Chemical Diversity of Eremophila Species and Screening Library Generation

Author Barnes, Emma Catherine

Published 2012

Thesis Type Thesis (PhD Doctorate)

School School of Biomolecular and Physical Sciences

DOI https://doi.org/10.25904/1912/908

Downloaded from http://hdl.handle.net/10072/366931

Griffith Research Online https://research-repository.griffith.edu.au

Chemical Diversity of Eremophila species and Screening Library Generation

Emma Catherine Barnes B.Sc. (Hons.)

School of Biomolecular and Physical Sciences Science, Environment, Engineering and Technology Griffith University

Submitted in fulfilment of the requirements of the degree of Doctor of Philosophy

October 2012

Abstract

This thesis explores two aspects of natural product (NP) chemistry. In part A, the use of NPs as scaffolds in the generation of screening libraries was explored as a valuable way to produce structurally diverse compounds with lead- or drug-like physicochemical parameters. Part B describes the chemical investigations of several species from the under-studied Australian endemic plant genus Eremophila, which was selected for examination as it had high potential to be a source of both new chemistry and of unique scaffolds for screening library production.

Part A.

The known plant NP 14-hydroxy-6,12-muuroloadien-15-oic acid (62) was identified from the Nature Bank compound repository as a unique scaffold that could be chemically elaborated to generate lead- or drug-like screening libraries.

Scaffold 62 was isolated from the leaves of Eremophila sturtii, then utilised in the parallel solution-phase synthesis of two series of analogues. The first library consisted of six semi-synthetic amide derivatives (82-84 and 91-93), whilst the second contained six carbamate analogues (103-108). Prior to synthesis a virtual library was generated and prioritised based on drug-like physicochemical parameters such as Log P,

Log D5.5, hydrogen bond donors/acceptors, and molecular weight. These semi-synthetic libraries have been evaluated for their antimalarial activity against a chloroquine- sensitive Plasmodium falciparum line (3D7). Several compounds displayed moderate activity in this screen with IC50 values ranging from 14 to 33 μM.

Further chemical investigations of the E. sturtii extract resulted in the isolation of three novel tetracyclic sesquiterpene lactones, mitchellenes A-C (68-70), two new sesquiterpene carboxylic acids, mitchellenes D and E (71 and 72), and the known compounds casticin (73), and centaureidin (74). A proposed biosynthetic pathway from mitchellene D to mitchellenes A-C is described.

During attempts to re-isolate scaffold 62 in order to generate further library members, it was found that the original plant source of NP 62 had been taxonomically misidentified. A number of Eremophila species were examined by analytical HPLC and off-line (+)-LRESIMS in order to assist in the identification of the sample, and after consultation with the Queensland Herbarium the specimen was re-classified from E. mitchellii to E. sturtii.

Part B.

While researching the Eremophila genus from which scaffold 62 had been obtained, it was observed that while this genus has been a source of structurally diverse compounds, a large number of its > 215 species remain chemically uninvestigated. This observation lead to a number of Eremophila species being chosen for chemical examination in this project.

Forty Eremophila specimens labelled as either 'leaf' or 'aerial parts' in the Nature

Bank biota library were examined by analytical HPLC. Three of these species with no reported chemical investigations in the literature were then selected for analysis.

E. eriocalyx and E. linsmithii were both found to predominantly contain the previously reported compounds verbascoside (110), mannitol (179), and geniposidic acid (155). E. eriocalyx additionally contained the known NPs mussaenoside (181), ladroside (182), 5,19-dihydroxy-3,14-viscidadien-20-oic acid (186) and the new cembrane 3,15-epoxy-19-hydroxycembra-7,11-dien-18-oic acid (188). The NPs from E. linsmithii also included the known NP 3,15-epoxycembra-7,11-dien-18-oic acid (191) and the new hemiterpene glycoside 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-

(1→6)-O-β-D-glucopyranoside (193).

Extraction and purification of a leaf sample of E. microtheca collected from south east Queensland led to the isolation of three new serrulatane type NPs, 3-acetoxy-

7,8-dihydroxyserrulat-14-ene (198), 3,7,8-trihydroxyserrulat-14-en-19-oic acid (199), and 3,19-diacetoxy-8-hydroxyserrulat-14-ene (200). The known compounds verbascoside (110) and jaceosidin (201) were also isolated. NP 199 was used in acetylation and methylation reactions to generate analogues for crystallisation and biological studies (205-207). The NPs and semi-synthetic analogues were analysed against a panel of nine Gram-positive and one Gram-negative bacterial strains. The serrulatanes were found to be moderately active (MICs 32-128 μg/mL) against

Streptococcus pyogenes (ATCC 12344). NP 198 demonstrated activity at MIC 128

μg/mL against the majority of bacterial strains. The flavonoid jaceosidin (201) had the greatest potency (MICs 8-32 μg/mL) against most Staphylococcus aureus strains.

In all, three novel, seven new, and ten known NPs were isolated from several

Eremophila species, and twelve screening library analogues were generated from the

Eremophila derived scaffold 62. All of these compounds were fully characterised using a combination of MS, IR, UV, [α]D, and 1D/2D NMR spectroscopic data analyses.

III

Statement of Originality

This work has not previously been submitted for a degree or diploma in any university. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made in the thesis itself.

______Emma Catherine Barnes Date

Table of Contents

Abstract ...... I Statement of Originality ...... IV Acknowledgements ...... VII List of Figures ...... IX List of Schemes ...... XI List of Tables ...... XII Abbreviations ...... XIII Publications Arising from this Thesis ...... XVI

Chapter 1. Introduction Part A ...... 1 1.1. Strategies for the generation of compound libraries in the search for new lead and/or drug molecules ...... 1 1.2. PhD research aims ...... 21

Chapter 2. Selection and isolation of the muurolane scaffold 62 along with the cyclic sesquiterpenes mitchellenes A-E from Eremophila sturtii ...... 22 2.1. NP scaffold selection process ...... 22 2.2. E. sturtii ...... 25 2.3. Isolation of scaffold 62 from E. sturtii ...... 26 2.4. Isolation of mitchellenes A-E from E. sturtii ...... 33 2.5. Proposed biosynthesis of mitchellenes A-E ...... 50 2.6. Conclusion ...... 53

Chapter 3. Screening library generation using scaffold 62 ...... 54 3.1. Introduction ...... 54 3.2. Selection of reaction partners ...... 55 3.3. Amide library generation ...... 55 3.4. Carbamate library generation ...... 64 3.5. Biological activity of the muurolane libraries ...... 67 3.6. Synthesis of a di-substituted analogue of scaffold 62 ...... 70 3.7. Conclusions ...... 72

Chapter 4. Where's my compound? Analytical strategies for the determination of taxonomy ...... 73 4.1. Introduction ...... 73 4.2. Analytical HPLC analysis of E. mitchellii samples ...... 74 4.3. Analytical HPLC analysis of Eremophila species from Nature Bank ...... 77 4.4. Analytical HPLC analysis of E. mitchellii samples collected from Kalbar, Queensland ...... 80 4.5. The difficulties associated with Eremophila taxonomy ...... 83 4.6. Conclusion ...... 89

Chapter 5. Introduction Part B ...... 90 5.1. The Eremophila genus – a source of potential medicines from the Australian desert ...... 90 5.2. Traditional use of Eremophila spp. by Australian Aboriginal people ...... 94 5.3. Chemical diversity of the Eremophila genus ...... 96 5.4. Modern era biological activity of Eremophila spp. extracts and pure NPs ...... 103 5.4.1. Bioactivity studies of Eremophila plant extracts ...... 103 5.4.2. Bioactivity studies of Eremophila pure NPs ...... 107 5.5. Conclusion ...... 113

Chapter 6. Chemical investigations of Eremophila linsmithii and Eremophila eriocalyx ...... 114 6.1. Introduction ...... 114 6.2. E. eriocalyx ...... 115 6.3. E. linsmithii ...... 128 6.4. Conclusion ...... 138

Chapter 7. Serrulatanes from Eremophila microtheca ...... 139 7.1. Introduction ...... 139 7.2. Extraction and purification of NPs from E. microtheca ...... 140 7.3. Semi-synthetic analogues of serrulatane 199 ...... 156 7.4. Bioactivity of isolated NPs and semi-synthetic analogues ...... 158 7.5. Conclusion ...... 161

Chapter 8. Experimental ...... 162 8.1. General experimental ...... 162 8.2. Chapter 2. Experimental ...... 163 8.3. Chapter 3. Experimental ...... 171 8.4. Chapter 4. Experimental ...... 196 8.5. Chapter 6. Experimental ...... 202 8.6. Chapter 7. Experimental ...... 211

References…………………………………………………………………………….219

Appendix I. Eremophila analytical HPLC chromatograms

Appendix II. CD NMR data list for thesis compounds

Acknowledgements

This project would not have been possible without the wonderful guidance of my supervisors, Dr Rohan Davis and Professor Ron Quinn.

To Dr Rohan Davis – thank you for always having the time to help me, your constant words of encouragement and advice and never ending patience are much appreciated.

And to Professor Ron Quinn – thank you for the wonderful opportunities and support you provide for students, you made possible many of the positive experiences I have been fortunate to have had while being a member of the Eskitis Institute.

I would like to thank the various collectors and taxonomists who have provided the plant specimens in this project, including Paul Forster and Gordon Guymer from the

Queensland Herbarium, and Barry Jahnke and Jan Glazebrook from the Society for

Growing Australian Plants (SGAP). I would also like to thank Brian Walters, Colin

Jennings, and Keith Townsend and the helpful team at the Australian National Botanic

Gardens for providing the beautiful photos of the plants.

I thank Dr Hoan Vu of the Eskitis Institute and Graham MacFarlane of the

University of Queensland for measuring the high-resolution mass spectral data. I would also like to thank Dr Brett Schwartz for undertaking the dimethyldioxirane reaction outlined in Chapter 3.

I gratefully acknowledge our collaborators for the biological screening undertaken in this project, including Vanida Choomuenwai and Dr Katherine Andrews for the anti-malarial screening and Michelle Liberio for the LnCap assay, all of whom are members of the Eskitis Institute, Griffith University. I'd also like to thank Angela

Kavanagh, Soumya Ramu, Dr Mark Blaskovich, and Professor Matthew Cooper of the

VII

Institute for Molecular Bioscience, University of Queensland, for the antibacterial testing.

To everyone past and present at the Eskitis Institute, thank you for being the best support team a student could ever have. Special thanks to Dr Tanja Grkovic and Dr

Karren Beattie, and those people who have come and gone but have had a huge impact on my learning and general well being as my project progressed – Dr Catherine

Roullier, Dr Frederic Martin, Dr Ozlem Demirkiran, Dr Daniela Muller, Lekha

Suraweera, Sujlesh Sharma, and Meredith Palframan just to name a few.

Thank you to all my fellow students who I have shared the ups and downs of our various projects with over the years.

Thank you to my parents and brother who have always encouraged me to do what I love in life and who have been there for me no matter what. To my extended family and friends, who may not really know what I do but who have supported me through it all, thanks.

Lastly, but most importantly, to my husband Brian, thank you for your constant love, patience, and support, I couldn’t have done this without you.

VIII

List of Figures

Figure 1. Representative NP scaffolds from the studies by Lee et. al. (A) and Hert et. al. (B) that were missing from commercial collections ...... 9 Figure 2. Images of E. sturtii. Photos © M. Fagg, Australian National Botanic Gardens…………………………………………………………………………………25 1 13 Figure 3. H and C NMR spectra for 62 in DMSO-d6 ...... 31 Figure 4. Key COSY, HMBC and ROESY correlations for 62 ...... 31 1 13 Figure 5. H and C NMR spectra for 68 in DMSO-d6 ...... 36 Figure 6. Expansion of COSY spectrum for 68 in DMSO-d6 ...... 38 Figure 7. HMBC spectrum for 68 in DMSO-d6 ...... 39 Figure 8. Key COSY, HMBC and ROESY correlations for 68 ...... 39 1 13 Figure 9. H and C NMR spectra for 69 in DMSO-d6 ...... 40 1 Figure 10. Comparison of H NMR spectra for 68 and 70 in DMSO-d6 ...... 42 1 13 Figure 11. H and C NMR spectra for 71 in DMSO-d6 ...... 44 Figure 12. Key COSY, HMBC and ROESY correlations for 71 ...... 46 Figure 13. Reaction apparatus for generation of DMDO ...... 53 1 Figure 14. H NMR spectrum for 82 in DMSO-d6 ...... 59 13 Figure 15. C NMR spectrum for 82 in DMSO-d6 ...... 60 1 13 Figure 16. H and C NMR spectra for 109 in DMSO-d6 ...... 71 Figure 17. Analytical HPLC chromatograms of 62, the original plant sample QID005836, and seven E. mitchellii samples archived within the Nature Bank library 75 Figure 18. Analytical HPLC chromatograms of 62, the original plant sample QID005836, and representative Eremophila samples from analysis of the 40 specimens ...... 79 Figure 19. Analytical HPLC chromatograms of 62, the original plant sample QID005836, the E. mitchellii leaf sample QID004133, and the three E. mitchellii samples collected from Kalbar, Queensland ...... 81 Figure 20. E. sturtii (left), photo © M. Fagg, Australian National Botanic Gardens, and E. mitchellii (right), photo courtesy of K. Townsend ...... 83 Figure 21. Analytical HPLC chromatograms of 62, the original plant sample QID005836, and the three E. sturtii samples ...... 86 Figure 22. Analytical HPLC chromatograms of 62, the original sample QID005836, the E. sturtii sample QID009030, verbascoside, casticin, and centaureidin ...... 87 Figure 23. E. mitchellii (left), E. gilesii (top right), and E. bowmanii (bottom right), photos © M. Fagg, Australian National Botanic Gardens ...... 91 Figure 24. E. maculata, photos courtesy of B. Walters ...... 92 Figure 25. E. nivea (right) photo courtesy of B. Walters. E. racemosa (top left) and E. warnesii (bottom left) photos © M. Fagg and L. Vallee, Australian National Botanic Gardens...... 93 Figure 26. E. longifolia (left), photo courtesy of B. Walters, and E. alternifolia (right), photo © M. Fagg, Australian National Botanic Gardens...... 96 Figure 27. Analytical HPLC UV chromatograms of E. linsmithii and E. eriocalyx extracts...... 114 Figure 28. E. eriocalyx, photo courtesy of C. Jennings ...... 115 Figure 29. E. linsmithii, photo © M. Fagg, Australian National Botanic Gardens. .... 128 1 13 Figure 30. H and C NMR spectra for 193 in DMSO-d6 ...... 133 Figure 31. Key COSY, HMBC and ROESY correlations for 193 ...... 134 Figure 32. E. microtheca, photos © M. Fagg, Australian National Botanic Gardens. .139 Figure 33. Analytical HPLC UV chromatogram for E. microtheca extract ...... 140 IX

1 13 Figure 34. H and C NMR spectra for 198 in DMSO-d6 ...... 143 Figure 35. Key COSY, HMBC and ROESY correlations for 198 ...... 143 Figure 36. COSY spectrum for 198 in DMSO-d6 ...... 144 Figure 37. HMBC spectrum for 198 in DMSO-d6 ...... 145 1 13 Figure 38. H and C NMR spectra for 199 in DMSO-d6 ...... 149 1 Figure 39. H NMR spectrum for 200 in DMSO-d6 ...... 152 13 Figure 40. C NMR spectrum for 200 in DMSO-d6 ...... 153 1 13 Figure 41. H and C NMR spectra for 206 in DMSO-d6 ...... 157

List of Schemes

Scheme 1. Generation of screening libraries using fredericamycin A (12) ...... 11 Scheme 2. Analogues of the fredericamycin A derived aldehydes 13 and 14 ...... 12 Scheme 3. Generation of semi-synthetic library from the NP scaffold FR901379 (37) .14 Scheme 4. Dipeptide library generation using D-cycloserine ...... 18 Scheme 5. Comparison of isolation procedures used to obtain scaffold 62 from E. sturtii ...... 27 Scheme 6. Conversion of 65 into 67 to achieve purification by Ghisalberti et. al...... 28 Scheme 7. Extraction and isolation procedure for E. sturtii ...... 35 Scheme 8. Conversion of arteannuin H (75) to 76 ...... 48 Scheme 9. Proposed biogenetic pathway for mitchellenes A-C (68-70) ...... 51 Scheme 10. Generation of DMDO and its subsequent use to oxidise mitchellene B (69)……………………………………………………………………………………...52 Scheme 11. Amine couplings with scaffold 62 using EDCI or DMTMM ...... 56 Scheme 12. Amine couplings with 85 using EDCI or DMTMM ...... 57 Scheme 13. Oxalyl chloride coupling between 89 and phenethylamine ...... 58 Scheme 14. Amine couplings with scaffold 62 using oxalyl chloride ...... 59 Scheme 15. TBDPS protection of scaffold 62, coupling with 4-methoxyphenethylamine using (COCl)2 to give 97, then de-protection to provide 92 ...... 62 Scheme 16. Coupling of cinnamyl alcohol (98) with isocyanates ...... 64 Scheme 17. Coupling of scaffold 102 with isocyanates to generate compounds 103- 108……………………………………………………………………………………...66 Scheme 18. Formation of the di-substituted analogue 109 ...... 70 Scheme 19. Extraction and isolation procedure for E. eriocalyx ...... 116 Scheme 20. Synthetic methodology used to establish the absolute configuration of 181...... 122 Scheme 21. Conversion of NP 189 to 188 by Ghisalberti et. al...... 126 Scheme 22. Extraction and isolation procedure for E. linsmithii ...... 129 Scheme 23. Extraction and isolation procedure for E. microtheca ...... 141 Scheme 24. Acetylation (i) and methylation (ii) of NP 199. (i) Ac2O:py (1:1), rt, 16 h. (ii) TMS-CH2N2, CH3OH:CH2Cl2 (1:1), rt, 20 min ...... 156

List of Tables

Table 1. NMR data for 14-hydroxy-6,12-muuroloadien-15-oic acid (62) ...... 32 Table 2. NMR data for mitchellene A (68) ...... 37 Table 3. NMR data for mitchellene B (69) ...... 41 Table 4. NMR data for mitchellene C (70) ...... 43 Table 5. NMR data for mitchellene D (71) ...... 45 Table 6. NMR data for mitchellene E (72) ...... 47 Table 7. NMR data for casticin (73) ...... 49 Table 8. NMR data for centaureidin (74)...... 50 Table 9. Results from DMTMM and EDCI amine couplings with 85 ...... 57 Table 10. Amine coupling yields with 62 using DMTMM, EDCI, and (COCl)2 ...... 64 Table 11. Isocyanate coupling yields with cinnamyl alcohol ...... 65 Table 12. Physicochemical profiling and biological activity of NPs and semi-synthetic analogues ...... 69 Table 13. E. mitchellii samples in the Nature Bank biota library ...... 74 Table 14. Eremophila 'leaf' and 'aerial parts' samples in Nature Bank biota library ..... 78 Table 15. E. sturtii samples archived in the Nature Bank library ...... 84 Table 16. Eremophila species utilised by Australian Aboriginal people ...... 95 Table 17. NMR data for geniposidic acid (155) ...... 118 Table 18. NMR data for verbascoside (110) ...... 120 Table 19. NMR data for mussaenoside (181) ...... 122 Table 20. NMR data for ladroside (182) ...... 123 Table 21. NMR data for 5,19-dihydroxy-3,14-viscidadien-20-oic acid (186) ...... 125 Table 22. NMR data for 3,15-epoxy-19-hydroxycembra-7,11-dien-18-oic acid (188)127 Table 23. NMR data for 3,15-epoxycembra-7,11-dien-18-oic acid (191) ...... 131 Table 24. NMR data for 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-D- glucopyranoside (193) ...... 136 Table 25. NMR data for 3-acetoxy-7,8-dihydroxyserrulat-14-en-19-oic acid (198) ... 147 Table 26. NMR data for 3,7,8-trihydroxyserrulat-14-en-19-oic acid (199) ...... 151 Table 27. NMR data for 3,19-diacetoxy-8-hydroxyserrulat-14-ene (200) ...... 153 Table 28. NMR data for jaceosidin (201) ...... 155 Table 29. Antibacterial activity of 198-201, 205 and 206……………………………161

