23, 1973-1976. 19. Murray, R. W.; Jeyaraman, R. J. Org. Chem. 1985, 50, 2847-2853. 20. Adam, W.; Bosio, S. G.; Wolff, B. T. Org. Lett. 2003, 5, 819-822.  Chapter 4: Novel Alkaloids from the Marine Sponge Stylissa  160  carteri.  21. http://www.nobel.se/chemistry/laureates/1979/. 22. Brown, H. C ; Rao, B. S. S. J. Org. Chem. 1957, 22, 1136-1137. 23. Brown, H. C ; Murray, K. J.; Murray, L. J.; Snover, J . A.; Zweifel, G. J. Am. Chem. Soc. 1960, 82, 4233-4241. 24. Brown, H. C ; Kulkarni, S. U.; Rao, C. G.; Patil, V. D. Tetrahedron 1986, 42, 55155522. 25. Brown, H. C.; Garg, C. P. J. Am. Chem. Soc. 1961, 83, 2951-2952. 26. Brown, H. C ; Garg, C. P. Tetrahedron 1986, 42, 5511-5514. 27. Brown, H. C ; Liotta, R.; Brener, L. J. Am. Chem. Soc. 1977, 99, 3427-3432. 28. Brown, H. C ; Gupta, S. K. J. Am. Chem. Soc. 1975, 97, 5249-5255. 29. Arase, A.; Nunokawa, Y.; Masuda, Y.; Hoshi, M. J. Chem. Soc. Chem. Comm. 1991,4,205-206. 30. Dess, D. B.; Martin, J . C. J. Am. Chem. Soc. 1991,  113, 7277-7287.  31. Swaringen, R. A., Jr.; Yeowell, D. A.; Wisowaty, J . C.; El-Sayad, H. A.; Stewart, E. L; Darnofall, M. E. J. Org. Chem. 1979, 44, 4825-4829. 32. Kinnel, R. B.; Gehrken, H.-P.; Swali, R.; Skoropowski, G.; Scheuer, P. J . J. Org. Chem. 1998, 63, 3281-3286. 33. Fedoreev, S. A.; Il'in, S. G.; Utkina, N. K.; Maksimov, O. B.; Reshetnyak, M. V.; Antipin, M. Y.; Struchkov, Y. T. Tetrahedron 1989, 45, 3487-3492. 34. D'Ambrosio, M.; Guerriero, A.; Debitus, C ; Ribes, O.; Pusset, J.; Leroy, S.; Pietra, F. J. Chem. Soc. Chem. Comm. 1993,  16, 1305-1306.  35. Tsukamoto, S.; Kato, H.; Hirota, H.; Fusetani, N. J. Nat. Prod. 1996, 59, 501-503. 36. Mourabit, A. A.; Potier, P. Eur. J. Org. Chem. 2001,  7, 237-243.  37. Koenig, G. M.; Wright, A. D. J. Nat. Prod. 1994, 5, 141-146. 38. Wright, A. E.; Chiles, S. A.; Cross, S. S. J. Nat. Prod. 1991, 54, 1684-1686. 39. Andrade, P.; Willoughby, R.; Pomponi, S. A.; Kerr, R. G. Tet. Lett. 1999, 40, 47754778. 40. Oki, M. Applications of dynamic NMR spectroscopy to organic chemistry; V C H Publishers: Deerfield Beach, FL, 1985. 41. Gutowsky, H. S.; Holm, C. H. J. Chem. Phys. 1959, 25, 1228-1234.  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  161  Chapter 5: Progress Towards the Synthesis of Isolatonduine. 5.1. Introduction. The vast majority of oroidin alkaloids isolated to date have been shown to possess one or more of a number of varied biological activities. The range and extent of activities shown by these metabolites make them one of the more diverse classes of compounds in terms of providing potential new leads for drug discovery. It was something of a surprise therefore to find that despite screening the latonduines as potential protein kinase inhibitors,' we were not able to demonstrate any evidence of biological activity for 66,  68 or 69.  Consideration of the spatial functional group  distribution for biologically active members of the oroidin family (vide infra) prompted us to propose that the reg/oisomer of latonduine A (66), henceforth termed isolatonduine (70), might prove to exhibit similar biological activity to that exhibited by a number of the biologically active oroidin alkaloids.  5.2. Overview of Oroidin Alkaloid Biological Activities. The biological activities reported for oroidin alkaloids include: anti-tumour; " 1  anti-bacterial; " 15  22  anti-viral;  4,17  anti-hypertensive;  23  anti-tubercular;  a-adrenoceptor blocking; actomyosin A T P a s e activating; 26  histaminic;  30  and anti-fouling  17  27,28  24  anti-muscarinic;  14  25  chitinase inhibiting; anti29  activities. The majority of compounds referenced in this  list were reported as novel structures, isolated as part of rational programs to screen for  ' Biological  screening  of 66,  68  and 69  performed  by Tamsin Tarling,  Department of Biochemistry, University of British Columbia.  Roberge  research group,  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  162  pharmacologically active agents. However, in a few cases, known structures have been re-reported because they have proven to be strongly active in a given biological assay. Of these biological responses, it is the anti-bacterial and anti-tumour activities that have received the majority of the investigative efforts to date. This is doubtless due in no small part to the overwhelming clinical need for new treatments for both of these illnesses. In the case of the anti-tumour activities of some members of the oroidin family, the number of papers reporting strongly selective cytotoxic behaviour towards cancerous cell types stands testament to the robust nature of this biological activity, and points towards this class of compounds as an excellent area for attempting rational drug design.  Both  hymenialdisine  (56)  and its debromo  analogue  55  have  been  demonstrated to have strong activity as inhibitors of an array of protein kinases. " 11  14,31  Indeed, the level of protein kinase inhibition is so strong for certain proteins that in a number of studies 56 has been used as a positive control against which the activities of newly discovered or developed inhibitors have been judged.  5,6  In addition, a number of  pyrroloazepinone compounds have already been found to be strongly active against a range of human carcinomas. " 5  7,31  Given the structural similarities between latonduine A  (66) and bioactive metabolites such as hymenialdisine (56) it was though that modifying the structure of 66 might lead to the generation of analogues that possessed similar pharmacological attributes.  5.3. Pharmacophore Model of Functional Group Spatial Distribution.  All of the oroidin alkaloids that have been reported as displaying anti-tumour activity against any cancer cell line are presented in figures 5.1 and 5.2. In addition, all compounds that have been tested for anti-tumour activity and proved to be inactive are presented in figure 5.3.  