XII

Abbreviations

(COCl) 2 oxalyl chloride [α]D specific rotation 13C NMR carbon NMR spectroscopy 1D one dimensional 1H NMR proton NMR spectroscopy 2D two dimensional 3D three dimensional Ac acetyl Ac2O acetic anhydride AcOH acetic acid aq aqueous Ar argon br broad C18 octadecyl bonded silica CC50 half maximal cytotoxic concentration CD circular dichroism CD3OD deuterated methanol CDCl3 deuterated chloroform CH2Cl2 dichloromethane CH2N2 diazomethane CH3CN acetonitrile CH3OH methanol CHCl3 chloroform conc concentrated d doublet DapRSA daptomycin resistant Staphylococcus aureus DIC diisopropylcarbodiimide diol 1,2-dihydroxypropyl bonded silica DMAP 4-dimethylaminopyridine DMDO dimethyldioxirane DMF N,N-dimethylformamide DMPK drug metabolism and pharmacokinetics DMSO dimethylsulfoxide DMSO-d6 deuterated DMSO DMTMM 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride DNP Dictionary of Natural Products, Chapman and Hall EC50 half maximal effective concentration EDCI N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride eq equivalent Et2O diethyl ether EtOAc ethyl acetate EtOH ethanol g gram gCOSY gradient correlation spectroscopy gHMBC gradient heteronuclear multiple-bond correlations

XIII gHSQC gradient heteronuclear single-quantum coherence GI50 half maximal growth inhibition GISA glycopeptide-intermediate Staphylococcus aureus GPS global positioning system h hour H2O water HBA hydrogen bond acceptor HBD hydrogen bond donor HCl hydrochloric acid HPLC high-pressure liquid chromatography HRESIMS high resolution electron spray ionisation mass spectrometry or spectrum HTS high throughput screening Hz Hertz IC50 half maximal inhibitory concentration i-PrOH isopropanol IR infrared J coupling constant KOH potassium hydroxide LC/MS liquid chromatography/mass spectrometry LDA lithium diisopropylamide Log D5.5 Logarithm of the distribution coefficient at pH 5.5 of ionised and un-ionised species between n-octanol and water Log P Logarithm of the partition coefficient of un-ionised species between n-octanol and water LRESIMS low resolution electron spray ionisation mass spectrometry or spectrum m multiplet m/z mass: charge ratio MBC minimum bactericidal concentration MDR multiple drug resistant Me methyl MIC minimum inhibitory concentration mL millilitre mmol millimole mol mole MRSA methicillin-resistant Staphylococcus aureus MS mass spectrometry Mw molecular weight (g/mol) N2 nitrogen NaOH sodium hydroxide NCE new chemical entity NMP N-methylpyrrolidine-2-one NMR nuclear magnetic resonance NP natural product PDA photo diode array ppm parts per million Py pyridine q quartet ROESY rotational nuclear Overhauser effect spectroscopy

XIV rt room temperature s singlet SAR structure activity relationship sp. species SPE solid-phase extraction t triplet TBAF tetrabutylammonium fluoride TBDPS tert-butyl(chloro)diphenylsilane TES triethylsilane TFA trifluroacetic acid THF tetrahydrofuran TIC total ion count TLC thin layer chromatography UV ultraviolet Van A vancomycin resistant VISA vancomycin intermediate resistant Staphylococcus aureus wt weight δ chemical shift μg microgram μL microlitre

Publications Arising from this Thesis

1. Design and synthesis of screening libraries based on the muurolane natural product scaffold

Emma C. Barnes, Vanida Choomuenwai, Katherine T. Andrews, Ronald J. Quinn and Rohan A. Davis. Org. Biomol. Chem., 2012, 10, 4015–4023.

Publication selected for front cover of Organic & Biomolecular Chemistry May 2012 issue.

2. Mitchellenes A-E, Cyclic Sesquiterpenes from the Australian Plant Eremophila mitchellii

Emma C. Barnes, Anthony R. Carroll, and Rohan A. Davis. J. Nat. Prod., 2011, 74, 1888–1893.

Publication selected for front cover of Journal of Natural Products issues from July to December 2012.

XVI

Chapter 1. Introduction Part A

1.1. Strategies for the generation of compound libraries in the search for new lead and/or drug molecules

The rapid increase in assay technologies and the development of high throughput screening (HTS) has given researchers the capability to screen vast numbers of compounds quickly and efficiently.1-4 Additionally, the advent of genome sequencing and an increasing understanding of molecular targets and processes has uncovered a plethora of potential therapeutic sites.2,5-7 With these advances there has been a perceived reciprocal requirement for the rapid production of compounds to be tested in these systems.4,6 There is also a growing need for small molecule probes that can be used to study the function of target sites and to validate their medicinal potential.5,8

In the 1990s combinatorial chemistry emerged with a promise to supply the thousands of compounds required to fill the HTS assays in a rapid manner, thereby leading to an increase in the number of hits and leads being produced.2-4,7,9-12 While this technique was relatively untested it swiftly overtook more classical methods such as natural product (NP) screening and rational drug design.13

However, during the early phases of combinatorial library development problems emerged that restricted the number of screening hits being generated. It was found that the combinatorial libraries being produced lacked design elements that would aid in target binding. Often both the synthetic methodology and reaction partners utilised were selected for their ease of use, compatibility, and reproducibility to ensure the generation of analogues and to maximise the size of the library. This was at the cost of chemical diversity, often resulting in topologically flat libraries with limited three dimensional diversity and library members that didn’t resemble small molecule drugs.6,10,11,13-16

Combinatorial chemistry started as a solid-phase synthetic method for peptide libraries, so early on there was a lack of translational methods for the use of complex organic scaffolds and reaction sequences in solution phase.4,10,11 The processes of synthesising, purifying, and characterising the members of large libraries were very time consuming, with complex deconvolution steps being required.10,13 Powerful diagnostic tools such as NMR spectroscopy were also difficult to use on the massive mixtures acquired from combinatorial synthesis.17

While potency and chemical diversity are often cited as requirements for a successful library, also of importance is that the library members being produced are biased towards drug- and lead-like properties18,19 to improve their drug metabolism and pharmacokinetics profile (DMPK) and toxicity.6,10,14,15,20 A balance between the potency and DMPK profiles of libraries at an early stage can streamline the drug discovery process and help in selecting hits and leads for progression.10,21 The emphasis on making combinatorial libraries through robust synthetic methods unfortunately did not provide compounds with drug- and/or lead-like properties.14 For example, functional groups that provide polar interactions for specificity in binding such as weakly acidic OH and NH bonds typically also interact during synthetic reactions, so were effectively removed by capping with other functionalities.14,15

To date just one de novo new chemical entity (NCE) arising from combinatorial chemistry has been reported and approved for drug use, the anti-tumour multikinase inhibitor compound sorafenib (Nexavar, 1). This compound is used to treat renal cell carcinoma and hepatocellular carcinoma.22

Although combinatorial chemistry appears to have had only a minimal impact on drug discovery to date in regards to marketable drugs, this field has contributed significantly to pharmaceutical research. Since its inception methods learnt through combinatorial chemistry have been adapted and modified to increase library quality in terms of biological relevancy and drug- and lead-like properties.2,10-13,16,18,19,23 The old adage 'quality over quantity' now resounds in the minds of contemporary library constructors,7,12,13,24,25 with chemists continuously refining compound library generation methods in order to build smarter libraries.2,10,26-29 The move has been towards the use of drug- and/or lead-like filters and DMPK profiling to produce smaller, higher quality libraries.10,13,18-20,29,30 Advances in the use of parallel synthesis and purification techniques have also benefited the production of compound libraries.10,13,31

Current synthetic library strategies include the selection of novel scaffolds, particularly those that adhere to lead- and drug-like physicochemical profiles for library generation.18,26,32-35 One main question still remains, however - how do researchers narrow down the seemingly infinite number of chemical entities to those that would make good scaffolds?24,29,30,36 Reducing the vast number of molecules found in chemical space to those that are biologically relevant, and then to those that have potential to be pursued for drug development is a significant challenge.29,36

There is a need to identify compound classes that are biologically validated, inherently possess drug- and/or lead-like properties,18,19 are synthetically accessible, and are amendable to modification for the generation of screening libraries for structure activity relationship (SAR) studies.24,30

Nature has provided humanity with medicines for thousands of years, with the earliest records of the use of natural extracts and oils dating from around 2600 BC from

Mesopotamia.37 The first pure drugs can be traced to the early 1800s when NPs such as morphine (2), atropine (3), and colchicine (4) came into use.37 NPs or NP derivatives have been used as treatments for a range of indications including cardiovascular, infectious (bacteria, viral, fungal), oncology, inflammatory, and immunological conditions.2,37-41 Many of the most well known drugs are in fact NPs or NP derivatives, including lovastatin (5), galantamine (6), quinine (7), erythromycin (8), and aspirin

(9).2,37

NPs continue to play an important role in providing new drugs.9,22,24 For example, in the area of cancer treatments from the 1940s to 2010, 48.6% are either NPs or are directly derived from them.22 In 2010, of the 20 small-molecule NCEs approved, half were NPs or NP derivatives.22

NPs are often cited as being structurally diverse, drug-like, and containing privileged motifs.6,12,26,28,32,34,42,43 The success of these molecules and their semi- synthetic derivatives as therapeutic agents is intrinsically linked to the fact that NPs have been biologically validated, since they have been selected during evolution to bind to both their biosynthetic enzymes and their biological targets.6,13,26,30-32,34,43-46 It has been hypothesised that this inherent capacity to bind in biological space allows NPs to also recognise human therapeutic targets.6,13,32,44,45 The biological relevancy of NPs also means that they often possess favourable pharmacokinetic properties, as they must journey through living systems in order to reach their target sites.15,24

NPs are often cited as being "privileged structures," a term which was originally defined by Evans et. al. to describe how benzodiazepine derivatives such as 10 could bind to a number of receptor types.47 The authors state that the modification of a privileged structure, which is capable of binding to multiple receptors, may lead to useful molecular probes and therapeutic agents.47 NPs are often classified under this definition, as they are synthesised by, modified by, and must bind to various proteins.6,12,24,30,31

Also of importance in drug discovery is the need for molecules that can be used as chemical probes for investigating macromolecular interactions and functions, information which can guide the discovery of new therapeutic agents.8,31,43,48-50 NPs have the ability to modulate macromolecular interactions, such as those between proteins, DNA, RNA, or carbohydrates, and so can be used to aid our understanding of both normal and dysfunctional dynamic biological processes.8,31,33,43,49-52 Rapamycin

(11), for example, played an integral part in the discovery of the serine/threonine protein kinase mammalian target of rapamycin (mTOR).48,53-55 Rapamycin has been utilised to study the role of mTOR in protein synthesis and cell growth, as well as in linking diseases to the dysregulation of mTOR.48,53-55

Despite these facts, NPs have fallen out of favour with pharmaceutical companies.2 There are valid reasons for why this has happened, including the costly and time consuming nature of NP programs, re-supply issues and seasonal variations in NP production, intellectual property concerns, and the redundancy of re-discovering known bioactive NPs.6,16,23,41,56 Some of the biggest concerns with NP drug discovery relate to difficulties in developing feasible, high yielding total synthesis routes for structurally complex NPs to provide the multi-gram quantities needed for SAR studies and clinical trials.2,11,41,43,49,57,58 Problems can also arise from low purity NP libraries and the amenability of NPs and complex extract mixtures to HTS.9,13,22,56,59 6

It is understandable why there was a move away from the use of NPs in light of the promises given by combinatorial chemistry.2,13 However, many of the issues listed above have at least in part been addressed. For instance, new strategies for extract selection, preparation, fractionation and HTS screening have allowed a number of groups to build up extensive pre-fractionated extract and NP libraries.16,40,59-61 The use of fully characterised NP libraries is advantageous to drug discovery since “hits” resulting from HTS can be quickly evaluated in an identical manner to those from synthetic libraries. Because the chemical structure of the hit compound is already known (unlike extract or pre-fractionated library screening), the potential for lead evaluation can be progressed more rapidly.

Improvements in purification and structure elucidation methods have shortened

NP program schedules and made them more compatible with HTS campaign timeframes.2,6,16,41,43,60-64 The development of dereplication methods using different combinations of HPLC, NMR spectroscopy, and MS analysis coupled with NP reference libraries such as DNP65 has also sped up the process of identification and helped with the issue of redundancy.6,9,16,60,62,64

These improvements have allowed NP analysis to move into the micro-scale.

The structure elucidation of complex NPs with < 1 mg of material is now possible.6,60

Researchers are rapidly moving away from the purification and identification of NPs as being rate limiting steps.6,9

The positive attributes of NPs give them great potential as scaffolds in the generation of screening libraries, and improvements made in the process of acquiring these compounds has made them more accessible for such projects.

The emerging paradigm is to combine the synthetic power of compound library generation methods with biologically validated and privileged NP scaffolds to expand the structural diversity and drug-likeness of screening libraries to increase the chance of obtaining a bioactive hit.6,20,23,56,58,66,67

Indeed, numerous scaffolds that have been identified in NPs have led to approved drugs or drug candidates for a range of diseases.27,28,41,68-70 Examples include antibacterials (-lactams, tetracyclines, erythromycins), antivirals (modified nucleosides), anticholesterolemics (lovastatin), and anti-tumour agents (paclitaxel, rapamycins, epothilones).68

Computational studies have shown that NP scaffolds occupy larger, complementary areas of chemical space compared with synthetic compounds, and can be used to increase the chemical complexity and drug-likeness of libraries.13,26,32,34,35,71,72

Lee et. al. found that the Bioscreen NP database (10,495 NPs) contained 1,748 different ring systems, compared to the Derwent World Drug Index (5,757 trade drugs) which comprised 807 different ring systems.71 The authors concluded that while there was a clear structural overlap between the two data sets, there were many NP central ring systems that were not present in marketed drugs (Figure 1) that had potential as scaffolds in combinatorial chemistry.71 In a more recent study, Hert et. al. found that

83% of core ring scaffolds found in the NPs they examined were absent from commercially available molecules (Figure 1).35

Figure 1. Representative NP scaffolds from the studies by Lee et. al. (A) and Hert et.

al. (B) that were missing from commercial collections

Feher and Schmid15 found that NPs typically possess a higher number of chiral centres, hydrogen bond acceptors (HBAs), hydrogen bond donors (HBDs), and oxygen atoms, and are more rigid. Combinatorial compounds were generally more hydrophobic and had a higher number of nitrogen atoms and rotatable bonds. The drug molecules analysed possessed properties between those of the combinatorial and NP sets. The increased rigidity and number of polar functional groups, particularly HBAs, in NPs gives them a favourable binding free energy, aiding in successful target binding.15 The authors also found that much of the plotted chemical space occupied by NPs and drugs contained no combinatorial compounds. They suggest that mimicking the distribution properties of NPs may result in libraries with a higher chance of containing biological hits.15

There has been a drop in the number of NCEs entering the market2,22,24 and there is a risk of there not being effective drugs to treat drug resistant and newly emerging pathogens and cancers.56 In a recent review on NPs as sources of new drugs from 1981-

2010, Newman and Cragg report that the number of all NCEs (including vaccines and biologics) hit a low in 2004 with just 25 being approved.22 This number rebounded to

54 in 2005, but subsequent years have seen it plateau at around 40.22 When just focusing on small molecule drugs, from 1989-2000 the number of those approved was generally between 40-70, but this dropped to less than 40 from 2001-2010, with just 20 being approved in 2010.22

The majority of the world's biodiversity remains unstudied, so Nature remains an untapped source of unique and desirable scaffolds for library production and subsequent drug discovery.26-28,32-35,41 As NPs have been highly successful as sources of new drugs, by incorporating these unique molecules as scaffolds into library production there is a chance to turn around the drop in NCEs entering the market.

Being biologically validated molecules, NPs often arise as hits in bioactivity screens. There have been a number of SAR studies undertaken using NPs as lead scaffolds to further improve the NPs potency and/or selectivity.

For example Abel et. al. used the NP fredericamycin A (12), which was isolated from the fermentation broth of Streptomyces griseus (FCRC-48), as a scaffold in the generation of small libraries for SAR studies aiming to improve the potency and selectivity of the NP.73-75 Fredericamycin A (12) possesses in vitro antibacterial, antifungal, and cytotoxic activities76 and demonstrated in vivo activity against P388 leukemic cells and reduced the median weight of CD8F mammary tumours in mice.76

NP 12 was shown to inhibit topoisomerases I and II and DNA polymerase in vitro.77

Numerous de novo synthesis studies have been undertaken on NP 12.78-80 Rather than go down the route of total synthesis, Abel et. al. utilised an optimised isolation and purification process81 to obtain multi-gram quantities of 12 from the fermentation broth of Streptomyces griseus ATCC49344 mutants.80 The group initially degraded the F-ring diene sidechain to an aldehyde (13) in order to lower the Mw and provide a chemical handle for synthetic transformations (Scheme 1). They then explored the halogenation of ring E of both compound 13 and NP 12 to give 14 and 15 and 16-19, respectively.

Two small libraries were then generated by replacing the A-ring methoxy group of NP

12 and the bromo-derivative 18 with primary and secondary amines or alcohols (20-

28).80 The aldehyde 13 and its bromo-derivative 14 were also used to generate hydrazone, oxime, and thiazole compounds (29-36, Scheme 2).

Scheme 1. Generation of screening libraries using fredericamycin A (12) 11

Scheme 2. Analogues of the fredericamycin A derived aldehydes 13 and 14

The semi-synthetic analogues were tested against a panel of 10 tumour cell lines.

The authors found that halogenation of ring E of NP 12 increased the potency of the compounds to the low nM range (16-19, IC70 8-20 nM compared to 517 nM for NP 12) and increased the selectivity. Replacement of the F-ring pentadiene resulted in lower potency. Replacement of the A-ring methoxy resulted in similar or higher potency in comparison to 12.80

While potency is an essential factor for drug development, also of significant importance is the DMPK and toxicity profile of any potential drug.6,10,14,18-20 If the design of a NP analogue library is given due consideration before the synthesis is undertaken, then adherence to important physicochemical parameters known to be associated with lead- or drug-like molecules such as Log P, Log D5.5, HBDs/HBAs, and

18,82 MW can be addressed, which ultimately has a positive impact on the progression of any potential hit.

The development of the lipopeptide FR901379 (37) into the marketed drug mycamine® is an example of the use of a NP scaffold that required modification of its

DMPK, toxicity, and potency in order to get it onto the market.

FR901379 (37) was isolated from the culture broth of Coleophoma empetri F-

11899.83 This NP is highly water-soluble and demonstrates strong antifungal activity

83,84 against Candida species. Compound 37 displayed IC50 values ranging from 0.008 to

0.025 μg/mL against four clinical isolates of Candida albicans and was found to inhibit

84 1,3--glucan synthase (IC50 0.7 μg/mL). This synthase only exists in fungal cell walls, not in mammalian cells, so is a useful target for antifungal drugs.84 While having positive activity and solubility properties, NP 37 was found to haemolyse mouse red blood cells in vitro at 62 μg/mL and was not effective against another clinically important fungus, Aspergillus fumigates.84 Thus there was a need to chemically modify

37 to improve the antifungal potency and to reduce the haemolytic activity of the compound.

Enzymatic deacylation of FR901379 followed by re-acylation gave 38, which displayed comparable antifungal activity to FR901379, but lacked the haemolysis activity associated with the NP (Scheme 3).85 Subsequently a library of acylated analogues of 37 were generated.86 Initially acyl side chains that provided improved lipophilicity were selected. The researcher's attention then turned to the enhancement of the antifungal activity through the introduction of a series of rigid side chain analogues.