163  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  The array of compounds that display anti-tumour activities encompasses the full range of structural complexity found within this class of natural products. Linear C n N  5  compounds are represented by oroidin (51) and slagenins B (90) and C (91). Cyclised compounds exhibiting the pyrroloazepinone carbon framework are present in the hymenialdisine  (56) and axinohydantoin  (92) families, and more highly cyclised  structures are also present in the form of the phakellins, phakellstatins and agelastatins. Both the symmetrical oroidin dimer sceptrin (93) and the asymmetrical dimer ageliferin (64) have shown activity, and even simple structures containing either the pyrrole or aminoimidazole fragments in isolation (e.g. 94 and 95) are reported as being active against human cancer cell lines.  Br.  Br,  Br,  NH NH  Br  2  N H  •  2  NH Br  N H  O  Br  \ O  62  94  Oroidin (51)  2  Br,  Br  H  H  3-Bromomaleimide (96)  3,4-Dibromomaleimide (97)  Figure 5.1. Oroidin alkaloids shown to exhibit anti-tumour activity (part 1).  164  Chapter 5 : Progress Towards the Synthesis of Isolatonduine. H  H  O  H  OCH  H  3  Slagenin C (91) Br  HoN  Dibromophakellin (71)  Dibromophakellstatin (73)  CI  12-Chloro-11-hydroxydibromoisophakellin (98)  N ^ O  N ^ o  HO  H N—\ 2  Br  H  HN^  > H  H N^N  H  2  CI  R = Br Agelastatin (99) R = H 100  H  O H  2  R = Br Z-Axinohydantoin (102) R = H Debromo-Z-axinohydantoin  cr (103)  6  R = Br Axinohydantoin (92) R = H Debromoaxinohydantoin (63)  H N  b  2  Palau'amine (19)  101  R = H Debromohymenialdisine (55) R = Br Hymenialdisine (56)  H  ''-NH  N  OEt  Hanishin  (104)  Girolline (95)  Figure 5.2. Oroidin alkaloids shown to exhibit anti-tumour activity (part 2).  165  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  NH  Hymenin (105)  Oroidin (51) H C  'H C  3  R  3Q  3  R3O  N.  R =R =H, R =Me, R =Br(99) R =R =R =Me, R =H (102) R =R =H, R =Ac, R =Br(100) 1  2  1  2  1  2  3  4  3  2  4  3  4  N  ^  R = R = R = M e (106) R =R =Me, R =H (107) R =R =H, R =Me (108) 1  2  1  2  1  2  3  3  3  109  O  Slagenin A (111)  C H  3  ir  OH  o A/-Methylmanzacidin C (112)  R = H Latonduine A (66) R = C 0 M e Latonduine B methyl ester (68) R = C 0 E t Latonduine B ethyl ester (69) 2  2  Figure 5.3. Oroidin alkaloids shown to be inactive against selected cancer cell lines.  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  166  The list of compounds that have been tested and found to have no effect on cancerous cells is shorter, yet equally diverse. Once again linear C n N compounds are 5  represented, this time by oroidin (51), slagenin A (111), A/-methylmanzacidin C (112) and hymenin (105). Pyrroloazepinones appear in the form of the axinohydantoins and the more highly cyclised C n N  5  alkaloids are also present, though only as a number of  semi-synthetic analogues of agelastatin (99). By comparison one can see that several compounds or classes of compounds appear in both lists. In the cases of the slagenins this is because of a straightforward variation in potency between the three isolated natural products.  1  The agelastatin  compounds presented were all generated by simple synthetic transformations of the natural product and tested in the same screen in order to probe the functional groups required for activity . By contrast, oroidin (51) is present in both lists because it has 9  been isolated numerous times by different groups and therefore been subject to biological assays in a number of different screens. 51 is active against NSCLC-N6 human non-small lung cancer, but shows no inhibition of protein tyrosine kinase or 32  mitogen-activated protein kinase kinase-1 (MEK-1),  33  both enzymes implicated in the  inhibition of proliferation of a number of cancer types including thyroid carcinoma. Oroidin (51) also shows no activity against L5178y mouse lymphoma cells.  7  The overall impression that one gets from these results is one of disorder. There are no obvious clear requirements for molecular size or functionality that define the parameters required for activity. Linear compounds, monocyclic, polycyclic, and even dimeric skeletons have all been reported as anti-tumour agents. However, one must always keep in mind that the reported biological activities are against a wide array of different cancers and that not all structures are active in all cases. It is possible to  167  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  somewhat simplify this question of functional group requirements for activity by limiting our discussion to just one sub-class of compounds; in this case the pyrroloazepinones. In 2002, a group of researchers led by Dr. Ireland from the University of Utah published an article in the Journal of Medicinal Chemistry  33  in which eight oroidin  alkaloids were examined for their abilities to inhibit the R a s - M A P K signalling cascade protein MEK-1. The Ras protein is involved in signalling pathways in over thirty percent of all cancers and has been implicated in the modulation of cytoplasmic events such as cell proliferation and cell recognition . In light of this, inhibitors of proteins such as 34  MEK-1 represent attractive targets for drug development. The eight compounds tested as inhibitors of MEK-1 (figure 5.4) were isolated from the marine sponge  Stylissa  massa. This set of compounds was well suited to an investigation of the structural requirements for biological activity. The set contained structures with and without the 2aminoimidazole ring, the bromine at position 2 of the pyrrole, and the pyrroloazepinone ring system, providing information about the impact of each of these functionalities on the level of MEK-1 inhibition. NH  49  55 R = H 56 R = Br  51  62  Figure 5.4. Oroidin alkaloids tested as inhibitors of the MEK-1 protein.  2  53 o,  60 R = H 61 R = Br  168  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Table 5.1. Kinase enzyme inhibitory activity of selected oroidin alkaloids (IC  50  nM).  Compound  MEK-1 to M A P K assay  LoVo in vitro results  49  1288  >10000  51  >2500  >10000  53  6  586  55  824  >10000  56  9  710  60  >2500  >10000  61  539  >10000  62  >2500  >10000  Table 5.1 shows that five of the tested compounds showed initial activity in the MEK-1 to M A P K assay, though only two of those (53 and 56) showed strong inhibitory activity. In vitro treatment of LoVo human colon cancer cells with the test compounds resulted in the inhibition of cell growth only when incubated in the presence of 53 or 56. Considering the structures of the active compounds, two functional group requirements become clear. Firstly, bromination at the 2-position of the pyrrole appears to be an important contributory factor for activity. 56 is strongly active, while the corresponding debromo compound 55 shows a significant decrease in biological activity. In addition, 61 is mildly active in the' MEK-1- to M A P K assay whereas the corresponding debromo compound 60 is completely inactive. Secondly, possession of a 2-amino group on the five membered ring appears to have significant bearing on relative biological activities. 53 and 56 exhibit statistically indistinguishable biological activities, however the authors dismiss the validity of the reported activity for 53 stating that "[ajlthough the initial samples were geometrically  pure, we believe that the activity observed for [53] is not  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  169  significant, because [53] is quite unstable and smoothly converts to [56] upon standing in dimethyl sulfoxide (DMSO), the solvent used as a vehicle in the current study". This irreversible interconversion has already been proposed,  35  both for the brominated  compounds 53 and 56 and their debromo analogues 52 and 55. If this argument for the interconversion of geometric isomers is accepted, then by comparing 56, where the five membered heterocycle is fixed in position by the geometry of the olefin at C-10/C-11, with 49, where the 2-aminoimidazole is free to rotate about C-10/C-11, we can offer the suggestion that the fixed position of a primary amine in plane with the bromopyrrole is a requirement for biological activity. By this rationale, we proposed that the construction of the geometric isomer of 66, with the position of the 2-aminopyrimidine ring reversed, could provide a compound possessing a novel carbon skeleton that could act as an inhibitor of one of a number of protein kinases, either as isolatonduine (70), or as the analogous compound debromoisolatonduine (113).  70  113  Figure 5.5. Isolatonduine A (70) and debromoisolatonduine A (113). In order to test the validity of the claim that in comparison to the pyrrole ring the primary amines of 70 and 113 would lie in the same spatial region as the primary amine for 56, lowest energy conformations were calculated in each case by performing computations using the MM2 force field."  MM2 minimum energy calculations performed using CambridgeSoft C h e m 3D Ultra v.7.0.0.  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  170  Figure 5.6. Calculated minimum energy conformations for 56, 70 and 113. The predictions provided by the MM2 calculations exhibit a reasonably high degree of conformational  similarity. While the construction of a comprehensive  pharmacophore model such as that offered for selected anti-mitotic agents by Snyder and co-workers  36  is outside the scope of this work, simple overlay models show that the  relative positioning of the bromopyrrole and the primary amine is similar in all cases. While this simple analysis was in no way a guarantee that 70 or 113 would show the desired  pharmacological  activity,  it  was  another  encouragement  that  either  Chapter 5 : Progress Towards the Synthesis of Isolatonduine.  171  isolatonduine (70) or debromoisolatonduine (113) might prove to be active as protein kinase inhibitors.  5.4. Retrosynthetic Analysis for Isolatonduine (70).  The synthesis of latonduine A (66) proceeded via olefinic intermediate 76. Hydroboration  chemistry,  coupled  with aqueous  alkaline  oxidative workup, was  employed to introduce an alcohol at C-9, p to the pyrrole. Our initial proposal for the formation of 70 (scheme 5.2) involved a divergent route from that used for the construction of 66. It was hypothesised that conversion of olefin 76 to compound 114, could be accomplished by addition of an alcohol at position C-10 by employing the oxymercuration chemistry  37,38  primarily developed by both Brown and Larock in the 1970s.  114 could then be converted to the ketone 115 in an analogous fashion to the conversion of 82 to 77. With ketone 115 in hand we hoped to introduce an orthoformate group at position C-9 under identical conditions to those used to convert 77 into 83. This transformation  would  again generate  a p-diketone,  and leave  116 set up for  condensation with guanidine to form the desired 2-aminopyrimidine 70 with the required geometrical configuration.  172  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  LiBH , THF 4  H  ii) N a O H , H 0 2  O  2  84%  82  76  74  Dess Martin periodinane THF 98%  NH JI  HN NH  (EtO) CH, T F A  NH  2  3  2  K C 0 , THF, H 0 reflux 2  3  2  B  reflux  r  35%  82%  Scheme 5.1. Synthetic route to latonduine A (66). i) H g ( O A c )  2  THF/H 0 2  ii) aq. N a O H  N ' l f H  NaBH  O  Br  4  74 Dess Martin periodinane THF  HoN OEt  NH  H N NH X  2  B r - X , / - ^ H  &  70  (EtO) CH, TFA  2  3  K C 0 , THF, H 0 2  3  r e f  2  '  U X  B  reflux  r  H  6  116  Scheme 5.2. Proposed route for the construction of isolatonduine A (70).  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  173  5.5. Synthetic Results.  Unfortunately, treatment of 76 under standard solvomercuration-demercuration conditions (scheme 5.2) showed no reaction by TLC. Repeating the reaction under more forcing conditions by refluxing the substrate in the presence of mercury (II) acetate prior to demercuration with an alkaline solution of sodium borohydride made no improvement to the reactivity of the system, with T L C again showing no signs of any reaction having occurred. In the original methodology paper describing the oxymercuration-demercuration of olefins presented by Brown and Geoghegan,  37  the effects of performing the reaction  with a wide variety of substrates is described. These results prove that it is possible to achieve excellent yields even with tetra-substituted olefins, or olefins adjacent to substituents with a high degree of steric bulk, alth.ough in the latter case hydroxyl will add at the position p to the more bulky functional group. It is possible to justify the lack of reactivity of 76 towards oxymercuration by considering steric constraints. Unlike hydroboration reactions, which can be considered to proceed via a four center cyclic transition state to give the syn addition product, oxymercurations are thought to proceed via a three membered ring mercurinium transition state, followed by backside attack of hydroxyl, giving the anti addition product. Molecular models show that the seven membered ring of 76  adopts a twisted  conformation, leaving one face of the olefin hindered to attack at either carbon, possibly preventing the addition of an alcohol at either C-9 or C-10 (figure 5.7).  