The compound with the best balance of antifungal and hemolytic activity was found to be 39c.86 Further optimisation led to the discovery of 45, which exhibited activity against Candida and Aspergillus species.87 The in vivo activity of 45 was found to be good when administered intravenously.87 Compound 45, known as micafungin or

FK463, is now marketed as the intravenous antifungal drug mycamine®.

Scheme 3. Generation of semi-synthetic library from the NP scaffold FR901379 (37)

Clearly, if NP analogue libraries are to become more common place in drug discovery, then the supply of suitable NP scaffolds is crucial. Improvements in both the solution88 and solid-phase synthesis of complex NPs and the generation of libraries12,30,89 have made even highly complex NP structures achievable targets. This has opened up a range of strategies for the use of NPs as scaffolds.42 For instance, NPs arising from de novo syntheses can be used directly as scaffolds, or core components of lead NPs can be synthesised and utilised as scaffolds, allowing for quicker and easier synthesis and SAR studies.

NP-like libraries which emulate the structural and stereochemical diversity of

NPs are also becoming more common place. Schreiber's 'diversity-orientated synthesis,' for instance, allows for the construction of intricate compounds through the use of carefully designed, multi-component reaction sequences.9,31,51,52,90

The chemical complexity and functional group density of NPs can in some cases be a negative factor for library construction, by requiring extensive protection chemistry and limiting the range of feasible reactions.63 There is a need for new synthetic methodologies which can be applied to architecturally intricate NPs to increase their use as scaffolds in library generation. Some inroads have been made into adapting synthetic methodologies for the use of NP scaffolds as described in the below examples.

Apoptolidin A (46) was isolated from the fermentation broth of the actinomycete

Nocardiopsis sp.. This complex macrolide was found to be a highly selective inducer of apoptosis in E1A-transformed cells, and it has been proposed that it operates through the inhibition of mitochondrial F0F1-ATPase. The promising bioactivity of apoptolidin

A in selectively inducing apoptosis in cancer cells led to a number of synthetic studies to form derivatives to explore the compounds SAR.

As apoptolidin A can be isolated in substantial quantities by fermentation (109 mg/L), Wender et. al. utilised the isolated NP to form a series of analogues through a progressive protection, functionalisation, and deprotection sequence aiming to design synthetic analogues that would share the NPs high selectivity, but would be more stable and be amenable to therapeutic use.91 The authors undertook the site-selective protection and modification of the eight hydroxyl functionalities of 46, to produce the acetylated compounds 47-51.91

The products generated from 46 were tested for cell proliferation and mitochondrial F0F1-ATPase inhibition. The authors found that the derivatives possessed

ATPase inhibition activity similar to 46. This suggests that these sites could be modified to investigate the mechanism and improve the DMPK profile of 46 for therapeutic development without compromising potency.91

While this method proved effective in producing derivatives of 46 for biological evaluation, the modification of 46 is limited by the reactivity of the compound.92 The

Wender group went on to explore the use of catalysts for the site-selective modification of 46.92 The authors also decided to develop this methodology for small scale use (~25 mg) to allow for the fact that complex NPs such as 46 are usually in scarce supply.

Peptide-based catalysts were used to provide three differentially acylated analogues (52-54) of apoptolidin A from only microgram quantities of the starting material. These new NP analogues were screened in growth inhibition assays against the

H292 human lung carcinoma cell line. Compounds 52-54 showed nanomolar potencies

(EC50 48-60 nM) that were slightly higher than that of the NP scaffold 46 (EC50 21 nM).

This study demonstrated a useful method for the selective modification of complex NPs that doesn’t require protection chemistry and can be achieved on a minute scale.92

The adaptation of solid-phase synthetic methods for the use of NP scaffolds has also met with success. In 1998 Gordeev et. al. developed methodology for the solid

93 phase synthesis of D- and L-cycloserine derivatives. D-Cycloserine (55) is available commercially as either a synthetic or as an isolate from microbial sources. NP 55 can be obtained from the fermentation broths of Streptomyces orchidaceus, Streptomyces garyphalus, and Streptomyces lavendulus. L-Cycloserine (56) is part of the core structure of the NP lactivicin (57) which is found in Empedobacter lactamgenus and

Lysobacter albus culture filtrates.

® Fmoc protected D- and L-cycloserine were immobilised on Sasrin or 2- chlorotrityl linker resins using Mitsunobu-based chemistry or direct tritylation, respectively. The Fmoc-cycloserine resins were then de-protected with piperidine in

DMF or CH2Cl2 to expose the primary amine to various synthetic additions.

Next the researchers conducted a series of reactions using electrophilic reagents on the immobilised cycloserines to afford carboxylic, activated formate, amino acid anhydride, activated heterocyclic chloride and sulfonyl chloride derivatives. The yields for these analogues ranged from 50-80%. The researchers then went on to use this methodology to undertake the parallel synthesis of 80 dipeptidic D-cycloserine derivatives in a reaction plate (Scheme 4).93 Greater than 80% of the compounds displayed the expected molecular ion after ESI MS analysis, and the average purity was found to be 76% by HPLC.

Scheme 4. Dipeptide library generation using D-cycloserine

While there is no one rule for scaffold selection, the adoption of validated strategies from both NP and synthetic chemistry to generate smarter compound libraries has the potential to provide new therapeutic agents and molecular probes for investigating biological functions.94

Nature has provided researchers with biologically relevant lead structures that inherently possess many of the desirable characteristics of valuable scaffolds.24 Indeed, a number of research groups have utilised NPs as scaffolds for the generation of screening libraries for SAR and/or DMPK studies, usually through the total synthesis of the molecule itself or of core components of bioactive NPs.

The isolation of NPs and their subsequent use as scaffolds is relatively rare in comparison to other strategies, mostly due to the low amounts typically obtained.67

However, as NP collections grow and purification procedures improve, the potential to use isolated NPs as scaffolds to produce screening libraries is becoming a possibility.

The example libraries produced from NP scaffolds described above, including from fredericamycin A (12), FR901379 (37), apoptolidin A (46), and D-cycloserine (55) are all examples where the NP has been isolated from its natural source and used in the production of analogues.

A major focus of our research is the design and synthesis of drug discovery libraries based on unique NP scaffolds67,95 that would complement, and potentially expedite, current NP drug discovery methods that typically involve the HTS of pre- fractionated or extract libraries.96

Since we have access to the Eskitis Institute’s Nature Bank,97 which includes over 45,000 plant and marine biota samples and 2,738 pure NP compounds, our approach has been to use this valuable resource to acquire relevant NP scaffolds, thus bypassing the de novo synthetic strategy for scaffold production.

Nature Bank is an integrated drug discovery resource based on NPs. It contains dried and ground biota samples, extracts, semi-purified fractions and pure compounds that have been archived to commercial standards.97 This resource is available for drug discovery partnerships with academic and industry groups. The biota library contains

45,000 plant and marine samples collected from Australia (tropical and temperate),

China, and Papua New Guinea. All samples have been collected in accordance with the

UN Convention on Biological Diversity.98 Each biota sample is collected by an expert taxonomist and is accompanied by GPS coordinates and other geographical information to make any required recollections straight forward. Since 1992 hundreds of screening campaigns have been undertaken at the Eskitis Institute using the Nature Bank extract and fraction libraries, resulting in the isolation of 2,738 pure NPs. This compound repository represents a plethora of NPs, with more molecules being constantly added. A large range of structure classes are represented in this collection, for example terpenes, alkaloids, polyketides, glycosides, and flavonoids.

1.2. PhD research aims

The focus of this research was to select a viable NP scaffold from the Nature

Bank collection that possessed physicochemical properties, such as Mw and Log P, which would assist the compliance of subsequent semi-synthetic library analogues to drug- and lead-like properties.

The majority of the example NP scaffolds discussed above possess Mw values greater than 500 Da, which exceeds the Mw threshold stated by Lipinski's "Rule of Five" for drug-like molecules (fredericamycin A [12, Mw 539], FR901379 [37, Mw 1197], and apoptolidin A [46, Mw 1127]). D-Cycloserine (55) is the only example given here of a small Mw NP scaffold (55, Mw 102). While NPs inherently possess properties which improve their DMPK profiles, it was decided that in this project the restriction of the physicochemical parameters of the library analogues would be explored as a way to improve the drug- and lead-likeness of the semi-synthetic compounds even further.16

The main aim of this process was not to produce combinatorial libraries of large numbers of molecules, but to explore the use of isolated NPs as scaffolds to form small sized libraries that could be screened against a range of targets and be used for SAR studies. This would also allow for the exploration of the concept of NPs as privileged scaffolds that have the potential to provide lead compounds against a range of targets.

It was also decided that during the re-isolation of any selected scaffolds from their biota sources, the samples would also be analysed for new chemistry and other

NPs that could potential be used as scaffolds.

Chapter 2. Selection and isolation of the muurolane scaffold 62 along with the cyclic sesquiterpenes mitchellenes A-E from Eremophila sturtii

2.1. NP scaffold selection process

The first step in this project was to narrow down the number of natural products

(NPs) available in the Nature Bank collection (2,738) to those that could be utilised as scaffolds in the generation of screening libraries. A few simple parameters were used to initially identify and prioritise potential scaffolds, which included:

 MW (< 400)

 Log P (< 5)

 Quantity available (> 50 mg)

 Inclusion of chemical handles for modification (e.g. -COOH, -OH, -CHO, -NH2,

Ar-Br)

Restricting the MW and Log P values of each scaffold aids in biasing the library members being generated towards preferential physicochemical parameters.16,18,19

Around 150 NPs were initially identified using this approach, examples of which included 58-61. The list of potential NP scaffolds was reduced further by giving priority to those compounds possessing stereogenic centres, as stereochemistry confers a unique

3D shape on a molecule and has the potential to improve target binding.15,32,49 Next the availability of biota for the re-extraction and re-isolation of each compound was taken into account.

After this process one NP stood out in particular, the previously reported

99,100 sesquiterpene 14-hydroxy-6,12-muuroloadien-15-oic acid (62). The low MW (250), favourable Log P (2.42), multiple stereogenic centres (4) and potential chemical handles

(i.e. the carboxylic acid101 and allylic hydroxyl group102-104) for synthetic elaboration made 62 a particularly attractive NP scaffold.

NP 62 was originally isolated from the Australian plant Eremophila virgata in

1989,99 then from Eremophila interstans in 1990.105 These two species were later re- classified as sub-species of E. interstans (E. interstans subsp. virgata and E. interstans subsp. interstans).106 Both samples were collected 60-80 km south of Coolgardie in

Western Australia and were of the leaves and branches of the plants.99,105

The compound repository within the Nature Bank library was found to contain ~

500 mg of 14-hydroxy-6,12-muuroloadien-15-oic acid (62, 80% purity) that was available for use in the generation of library members. Some of the original plant sample (27.5 g) from which scaffold 62 had been isolated was still available in the

Nature Bank biota collection. This remaining material was therefore taken for extraction to supplement the amount of 62 already available, resulting in the isolation of another

201.7 mg of 62 (0.73 % dry wt, ≥ 90% purity).

There came a time during this research that the stock amount of 62 (~ 700 mg) was becoming low, yet there was still interest in producing further semi-synthetic analogues from this scaffold for biological screening. The biota specimen from which

62 had been obtained had been taxonomically identified as Eremophila mitchellii, however, while trying to procure more of NP 62 from other samples of this species it came to our attention that the taxonomy of the original plant sample had been incorrectly identified by the Queensland Herbarium.107 The strategies developed to investigate the issue of the incorrect taxonomy, which led to the specimen being re- classified as Eremophila sturtii, are discussed in Chapter 4. This chapter describes the extraction and purification of scaffold 62 from E. sturtii, as well as the isolation of a number of novel and new NPs from this species.

2.2. E. sturtii

E. sturtii, named after the explorer Sir Charles Sturt, grows as a shrub up to 3 m tall with resinous leaves and white or light pink flowers (Figure 2).106 This species can be found widespread across south-western Queensland, western New South Wales, eastern South Australia, and in the south-central part of the Northern Territory.108 E. sturtii was traditionally used by Australian indigenous people to treat coughs, respiratory infections, cuts and sores, eye complaints, and as a general healing agent.106,108-110 The foliage of E. sturtii is also a good fly repellent, and it was used by white settlers to line meat-safes until the early twentieth century.106 Extracts of E. sturtii plants have been assessed for their antiviral and antibacterial activity.108-110

Figure 2. Images of E. sturtii. Photos © M. Fagg, Australian National Botanic

Gardens111

On finding that a leaf extract of E. sturtii possessed antibacterial activity against

Staphylococcus aureus and inhibited the inflammation pathway enzymes cyclooxygenase 1 (COX-1) and COX-2, Liu et. al. undertook bioassay-guided fractionation and isolated two serrulatane diterpenes, 3,8-dihydroxyserrulatic acid (63) and serrulatic acid (64).108 Compound 64 possessed bactericidal activity against S. aureus (MBC 15 μg/mL) and strongly inhibited both COX-1 and COX-2 at 1 mg/mL with 99% and 97% inhibition, respectively.108

2.3. Isolation of scaffold 62 from E. sturtii

The air-dried and ground leaves of E. sturtii (27.5 g) were exhaustively extracted with sequential washes of n-hexane, CH2Cl2, and CH3OH (Scheme 5). The n-hexane extract was not investigated further as it only contained highly lipophilic material. The

CH2Cl2 and CH3OH extracts were combined and evaporated to give a dark brown gum

(5.76 g).

E. sturtii (27.5 g)

n-hexane

CH2Cl2/CH3OH

CH2Cl2/CH3OH (5.76 g)

Method A Method B 0.79 g 0.82 g

diol-bonded silica C18-bonded silica flash column, flash column, n-hexane/EtOAc CH3OH/H2O

Fraction 5 Fraction 4 160.3 mg 154.2 mg

diol-bonded silica HPLC, C18-bonded silica HPLC, i-PrOH/n-hexane CH3OH/H2O

Fractions 32-33 Fraction 46-47 (62, 53.0 mg, ≥ 90% pure) (62, 27.0 mg, ≥ 95% pure)

Scheme 5. Comparison of isolation procedures used to obtain scaffold 62 from E. sturtii

Two portions of this extract were used to investigate different purification methods to see which would afford 62 in the highest yield and purity (Scheme 5). One portion of crude (0.79 g) was first purified using a diol-bonded silica flash column and a n-hexane/EtOAc gradient. 1H NMR spectroscopic and (+)-LRESIMS analysis of the flash column fractions indicated that the 20% n-hexane/80% EtOAc fraction contained

62. Further purification of this fraction by diol-bonded silica HPLC (i-PrOH/n-hexane) resulted in 62 eluting in fractions 32-33 at around 90% purity by 1H NMR spectroscopic analysis.

A similar process was undertaken on the second portion of crude (0.82 g) using a C18-bonded silica flash column and a H2O/CH3OH gradient. The 20% H2O/80%

CH3OH fraction was further purified by C18-bonded silica HPLC (H2O/CH3OH).

Fractions 46-48 were found to contain 62 in 95% purity following 1H NMR spectroscopic analysis.

The diol-bonded silica method resulted in a higher yield of 62, but the C18- bonded silica method gave a purer product. However, the purity of scaffold 62 achieved by either method was not ideal. 1H NMR spectroscopic and (+)-LRESIMS data indicated that 62 eluted off the HPLC columns at the same time as the related known sesquiterpene acid 65, which only differs from 62 in that the exocyclic double bond has been reduced.

Scheme 6. Conversion of 65 into 67 to achieve purification by Ghisalberti et. al.99

Compound 65 was originally isolated from E. virgata (later re-classified as E. interstans subsp. virgata)106 in the same study as that of 62.99 The authors discuss the difficulties they had in separating these closely related analogues. They employed synthetic methods to achieve separation of a mixture of NPs 62, 65 and additional sesquiterpene compounds. NP 65 was methylated to give 66 followed by the formation of the ether derivative 67 (Scheme 6).99 After separation, reopening of the ring structure

67 using LDA afforded the dihydro hydroxy ester 66 (12 mg), which could be characterised. In this study NP 65 was only ever obtained pure in miniscule amounts (<

0.5 mg), so its structure and relative configuration were never conclusively established.

During this investigation some time was spent varying the solvents, gradients, and collection times used during HPLC in order to get a cleaner separation. Some pure fractions were obtained by using diol-bonded silica then C18-bonded silica HPLC, however, most fractions obtained were still enriched mixtures of 62.

While a high purity of 62 could be achieved over two to three HPLC steps, there was a low recovery of the NP off both HPLC columns. The loss of the valuable scaffold was deemed too high, thus it was decided that purity levels of ≥ 90% for 62 would be acceptable for the synthetic reactions, even though this would affect the yield calculations. It was envisaged that purification would be easier after the reaction products had been formed.

The diol-bonded silica purification had proven to be the highest yielding, therefore, this method was utilised to fractionate the remaining crude extract. The extract was divided into ~ 500 mg portions, which were loaded onto several diol-bonded silica flash columns (n-hexane/EtOAc gradient). Fractions 5 and 6 from these columns were dried down, weighed, divided (~ 500 mg), and pre-adsorbed to diol-bonded silica, before being purified by diol-bonded silica HPLC (i-PrOH/n-hexane). Overall 201.7 mg of 62 (0.73% dry wt, ≥ 90% purity) was isolated. This fraction was used in addition to the ~ 500 mg of 62 already available in the Nature Bank compound library in the generation of two screening libraries as discussed in Chapter 3.

The spectroscopic data for compound 62 had only been partially reported in the literature,99 so MS and 1D/2D NMR spectroscopic studies were undertaken to confirm its structure. Scaffold 62 was assigned the molecular formula C15H22O3 on the basis of

(+)-HRESIMS and NMR data (Table 1). The 1H NMR spectrum of 62 (Figure 3, Table

1) displayed distinctive signals for the olefinic proton at H-6 (δH 6.77), the exocyclic double bond hydrogens (δH 5.05, 4.90, H-13), the methylene moiety adjacent to the

13 allylic hydroxyl (δH 3.80, 3.86, H-14), and the methyl group (δH 0.91, H-11). The C

NMR spectrum of 62 (Figure 3, Table 1) showed characteristic signals for the -CH3 (δC

19.3, C-11), -CH2OH- (δC 63.2, C-14), endocyclic -C=C- (δC 142.0 and 129.7, C-6 and

C-7, respectively), exocyclic -C=CH2 (δC 152.5 and 107.4, C-12 and C-13,

112 respectively), and –COOH (δC 168.4) moieties.

Analyses of the 1H-1H COSY, HSQC and HMBC spectra allowed for the planar bicyclic structure of the sesquiterpene 62 to be constructed (Figure 4). To briefly elaborate on this data, the methyl protons at H-11 (δH 0.91) possessed HMBC correlations to C-1 (δC 34.2), C-2 (δC 29.0), and C-10 (δC 38.6). The positioning of the methyl group was also supported by a COSY correlation between H-1 (δH 1.72) and H-

11 (δH 0.91). HMBC signals were identified from the carbonyl carbon at C-15 (δC

168.4) to H-6 (δH 6.77) and H-8 (δH 2.35). The positioning of the allylic alcohol side chain at C-4 was confirmed by HMBC correlations from H-4 (δH 1.82) to C-12 (δC

152.5), C-13 (δC 107.4) and C-14 (δC 63.2).