174  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  114 Scheme 5.3. Transition states for the hydroboration and oxymercuration of 76.  Faced with the difficulty in converting 76 to 115 under our initial conditions, we elected instead to reduce 76 by hydrogenation with the aim of introducing the desired ketone either via bromination at C-10 or by direct conversion of the C-10 methylene to the corresponding ketone under strongly oxidative conditions.  175  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  NBS, AIBN  H , Pd on C 2  OH",  H0 2  Dess Martin periodinane  Scheme 5.4. Alternative methods for the conversion of 76 to 115.  Hydrogenation of 76 at 20 p.s.i. in the presence of 10% palladium on carbon led to a 1:2:2 mixture of the desired material 117, the mono-debrominated product 118 and the fully debrominated product 119. This mixture of products was treated in situ with bromine  in acetic acid to re-brominate  dibromopyrrole 117 in 70% overall yield.  39  118  and  119,  giving the desired  2,3-  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  176  117  Scheme 5.5. Conversion of 76 to 117. There are reports in the literature of the successful bromination of ct,Bunsaturated amides and ketones in the y position under radical conditions.  40  76 may be  considered as an a,|3-unsaturated amide, and as such is theoretically amenable to photochemically assisted radical bromination. In this case, irradiation of a solution of 115,  /V-bromosuccinimide (NBS) and a catalytic quantity of 2,2'-azobisisobutyronitrile  (AIBN) with a 100W halogen light source afforded a complex mixture of inseparable products with no evidence of the formation of one major product. Given that the formation of complex mixtures has been cited as a potential pitfall with the use of radical bromination chemistry, no further effort was directed towards this line of inquiry. Attempts to convert 117 directly into 115 using a variety of oxidants also proved unsuccessful, returning pure starting material in all cases. These conditions included: treatment with dichlorodicyanobenzoquinone chlorochromate ( P C C ) ;  43,44  (DDQ);  41,42  treatment with pyridinium  Jones chromic acid oxidation;  45,46  and treatment with  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  177  molecular oxygen in the presence of a catalytic quantity of N-hydroxyphthalimide.  47,48  In  all cases, starting material was returned without evidence by T L C of the formation of any other products. Given the difficulties encountered in our attempts to generate 70 via a common intermediate to that used in the synthesis of 66, our attention next turned to the design of an independent route to 70.  There have been several reports of the use of  intermediate 120 in the construction of debromo oroidin a l k a l o i d s  23,49,50  and we aimed to  utilise this methodology by effecting a similar transformation on 121, where the pyrrole sub-unit bears bromine atoms at positions C-2 and C-3 in order to construct the desired 2,3-dibromopyrroloazepinone  115.  79  121  115  Scheme 5.6. Revised route to the synthesis of 115.  121 was constructed in a straightforward fashion by stirring a solution of 79, |3alanine and triethylamine in T H F / H 0 (3:1) at room temperature overnight. Attempts to 2  form the fused 5,7 ring system by treatment of 121 with polypohosphoric acid (PPA) and phosphorus pentachloride (PCI ) at 100°C gave no further reaction, in contrast to the 5  178  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  reported result observed for 123 where treatment under the same conditions led cleanly to the formation of the cyclised product 120. It appears clear from this sequence of results that the presence of bromine atoms at C-3 and/or C-2 severely impedes the reactivity in the position a to C-4. Attempts to brominate or oxidise at C-10 of 2,3-dibromopyrroloazepinones has met with universal failure, and attempts to convert 121 to 115 by the creation of a new carbon-carbon bond between C-4 and C-10 have also proved fruitless. With these results in hand it was deemed necessary to delay the introduction of bromines at C-2 and C-3 until a late stage in the synthesis. The conclusions drawn in section 5.3 by comparing the biological activities of a number of related oroidin alkaloids suggested debromoisolatonduine (113)  as an  equally attractive target to 70. Results from previous syntheses have reported both mono and dibrominations at C-2 and C-3 of similar pyrrolo s y s t e m s .  39,50  Armed with this  information we elected to construct the analogous 2,3-dihydroisolatonduine (122) with the intention of generating both 70 and 113 from 122 by utilising different conditions for bromination as the final steps in each synthesis. H N. 2  N  H  o 122  Figure 5.8. 2,3-Dihydroisolatonduine (122). Starting with pyrrole-2-carboxylic acid (82) and employing a standard literature procedure  23  generated 120 in modest yield (scheme 5.7). Treatment of this material with  triethylorthoformate and a catalytic amount of trifluoroacetic acid (TFA) at reflux did not  179  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  provide the expected addition product 124, but instead generated the unexpected Nalkylated product 125. OEt 1 )H N'^^OEt, 2  DCC, Et N, MeCN/ H 0 3  //  2  2) 1 0 % N a O H , H 0(2:1) 3) PPA, P O 6% overall yield  n 0  2  122  2  s  120  (EtO) CH, 3  TFA, reflux  52%  O EtO"  ^OEt  125 Scheme 5.7. Acid catalysed addition of triethylorthoformate to 120.  One can rationalise this difference in reactivity between 120  and 77  by  considering the possible tautomeric forms in each case (scheme 5.8). 77 will undergo acid catalysed enolisation as part of the mechanism in the conversion to 80. Equivalent enolisation is also possible for 120,  however in this case there is also a competing  tautomer that results in deprotonation of the pyrrole nitrogen (N-1). This second tautomer provides an alternative site for alkylation that is not available to 77, and helps to explain the difference in reactivity of these two species.  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Scheme 5.8. Mechanisms for the conversion of 77 to 83 and 120 to 125.  180  181  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  An alternative to introducing an orthoformyl group at C-9 in an acid catalysed fashion is to employ strongly basic conditions to abstract a proton at C-9 and to treat the resulting anion with ethyl formate in order to introduce a formyl group at C-9. Initial investigations of this method by treating 120 with lithium diisopropylamine (LDA) in THF at -78°C followed by treatment with ethyl formate gave no reaction after a variety of reaction conditions. It was thought that deprotonation at N-1 may be competing with deprotonation at the desired site, and so the reaction was repeated, treating the anion with triisopropylsilyl chloride (TIPSCI) in order to protect N-1. This reaction proceeded cleanly with complete consumption of starting material to give one less polar spot by TLC. Unexpectedly, N M R analysis of the resultant product (126) showed that C-silation had occurred at C-9, introducing a triisopropylsilyl group a to the ketone at C-10. o  r—f~ jf j|P5 U N  o }  1) L D A , T H F , - 7 8 ° C 2)TIPSCI, 0 ° C  127  o ,  TIPS f  N  1 1  )  )\ U  1) L D A , T H F , - 7 8 ° C 2)TIPSCI,0°C  r—f H  "  120  )  jf  u 126  Scheme 5.9. C-Silation of 120 with TIPSCI under strongly basic conditions.  This result suggests that the use of a more electrophilic formylating agent would allow for the generation of an intermediate that possesses a formyl or formyl equivalent group at C-9, which would possess the 1,3-diketone subunit required for condensation with guanidine to afford 122. In the course of this project there were many instances in which several potential synthetic routes were investigated simultaneously. The unexpected generation of 126 was one of those occasions, with an alternative route to 122  being explored  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  182  concurrently (scheme 5.10). Treatment of 127 with dimethylformamide dimethyl acetal (DMF-DMA)  proved  moderately  successful  intermediate. The resultant product (128)  in generating  a  synthetically  contained a new exocyclic  useful  methylene  attached at C-9, with a dimethylamine attached at the C-11 terminal position. This intermediate is also theoretically amenable to condensation with guanidine to provide the desired 2-aminopyrimidine of the target structure, however the reaction with DMFDMA also resulted in the unwanted methylation at N-1 of the pyrrole subunit, making the final product no longer easily convertible to 81.  120  128  Scheme 5.10. Conversion of 120 to 128 with DMF-DMA.  At this point in the project further method development was hampered by the lack of available material. The conversion of 120 to 128 was performed on only 16 mg of material, and in its unoptimised state afforded a yield of only 19% to provide just 4 mg of 128. In order to test the potential of this compound as an intermediate in the synthesis of the isolatonduine core, 128 was treated with guanidine hydrochloride in the presence of potassium carbonate at reflux in THF/water for one hour. T L C showed complete absence of starting material, and subsequent workup provided 0.9 mg of material with ^  m a x  at 210 and 280 nm that was preferentially soluble in the aqueous phase when  partitioned between water and ethyl acetate. Purification by reversed phase high performance liquid chromatography, followed by extensive analyses utilising both  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  183  nuclear magnetic resonance spectroscopy and mass spectrometry failed to shed conclusive light upon the structure of this new material, but this is of little surprise when one considers the paucity of signals in the H NMR, and that the small quantity of 1  material made it impossible to obtain complete  1 3  C or H M B C N M R spectra for this new  compound. 5.6. Future Directions. With much of the methodology now in place for the construction of the isolatonduine skeleton a number of modifications can be suggested that may prove of use to those involved in the continuation of this project. In light of the numerous difficulties encountered with the formation of compound 77 where positions C-2 and C-3 of the pyrrole ring are brominated, and a ketone has been introduced at C-10, it is recommended that bromination at C-2 and C-3 be performed as the final stage of the synthesis. It is not clear whether steric effects will allow for dibromination at C-2 and C3, but the pharmacophore  model  monobrominated  113  compound  proposed  may also  in section  5.3 suggests  exhibit the desired  that the  pharmacological  response. Difficulties were also encountered in generating an intermediate suitable for the construction of the 2-aminopyrimidine ring via condensation with guanidine. Of the methods pursued, reaction of 120 with DMF-DMA looks to be the most promising, although the pyrrole nitrogen would have to be protected with a suitable protecting group at an earlier stage in the synthesis in order to avoid the unwanted methylation at N-1. Finally, reacting 128 with guanidine hydrochloride appears to offer hope as a method for the creation of the 2-aminopyrimidine functionality, although the reaction would have to be repeated on a larger scale in order to unequivocally identify the product of this last reaction.  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  184  5.7. Experimental. Preparation of bromopyrrole 117.  