1 13 Figure 3. H and C NMR spectra for 62 in DMSO-d6

Figure 4. Key COSY, HMBC and ROESY correlations for 62

Table 1. NMR data for 14-hydroxy-6,12-muuroloadien-15-oic acid (62)a b c Position δC δH COSY HMBC ROESY 1 34.2, CH 1.72 (m) 2α, 2β, 10, 11 2, 3, 5, 11 2α, 3α, 5, 11 2α 29.0, CH2 1.37 (m) 1, 2β, 3β 3, 11 1, 2β, 11 2β 1.11 (dddd, 13.2, 1, 2α, 3α, 3β 3, 4, 10, 11 2α, 3β, 4w, 11 12.6, 12.6, 3.0) 3α 33.8, CH2 1.32 (dddd, 12.6, 2β, 3β, 4 1, 2, 5, 12 1, 3β, 5, 13, 14 12.6, 11.4, 3.0) 2β, 3α, 4 1, 2, 12 2β, 3α, 4, 14 3β 1.62 (m) 4 42.4, CH 1.82 (ddd, 11.4, 3αw, 3β, 5, 12 2, 3, 5, 6, 12, 13, 2βw, 3β, 9β, 11.4, 3.0) 14 13, 14 5 41.1, CH 2.28 (ddd, 11.4, 4, 6, 10 3, 4, 6, 7, 9, 12 1, 3α, 6, 10, 5.4, 4.8) 13, 14 6 142.0, CH 6.77 (brd, 4.8) 5, 8αw, 8βw 7w, 8, 10, 15 5, 13, 14 7 129.7, C w w 8α 25.1, CH2 1.98 (m) 6 , 8β, 9α, 9β 6, 7, 9, 10 8β, 9α, 10 8β 2.35 (brdd, 18.0, 6w, 8α, 9αw, 9β 6, 7, 9, 10, 15 9αw, 8α, 9β 5.4) w 9α 16.1, CH2 1.57 (brdd, 13.2, 8α, 8β , 9β, 10 1, 5, 7, 8 8α, 9β, 11 6.6) 8α, 8β, 9α, 10 8 4, 8β, 9α 9β 1.35 (m) 10 38.6, CH 1.63 (m) 1, 5, 9α, 9β 4, 6, 8, 11 5, 11 11 19.3, CH3 0.91 (d, 7.2) 1 1, 2, 10 1, 2α, 2β, 10 12 152.5, C w 13 107.4, CH2 5.05 (s), 4.90 (s) 4 , 14 4, 12, 14 3α, 4, 5, 6, 14 14 63.2, CH2 3.80 (d, 15.0), 13 4, 12, 13 3α, 3β, 4, 5, 6, 3.86 (d, 15.0) 13 15 168.4, C a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

The relative configuration of 62 was assigned after analyses of the ROESY and

1H-1H coupling constant data (Figure 4, Table 1). ROESY correlations between H-5/H-

10 and H-5/H-1 established the cis configuration of these protons. The 1H-1H coupling constant between H-5 and H-10 (J5α,10α = 5.4 Hz) supported the cis orientation of these bridgehead protons.113,114

H-4 showed ROESY correlations to H-3β and H-9β, which placed the allylic hydroxyl side chain at C-4 on the same face as the bridgehead protons. The 1H-1H coupling constant data between H-4 and H-5 (J4β,5α = 11.4 Hz), H-3α and H-4 (J3α,4β =

11.4 Hz) and H-3α and H-2β (J3α,2β = 12.6 Hz) verified the trans relationship of these protons.113,114 The combination of the bridgehead protons (H-5 and H-10) being cis to each other and both trans to H-4 places scaffold 62 in the muurolane-type sesquiterpene class.115 Hence the structure and relative configuration of 62 were verified, with the data obtained being consistent with the information reported in the literature.99

2.4. Isolation of mitchellenes A-E from E. sturtii

During the large-scale isolation of scaffold 62 from E. sturtii, several other NPs were also obtained. These included three novel tetracyclic sesquiterpene lactones, namely, mitchellenes A-C (68-70), two new sesquiterpene carboxylic acids, mitchellenes D (71) and E (72), and the known flavonoids casticin (73)116,117 and centaureidin (74).118,119 This section describes the isolation, structure elucidation, and proposed biosynthesis of these compounds.

 This work was published before the taxonomy of the plant was changed from E. mitchellii to E. sturtii, hence the use of the trivial names mitchellenes A-E.

The diol-bonded silica flash column fractions obtained from the crude extract of

E. sturtii were purified by either diol-bonded silica HPLC (i-PrOH/n-hexane) or C18- bonded silica HPLC (CH3OH/H2O) (Scheme 7). (+)-LRESIMS was carried out on all the HPLC fractions obtained, then those samples containing strong, distinct ions were analysed by 1H NMR spectroscopy. Promising fractions were then either further purified by HPLC, or investigated by 1D/2D NMR (1H NMR, 13C NMR, COSY,

HSQC, HMBC, and ROESY). This resulted in the isolation of NPs 68-74, as described below.

E. sturtii (27.5 g)

CH2Cl2/CH3OH (4.94 g)

Fraction 3 Fraction 4 Fractions 5-6

b b

Fractions 25-26 Fraction 45 (56.5 mg) (5.4 mg) c c c

mitchellene B (69) mitchellene C (70) (28.3 mg, 0.103 % dry wt) (2.2 mg, 0.008% dry wt)

mitchellene A (68) mitchellene E (72) centaureidin (74) (10.5 mg, 0.038% dry wt) (8.4 mg, 0.031% dry wt) (9.7 mg, 0.035% dry wt)

mitchellene D (71) casticin (73) (9.6 mg, 0.035% dry wt) (9.0 mg, 0.033% dry wt)

a diol-bonded silica flash column, n-hexane/EtOAc

b C18-bonded silica HPLC, CH3OH/H2O c diol-bonded silica HPLC, i-PrOH/n-hexane

Scheme 7. Extraction and isolation procedure for E. sturtii

Mitchellene A (68) was isolated as an optically active brown gum. The 1H NMR spectrum of 68 (Figure 5, Table 2) showed six aliphatic signals between δH 0.76 and

4.86, one olefinic resonance at δH 6.57, three methylene moieties (δH 2.37/1.84,

13 1.39/1.37 and 1.40/1.21), and two methyl resonances (δH 0.82 and 1.08). The C NMR spectrum of 68 (Figure 5, Table 2) contained signals that indicated the presence of an oxygenated carbon (δC 70.9), an olefin (δC 136.8 and 127.8), and a carbonyl moiety (δC

170.7) within the molecule.

1 13 Figure 5. H and C NMR spectra for 68 in DMSO-d6

Table 2. NMR data for mitchellene A (68)a b c Position δC δH COSY HMBC ROESY 1 70.9, C 2α 34.0, CH2 1.39 (m) 3, 4 11 2β 1.35 (ddd, 13.8, 3α 1, 3, 4 9βw 12.0, 4.2) 3α 23.9, CH2 1.21(dddd, 12.0, 2β, 3β 1, 2, 12 3β, 5, 12 11.4, 11.4, 4.2) 3α 1, 2 3α, 4 3β 1.40 (m) 4 41.4, CH 0.76 (dddd, 11.6, 3α, 3βw, 5, 12 2, 3, 5, 12, 13 3β 11.4, 11.4, 3.0) 5 39.0, CH 2.21 (ddd, 11.6, 4, 6, 10 1w, 4, 6, 7, 9 3α, 6, 10, 12, 4.8, 4.2) 14w 6 41.7, CH 3.07 (m) 5, 8w, 9αw, 14 4, 7, 8, 14 5, 10w, 14 7 127.8, C 8 136.8, CH 6.57 (ddd, 4.8, 6w, 9α, 9β 6, 9, 10, 15 9α, 9β 3.0, 2.0) w 9α 25.4, CH2 2.37 (dddd, 21.0, 6 , 8, 9β, 10 7, 8, 10 8, 9β, 10 8.4, 3.6, 3.0) 6w, 8, 10 1, 7, 8, 10 2βw, 8, 9α, 11 9β 1.83 (dddd, 21.0, 9.0, 5.4, 4.8) 10 38.9, CH 1.91 (ddd, 9.0, 5, 9α, 9β 1, 2, 5, 9 9α, 5, 6w, 11 8.4, 4.2) 11 27.9, CH3 1.08 (s) 1, 2, 10 2α, 9β 12 43.2, CH 1.63 (ddq, 11.4, 4, 13, 14 3, 6, 13, 14 3α, 5, 13, 14 7.2, 7.2) w 13 11.6, CH3 0.82 (d, 7.2) 12, 14 4, 12, 14 12 14 83.2, CH 4.86 (dd, 7.2, 6, 12, 13w 6, 7, 15 5w, 6, 12 7.2) 15 170.7, C a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

Investigation of the 1H-1H COSY (Figure 6), HSQC, and HMBC (Figure 7) spectra allowed the planar structure of 68 to be constructed (Figure 8). HMBC correlations from H-3, H-5, and H-11 to C-1 allowed this carbon to be assigned as δC

70.9. The downfield chemical shift of C-1 indicated that there was a hydroxyl group located at this position.112

The methyl signal at δH 1.08 (H-11) showed HMBC correlations to C-2 (δC

34.0) and C-10 (δC 38.9), establishing that it was also attached to C-1. A second methyl group resonating at δH 0.82 possessed HMBC correlations to C-4 (δC 41.4) and C-14 (δC

83.2), allowing it to be positioned at C-12 (δC 43.2). The chemical shift of C-14 (δC

112 83.2) indicated that it was next to an oxygen atom, and both H-8 (δH 6.57) and H-14

(δH 4.86) showed HMBC correlations to a carbon at δC 170.7 (C-15). This allowed a γ- lactone ring to be constructed, giving the planar tetracyclic ring system shown in Figure

8. The conjugated γ-lactone moiety was supported by an absorption at 1743 cm-1 in the

IR spectrum of 68.112

Figure 6. Expansion of COSY spectrum for 68 in DMSO-d6

Figure 7. HMBC spectrum for 68 in DMSO-d6

Figure 8. Key COSY, HMBC and ROESY correlations for 68

The relative configuration of 68 was assigned after analyses of the ROESY and

1H-1H coupling constant data (Figure 8, Table 3). ROESY correlations between H-

10/H-5, H-5/H-6, H-6/H-14, and H-14/H-12 established that these protons all had cis orientations around the tetracyclic ring system. The 1H-1H coupling constants between

H-5 and H-10 (J5α,10α = 4.2 Hz) and between H-6 and H-14 (J6α,14α = 7.2 Hz) further supported the cis orientation of these bridgehead protons.113,114

In a similar manner the 1H-1H coupling constant data between H-4 and H-5

(J4β,5α = 11.6 Hz) and H-3α and H-4 (J3α,4β = 11.4) established the trans relationship of these protons.113,114 With the relative configuration determined, structure 68 was assigned the trivial name mitchellene A.

The molecular formula C15H20O2 was assigned to mitchellene B (69) on the basis of HRESIMS and NMR data (Table 3). The NMR data of 69 was similar to that of

68, with the only major differences being that 69 contained one extra proton signal at δH

1.72, and was missing the oxygenated quaternary carbon at δC 70.9, which had been replaced by an up-field signal at δC 34.3 (Figure 9).

1 13 Figure 9. H and C NMR spectra for 69 in DMSO-d6

Table 3. NMR data for mitchellene B (69)a b c Position δC δH COSY HMBC ROESY 1 34.5, CH 1.72 (m) 2αw, 2β, 10, 11 3, 5, 9, 10, 11 w w w 2α 28.8, CH2 1.43 (dddd, 13.2, 1 , 2β, 3α, 3β 1, 3, 4, 10, 11 1 , 2α , 2β, 4.2, 3.6, 3.0) 1, 2α, 3α, 3β 1, 3, 4, 11 3α 2β 1.07 (dddd, 13.2, 2α, 3β, 9βw 13.2, 12.6, 3.6) 3α 28.0, CH2 0.91 (dddd, 13.2, 2α, 2β, 3β, 4 1, 2, 12 2α, 3β, 5 12.0, 11.4, 3.6) 3α, 4, 2α, 2β 1, 5, 12 2β, 3α, 4 3β 1.65 (m) 4 41.4, CH 0.79 (dddd, 12.6, 3α, 3β, 5, 12 10, 12, 13 3β 12.3, 12.0, 3.0) 5 45.2, CH 1.68 (m) 4, 6, 10 3, 6, 7, 9, 10 3α, 6, 10 6 42.0, CH 3.14 (m) 5, 8, 9α, 9β, 14 4, 5, 7, 8, 9, 14 5, 10w, 14 7 127.9, C 8 136.7, CH 6.60 (ddd, 4.2, 6, 9α, 9β 6, 9, 10, 15 9α, 9β 3.0, 2.4) w 9α 22.6, CH2 2.24 (dddd, 21.0, 6, 8, 9β, 10 5, 7, 8, 10, 15 9β, 8, 11 8.4, 4.2, 2.4) 6, 8, 9α, 10 1, 5, 7, 8, 10 9β 1.94 (dddd, 21.0, 2βw, 8, 9α, 8.4, 4.2, 4.2) 11 10 32.2, CH 2.11 (m) 1, 5, 9α, 9β 1, 2, 5, 9, 11 1, 5, 6w, 11 w 11 18.5, CH3 0.88 (d, 7.2) 1 1, 2, 10 1, 9β, 10 12 43.3, CH 1.63 (m) 4, 13, 14 3, 4, 13, 14 13, 14 13 11.5, CH3 0.83 (d, 7.2) 12 4, 12, 14 12 14 82.9, CH 4.85 (dd, 7.2, 6, 12 5, 6, 7, 15 6, 12 7.2) 15 170.6, C a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

The (+)-LRESIMS spectrum of 69 showed an ion at m/z 233 [M + H]+, which identified a MW difference between 69 and 68 of 16 Da. HMBC correlations from H-3,

H-9, and H-10 to the carbon signal at δC 34.3, in addition to a HSQC correlation from this carbon to δH 1.72, positioned these signals at C-1 and H-1, respectively. These data indicated that the 1-OH in 68 had been replaced with a H atom. The ROESY and 1H-1H coupling constant data for 69 was essentially identical to that of 68, hence structure 69 was assigned the trivial name mitchellene B.

The minor NP 70 was isolated as a stable opaque gum. Comparison of the MS and NMR spectroscopic data of 70 and 68 indicated that these two molecules were isomers. The 1H NMR spectrum of 70 lacked the methylene signals at H-9 seen in 68, however, additional signals for two olefinic protons at δH 5.94 and 5.72, a hydroxyl resonance at δH 6.02, and an aliphatic proton at δH 1.63 were observed (Figure 10, Table

4). C-1 in 70 resonated at δC 34.4 compared to δC 70.9 in 68.

1 Figure 10. Comparison of H NMR spectra for 68 and 70 in DMSO-d6

A correlation in the HSQC spectrum allowed the proton at δH 1.63 to be placed at C-1. HMBC correlations were seen between H-6/H-8/H-9/H-14 and the oxygenated carbon at C-7 (δC 69.6). HMBC signals from the hydroxyl proton (δH 6.02) to C-6, C-7,

C-8, and C-15 allowed this group to be placed at C-7.

HMBC correlations from H-8 (δH 5.72) to C-6, C-10, and C-15 and from C-9 (δH

5.94) to C-1, C-5, and C-7 indicated that the endocyclic double bond seen in 68 had migrated to C-8/C-9 in 70. In a similar manner to 68 and 69, the relative configuration of 70 was assigned based on ROESY and 1H-1H coupling constant data (Table 4) and found to be the same as that of mitchellenes A and B. With the relative configuration determined, structure 70 was assigned the trivial name mitchellene C.

Table 4. NMR data for mitchellene C (70)a b c Position δC δH COSY HMBC ROESY 1 34.4, CH 1.63 (m) 2α, 2β, 10, 2, 9, 10, 11 10, 11 11 2α 30.7, CH2 1.47 (m) 1, 2β, 3α, 3β 3, 4, 10, 11 2β 2β 0.80 (m) 1, 2α, 3β 1, 3, 11 2α 3α 27.7, CH2 0.89 (m) 2α, 3β 1, 2 3β 3β 1.68 (m) 2α, 2β, 3α, 4 1, 2, 4, 5 3α, 4 4 39.0, CH 0.81 (m) 3β, 5 13, 14 3α, 13 5 47.0, CH 1.71 (m) 4, 6, 10 3, 6, 7, 9, 10, 12 6, 10, 12 6 47.0, CH 2.83 (dd, 7.8, 7.8) 5, 14 4, 5, 7, 8, 15 5, 7-OHw, 14 7-OH 69.6, C 6.02 (brs) 6, 7, 8, 15 6w 8 127.4, CH 5.72 (brdd, 10.2, 9, 10 6, 7, 10, 15 9 3, 3) 9 129.9, CH 5.94 (brd, 10.2) 8, 10 1, 5, 7, 10 8, 10w, 11w 10 36.8, CH 2.45 (m) 1, 5, 8, 9 1, 2, 4, 6, 8, 11 1, 5, 9w, 11w w w 11 18.3, CH3 1.00 (d, 7.2) 1 1, 2, 10 1, 9 , 10 12 42.5, CH 1.46 (m) 13, 14 3, 4, 5, 13 5, 13, 14 13 11.2, CH3 0.90 (d, 7.2) 12 12, 14 4, 12 14 83.9, CH 4.90 (dd, 7.8, 6.0) 6, 12 4, 5, 7w, 15 6, 12 15 177.6, C a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

Compound 71 was isolated as a light brown gum and was assigned the molecular formula C15H22O4 on the basis of HRESIMS and NMR data (Table 5). The

1 H NMR spectrum of 71 (Figure 11, Table 5) showed five aliphatic signals between δH

1.61 and 2.53, one olefinic signal at δH 6.88, four methylene moieties (δH 2.37/2.03,

13 1.58/1.40, 1.38/1.09 and 1.32/1.16), and two methyl signals (δH 0.97 and 0.88). The C

NMR spectrum of 71 (Figure 11, Table 5) suggested that the molecule contained an

112 olefin (δC 140.1 and 131.4) and two carbonyl moieties (δC 168.5 and 176.8).

1 13 Figure 11. H and C NMR spectra for 71 in DMSO-d6

Analyses of the 1H-1H COSY, HSQC, and HMBC spectra allowed for the planar bicyclic structure of the sesquiterpene 71 to be constructed (Figure 12). The proton at

H-4 (δH 1.84) showed a HMBC correlation to the methyl group at H-13 (δH 0.97). It was also found that H-4 (δH 1.84), H-12 (δH 2.53), and H-13 (δH 0.97) all showed HMBC correlations to a carbon at δC 176.8, allowing a carboxyl group to be positioned at C-14.

HMBC correlations from both H-6 and H-8 to C-15 (δC 168.5) enabled a second carboxylic acid side chain to be positioned at C-7 (δC 131.4).

Table 5. NMR data for mitchellene D (71)a b c Position δC δH COSY HMBC ROESY 1 34.2, CH 1.63 (m) 2α, 2β, 11 11 2α, 3α, 5, 11 w 2α 28.6, CH2 1.38 (m) 1, 2β, 10 1, 3, 4 1, 2β, 3α 2β 1.09 (m) 1, 2α 1, 10 2α, 3β, 4 w 3α 26.6, CH2 1.16 (brddd, 3β, 4 1, 4 1, 2α, 3β, 5 13.2, 13.2, 11.4) 3α, 4 1, 2, 5 2β, 3α, 4 3β 1.32 (brd, 13.2) 4 40.2, CH 1.84 (brdd, 12.0, 3αw, 3β, 5, 12 3, 6w, 10, 13, 14 2β, 3β, 9β, 11.4) 12 5 38.7, CH 2.05 (m) 4, 6, 10 3, 6, 7 1, 3α, 6, 10, 12w, 13 6 140.1, CH 6.88 (brd, 4.2) 5, 8αw, 8βw 7, 8, 10, 15 5, 12 7 131.4, C w w 8α 25.2, CH2 2.03 (m) 6 , 8β, 9α, 9β 6, 7, 9, 10 8β, 9α 8β 2.37 (brdd, 18.6, 6w, 8α, 9αw, 9β 6, 7, 9, 10, 15 8α, 9β 4.8) w 9α 16.2, CH2 1.58 (m) 8α, 8β , 9β 7, 8, 10 8α, 9β 9β 1.40 (m) 8αw, 8β, 9α, 10 8, 10 4, 8β, 9α 10 38.5, CH 1.61 (m) 5, 9β 5 5, 11 11 19.2, CH3 0.88 (d, 6) 1 1, 2, 10 1, 10 12 39.0, CH 2.53 (dq, 3.6, 4, 13 3, 5, 13, 14 4, 5w, 6, 13 7.2) 13 9.6, CH3 0.97 (d, 7.2) 12 4, 14 5, 12 14 176.8, C 15 168.5, C a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

Figure 12. Key COSY, HMBC and ROESY correlations for 71

The relative configuration of 71 was obtained following investigation of the

ROESY spectrum and 1H-1H coupling constants (Figure 12, Table 5). This assignment initially proved difficult due to overlapping or broadened signals in the 1H NMR spectrum, however, comparison with data obtained on 62 and 68-70 identified that 71 also possessed a muurolane skeleton.115 The relative configuration of the methyl group at C-13 was assigned as having a  orientation on biosynthetic grounds, as it was postulated that 71 is a biosynthetic intermediate for mitchellenes A-C. Hence structure

71 was assigned the trivial name mitchellene D.

The NMR data of mitchellene E (72) (Table 6) was very similar to that of 71, with the only differences being that 72 contained an additional oxygenated methylene

13 signal (δH 3.77, δC 67.4), and was missing one carboxylic acid C resonance. HMBC correlations from δH 3.77 (H-15) to C-8 (δC 26.2) and C-6 (δC 122.6) allowed this –

OCH2– moiety to be attached to C-7. These data, along with a difference of 14 Da in the

(+)-LRESIMS spectra of 71 compared to 72, indicated that the carboxylic acid at C-15 in 71 had been replaced with a di-hydro derivative in 72.

Table 6. NMR data for mitchellene E (72)a b c Position δC δH COSY HMBC ROESY 1 34.8, CH 1.61 (m) 2β, 11 2, 10 2β, 11 2α 28.9, CH2 1.37 (m) 2β, 3α, 3β 1, 3, 10 2β, 11 2β 1.09 (brddd, 13.2, 1, 2α, 3β 1, 3 1, 2α, 4, 11 13.2, 11.4) 3α 26.4, CH2 1.10 (m) 2α, 3β, 4 2, 5 3β, 13 3β 1.28 (m) 2α, 2β, 3α, 4 1, 2, 12 3α, 4 4 38.0, CH 1.81 (m) 3α, 3β, 12 3 2β, 3β, 9β, 12, 13 5 40.4, CH 1.82 (m) 6, 10 10 1, 6, 10 6 122.6, CH 5.67 (brd, 3.6) 5, 8αw, 8βw, 4, 8, 15 5, 12, 15 15w 7 138.8, C w w 8α 26.2, CH2 1.88 (m) 6 , 8β, 9α, 6, 7, 9, 10 , 8β, 9α, 15 8β 2.02 (brdd, 18.0, 9β w, 15w 15w 9αw, 9β, 8α, 5.4) 6w, 8α, 9αw, 6, 7, 9, 10, 15 15 9β, 15w w w 9α 16.4, CH2 1.53 (brdd, 12.6, 8α, 8β , 9β, 7, 8 8α, 8β , 9β, 6.6) 10 7, 8, 10 11 9β 1.40 (m, 1H) 8αw, 8β, 9α, 4, 8β, 9α 10 10 39.0, CH 1.62 (m) 5, 9α, 9β 1 5 11 19.3, CH3 0.87 (d, 6.6) 1 1, 2, 10 1, 2α, 2β, 9α 12 38.6, CH 2.62 (dq, 3.0, 7.2) 4, 13 3, 4, 5, 13, 14 4, 6, 13 13 9.2, CH3 0.93 (d, 7.2) 12 4, 14 3α, 4, 12 14 177.1, C w w w 15 64.9, CH2 3.77 (s) 6 , 8α , 8β 5, 6, 7, 8 6, 8α, 8β a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

After analyses of the ROESY spectrum and 1H-1H coupling constants (Table 6) the relative configuration of 72 was found to be the same as that of 71. As in 71, the relative configuration of the methyl group at C-13 was assigned as having a  orientation based on biosynthetic reasoning. Structure 72 was thus assigned the trivial name mitchellene E. Compound 72 was found to be a new NP, however, a diastereoisomer of this molecule (76) has been produced in a degradation study undertaken on the related NP arteannuin H (75) by Sy et al (Scheme 8).120

Scheme 8. Conversion of arteannuin H (75) to 76120

The known compounds casticin and centaureidin were assigned as structures 73 and 74, respectively, after MS and 1D/2D NMR spectroscopic data analyses and comparison with literature values ( Table 7 and Table 8, respectively).116-119

Casticin (73) is found predominantly within plants of the genus Vitex, which has traditionally been used across Asia and Europe for treating complaints such as inflammation of the respiratory tract,121 premenstrual syndrome and menopause,122-124 pain (e.g. headaches, migraines, eye pain), colds, asthma, and gastrointestinal infections.123,124 As a major constituent of this plant genus, casticin has been investigated for a range of biological activities including antioxidant,122 antiinflammatory,121 antimicrobial,125 and for estrogenic activity for the relief of premenstrual syndrome.123,124,126 Casticin has also been examined for anti-malarial activity,127 and it has been suggested that it may act synergistically with the well known anti-malarial drug artemisinin to increase its activity against the malaria parasite.128,129

The anti-cancer activity of compound 73 has also been well explored;130-135 it has been found to be active against a range of cell lines including breast cancer (MCF-7),131,132 cervix epithelial adenocarcinoma (HeLa),132 human epidermoid carcinoma (KB and

A431),132,134 human lung cancer (PC-12),133 human colon cancer (HCT116),133 and leukemia (K562, Kasumi-1 and HL-60).134,135

Table 7. NMR data for casticin (73)a b c Position δC δH COSY HMBC ROESY 2 146.4, C 3 138.0, C 4 178.2, C 5 151.7, C 6 131.6, C 7 158.7, C w 8 91.3, CH 6.87 (s) 7-OCH3 4, 6, 7, 9, 10 7-OCH3 9 151.6, C 10 105.6, C 1' 122.2, C 2' 115.1, CH 7.58 (m) 2, 3', 4', 6' 3-OCH3 3' 155.6, C 4' 150.3, C w 5' 111.9, CH 7.10 (d, 9.0) 6', 4'-OCH3 1', 2, 3', 4' 6', 4'-OCH3 6' 120.4, CH 7.59 (m) 5' 2, 2', 4' 3-OCH3, 5' 3-OCH3 59.7, CH3 3.79 (s) 3 2', 6' 5-OH 12.60 (s) 6-OCH3 60.0, CH3 3.72 (s) 6 w 7-OCH3 56.5, CH3 3.91 (s) 8 7, 8 8 3'-OH 9.45 (brs) w 4'-OCH3 55.6, CH3 3.86 (s) 5' 4', 5' 5' a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

Centaureidin (74) has also been investigated for cytotoxicity towards a number of cell lines, including cervix epithelial adenocarcinoma (HeLa),132,136 human ovarian cancer (A2780),137 human myelogenous leukemia (K562),136 human melanoma (Fem-

X),136 breast epithelial adenocarcinoma (MCF-7),132 and skin epidermoid carcinoma

(A431).132 It has been found that the cytotoxicity of centaureidin comes from its ability to interfere with tubulin polymerization and its inhibition of [3H] colchicine binding to tubulin.138,139 Compound 74 has also demonstrated antiinflammatory140,141 and antimicrobial142 activity and been found to effect IFN- expression, which modulates pathogen clearance, T-cell activation, and inflammatory responses.118

Table 8. NMR data for centaureidin (74)a b c Position δC δH COSY HMBC ROESY 2 146.3, C 3 137.5, C 4 178.0, C 5 151.8, C 6 131.6, C 7 152.1, C 8 92.6, CH 6.45 (s) 4, 6, 7, 9, 10 9 151.6, C 10 106.8, C 1' 122.4, C 2' 114.8, CH 7.51 (s) 2, 3', 4', 6' 3-OCH3 3' 155.6, C 4' 150.1, C w 5' 111.9, CH 7.08 (d, 9.0) 6', 4'-OCH3 1', 2, 3', 4' 6', 4'-OCH3 6' 120.2, CH 7.52 (d, 9.0) 5' 2, 2', 4', 5' 3-OCH3, 5' 3-OCH3 59.7, CH3 3.76 (s) 3 2', 6' 5-OH 6-OCH3 59.7, CH3 3.72 (s) 6 7-OH 3'-OH w 4'-OCH3 55.6, CH3 3.84 (s) 5' 4' 5' a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); Weak correlations.

2.5. Proposed biosynthesis of mitchellenes A-E

The biosynthesis of bicyclic sesquiterpenes such as 62, 71 and 72 has been well studied and several detailed investigations have been reported in the literature. These

NPs are cyclised from the common sesquiterpene precursor farnesyl diphosphate

(FPP).143-146 Mitchellenes A-C are the first tetracyclic sesquiterpene lactones to be reported. The only other NPs with a similar tetracyclic lactone skeleton are the liverwort diterpenes pallavicinolides A-C (77-79), which were proposed to be biosynthetically derived from the reconstruction of labdane-type diterpenoids.147,148

Scheme 9 shows a proposed biosynthetic pathway for 68-70 starting from NP

71. Similar pathways for the generation of 68-70 would also be possible from compound 72 and scaffold 62. Reduction of the C-12 carboxylic acid in 71 to form the aldehyde i, followed by a proton abstraction at C-8 then an intramolecular cyclisation generates the first five membered ring as shown in ii. Attack by the hydroxyl group at the conjugated carboxylic acid, aided by an unspecified co-enzyme, allows for the formation of the lactone ring. Loss of hydrogen then gives 69. Addition of a hydroxyl group at either C-1 or C-7 facilitates the formation of 68 and 70, respectively.

Scheme 9. Proposed biogenetic pathway for mitchellenes A-C (68-70)

In order to explore possible synthetic routes between mitchellene B (69) and mitchellene A (68), the oxidising reagent dimethyldioxirane (DMDO, 80) was employed in an attempt to introduce a hydroxyl group at position C-1 of 69 (Scheme

10).*149 Dioxiranes have been used in the literature to oxidise unactivated C-H bonds of alkanes under mild conditions.150 DMDO was generated from acetone and potassium peroxymonosulfate using the reaction set up in Figure 13. This apparatus facilitates the slow addition of potassium peroxymonosulfate and acetone to a mixture of H2O, acetone, and NaHCO3 in the main reaction vessel, and allows for the collection of the volatile DMDO in acetone as it is generated.149

Scheme 10. Generation of DMDO and its subsequent use to oxidise mitchellene B (69)

DMDO is also used to epoxidise alkenes, so it was suspected that the endocyclic double bond at C-8 of 69 may undergo oxidation to form 81 (Scheme 10). Just 5 mg of mitchellene B (69) was therefore used in a small scale concept experiment.149 Initially 1 eq of DMDO was added to the reaction vessel. After 2 h the reaction had not proceeded by TLC, so an excess (~ 10 eq) of DMDO was added and the reaction left to stir overnight.

* DMDO reaction was undertaken by Dr Brett Schwartz of the Eskitis Institute, Griffith University.

Figure 13. Reaction apparatus for generation of DMDO149

Analysis of the reaction mixture by (+)-LRESIMS and 1H NMR spectroscopy showed that it was a complex mixture (5 mg). Purification attempts were unsuccessful.

1H NMR and (+)-LRESIMS data (m/z 249 [M + H]+) obtained on the crude mixture did indicate that the epoxide 81 had formed, however, as it was only present within the mixture in a small amount further purification was not pursued. The presence of 68 was not detected.

2.6. Conclusion

Seven compounds, including three novel tetracyclic sesquiterpene lactones, mitchellenes A-C (68-70), two new sesquiterpene carboxylic acids, mitchellenes D (71) and E (72), and the known flavonoids casticin (73) and centaureidin (74) were isolated alongside scaffold 62 from the leaves of E. sturtii. A feasible biogenetic pathway from

NP 71 to give 68-70 is suggested.

Chapter 3. Screening library generation using scaffold 62

3.1. Introduction

With scaffold 62 isolated in sufficient quantity (~ 700 mg), the project proceeded to the synthesis of library analogues. It was envisaged that both the carboxylic acid and allylic hydroxyl of 62 would be utilised to create library members in order to increase the structural diversity of the analogue series. This chapter describes the production of two screening libraries using scaffold 62. The first library consisted of six amides that were generated by elaboration at the carboxylic acid side chain of 62, and the second library of six carbamates that were synthesised by reacting the allylic alcohol with a series of isocyanates. The anti-plasmodial activity of the semi-synthetic compounds and the structure activity relationships (SAR) obtained after testing against the chloroquine sensitive Plasmodium falciparum line 3D7 are also discussed.

3.2. Selection of reaction partners

In order to ensure the drug-likeness of the library members being generated from scaffold 62, the commercially procured reaction partners were selected by first generating virtual libraries of potential products using the software packages Reactor151 and Instant JChem.152 A number of physicochemical parameters including Lipinski's

18 “Rule of Five” for drug-like molecules (MW < 500, HBD < 5, HBA < 10, Log P < 5) and Log D5.5 were calculated in an effort to select the most desirable molecules for subsequent synthesis.

The distribution coefficient at pH 5.5 (Log D5.5) was included since it has been proposed that it is a better descriptor (c.f. Log P) of the lipophilic nature of drug-like molecules, particularly NPs,153 in the small intestine, where the majority of oral drug absorption occurs.82 Whilst Log P calculations specifically predict the partitioning of neutral (i.e. un-ionised) species between n-octanol and water, Log D is a more appropriate measure as it considers the distribution of both ionised and un-ionised species at a given pH. On the basis of the in silico data, coupling reagents were subsequently selected that would ensure the drug-likeness of the products generated from scaffold 62.

3.3. Amide library generation

Initially 62 was modified at the carboxylic acid side chain by generating a small library of amides. The reaction partners were chosen as described above using a list of commercially available amines (895 in total)154 and the Reactor151 and Instant JChem software packages.152

Due to the low quantity of starting material, only ~ 10 mg of 62 was used in each reaction. Three standard coupling reagents were initially investigated to examine which would give optimal yields. These reagents included 4-(4,6-dimethoxy-1,3,5- triazin-2-yl)-4-methylmorpholinium chloride (DMTMM),101 N-(3- dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDCI),95 and oxalyl

155 chloride [(COCl)2].

Scheme 11. Amine couplings with scaffold 62 using EDCI or DMTMM

Reactions using either DMTMM or EDCI (Scheme 11) afforded low yields (82-

84, < 35%) that were believed to be due to a combination of the hygroscopic nature of the NP scaffold and the small reaction scale employed. The small proportion of minor impurities (< 10%) in the starting material would have also had an impact on the yields obtained.

As the stock amount of 62 was low, test reactions using DMTMM or EDCI and a commercially available carboxylic acid were carried out to assess the effectiveness of these coupling reagents.

3-Chloro-4-hydroxyphenylacetic acid (85) was chosen as the model carboxylic acid as it had proven to be a good reaction partner in previous work undertaken within the group.95 Compound 85 was reacted with three amines using the same reaction scale and timeframe as the amine couplings with scaffold 62 using DMTMM or EDCI

(Scheme 12, Table 9). The structures of the products generated (86-88) were confirmed by MS and 1D/2D NMR spectroscopic analyses and comparison with literature values.95 This is the first report of compound 86, structural data for this new compound can be found in Chapter 8 (Experimental).

Scheme 12. Amine couplings with 85 using EDCI or DMTMM

Table 9. Results from DMTMM and EDCI amine couplings with 85 Coupling Reaction Amine Product Yield (%) Reagent Time (h) EDCI 16 histamine 86 23 EDCI 16 phenethylamine 87 92 EDCI 16 N-(2-aminoethyl)acetamide 88 0 DMTMM 5 histamine 86 15 DMTMM 5 phenethylamine 87 0 DMTMM 5 N-(2-aminoethyl)acetamide 88 0

Only when 85 was reacted with phenethylamine using EDCI was a high yield achieved (92%). It was determined that the couplings using either DMTMM or EDCI were not consistent enough, with no product being generated in a number of the reactions. It was therefore decided that another coupling reagent would be explored to try and improve the yields.

Oxalyl chloride (COCl)2 was selected as the next coupling reagent to investigate. On using this chemical for the coupling of the model compound 3- chlorophenylacetic acid (89) with phenethylamine to give 90, a yield of 83% was achieved (Scheme 13). The structure of 90 was confirmed by MS and 1D/2D NMR spectroscopic analyses and comparison with literature values.156

Scheme 13. Oxalyl chloride coupling between 89 and phenethylamine

The first reaction utilising (COCl)2 to couple phenethylamine with scaffold 62 gave a high yield of 77% (82, Scheme 14, Figures 14 and 15), however, subsequent reactions employing other amines resulted in yields ranging from 5 to 48% (84 and 91-

93, Scheme 14).

Scheme 14. Amine couplings with scaffold 62 using oxalyl chloride

1 Figure 14. H NMR spectrum for 82 in DMSO-d6

13 Figure 15. C NMR spectrum for 82 in DMSO-d6

In all (COCl)2 coupling reactions minor side products were observed by (+)-

LRESIMS and 1H NMR spectroscopy. These data suggested some of the minor products included the substitution of the primary hydroxyl (14-OH) with a chlorine atom.157,158 Due to the low amounts obtained (< 0.5 mg) and difficulty associated with purifying these potentially unstable minor side-products, full spectroscopic characterisation was not possible.

When scaffold 62 was reacted with 4-methoxyphenethylamine using (COCl)2, formation of the oxalic acid derivatives 94 and 95 was observed following (+)-

LRESIMS and 1H NMR spectroscopic data analysis.159,160 Compound 94 showed an ion at m/z 456 [M + H]+ in the (+)-LRESIMS spectrum, which corresponds to the oxalic acid structure shown. However, this compound was found to be unstable, with it degrading back to starting material 62 over time. Conclusive 2D NMR or HRESIMS data was not obtained.

Compound 95 was initially identified by analysis of the (+)-LRESIMS data (m/z

589 [M + H]+). This compound was found to be stable, so 1D/2D NMR data was obtained. Compound 95 has not been reported in the literature before, therefore its 1H and 13C NMR data is detailed in Chapter 8 (Experimental). This sample was < 50% pure and was obtained in such low amounts (0.4 mg), that purification was not pursued.

Oxalyl chloride has been previously reported to form similar oxalic acid moieties.159-162

It was decided that protection chemistry would be investigated in an attempt to increase the reaction yields by preventing the allylic primary hydroxyl group in 62 from undergoing side reactions.

Tert-butyl(chloro)diphenylsilane (TBDPS-Cl) was the protection group of choice as the reaction conditions required for protection and de-protection are simple and mild.163 After reaction of 62 with TBDPS-Cl to give 96 (Scheme 15), the crude sample obtained was purified by silica flash column. It was found that both the product and un-reacted TBDPS co-eluted over many fractions, and there was a substantial loss of the protected starting material. When one quantity of 96 (112 mg) was purified using a silica flash column only 35 mg (31%) of 96 was recovered.

Scheme 15. TBDPS protection of scaffold 62, coupling with 4-methoxyphenethylamine

using (COCl)2 to give 97, then de-protection to provide 92

The protected scaffold 96 was reacted with (COCl)2 and 4- methoxyphenethylamine to give 97 (4.7 mg, 56%). Removal of the TBDPS group was carried out using tetrabutylammonium fluoride (TBAF, Scheme 15). After 1H NMR analysis of the crude compound 92 was identified, however, two flash columns were required in order to remove the TBDPS, and the yield obtained was low (1 mg, 19% over two steps).

The coupling between 96 and 4-methoxyphenethylamine was repeated without chromatographic steps in between the amine coupling and de-protection stages. This resulted in a higher yield of 44% over the two steps.

While the yield obtained for the formation of 92 was improved (44%) in comparison to when 62 was unprotected (5%), there was a substantial loss of the scaffold during the protection and purification steps. Due to the limited quantities of scaffold 62, it was deemed that the loss was too high for this form of protection chemistry to be worthwhile, and further protection chemistries were not attempted.

A summary of the yields obtained using the different coupling reagents to form the library of six amides (82-84 and 91-93) can be seen in Table 10.