To a stirred solution of 76 (200 mg, 0.61 mmol) in dry T H F (10ml_) at room temperature was added a catalytic amount of palladium on carbon (10% Pd-on-carbon) and the resulting black suspension placed under a positive pressure of hydrogen (20 p.s.i.). After stirring for 4 hours the reaction mixture was filtered and concentrated to dryness in vacuo. The resulting cream solid was dissolved in A c O H (20ml_) and treated with neat Br (1.2 mL, 23.7 mmol). This orange solution was stirred at room temperature in air for 2  48 hours, then concentrated to dryness in vacuo and partitioned between water (100 mL) and EtOAc (3 x 75 mL). The combined organic phases were concentrated to dryness in vacuo and the crude product purified by flash chromatography (40 x 180 mm, 100% EtOAc). Removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 141 mg (70%) of the bromopyrrole 117 as an orange solid.  1  H NMR (400 MHz, MeOD-d4) 8 2.01 (m, 2H), 2.72 (t, 2H, J = 6.6 Hz), 3.31 (m, 2H).  1 3  C NMR (100 MHz, MeOD-d4) 5 27.7, 29.4, 43.1, 102.1, 107.3, 128.6, 129.9, 164.8.  Exact mass calculated for C H N 0 ' B r B r : 307.8983; found: 307.8986. y  8  8  2  a i  185  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Preparation of amide 121.  To a solution of 4,5-dibromopyrrol-2-yl trichloromethyl ketone (79)  (260 mg, 0.702  mmol) in M e C N (10 mL) was added (3-alanine (50 mg, 0.495 mmol), Et N (200 uL, 2.32 3  mmol) and water (50 ul). After stirring for 18 hours the reaction mixture was concentrated to dryness in vacuo and partitioned between water (10 mL) and C H C I (8 2  mL). The phases were separated and the aqueous phase washed with C H C I 2  2  2  (2x8  mL), then concentrated to dryness (freeze drier) to give 120 mg (71%) of the amide 121 as a white solid.  1  H NMR (400 MHz, MeOD-c/4) 5 2.46 (t, 2H, J = 6.5 Hz), 3.07 (t, 2H, J = 6.5 Hz), 6.62  (s, 1H).  13 C NMR (100 MHz, MeOD-c/4) 6 34.2, 38.1, 99.4, 103.4, 115.7, 132.1, 167.1, 177.9.  Exact mass calculated for C H N 0 B r : 252.8384; found: 252.8390 8 1  5  2  Exact mass calculated for C H N 0 : 89.0477; found: 89.0476 3  7  2  Br,  89 -OH  Br H 251  O  O  186  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Preparation of pyrrol-2-carboxylic acid (122).  A solution of pyrrol-2-yl trichloromethyl ketone (129)  (15.0 g, 70.6 mmol) in 10%  NaOH(gq) (60 mL) and MeOH (30 mL) was stirred for 18 hours, then concentrated to dryness in vacuo. Purification of the crude product by flash chromatography (45 x 220 mm, EtOAc then MeOH) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 5.90 g (75%) of the acid 122 as a white solid.  1  H NMR (400 MHz, D 0 ) 5 6.27 (bs, 1H), 6.76 (bs, 1H), 7.00 (bs, 1H). 2  1 3  C NMR (100 MHz, D 0 ) 5 111.8, 115.4, 124.6, 130.9, 172.1. 2  Exact mass calculated for C H N 0 : 111.0320; found: 111.0322. 5  5  2  187  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Preparation of amide 130.  To a solution of acid 122 (1.00 g, 9.0 mmol) in M e C N (30 mL) and water (10 mL) was added p-alanine ethyl ester (1.66 g, 10.8 mmol), Et N (880 u l , 10.8 mmol) and D C C 3  (2.23 g, 10.8 mmol). Following stirring for 18 hours the reaction mixture was concentrated to dryness in vacuo and partitioned between water (200 mL) and EtOAc (200 mL). The phases were separated and the aqueous phase extracted with EtOAc (2 x 200 mL). The combined organic phases were concentrated to dryness in vacuo. Purification of the crude product by flash chromatography (40 x 150 mm, 1:1 Hexanes/EtOAc) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 365 mg (34%) of the amide 130 as a white solid.  1  H N M R (400 MHz, MeOD-cf4) 5 1.21 (t, 3H, J = 7.1 Hz), 2.60 (t, 2H, J = 6.8 Hz), 3.57 (t,  2H, J = 6.8 Hz), 4.11 (q, 2H, J = 7.1 Hz), 6.13 (dd, 1H, J = 2.5, 3.7 Hz), 6.73 (dd, 1H, J = 1.4, 3.7 Hz), 6.88 (dd, 1H, J = 1.4, 2.5 Hz).  1 3  C NMR (100 MHz, MeOD-c/4) 5 14.4, 35.3, 36.3, 61.6, 110.2, 111.7, 122.9, 126.7,  163.8, 173.5.  Exact mass calculated for C i o H N 0 3 : 210.1004; found: 210.003. 14  2  188  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Preparation of a-B-unsaturated ketone 120. o,  A solution of amide 130 (365 mg, 1.738 mmol) in 10% N a O H  ( a q )  (20 mL) and MeOH (10  mL) was stirred for 18 hours. The reaction mixture was concentrated to dryness in vacuo and stirred as a suspension in P P A (100 mL) at 100°C for 1 hour. The reaction mixture was poured into cold (0°C) water and the aqueous phase partitioned against EtOAc (200 mL). The phases were separated and the aqueous phase was extracted with EtOAc (2 x 200 mL). The combined organic phases were concentrated to dryness in vacuo. Purification of the crude product by flash chromatography (40 x 200 mm, 2% MeOH in EtOAc to 4% M e O H in EtOAc) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 55 mg (19%) of the a-B-unsaturated ketone 120 as a white solid.  1  H NMR (400 MHz, DMSO-d6) 5 2.68 (m, 2H), 3.34 (m, 2H), 6.54 (bs, 1H), 6.97 (bs,  1H), 8.31 (bs, 1H), 12.15 (bs, 1H).  1 3  C NMR (100 MHz, MeOD-d4) 5 38.3, 44.5, 111.4, 123.8, 125.7, 129.9, 165.0, 197.3.  Exact mass calculated for C H N 0 2 : 164.0586; found: 164.0587. 8  8  2  189  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Preparation of hemiketal 125.  NH  Eta  A solution of a-p-unsaturated ketone 120 (25 mg, 0.152 mmol) in (EtO) CH (10 mL) in 3  the presence of T F A (200 u l ) was stirred at reflux for 8 hours. The reaction mixture was concentrated to dryness in vacuo (vacuum pump) and the crude product purified by flash chromatography (20 x 200 mm, EtOAc to 4% M e O H in EtOAc). Removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 11 mg (27%) of the hemiketal 125 as a white solid and 12 mg (48%) of the starting material 120.  1  H NMR (400 MHz, MeOD-c/4) 5 1.22 (t, 6H, J = 14.2 Hz), 2.77 (m, 2H), 3.48 (m, 2H),  3.57-3.70 (m, 4H), 6.65 (d, 1H, J = 3.2 Hz), 7.03 (s, 1H), 7.35 (d, 1H, J = 3.2 Hz).  1 3  C NMR (100 MHz, MeOD-c/4) 8 15.1, 38.2, 45.0, 63.7, 103.4, 109.9, 124.4, 127.4,  128.4, 164.8, 197.8.  Exact mass calculated for C H N 0 4 : 266.1267; found: 266.1266. 1 3  1 8  2  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  190  Preparation of silyl ether 126. o  JIPS  To a solution of cc-B-unsaturated ketone 120 (5 mg, 0.03 mmol) in dry THF (10 mL) was added LDA (1.4M, 47 u.L, 0.066 mmol) dropwise over 15 minutes at room temperature and the reaction mixture stirred for 15 minutes. A solution of TIPSCI (126 u l , 0.60 mmol) in THF (774 u,L) was added dropwise and the reaction mixture stirred at room temperature for 40 minutes. Following stirring the reaction mixture was quenched with water (1 mL), concentrated to dryness in vacuo and partitioned between 1N HCI (10 mL) and EtOAc (10 mL). The phases were separated and the aqueous phase was extracted with EtOAc ( 2 x 1 0 mL). The combined organic phases were concentrated to dryness in vacuo. Purification of the crude product by flash chromatography (18 x 260 mm, 8:2 EtOAc/hexanes then EtOAc) and removal of trace amounts of solvent (vacuum pump) from the resulting solid provided 2.2 mg (23%) of the silyl ether 126.  1  H NMR (400 MHz, DMSO-c/6) 5 1.04 (d, 18H, J = 7.6 Hz), 1.22 (septet, 3H, J = 7.6 Hz),  3.36 (m, 2H), 5.05 (t, 1H, J = 6.8 Hz), 6.25 (t, 1H, J = 2.7 Hz), 6.94 (t, 1H, J = 2.7 Hz), 7.57 (t, 1H, J = 5.1 Hz), 11.71 (bs, 1H).  1 3  C NMR (100 MHz, DMSO-c/6) 5 12.1, 17.8, 36.8, 100.5, 106.4, 121.6, 123.8, 124.9,  151.3, 163.1.  Exact mass calculated for C H 2 6 N 0 2 S i : 320.1920; found: 320.1921. 18  2  191  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  Preparation of ene-amine 128. CH  H  3  N CH  3  NH CH  3  O  To a solution of a-p-unsaturated ketone 120 (16 mg, 0.098 mmol) in dry THF (10 mL) was added DMF-DMA (39 ul, 0.292 mmol) and the reaction stirred at reflux for 18 hours. No reaction was observed, so a further portion of DMF-DMA (400 ul, 2.95 mmol) was added and the reaction mixture stirred at reflux for a further 48 hours. The reaction mixture was concentrated to dryness in vacuo (vacuum pump) and the crude product purified by flash chromatography (10 x 250 mm, EtOAc to 9:1 EtOAc/MeOH) followed by reversed phase HPLC (25% MeOH/75% H 0 ) . Removal of trace amounts of solvent 2  (vacuum pump) from the resulting solid provided 4 mg (19%) of the ene-amine 128 as a yellow solid.  1  H NMR (400 MHz, DMSO-c/6) 5 3.12 (s, 6H), 3.81 (s, 3H), 6.39 (d, 1H, J = 2.5 Hz),  6.97 (d, 1H, J = 2.5 Hz), 7.37 (s, 1H), 8.35 (t, 1H, J = 5.9 Hz).  1 3  C NMR (100 MHz, DMSO-c/6) 5 36.1, 36.9, 43.5, 103.9, 108.2, 125.4, 127.5, 128.6,  148.6, 162.5, 185.2.  Exact mass calculated for C H N 3 0 2 : 234.1243; found: 234.1237. 12  16  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  192  5.8. References. 1. Tsuda, M.; Uemoto, H.; Kobayashi, J . Tet. Lett. 1999, 40, 5709-5712. 2. Mancini, I.; Guella, G.; Amade, P.; Roussakis, C ; Pietra, F. Tet. Lett. 1997, 38, 62716274. 3. Boyd, M. R.; Pettit, G. R.; McNulty, J.; Herald, D. L; Doubek, D. L; Chapuis, J . - C ; Schmidt, J . M.; Tackett, L. P. J. Nat. Prod. 1997, 60, 180-183. 4. Tsukamoto, S.; Tane, K.; Ohta, T.; Matsunaga, S.; Fusetani, N.; van Soest, R. W. M. J. Nat. Prod. 2001, 64, 1576-1578. 5. Pettit, G. R.; Herald, C. L; Leet, J . E.; Gupta, R.; Schaufelberger, D. E.; Bates, R. B.; Clewlow, P. J.; Doubek, D. L; Manfredi, K. P.; al., e. Can. J. Chem. 1990, 68, 16211624. 6. Groszek, G.; Kantoci, D.; Pettit, G. R. Liebigs Ann. 1995, 4, 715-716. 7. Supriyono, A.; Schwarz, B.; Wray, V.; Witte, L; Mueller, W . E. G.; Van Soest, R. W. M.; Sumaryono, W.; Proksch, P. Z. Naturforsch C 1995, 50, 669-674. 8. Patil, A. D.; Freyer, A. J.; Killmer, L; Hofmann, G.; Johnson, R. K. Nat. Prod. Lett. 1997,9,201-207. 9. D'Ambrosio, M.; Guerriero, A.; Ripamonti, M.; Debitus, C.; Waikedre, J.; Pietra, F. Helv. Chim. Acta 1996, 79, 727-735. 10. Commercon, A.; Gueremy, C. Tet. Lett. 1991, 32, 1419-1422. 11. Meijer, L; Thunnissen, A. M. W. H.; White, A. W.; Gamier, M.; Nikolic, M.; Tsai, L. H.; Walter, J.; Cleverley, K. E.; Salinas, P. C ; Wu, Y. Z.; Biernat, J.; Mandelkow, E. M.; Kim, S. H.; Pettit, G . R. Chem. Biol. 2000, 7, 51-63. 12. Badger, A. M.; Cook, M. N.; Swift, B. A.; Newman-Tarr, T. M.; Gowen, M.; Lark, M. J. Pharmacol. Exp. Ther. 1999, 290, 587-593. 13. Breton, J . J.; Chabot-Fletcher, M. C. J. Pharmacol. Exp. Ther. 1997, 282, 459-466. 14. Roshak, A.; Jackson, J . R.; Chabot-Fletcher, M.; Marshall, L. A. J. Pharmacol. Exp. Ther. 1997, 283, 955-961. 15. Kobayashi, J.; Ohizumi, Y.; Nakamura, H.; Hirata, Y.; Wakamatsu, K.; Miyazawa, T. Experientia 1986, 42, 1064-1065. 16. Sharma, G.; Magdoff-Fairchild, B. J. Org. Chem. 1977, 42, 4118-4124. 17. Keifer, P. A.; Schwartz, R. E.; Koker, M. E. S.; Hughes, R. G., Jr.; Rittschof, D.; Rinehart, K. L. J. Org. Chem. 1991, 56, 2965-2975.  193  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  18. Eder, C ; Proksch, P.; Wray, V.; van Soest, R. W. M.; Ferdinandus, E.; Pattisina, L. A.; Sudarsono J. Nat. Prod. 1999, 62, 1295-1297. 19. Peng, J.; Shen, X.; El Sayed, K. A.; Dunbar, D. C ; Perry, T. L ; Wilkins, S. P.; Hamann, M. T.; Bobzin, S.; Huesing, J.; Camp, R.; Prinsen, M.; Krupa, D.; Wideman, M. A. J. Agr. Food Chem. 2003, 51, 2246-2252. 20. Urban, S.; Leone, P. d. A.; Carroll, A. R.; Fechner, G. A.; Smith, J.; Hooper, J . N. A.; Quinn, R. J . J. Org. Chem. 1999, 64, 731-735. 21. Walker, R. P.; Faulkner, D. J.; Van Engen, D.; Clardy, J . J. Am. Chem. Soc. 1981, 103, 6772-6773. 22. Bernan, V. S.; Roll, D. M.; Ireland, C. M.; Greenstein, M.; Maiese, W. M.; Steinberg, D. A. J. Antimicrob. Chemoth. 1993, 32, 539-550. 23. Mizuno, A.; Inomata, N.; Miya, M.; Kamei, T.; Shibata, M.; Tatsuoka, T.; Yoshida, M.; Takiguchi, C ; Miyazaki, T. Chemical & Pharmaceutical  Bulletin 1999, 47, 246-256.  24. Shen, X.; Perry, T. L; Dunbar, C. D.; Kelly-Borges, M.; Hamann, M. T. J. Nat. Prod. 1998,67,1302-1303. 25. Rosa, R.; Silva, W.; Escalona de Motta, G.; Rodriguez, A. D.; Morales, J . J.; Ortiz, M. Experientia 1992, 48, 885-887. 26. Kobayashi, J.; Nakamura, H.; Ohizumi, Y. Experientia 1988, 44, 86-87. 27. Kobayashi, J.; Tsuda, M.; Ohizumi, Y. Experientia 1991, 47, 301-304. 28. Kobayashi, J.; Tsuda, M.; Murayama, T.; Nakamura, H.; Ohizumi, Y.; Ishibashi, M.; Iwamura, M.; Ohta, T.; Nozoe, S. Tetrahedron 1990, 46, 5579-5586. 29. Kato, T.; Shizuri, Y.; Izumida, H.; Yokoyama, A.; Endo, M. Tet. Lett. 1995, 36, 21332166. 30. Cafieri, F.; Carnuccio, R.; Fattorusso, E.; Taglialatela-Scafati, O.; Vallefuoco, T. Bioorg. Med. Chem. Lett. 1997, 7, 2283-2288. 