Table 10. Amine coupling yields with 62 using DMTMM, EDCI, and (COCl)2 Yield using Yield using Yield using Product Amine Partner DMTMM (%) EDCI (%) (COCl)2 (%) 82 phenethylamine 19 12 77 83 ammonia 27 0 * 84 N-(2-aminoethyl)acetamide 31 22 5 91 2-(4-chlorophenyl)ethylamine * * 48 92 4-methoxyphenethylamine * * 5 93 tyramine * * 8 *Reaction not undertaken

3.4. Carbamate library generation

At this point, due to the scarcity of the scaffold and our desire to generate other

analogues based on the muurolane NP 62, it was decided that the allylic hydroxyl group

of 62 would be utilised in reactions with commercially available isocyanates to generate

a library of carbamates.102

Commercially available cinnamyl alcohol (98) was first used in reactions with

benzyl isocyanate and butyl isocyanate at room temperature and 40 °C in order to

optimise the reaction conditions (Scheme 16, Table 11). The structures of compounds

99 and 100 were confirmed by (+)-LRESIMS and 1D/2D NMR spectroscopic analyses.

Scheme 16. Coupling of cinnamyl alcohol (98) with isocyanates

Table 11. Isocyanate coupling yields with cinnamyl alcohol Product Yield (%) at rt Yield (%) at 40 °C 99 83 68 100 36 20

The known imidodicarbonic diamide side product 101, arising from reaction of the excess benzyl-isocyanate with H2O, was detected by (+)-LRESIMS and its structure confirmed by 1D/2D NMR spectroscopic analyses and comparison with literature values.164 This side product was obtained in higher amounts when the reaction was heated at 40 °C. Higher yields of the desired products 99 and 100 were obtained at room temperature (Table 11), therefore these same conditions were selected for the reaction of scaffold 62 with a series of isocyanates.

In an identical manner to the amide library, a virtual collection of potential carbamates was initially generated using the software Reactor151 and Instant JChem152 coupled with a list of commercially available isocyanates (315 in total).154 Analysis of the in silico physicochemical properties (MW, HBD, HBA, Log P, Log D5.5) of the virtual library identified those members that were compliant with drug-like properties.18,82

Literature reports indicated that the carboxylic acid of 62 would react with the

isocyanate reagents to form anhydrides or amides,165 hence this was protected by

methylation using TMS-diazomethane166 prior to carbamate library synthesis. In

contrast to the unfavourable attempt at protection chemistry in regards to the amide

library, the TMS-diazomethane methylation of scaffold 62 could be achieved in high

yields (> 90%), with no de-protection or purification required. Subsequent reactions

of the methyl ester scaffold 102 with various isocyanates resulted in six carbamates

with yields ranging from 31 to 76% (Scheme 17, 103-108).

Scheme 17. Coupling of scaffold 102 with isocyanates to generate compounds 103-108

Typically all of the amides and carbamates produced in this study were first purified by fractionation using a silica solid-phase extraction (SPE) cartridge employing either a n-hexane/EtOAc or n-hexane/i-PrOH gradient. Further purification was achieved by semi-preparative diol-bonded silica HPLC (i-PrOH/n-hexane) or C18- bonded silica HPLC (CH3OH/H2O) when required. The structures of all library members (82-84, 91-93, 102-108) were determined following (+)-HRESIMS and 1D

(1H, 13C) and 2D NMR (COSY, HSQC, HMBC and ROESY) spectroscopic data analysis. This data can be seen in Chapter 8 (Experimental).

Prior to submitting the library compounds (82-84, 91-93, 102-108) and the NPs isolated from E. sturtii (62 and 68-74) for biological evaluation, purity studies were performed using analytical C18 HPLC. Analysis of the HPLC data and integration of all

UV peaks at 210 nm determined that all compounds had purities > 90%.

3.5. Biological activity of the muurolane libraries

To date, the library analogues (82-84, 91-93, 102-108) have been screened for anti-plasmodial activity in an in vitro growth inhibition assay using a chloroquine sensitive Plasmodium falciparum line (3D7). The NPs isolated from E. sturtii, mitchellenes A-E (68-72), casticin (73) and centaureidin (74), were also tested in this screen. Initially all compounds were screened at 50 μM, and those displaying a percent inhibition of ≥ 50 had their IC50 values determined (Table 12). All compounds were also screened for cytotoxicity towards a normal mammalian cell line [neonatal foreskin fibroblasts (NFF)] in order to determine the selectivity of the compounds towards the malaria parasite.

Scaffold 62 and those library members with simple modifications at the carboxylic acid, such as the amide 83 and the methyl ester 102, exhibited no activity against P. falciparum 3D7. The related NPs 71 and 72 also demonstrated no inhibition in this assay. In contrast, the introduction of larger substituents at the carboxylic acid increased the antimalarial activity, with the aromatic analogues eliciting a low to moderate response against the parasite (IC50 14-19 μM). Of the phenethylamine derivatives (82, 91-93) only compound 93 was inactive, suggesting that the polar hydroxyl group on the benzene system was responsible for the reduced activity.

All members of the carbamate library demonstrated low to moderate activity against the P. falciparum 3D7 parasites (IC50 16-33 μM). These compounds only showed minor variations in their activity, with no particular carbamate substituent eliciting a greater response against the parasite than the others.

The tetracyclic NPs 68-70 also showed low to moderate anti-plasmodial inhibitory activity (IC50 20-30 μM). While being structurally similar and possessing almost identical MW values to NPs 71 and 72, the rigid structures of 68-70 may allow them to bind more effectively to the target than compounds 62, 71 and 72, which possess a greater number of rotatable bonds.15 However, 68-70 do not possess chemical handles that would be useful in the synthesis of derivatives for the exploration of structure activity relationships (SAR). As indicated in the literature,127-129 casticin (73) was found to possess activity against the malaria parasite (IC50 10 μM), as did the closely related flavonoid centaureidin (74, IC50 10 μM).

The two screening libraries and E. sturtii NPs were also tested against the prostate cancer cell line LNCaP at a final concentration of 10 µM using a real time cell analyser (xCELLigence). However, none of the compounds displayed activity against this cell line.

Additionally, all library members generated from scaffold 62 and the NPs from

E. sturtii have been added to the open-access small molecule repository at the

Queensland Compound Library (QCL)167 located at the Eskitis Institute where they are available for drug discovery and chemical biology research.

Table 12. Physicochemical profiling and biological activity of NPs and semi-synthetic analogues† physicochemical parametersa biological activity

b c compound Mw Log P Log D5.5 HBA HBD 3D7 (IC50 ± SD μM) NFF (% ± SD) 62 250 2.42 1.70 3 2 > 50 d 68 248 1.78 1.78 2 1 24 ± 8 49 ± 10 69 232 3.09 3.09 1 0 30 ± 7 53 ± 15 70 248 1.94 1.94 2 1 20 ± 7 e 71 266 2.98 1.14 4 2 > 50 d 72 252 2.51 1.73 3 2 > 50 d 73 374 2.40 2.40 8 2 10 ± 8 74 360 2.26 2.24 8 3 10 ± 1 66 ± 5 82 353 3.85 3.85 2 2 14 ± 4 32 ± 16 83 249 1.61 1.61 2 2 > 50 d 84 334 0.86 0.86 3 3 > 50 d 91 387 4.45 4.45 2 2 19 ± 2 86 ± 5 92 383 3.69 3.69 3 2 19 ± 1 16 ± 5 93 369 3.54 3.54 3 3 > 50 d 102 264 2.80 2.80 2 1 > 50 d 103 427 4.77 4.77 5 1 33 ± 8 e 104 397 4.85 4.85 2 1 28 ± 7 44 ± 10 105 363 4.45 4.45 2 1 20 ± 3 15 ± 4 106 389 4.93 4.93 2 1 16 ± 1 41 ± 15 107 427 4.69 4.69 3 1 27 ± 3 45 ± 9 108 457 4.54 4.59 4 1 18 ± 3 86 ± 5 chloroquine 319 3.93 -0.76 3 1 0.03 ± 0.01 f a 152 In silico calculations performed using Instant JChem software; MW = molecular weight; Log P = partition coefficient; Log D5.5 = distribution coefficient at pH 5.5; HBA = H bond acceptors; HBD = H bond donors; b3D7 = Plasmodium falciparum (chloroquine sensitive) strain; cNFF = neonatal foreskin d e f fibroblast cells, percent inhibition measured at 100 μM; not tested; not active; chloroquine IC50 = 34 ± 3 µM.

†Antimalarial and NFF cytotoxicity assays were undertaken by Vanida Choomuenwai and Dr. Katherine Andrews, and the LNCap cytotoxicity assay by Michelle Liberio, all of whom are members of the Eskitis Institute, Griffith University.

3.6. Synthesis of a di-substituted analogue of scaffold 62

While only low anti-malarial activity was obtained from the amide and carbamate analogues of scaffold 62, it was observed that some activity had been generated compared to that of the scaffold itself. It was hypothesised that generating an analogue that had been chemically elaborated at both the carboxylic acid and allylic hydroxyl group of 62 may result in a compound with higher activity. The most active amide and carbamate were 82 and 106, respectively. Hence the side chains of these molecules were used in the synthesis of compound 109, which was predicted to display greater potency in the malaria assay.

Scaffold 62 was first coupled with phenethylamine at the carboxylic acid side chain using (COCl)2 (Scheme 18, 82, yield 36%). Compound 82 was then reacted with cyclohexyl isocyanate and the crude product obtained purified by silica flash column (n- hexane/EtOAc) to give 109 in a yield of 10% over two steps. The structure of 109 was confirmed by (+)-HRESIMS and 1D/2D NMR spectroscopic data analysis (Figure 16).

Scheme 18. Formation of the di-substituted analogue 109

1 13 Figure 16. H and C NMR spectra for 109 in DMSO-d6

Compound 109 was screened against the chloroquine sensitive P. falciparum line 3D7 for anti-plasmodial activity, and for cytotoxicity towards the normal mammalian cell line NFF. An IC50 value of 20 μM against the malaria parasite was obtained. Compound 109 therefore retained the same moderate level of activity as its smaller MW analogues, 82 and 106. It is noteworthy that compound 109 has a number of physicochemical parameters, including Mw and Log P, that are now very close to or out side of the ranges suggested by Lipinski's “Rule of Five” for drug-like molecules.18 This could effect how the compound reaches the target site and its binding.

Compound 109 displayed the greatest toxicity of all the semi-synthetic analogues (Table 12) with an IC50 value of 86 μM towards the NFF cells. The poor selectivity of 109 towards the malaria parasite indicates further investigation of di- substituted analogues of 62 is not warranted.

3.7. Conclusions

The NP scaffold 14-hydroxy-6,12-muuroloadien-15-oic acid (62), isolated from the plant E. sturtii, was utilised in the generation of two screening libraries. Initial screening data suggests that scaffold 62 may be a valid starting point for the generation of bioactive compounds, as although the scaffold was inactive against P. falciparum

3D7, many of the library members generated from this molecule demonstrated improved activity. Furthermore, NP scaffolds such as the one utilised in this study have the potential to be used as tools for investigating biological pathways and developing novel therapeutics.33

Chapter 4. Where's my compound? Analytical strategies for the determination of taxonomy

4.1. Introduction

The positive results obtained from the use of the sesquiterpene 62 as a scaffold for analogue generation indicated that it would be worth pursuing its use in the formation of additional screening libraries. It was also deemed worthwhile to ascertain whether this natural product (NP) is a privileged structure. However, the stock amount of this compound was nearly depleted during the initial synthetic investigations undertaken in this project, and all the raw plant material housed within the Nature Bank library had been exhausted. The re-isolation of this interesting scaffold for further studies initially proved elusive.

The original plant sample from which scaffold 62 was obtained had been identified as Eremophila mitchellii by the Queensland Herbarium. This plant was collected from Currawinya National Park, which is located in a remote area of western

Queensland, on the border with New South Wales. A re-collection of the plant by the

Queensland Herbarium was requested, however, costs were prohibitive ($15,000).

Therefore, the Nature Bank biota library was searched for additional specimens of E. mitchellii, as well as other Eremophila sp. leaf samples that might contain scaffold 62.

Within this chapter, the plant samples studied will be referred to by their QID codes, which are the codes assigned to them for storage and tracking within the Nature

Bank library. For example, the original plant sample thought to be E. mitchellii will be referred to as QID005836.

4.2. Analytical HPLC analysis of E. mitchellii samples

Seven additional samples of E. mitchellii (Table 13) collected from three different sites in Queensland, Australia, were archived in the Nature Bank biota library.

These samples were of different parts of the plant, one consisting of the leaves, three of the wood, two of the bark and one of the roots.

Table 13. E. mitchellii samples in the Nature Bank biota library Sample code Plant part Mass available (g) QID001000 bark 111 QID000948 wood 69 QID005817 bark 50 QID005820 roots 52 QID005823 wood 44 QID004133 leaves 116 QID004138 wood 85

Small scale extraction (CH2Cl2/CH3OH) of 200 mg of each plant sample gave seven crude extracts that were examined by analytical HPLC on a C18 Onyx monolithic column (H2O/CH3OH/0.1% TFA). A portion of the original QID005836 plant extract and a sample of pure NP 62 were also injected for comparison.

Figure 17 shows the UV traces obtained for each analytical HPLC run. Scaffold

62 eluted at 6.20 min. Only QID004133 possessed a UV peak at around 6.20 min in its

HPLC chromatogram that we postulated could correspond to scaffold 62, though its retention time varied slightly. Like the original biota QID005836, sample QID004133 was also leaf material. (+)-LRESIMS was conducted on the fractions collected between

6-7 min from each injection and a very weak ion that potentially corresponded to 62 could be seen for QID004133. The unconvincing MS data and the distinct differences seen between the analytical HPLC chromatograms of the two leaf samples hinted that

QID004133 may not produce NP 62.

scaffold 62

QID005836, leaves, original sample

QID004133, leaves

QID005820, roots

QID001000, bark

QID005817, bark

QID005823, wood

QID004138, wood

QID000948, wood

UV wavelength key

Figure 17. Analytical HPLC chromatograms of 62, the original plant sample

QID005836, and seven E. mitchellii samples archived within the Nature Bank library

It was initially suspected that the production of 62 in the second leaf sample was just much lower than that of the original sample. An extraction and purification was therefore undertaken on 5 g of QID004133 using the same protocol as that of the original plant sample (Scheme 5). A number of the HPLC fractions were analysed by

(+)-LRESIMS and 1H NMR spectroscopy, however, the presence of scaffold 62 in

QID004133 was not confirmed.

These results raised issues that sometimes occur during NP research. Firstly, there was the possibility that the taxonomy of the sample had been incorrectly identified. Secondly, seasonal variations in the production of the desired secondary metabolite may have accounted for the discrepancy. The second Eremophila

(QID004133) was collected in a similar region in Queensland as the first, however, they were obtained at different times of year, the first in March, 1996, and the second in

October, 1993. It was initially suspected that seasonal variation was most probably the cause of the unsuccessful re-isolation of 62, as Eremophila plants are known to behave differently depending on the amount of water they are receiving.106 The long, sporadic drought periods experienced in Australia's arid regions, where these plants predominantly grow, can affect, for instance, the amount of leaf coverage and flowering patterns of Eremophila species.106 It is therefore conceivable that there are seasonal differences in chemical production of these plants, particularly in their leaves and outer branches.

Whilst the cost of recollecting the plant from the original collection point was prohibitive, it was also known that these plants grow across large regions of western

Queensland and New South Wales. Whilst waiting for information on new sources of E. mitchellii to study, it was decided that other Eremophila samples would be investigated in an attempt to find new sources of scaffold 62.

4.3. Analytical HPLC analysis of Eremophila species from Nature Bank

To date, scaffold 62 had been isolated from the two sub-species of E. interstans

(E. interstans subsp. virgata and E. interstans subsp. interstans) as well as the species studied at the Eskitis Institute.105,113 It was therefore hypothesised that it may also be found in other Eremophila plants. The Nature Bank biota library was found to contain

184 specimens from the Eremophila genus. Thirty-nine of these were labelled as 'leaves' or 'aerial parts' (Table 14). A portion (200 mg) of each of these 40 samples was extracted with CH2Cl2/CH3OH before being evaluated by analytical HPLC (Figure 18) and (+)-LRESIMS to see whether they contained scaffold 62.

Of the 40 samples, two appeared to contain 62 after (+)-LRESIMS analysis of the fractions eluting between 6-7 min (samples QID027518 and QID032241). A larger portion (1 g) of each of these samples was therefore taken for extraction

(CH2Cl2/CH3OH) and purification by diol-bonded silica HPLC. After purification, however, only tiny amounts (< 0.5 mg) of scaffold 62 could be seen in QID027518 and none at all in QID032241 by 1H NMR spectroscopy. Diastereoisomers of 62 have been found in previous studies,113 and with such a small amount being isolated we could not definitively say whether the NP from QID027518 was NP 62 or an isomer of 62. Figure

18 shows the analytical HPLC traces for QID027518 and QID032241, as well as some further examples of the UV profiles obtained from the 40 specimens. The HPLC chromatograms for all 40 samples can be seen in Appendix I.

Table 14. Eremophila 'leaf' and 'aerial parts' samples in Nature Bank biota library Sample code Plant Taxonomy Plant part Mass available (g) QID001037 Eremophila goodwinii subsp. goodwinii leaves 26 QID006314 Eremophila bignoniiflora leaves 44 QID005814 Eremophila longifolia leaves 86 QID005067 Eremophila latrobei subsp. glabra leaves 44 QID004249 Eremophila maculata subsp. maculata leaves 39 QID005257 Eremophila dalyana leaves 25 QID006441 Eremophila gilesii subsp. gilesii leaves 23 QID006165 Eremophila bowmanii subsp. bowmanii leaves 49 QID008865 Eremophila bignoniiflora leaves 56 QID009016 Eremophila bowmanii subsp. latifolia leaves 61 QID009293 Eremophila longifolia leaves 117 QID001052 Eremophila latrobei subsp. latrobei leaves 23 QID003102 Eremophila longifolia leaves 74 QID006094 Eremophila glabra subsp. glabra leaves 38 QID009233 Eremophila obovata leaves 27 QID001081 Eremophila gilesii subsp. gilesii leaves 27 QID026927 Eremophila eriocalyx aerial parts 124 QID026930 Eremophila oldfieldii subsp. angustifolia aerial parts 64 QID026806 Eremophila eriocalyx aerial parts 49 QID027996 Eremophila linsmithii aerial parts 44 QID1987814 Eremophila longifolia aerial parts 93 QID032416 Eremophila sp. aerial parts 91 QID032146 Eremophila longifolia aerial parts 74 QID030822 Eremophila platycalyx aerial parts 142 QID030825 Eremophila platycalyx aerial parts 76 QID032239 Eremophila spectabilis aerial parts 81 QID031881 Eremophila longifolia aerial parts 180 QID032240 Eremophila rotundifolia aerial parts 60 QID032241 Eremophila drummondii aerial parts 27 QID011745 Eremophila latrobei subsp. latrobei leaves 25 QID013779 Eremophila gilesii subsp. gilesii leaves 31 QID026862 Eremophila calorhabdos aerial parts 132 QID027518 Eremophila sp. aerial parts 68 QID027538 Eremophila glabra aerial parts 128 QID028082 Eremophila pantonii aerial parts 74 QID028102 Eremophila resinosa aerial parts 76 QID026919 Eremophila fraseri aerial parts 65 QID026921 Eremophila glabra aerial parts 28 QID026924 Eremophila oldfieldii subsp. angustifolia aerial parts 113 RAD039 Eremophila microtheca aerial parts 45

scaffold 62

QID005836, original sample

QID027518

QID032241

QID005257

QID026927

QID008865

UV wavelength key

Figure 18. Analytical HPLC chromatograms of 62, the original plant sample

QID005836, and representative Eremophila samples from analysis of the 40 specimens

4.4. Analytical HPLC analysis of E. mitchellii samples collected from Kalbar, Queensland

By this time three E. mitchellii samples (leaves and terminal branches) collected from Kalbar, Queensland, had been sourced from a member of the Australian Native

Plant Growers Society. These three samples were therefore also examined for scaffold

62. Small scale extraction and analytical HPLC analysis (Figure 19) revealed that these samples possessed HPLC chromatograms quite similar to QID004133, the only other E. mitchellii leaf sample that was found in the Nature Bank biota library (Figure 19).