31. Curman, D.; Cinel, B.; Williams, D. E.; Rundle, N.; Block, W. D.; Goodarzi, A. A.; Hutchins, J . R.; Clarke, P. R.; Zhou, B.-B.; Lees-Miller, S. P.; Andersen, R. J.; Roberge, M. J. Biol. Chem. 2001, 276, 17914-17919. 32. Podtcheko, A.; Ohtsuru, A.; Tsuda, S.; Namba, H.; Saenko, V.; Nakashima, M.; Mitsutake, N.; Kanda, S.; Kurebayashi, J.; Yamashita, S. J. Clin. Endocr. Metab. 2003, 88, 1889-1896.  194  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  33. Tasdemir, D.; Mallon, R.; Greenstein, M.; Feldberg, L. R.; Kim, S. C ; Collins, K.; Wojciechowicz, D.; Mangalindan; G. C ; Concepcion, G. P.; Harper, M. K.; Ireland, C. M. J. Med. Chem. 2002, 45, 529-532. 34. Kolch, W. Biochem. J. 2000, 351, 289-305. 35. Eder, C ; Proksch, P.; Wray, V.; Steube, K.; Bringmann, G.; Van Soest, R. W. M.; Sudarsono; Ferdinandus, E.; Pattisina, L. A.; Wiryowidagdo, S.; Moka, W. J. Nat. Prod. 1999, 62, 184-187. 36. Wang, M.; Xia, X.; Kim, Y.; Hwang, D.; Jansen, J . M.; Botta, M.; Liotta, D. C ; Snyder, J . P. Organic Letters 1999, 1, 43-46. 37. Brown, H. C ; Geoghegan, P. J., Jr. Journal of Organic Chemistry 1970, 35, 18441850. 38. Larock, R. C. Solvomercuration/demercuration  reactions in organic synthesis;  Springer-Verlag: Berlin, 1986. 39. Xu, Y.-z.; Yakushijin, K.; Home, D. A. J. Org. Chem. 1997, 62, 456-464. 40. Bedel, S.; Ulrich, G.; Picard, C. Tetrahedron Letters2002,  43, 1697-1700.  41. Pettit, G. R.; Knight, J . C ; Herald, D. L; Davenport, R.; Pettit, R. K.; Tucker, B. E.; Schmidt, J . M. Journal of Natural Products 2002, 65, 1793-1797. 42. Magnus, P.; Sear, N. L ; Kim, C. S.; Vicker, N. Journal of Organic Chemistry 1992, 57, 70-78. 43. Rathore, R.; Saxena, N.; Chandrasekaran, S. Synthetic Communications  1986, 16,  1493-1498. 44. Parish, E. J.; Chitrakorn, S.; Wei, T. Y. Synthetic Communications  1986, 16, 1371-  1375. 45. Rangarajan, R.; Eisenbraun, E. J . Journal of Organic Chemistry 1985, 50, 24352438. 46. Galli, R.; Dobler, M.; Guller, R.; Stahl, R.; Borschberg, H.-J. Helvetica Chimica Acta 2002, 85, 3400-3413. 47. Ishii, Y.; Nakayama, K.; Takeno, M.; Sakaguchi, S.; Iwahama, T.; Nishiyama, Y. Journal of Organic Chemistry 1995, 60, 3934-3935. 48. Hirai, N.; Sawatari, N.; Nakamura, N.; Sakaguchi, S.; Ishii, Y. Journal of Organic Chemistry 2003, 68, 6587-6590.  195  Chapter 5: Progress Towards the Synthesis of Isolatonduine.  49. Mizuno, A.; Miya, M.; Kamei, T.; Shibata, M.; Tatsuoka, T.; Nakanishi, K.; Takiguchi, C ; Hidaka, T.; Yamaki, A.; Inomata, N; Chemical & Pharmaceutical  Bulletin 2000, 48,  1129-1137. 50. Annoura, H.; Tatsuoka, T. Tetrahedron Letters 1995, 36, 413-416.  196  Summary.  Summary. The work presented within this thesis has sought to explore two separate areas of marine natural products chemistry. The first, presented in chapters two and three, utilised a novel screening method to probe for marine natural products that could act as antibacterial agents via disruption of the type III secretion system. The second, presented in chapters four and five, confirmed the proposed structure of a novel marine alkaloid by the completion of a total synthesis, and made progress towards the generation of analogues for further S A R studies. The proposal that the secondary metabolite chemistry of marine invertebrates can provide lead structures for drug development  has been borne out by the  investigation of the chemistry of the marine sponge Caminus sphaeroconia.  Bioassay  guided fractionation led to the discovery of the structurally novel glycolipids caminosides A - D, which showed initial activity in a screen designed to selectively identify type III secretion inhibitors. As marine natural products chemistry develops into a mature field, and a number of the more common structural classes become known, it is likely that chemical prospecting will prove less and less successful at identifying marine secondary metabolites with previously unreported carbon skeletons. It is gratifying that in this case, the approach of screening a library of extracts to look for compounds that exhibit a particular biological function has proved successful in the isolation of a novel family of natural products with several unusual structural features. This investigation has acted as a proof of principle for the concept of tackling bacterial resistance to antibiotic therapy by targeting the type III secretion system, and it is likely that this approach will see further research efforts as the need for alternative methods for the treatment of bacterial infections becomes more acute.  Summary.  197  The confirmation of the assignment of natural product structures by independent synthesis is seen by many as the ultimate proof of the validity of the original structural assignment. In the case of the latonduines, the information available from spectroscopic analyses was insufficient to differentiate between two possible structures (66 and 70). Total synthesis of latonduine A (66) confirmed the structure of the natural product, and provided material for further biological testing. The final synthesis was both concise and high yielding, and afforded material that was spectroscopically indistinguishable from that of the natural product. The final portion of the thesis presents the attempts towards the generation of the synthetic analogue of latonduine A, isolatonduine (70). Consideration of the biological activities of a number of the oroidin alkaloids suggested that in contrast to 66, 70 might prove active as a protein kinase inhibitor against one or more of a number of protein kinases. Development of the synthetic methodology has provided a route that, while not complete, contains much of the information required for the ultimate construction of isolatonduine. The biological activities of the oroidin alkaloid family continues to be a topic of interest in both the biological and chemical literature and, were isolatonduine to show activity as predicted, it could prove to be well suited for further drug development due to its structural simplicity and potential for further structural adaptation at sites remote from the aminopyrimidine ring.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0061160/manifest

Comment

Related Items