Inspection and comparison of the UV traces of the original leaf sample

QID005836 to QID004133 and the Kalbar samples showed that while the latter samples possessed a UV peak near 6.20 min and with an absorbance at 254 nm that could correspond to scaffold 62, the intensity of this peak was higher for these samples. On comparing the UV chromatograms it could also be seen that sample QID004133 and the

Kalbar plants have striking similarities to each other, such as the UV peaks between

5.60 and 7.10 min, which are lacking in the original plant sample QID005836. (+)-

LRESIMS was conducted on the fractions collected between 6-7 min for each of the

Kalbar samples, and a very weak ion that could correspond to 62 was seen in each of them. Much like in the case of QID004133, the weakness of these ions suggested that

62 was not being produced in high amounts, if at all, in these plants.

scaffold 62

QID005836, original sample

QID004133

Kalbar collection, sample 1

Kalbar collection, sample 2

Kalbar collection, sample 3

UV wavelength key

Figure 19. Analytical HPLC chromatograms of 62, the original plant sample

QID005836, the E. mitchellii leaf sample QID004133, and the three E. mitchellii

samples collected from Kalbar, Queensland

To definitively find out whether these samples contained 62, 5 g of one of the

Kalbar collected specimens (sample 1) was extracted with CH2Cl2/CH3OH and the crude obtained purified by diol-bonded silica HPLC. After analysis by 1H NMR spectroscopy, NP 62 was found in this sample, however, in very low yields (7.5 mg, ~

60% pure, 0.15% dry wt) compared to that of the original sample QID005836 (201.7 mg, > 90% pure, 0.73 % dry wt).

This data seems to suggest that if the original leaf sample QID005836 was definitely E. mitchellii, there is seasonal variation in regards to the production of 62 within this species, with one collection being found to produce a high amount of 62

(201.7 mg, 0.73 % dry wt) while others were found to produce it in low amounts or not at all. However, it is almost impossible to definitively make the statement that there is seasonal variation within this species without analysing a much larger cross section of these plants collected from different locations and at different times of year.

Of additional note, the NPs mitchellenes A-E (68-72), casticin (73) and centaureidin (74) were also isolated from QID005836, however, these compounds were not found in the Kalbar sample. This suggested that seasonal variation was not actually the problem, and that the original sample QID005836 was a different species that produces 62 in higher amounts than that seen for the E. mitchellii samples. This dilemma is compounded by the fact that many of the Eremophila plants are closely related and hard to distinguish.106 The data presented here suggests that NP 62 is a common biosynthetic intermediate found in various Eremophila spp., and that it cannot be used as an indicator of the taxonomy of the Eremophila plant sample QID005836.

The next section explores whether the taxonomy of the original leaf sample

(QID005836) was incorrectly identified.

4.5. The difficulties associated with Eremophila taxonomy

In 2007 R.J. Chinnock released a comprehensive monograph on the Eremophila genus.106 While reading the section on E. mitchellii in this book an interesting quote was found, stating that "E. mitchellii is very close to E. sturtii and it is often difficult to distinguish between these two species."106

On reading further, it was found that while these two species can be distinguished on characteristics such as flower colour, corolla pubescence, and leaf shape, none of these are certain discriminators. The flowers are usually a good indicator, with E. sturtii possessing pinkish-mauve and E. mitchellii white flowers (Figure 20). In some cases these colours can be switched, however. E. sturtii typically has one flower per axil, while E. mitchellii has two to three. The two plants can also be separated on the basis of their bark, however, this relies on them having formed a trunk, with both species ranging from shrub to tree form.

Figure 20. E. sturtii (left), photo © M. Fagg, Australian National Botanic Gardens,111

and E. mitchellii (right), photo courtesy of K. Townsend168

E. mitchellii also possesses a distinctive scent which is reminiscent of sandalwood (Santalum sp.).106,169,170 It can be noted that the original leaf sample

(QID005836) studied at the Eskitis Institute was observed not to possess the sandalwood-like scent of the other E. mitchellii samples analysed.

During his re-classification of the Eremophila genus, Chinnock concluded that there weren’t any well defined discontinuities that would allow for the subdivision of

Eremophila into sub-genera, so instead separated it into 25 sections.106 Of interest is that the two sub-species of E. interstans, from which 62 was originally isolated,105,113 as well as E. mitchellii and E. sturtii are all classified by Chinnock as being part of the

Crustaceae section.106 The geographical distribution of E. mitchellii and E. sturtii also overlaps within north-central to north-western New South Wales and south-western

Queensland.106

It was therefore decided that it would be worth analysing samples of E. sturtii to determine whether the sample QID005836 was in fact this species. Three samples of E. sturtii were identified in the Nature Bank biota library (Table 15). All three had been gathered near Currawinya National Park, Queensland - the same National Park from which the original plant sample (QID005836) that yielded 62 was collected.

Table 15. E. sturtii samples archived in the Nature Bank library Sample code Plant part Mass available (g) QID009029 roots 58 QID009030 mixed 87 QID009031 wood 37

All three E. sturtii samples (1 g each) were extracted with successive washes of

CH2Cl2/CH3OH. A portion of each crude was analysed by C18 analytical HPLC (Figure

21). A weak UV peak at around 6.20 min could be seen in the analytical trace for the mixed E. sturtii sample (QID009030) that could correspond to scaffold 62. The remaining crude obtained from each plant sample was purified by diol-bonded silica

HPLC. (+)-LRESIMS of fractions 25-60 showed that the root and mixed plant extracts contained the ion for 62 at m/z 273 [M+Na]+ in fractions 36-38. Furthermore, analysis by 1H NMR spectroscopy confirmed that 62 was within these fractions, with the mixed plant sample containing higher amounts of scaffold 62 (6.5 mg) compared to that of the roots (1 mg). The percent yield of pure 62 obtained from the mixed E. sturtii sample

(6.5 mg, 0.63% dry weight) was similar to that obtained from the original plant sample

QID005836 (201.7 mg, 0.73 % dry wt).

Another clear indicator that the taxonomy of QID005836 was in fact E. sturtii was that mitchellenes B-E (69-72) were detected in QID009030 by (+)-LRESIMS and

1H NMR spectroscopy, but not in any of the E. mitchellii samples. Mitchellenes A-E

(68-72) possess low UV absorption profiles, so it wasn't possible to compare them in the

HPLC chromatograms obtained. The absence of mitchellene A (68) and some of the discrepancies in the analytical chromatograms of samples QID005836 and QID009030 may be explained by seasonal differences between the two plants. The biota QID009030 was collected in August 1995, and QID005836 in March, 1996. A much larger sample size of E. sturtii plants collected at different locations and at different times of year would be needed to conclusively point out the reasons for these disparities.

scaffold 62

QID005836, leaves, original sample

QID009030, mixed

QID009029, roots

QID009031, wood

UV wavelength key

Figure 21. Analytical HPLC chromatograms of 62, the original plant sample

QID005836, and the three E. sturtii samples

The flavonoids casticin (73) and centaureidin (74) were also isolated from the original biota sample QID005836. These compounds were injected onto the C18 analytical HPLC column, with casticin eluting at 5.90 min and centaureidin at 5.80 min

(Figure 22). Both flavonoids can be seen in the chromatogram of QID005836, but only centaureidin was identified in QID009030. This is not surprising seeing as these two

NPs only differ by one methoxy group, the small conversion between the two NPs may be subject to seasonal changes or certain requirements of the plant itself.

QID005836, original sample

QID009030

casticin (73)

centaureidin (74)

verbascoside (110)

UV wavelength key

Figure 22. Analytical HPLC chromatograms of 62, the original sample QID005836, the

E. sturtii sample QID009030, verbascoside, casticin, and centaureidin

The more polar fractions obtained from QID005836 were not initially analysed

(Chapter 2). However, chemical investigations of these fractions from QID009030 were undertaken in order to ascertain the identity of the major compound eluting at 3.90 min in the HPLC chromatograms of these samples (Figure 22). After 1D/2D NMR and MS data analysis, comparison to literature values,171 and re-injection of the pure NP onto the analytical HPLC column, it was determined that this NP was the known caffeic acid glycoside verbascoside (110).

Verbascoside (110) is found in a range of plants (> 30 spp.), and has been isolated from at least 12 Eremophila spp., not including those investigated in this project.172-177 When comparing the analytical HPLC chromatograms obtained for the 40

Eremophila samples analysed above (Appendix I) to that of verbascoside, it can be seen that the majority (30 samples) contain a UV peak that corresponds with this NP. The level of production of 110 in some Eremophila spp. has been linked to stress in these plants.176 This may explain the different amounts of this compound in the analytical

HPLC chromatograms of sample QID009030 and QID005836.

The similar chemical profiles obtained for the E. sturtii mixed sample

(QID009030) and the original sample QID005836, in terms of the yield of scaffold 62 and the NPs isolated, indicated that the initial taxonomy of QID005836 was erroneous.

The Queensland Herbarium was asked to re-examine the taxonomy of the voucher specimen, and they confirmed that their initial taxonomic assignment had been incorrect. The sample QID005836 has now been re-classified as E. sturtii.

4.6. Conclusion

It was discovered that the taxonomy of a plant sample thought to be E. mitchellii

(QID005836) had been incorrectly identified, and after chemical analysis of a number of further Eremophila samples the re-classification to E. sturtii was undertaken. Whilst the discovery of the taxonomic error and its subsequent ratification came too late for more substantial work to be completed using scaffold 62, it does ensure that this appealing NP can now be obtained in larger quantities for future chemistry projects stemming from the initial studies undertaken here. This study also underpins the value of using chemical means to aid in taxonomic identification. A combination of chemical and taxonomical indicators would ideally be the best method for distinguishing samples that do not possess definitive botanical characteristics.

Chapter 5. Introduction Part B

5.1. The Eremophila genus – a source of potential medicines from the Australian desert

While researching the Eremophila genus from which scaffold 62 was obtained, it became apparent that these plants produce a wide range of structurally unique

NPs.169,170 A number of Eremophila species were used as traditional medicines by

Australian indigenous people169,170,178-182 and in contemporary times numerous crude extracts of these species have been shown to possess biological activity in predominantly anti-bacterial and anti-inflammatory screening campaigns. It was also evident on doing a literature review that while a number of the > 215 Eremophila species had been chemically analysed, there still remained a vast number of these plants that had never been studied. These observations suggested that the Eremophila genus could prove to yield a range of interesting potential scaffolds for further screening library production, and be a source of new compounds. The second part of this project therefore involved the chemical investigations of a number of Eremophila plants in the search for new chemistry and potential scaffolds to be used in the generation of screening libraries.

The taxonomic classification of the family Myoporaceae has had a long and confused history since its inception in 1810 by Robert Brown, most likely due to the constant discovery of undescribed species and the complex variation patterns seen for these plants.106 In 2007 Chinnock published a comprehensive monograph that described a number of unclassified species, and re-categorised the plants of this family into seven genera including Bontia, Pentacoelium, Myoporum, Diocirea, Glycocystis,

Calamphoreus, and Eremophila.106

While members of the other genera can be found in China, Japan, the Caribbean, and islands of the Indian and Pacific oceans, Eremophila is endemic to

Australia.106,182,183 These plants can be found primarily in Western Australia, with 181 of the 215 species described by Chinnock106 being found in this state, however, they grow right across the country.106,182,183

Eremophila plants range from small woody shrubs to small trees (Figure 23), and possess a great diversity of flowers and fruit.106,181-183 The fact that these plants grow predominantly in semi-arid to arid regions of Australia is reflected in the Greek meaning of the term Eremophila: eremos, desert; phileo, to love.106,178,183-185 Eremophila spp. predominantly grow in areas were the annual rainfall is < 250 mm, and are typically drought tolerant for long periods.106,183 Very few species can be found in high rainfall areas such as the northern tropics, and none at all grow in Tasmania.106

Figure 23. E. mitchellii (left), E. gilesii (top right), and E. bowmanii (bottom right),

photos © M. Fagg, Australian National Botanic Gardens111 91

Some Eremophila plants are highly localised and can only be found in small populations.106 Encroaching human developments have placed a number of these species on the rare and endangered flora listing in Australia.182

In contrast, a number of other Eremophila species are classified as weeds, particularly in areas of overgrazing where they opportunistically take over.182 Many species can be considered as fodder for grazing animals (E. longifolia, E. latrobei, E. forrestii, E. scoparia, E. oldfieldii and E. maculata) but at the same time there have been several accounts of stock poisonings, particularly in the case of stressed animals that are hungry and/or being driven.169,170,182 Furanoid sesquiterpenes such as (-)- ngainone (111), found in E. latrobei,186-188 and the cyanogenic glycoside prunasin (112) from E. maculata189 (Figure 24) have been identified as some of the toxic agents found in these plants.

Figure 24. E. maculata, photos courtesy of B. Walters190

This genus had experienced a gradual increase in popularity within the horticultural industry due to the drought resistant and salt tolerant nature and the attractive foliage and flowers of its members (Figure 25).106,182,183 A number of these species are now available from nurseries both in Australia and internationally.182

Figure 25. E. nivea (right) photo courtesy of B. Walters.190 E. racemosa (top left) and

E. warnesii (bottom left) photos © M. Fagg and L. Vallee, Australian National Botanic

Gardens.191

5.2. Traditional use of Eremophila spp. by Australian Aboriginal people

Given the number of Eremophila species (< 215) and their wide distribution across mainland Australia, these plants were frequently used for treating ailments and in ceremonial rituals by Australian Aboriginal people.169,170,181 Being nomadic, Aboriginal tribes could not carry large numbers of plants as they travelled, so the most common remedies were those that could be found across larger areas.170,180,181 Eremophila plants were therefore good candidates for inclusion in Aboriginal pharmacopoeias.170,181

Eremophila species are also typically highly aromatic, with many secreting scented oils and resins from their leaves and wood.169,181 Australian Aboriginal people discovered that many plants possessing strong, aromatic scents such as eucalypts and tea trees could be used to treat a number of diseases, including colds, throat and chest complaints, coughs, fever, headaches, and eye infections.170,181,192 They found that inhaling the smoke or vapours of these plants gave relief for many of these ailments.

Aromatic oils are now known to act by irritating respiratory cells leading to increased mucus secretion, which in turn eases dry throat coughing and dislodges phlegm.181

Aromatic oils are thought to be produced by otherwise defenceless, sedentary plants to discourage the growth of bacteria and fungi, to ward off grazing animals, and as protection against harsh arid environments.181,183,184 A number of anti-bacterial or anti-fungal compounds have been identified in aromatic oils such as eucalyptol, limonene, pinene, camphor, geraniol, and menthol.181 Aromatic resins have also been shown to possess anti-inflammatory properties and can relieve muscle pain.181 Being mildly irritating, they stimulate the release of anti-inflammatory agents and draw blood to the effected area.181

Examples of Eremophila plants and their traditional usage can be seen in Table

16. As can be seen in this table the most common uses of Eremophila species were as decongestant, expectorant, anti-diarrheal, antiseptic, and analgesic agents.

Table 16. Eremophila species utilised by Australian Aboriginal people169,170,178-182 Species Usage E. alternifolia Coughs, colds, fever, internal pain, to induce sleep and for general malaise E. duttonii Sores, wounds, inflamed eyes, sore throats, scabies E. freelingii Sores, cuts, scabies, colds, influenza, fever, headaches, diarrhoea, chest pains E. longifolia Counter irritant, scabies, sores, inflamed eyes, cuts, boils, muscle and joint pain, colds, fever, insomnia, headaches E. maculata Colds E. sturtii Colds, chest complaints, backaches, sores, cuts, diarrhoea E. mitchellii Rheumatism E. bignoniiflora Colds, chest complaints, laxative E. latrobei Scabies, colds, influenza E. dalyana Chest complaints, colds, scabies E. gilesii Chest complaints, colds, scabies, sores, headaches E. paisleyi Scabies E. cuneifolia Colds E. fraseri Colds, rheumatism, toothache

E. alternifolia (Figure 26) was of particular importance, being one of the few plants that Australian Aboriginal people dried, stored, and carried with them.170,179,181,182

It was used as a general remedy, but mainly as a decongestant, expectorant and analgesic.170,179-182

Eremophila species are also regarded as being sacred plants by Australian

Aboriginal people. For example, E. longifolia (Figure 26) is of particular importance in

Aboriginal culture.180,182 This plant is used to line graves and to shroud the deceased, in circumcision rites, and to brush sacred objects and the bodies of men during rituals.170,180-182 E. longifolia is also used as a smoking treatment for mothers and their new born children.170,179,180 Inhaling the fumes from the burning wood is thought to strengthen the baby, increase the mother's milk supply, and stop her bleeding.180-182

Figure 26. E. longifolia (left), photo courtesy of B. Walters,190 and E. alternifolia

(right), photo © M. Fagg, Australian National Botanic Gardens.111

For a more extensive insight into the use of Eremophila spp. by Australian

Aboriginal people, the reader is referred to the references by Richmond 1994,182

Ghisalberti 1994,170 Barr 1988,179 and Low 1990.181 Richmond and Ghisalberti give an extensive listing of the medicinal, cultural, horticultural, and phytochemical uses of

Eremophila species.182

5.3. Chemical diversity of the Eremophila genus

The majority of the NPs isolated from Eremophila species fall into the sesquiterpene and diterpene compound classes,183 however, within these groupings the structures produced are quite varied. The following section gives examples of some of the NPs isolated from the Eremophila genus.

Furanosesquiterpenes173,193,194 such as (-)-ngaione (111),186,187,195 freelingii

(113),196-198 freelingnite (114),199 and eremoacetal (115),200 have been implicated in the toxicity of some Eremophila plants and species of the related genera Myoporum to stock.186,194 Freelingii (113) was the first acetylenic terpenoid NP to be isolated.196,197

The isolation of eremophilone type sesquiterpenes such as 116-119201-207 from E. mitchellii201-206 and E. rotundifolia207 created some debate as they did not seem to obey the 'isoprene rule,' which had been established around the same time.205,208-211 It was later suggested that these compounds arise from eudesmane precursors such as 120,212-

216 which possess the isoprenoid skeleton, by methyl migration.169,205,209 This proposal was significant as it opened up the possibility of skeletal rearrangement and methyl migrations in the biosynthesis of terpenes.169 The unusual eremophilone dimer 121 was also isolated from E. mitchellii, and is thought to be derived from the dehydrated eremophilone 119 by Diels-Alder cycloaddition of the conjugated enone.169,217,218

Other sesquiterpenes isolated from Eremophila spp. include compounds of the zizaene series such as 122 and 123,219,220 the aromatic calamenene structures including

124 and 125,221,222 the keto-alcohol (+)-oplopanone (126),223 the cadinene sesquiterpenes such as 127 and 128,99,105,113 and spathulenol (129).213 Of interest is that many of these NPs are produced by Eremophila plants as the opposite enantiomer to other terrestrial sources.169

Many Eremophila spp. possess resin coatings on their leaves and terminal branches, which on analysis have been shown to consist of rich mixtures of oxygenated diterpenes.169 The most numerous diterpene NP type to be isolated from Eremophila spp. is that of the serrulatane class.207,214,224-232 In general the serrulatanes only differ in the position and level of oxygenation of the core skeleton, as shown in examples 130-

132. Many serrulatane compounds have been found to possess anti-bacterial and anti- inflammatory activity, as discussed in section 5.4, below.

The viscidanes are another class of bicyclic diterpenes which are common in

Eremophila spp., again they typically only differ in the position and degree of oxidation.213,233-235 All viscidanes, such as 133-135, possess allylic oxygenation on the cyclohexane ring.169 Syah et. al. isolated an unusual viscidane derived tricyclic lactone

NP from E. georgei (136).236 The bisabolene isoprenologues 137-139, isolated from E. foliosissima, gave insights into the cyclisation process leading to the polycyclic

Eremophila diterpenes.169,237

The 14-membered macrocyclic diterpenes of the cembrane class can be found in both terrestrial and marine environments.169 A number of cembranes have been isolated from Eremophila spp., including 140-142.238-243 Eremophila-derived cembranes are unique in that they possess a cis configuration for all their residual double bonds with respect to the carbon ring. Cembranes from other natural sources typically possess trans-internal double bonds. A number of acyclic diterpenes possessing internal cis double bonds, including 143 and 144, have been isolated from E. exilifolia,244 E. petrophilia169 and E. glutinosa.244 This again goes against the trend with the majority of acyclic diterpenes possessing a trans double bond configuration.244 This would suggest these compounds are biogenetic precursors to the cembrane molecules, and may also give rise to the dissimilar stereochemical configurations seen in a number of

Eremophila NPs compared to similar structures from other natural sources.244

100

The diterpenes from Eremophila spp. also include a number of unusual tricyclic compounds, including examples from the cedrane (145),245-247 eremane (146),214,248-254 and decipiane (147 and 148)255-259 compound classes.

101

Just one sterol (149)205 and a small grouping of triterpenes, including 150, have been isolated from Eremophila spp.. Other compound classes represented in

Eremophila plants include monoterpenes (e.g. 151),215,240,241 fatty acids (e.g. 152),260,261 flavonoids (e.g. 153),100,214,251 lignans (e.g. 154),215,228,230,262,263 and iridoids (e.g.

155).169

102

5.4. Modern era biological activity of Eremophila spp. extracts and pure NPs

Despite the varied and unique NPs obtained from the Eremophila genus thus far, many of the compounds isolated have not been evaluated for biological activity. Most of the biologically orientated investigations of Eremophila spp. have been undertaken on extracts of these plants. The following section is therefore divided into two parts, the first summarising biological studies undertaken on Eremophila extracts and the second on bioactive pure NPs from this genus.

5.4.1. Bioactivity studies of Eremophila plant extracts

The majority of the studies summarised here on bioactive extracts of Eremophila specimens based their plant selection on traditional medicinal usage by Australian

Aboriginal people. Therapeutic areas include anti-bacterial, anti-viral, anti- inflammatory, and the treatment of neurological disorders such as migraine.

Ndi et. al. found that a number of the 72 Eremophila spp. they had chosen for analysis showed selective activity against Gram-positive bacteria.184 The minimum inhibitory concentrations (MICs) for the most active species ranged from 16 to 62

μg/mL against Streptococcus spp., and 62 to 250 μg/mL for standard strains of

Staphylococcus aureus. E. virens was found to inhibit the growth of 68 methicillin- resistant S. aureus (MRSA) clinical isolates at the minimum tested concentration of 31

μg/mL.184

103

Thirty-nine plants used by Australian Aboriginal people, including six

Eremophila spp., were examined by Palombo et. al. against four Gram-positive bacterial strains (Bacillus cereus, Enterococcus faecalis, Escherichia coli, S. aureus, and Streptococcus pyogenes).109 Eremophila spp. leaf extracts were found to be the most active, with E. duttonii exhibiting the greatest activity.109 Palombo et. al. went on to investigate whether some of the most active extracts could inhibit clinical isolates of methicillin-resistant S. aureus (MRSA) and vancomycin-resistant enterococci (VRE) using plate-hole diffusion assays.264 E. alternifolia showed activity against MRSA and

E. duttonii was active against both types of bacteria. In a time-kill assay E. duttonii was found to reduce the number of viable cells of MRSA and VRE to undetectable levels within 1 hour.264 Furthermore, the group found that the active constituent of the E. duttonii extract was most likely a terpene using TLC studies.265 They then investigated the mechanism of action of the extract using propidium iodide (PI) uptake and salt tolerance assays.266 It was found that the extract compromised the integrity of the cytoplasmic membrane of S. aureus, as indicated by an increase in the uptake of PI and a decrease in the bacteria's ability to exclude NaCl. The mechanism of antibacterial action of the E. duttonii extract was therefore concluded to be due to its effect on the bacterial membrane.266

In 1951 Kerr discovered that the wood oil of E. mitchellii acted as an effective fly spray when used as an adjuvant with pyrethrins.267 Wilkinson and Cavanagh found that an essential oil sample of E. mitchellii demonstrated antimicrobial activity in standard disc diffusion assays against four bacteria (Escherichia coli, Salmonella typhimurium, S. aureus and Alcaligenes faecalis) and the yeast Candida albicans.268

104

Owen et. al. reported that extracts from E. alternifolia and E. duttonii inhibited the growth of the food-borne pathogen Listeria monocytogenes in time-kill experiments at both 4 °C and 37 °C.269

Forty plant extracts, including six from Eremophila spp., were investigated for antiviral activity by Semple et. al..110 The extracts were tested against one DNA virus, human cytomegalovirus (HCMV), and two RNA viruses, Ross River virus (RRV) and poliovirus type 1, at non-cytotoxic concentrations. The leaves of E. latrobei subsp. glabra inhibited the development of RRV-induced cytopathic effect (CPE) by more than 25%.110

The enzyme xanthine oxidase (XO) can cause gout and oxidative damage to living tissue. Sweeney et. al. investigated 17 native Australian plant species and found that an extract of E. maculata inhibited this enzyme by 61% at 50 μg/mL (IC50 30.9

μg/mL).270

In order to identify potential migraine therapeutics, Rogers et. al. screened the extracts of eighteen plants to see how they affected ADP induced platelet aggregation and [14C]5-hydroxytryptamine (5-HT) release, which have been implicated in the occurrence of migraines.271 They found that E. freelingii and E. longifolia potently inhibited ADP induced human platelet [14C]5-HT release in vitro, with levels ranging from 62 to 95% inhibition.271 In another study Rogers et. al. found that E. gilesii inhibited platelet 5-HT release (56% inhibition).272

Rogers et. al. also found that an extract of E. bignoniiflora caused dose-

+ 2+ 273 dependent inhibition of K -depolarised Ca influx (IC50 234 μg/mL). Compounds capable of blocking voltage-gated Ca2+ channels (VGCCs) can be utilised to treat neurological disorders such as episodic ataxia type 2, spinocerebellar ataxia type 6, hypokalemic periodic paralysis, and familial hemiplegic migraine.273

105

Pennacchio et. al. selected E. alternifolia as part of a study into the cardioactivity of plant extracts.274 They found that injection of the extract into the isolated hearts of normotensive rats induced cardioactivity, consisting of a short initial increase in force of contraction (positive inotropism), followed by a decrease in the force (negative inotropism), with simultaneous increase in heart rate (positive chronotropism), and in coronary perfusion rate.274 Pennacchio et. al. also reported that an extract of E. alternifolia significantly increased heart rate in spontaneously hypertensive rats.275 In this same publication they found that E. maculata dose- dependently inhibited the growth of three Gram-positive bacteria, S. aureus, S. pyogenes and B. cereus.275

Beattie evaluated the methanolic extracts of 20 Eremophila spp. against a panel of cell lines (mammary adenocarcinoma [MCF7], hepatoblastoma [Hep G2], ovarian carcinoma [A2780], malignant melanoma [A-375], mouse lymphoblasts [P388D1], and prostate cancer [PC-3]).276 The majority of extracts exhibited cytotoxicity towards at least one cell line at 0.1 mg/mL. Five species demonstrated cytotoxicity across all the cell lines tested at 0.1 mg/mL (14-100% inhibition).276

106

5.4.2. Bioactivity studies of Eremophila pure NPs

The majority of the work undertaken on bioactive pure compounds from

Eremophila spp. has focused on the antibacterial activity obtained from NPs of the serrulatane structure class. To date, the majority of serrulatane NPs have demonstrated selective antimicrobial activity towards Gram-positive bacteria.108,277-279 The lack of inhibition towards Gram-negative strains may be linked to the bicyclic nature and molecular size of the serrulatanes, as Gram-negative species have been found to effectively exclude compounds via size selective porin channels in their outer membranes.279

3,8-Dihydroxyserrulatic acid (63) and serrulatic acid (64) were isolated from an

E. sturtii leaf extract by bioassay-guided fractionation by Liu et. al..108 These NPs exhibited bactericidal activity against S. aureus with minimum bactericidal concentrations (MBC) of 200 μg/mL and 15 μg/mL, respectively.108 Serrulatic acid (64) also strongly inhibited the inflammation pathway enzymes cyclooxygenase 1 (COX-1)

108 and COX-2 at 1 mg/mL with 99% and 97% inhibition, respectively. IC50 values of 27

μg/mL for COX-1 and 73 μg/mL for COX-2 were obtained.108

107

Leading on from their 2007 report on the antimicrobial activity of 72

Eremophila spp. extracts,184 Ndi et. al. completed bioassay-guided fractionation of both

E. serrulata and E. neglecta. E. serrulata was found to contain naphthoquinone (156) and three serrulatane-type diterpenoids (157-159).230,278 All four compounds demonstrated antimicrobial activity against S. aureus with MICs ranging from 15.6 to

250 μg/mL. Compound 158 was the most active with an MIC of 15.6 μg/mL and a

MBC of 125 μg/mL.278 NP 158 also inhibited S. pyogenes and Streptococcus pneumonia with a MIC of 7.8 μg/mL against both strains.

Bioassay-guided fractionation of a leaf sample of E. neglecta led to three new serrulatane-type diterpenoids (160-162) and the known compound biflorin (163). 277,280

NPs 161-163 showed antimicrobial activity against S. aureus, S. pyogenes, and S. pneumonia with MIC values ranging from 6.5 to 101.6 μM, while 160 was found to be inactive.277 Anakok et. al. explored the antibacterial activity of NPs 161 and 162 against a wider range of Gram-positive (24) and Gram-negative (5) bacterial strains, including clinical and multidrug-resistant isolates.281

108

Serrulatanes 161 and 162 exhibited antibacterial activity towards all the Gram- positive strains with MIC values ranging from 3.1-100 μg/mL. Compound 161 also demonstrated activity against the Gram-negative strain Moraxella catarrhalis (ATCC

49143, MIC 3.1 μg/mL). Serrulatanes 161 and 162 showed moderate cytotoxicity

281 towards Vero mammalian cells (CC50 9.2 μg/mL and 20 μg/mL, respectively).

Two serrulatane diterpenes, serrulat-14-en-7,8,20-triol (164) and serrulat-14-en-

3,7,8,20-tetraol (165), were isolated and tested for antibacterial activity by Smith et. al. from E. duttonii.226,232,279 The MIC values for both compounds were obtained using a micro-titre plate broth dilution assay and ranged from 23–94 μg/mL for 164 and from

375–1500 μg/mL for 165 against the bacteria strains S. aureus, S. epidermidis, and S. pneumonia.279 In a separate study Ndi et. al. found that acetylation of the hydroxyl groups of 164 and 165 to give 166 and 167, respectively, resulted in a loss of antibacterial activity against S. aureus (ATCC 29213).282

109

The serrulatane related seco-pseudopterosins and pseudopterosins, represented by 168 and 169, respectively, were isolated from sea whips of the Pseudopterogorgia genus and have been found to possess potent inflammatory and analgesic activity.283-286

These compounds are diastereoisomers of the serrulatanes, with different configuration at C-1, and occur as glycosides.228,232,283 The serrulatanes from Eremophila generated some interest as synthetic precursors to the highly active seco-pseudopterosins, as they could be obtained in much higher quantities for synthetic and biological studies.232

It was observed that timber fence posts constructed from the wood of E. mitchellii were resistant to termites.185 The toxicity of the steam distilled oil and extracts of E. mitchellii towards termites was later confirmed by Leach et. al..287 Beattie et. al. went on to undertake a comprehensive survey of the distribution, yield, and variation of the chemical composition of oils from different parts of an E. mitchellii plant and describe the major constituents of the wood and root oils.276 The E. mitchellii oils were found to be complex mixtures of mono- and sesqui-terpene NPs, and the leaf, wood, and root oils possessed significant termicidal activity.276 The most active pure NPs isolated were found to be eremophilone (116, LC50 744 ppm) and 8-hydroxyeremophilone (170,

LC50 1983 ppm) in a direct contact assay, compared to the whole wood oil (LC50 811 ppm).276 110

Beattie et. al. also found that the eremophilane NPs 116, 118, 170 and 171 possessed IC50 values ranging from 42-105 μg/mL in a cytotoxicity assay against

204 P388D1 mouse lymphoblast cells. In a separate study Beattie tested luteolin (172) and phillygenin (173), isolated after bioassay guided fractionation of E. racemosa, against

P388D1 cells and found they possessed IC50 values of 160 μM and 430 μM,

276 respectively.

Grice et. al. fractionated the E. gilesii extract that had been found to inhibit platelet 5-HT release (56% inhibition) in the publication by Rogers et. al.272 and isolated verbascoside (110) and the related phenylethanoid glycoside poliumoside (174). They found that verbascoside (1 mg/mL) possessed 94% inhibition of ADP-induced platelet

5-HT release, and 17% inhibition of ADP-induced platelet aggregation.174

111

In their search for cardioactive compounds Pennacchio et. al. identified verbascoside (110) from E. alternifolia and the iridoid glycoside geniposidic acid (155) from E. longifolia as active constituents. Verbascoside had a stimulatory effect on isolated rat heart, it mediated significant increases in chronotropism, inotropism and coronary perfusion rate.175 Geniposidic acid had an opposite, inhibitory effect on the

Langendorff rat heart model used.175 Pennacchio et. al. went on to investigate the cardioactivity of four iridoid glycosides, melampyroside (175) and ferruloylajugol (178) from E. pantonii, verminoside (177) from E. ionantha and catalpol (176) from E. maculata.288 They found that NPs 175-178 all altered myocardial activity in

Langendorff rat hearts. Iridoids 175 and 177 had a stimulatory effect, while 178 was biphasic and 176 inhibitory.288

112

5.5. Conclusion

Historically the Eremophila genus has displayed both a wide range of structural diversity within its compounds and high levels of bioactivity from both its extracts and pure NPs. A number of Eremophila species were used as traditional medicines by

Australian indigenous people. Despite these interesting facts a large number of the >

215 species contained within this genus remain chemically un-investigated. It is evident that the Eremophila genus has high potential to be a source of both new chemistry and of unique potential scaffolds for screening library production. This observation lead to a number of Eremophila species being chosen for chemical investigation in this project.

113

Chapter 6. Chemical investigations of Eremophila linsmithii and Eremophila eriocalyx

6.1. Introduction

As a library of 40 analytical HPLC UV chromatograms of Eremophila spp. extracts had been generated in Chapter 4 during the search for scaffold 62 (Appendix I), it was decided that a number of these plant specimens would be chemically investigated for new NPs and potential scaffolds.

Eremophila linsmithii and Eremophila eriocalyx were chosen from the 40 samples for two main reasons; firstly, searches in both SciFinder289 and DNP65 identified no references for either of these plants, and secondly, the analytical HPLC

UV chromatograms (Figure 27) of these samples contained several minor and major UV peaks which we postulated could correlate to new NPs.

E. linsmithii

E. eriocalyx

UV wavelength key

Figure 27. Analytical HPLC UV chromatograms of E. linsmithii and E. eriocalyx

extracts.

114

6.2. E. eriocalyx

The species E. eriocalyx (Figure 28) can be found in central and southern

Western Australia.106 The name eriocalyx means woolly calyx, referring to the dense hairs on the calyx of the flowers of these plants.106

Figure 28. E. eriocalyx, photo courtesy of C. Jennings290

CH2Cl2 and CH3OH were used to extract a 10 g sample of a whole E. eriocalyx plant (Scheme 19). The extract was then fractionated using a C18-bonded silica flash

1 column (H2O/CH3OH), followed by analysis of the fractions by H NMR spectroscopy.

115

E. eriocalyx (10 g)

CHCl /CH OH 2 3

CH2Cl2/CH3OH (2.51 g)

C18-bonded silica flash column, H O/CH OH 2 3

Fraction 3 Fraction 4 Fraction 5 a a a a a a a a

verbascoside Fraction 32 Fraction 29 Fraction 45 (110, 36.3 mg, 0.363% dry wt) a Fractions 49-51

geniposidic acid b b (155, 37.2 mg, 0.372% dry wt ) and geniposidic acid-Na Fraction 46 c viscidane (180, 15.9 mg, 0.159% dry wt ) (186, 0.6 mg, 0.006% dry wt) c

ladroside mussaenoside (182, 4.1 mg, 0.041% dry wt) (181, 7.6 mg, 0.076% dry wt)

a C18-bonded silica HPLC, CH3OH/H2O cembrane b diol-bonded silica HPLC, i-PrOH/n-hexane (188, 0.6 mg, 0.006% dry wt) c Sephadex LH-20, CH3OH/ CH2Cl2

Scheme 19. Extraction and isolation procedure for E. eriocalyx

The 1H NMR spectroscopic data showed that fractions 1-2 predominantly contained highly polar sugar moieties. The two major components of these fractions were identified as mannitol (179) and geniposidic acid (155)291,292 by (+)-LRESIMS and

1H NMR spectroscopy. Mannitol has been identified in a number of Eremophila species and is considered to be a common constituent of these plants.169

116

Fractions 3-5 were purified using C18-bonded silica HPLC (CH3OH/H2O) and the fractions obtained analysed by (+)-LRESIMS and 1H NMR spectroscopy. These data suggested that the same NP was eluting in both fractions 5-7 and 19-21 of the

HPLC runs, which was initially confusing. This compound was identified as the iridoid glycoside geniposidic acid (155) by MS and 1D/2D NMR spectroscopic (Table 17) data analysis and comparison with literature values.291,292

Initially it was suspected that diastereoisomers of 155 had been isolated, which could potentially account for the differing HPLC retention times. However, no discernible differences could be seen in the ROESY spectra of the compounds, and the

27 [α]D values obtained for each sample were similar (fractions 5-7 [α]D + 10 [c 0.03,

25 CH3OH] and fractions 19-21 [α]D + 8 [c 0.04, CH3OH]).

It was found that geniposidic acid has previously been isolated as its Na carboxylate salt (180).293,294 On analysis of fractions 5-7 and 19-21 by LC/MS, it was determined that the compounds eluted at the same time with the same mass ion (m/z 373

[M-H]-), indicating that the formic acid that was an additive in the LC/MS solvents was protonating the carboxylate 180 to form 155. Additionally, Konig et. al. had witnessed a broadening of H-3 in the 1H NMR spectrum of 180, thought to be due to the lack of neighbouring protons for dipolar relaxation.293 This same effect was seen in this study for the compound eluting in fractions 5-7, indicating that it was the carboxylate form

(180).

117

Konig et. al. determined that the counterion of 180 was Na by atomic absorption spectroscopy.293 Although atomic absorption spectroscopy experiments were not performed during this project, it was assumed that Na was the most probable counter ion in our case. It was therefore concluded that fractions 5-7 contained geniposidic acid-

Na (180)293 and fractions 19-21 geniposidic acid (155).291,292

Table 17. NMR data for geniposidic acid (155)a b c Position δC δH COSY HMBC ROESY 1 95.7, CH 5.07 (d, 8.4) 9 1', 3, 5, 8, 9 1', 5w, 6b, 9, 10 3 150.9, CH 7.38 (s) 1, 4, 5, 11 4 111.8, C 5 34.7, CH 3.02 (ddd, 9.6, 9.0, 8.4) 6a, 6b, 9 1, 3, 4, 6, 9, 11 1w, 6a, 6bw, 9 6a 38.1, CH2 2.68 (dd, 19.2, 9.6) 5, 6b, 7 5, 7, 8, 9 5, 6b, 7 6b 2.03 (dd, 19.2, 8.4) 5, 6a, 7 4, 5, 7, 8 1, 5w, 6a, 7 7 125.6, CH 5.67 (brs) 6a, 6b, 10 5, 6, 8, 9, 10 6a, 6b, 10 8 144.2, C 9 45.9, CH 2.59 (dd, 9.0, 8.4) 1, 5 1, 5, 6, 7, 8 1, 5, 10 10 59.4, CH2 4.13 (d, 17.4) 7 7, 8 1, 7, 9 3.97 (d, 17.4) 11 168.2, C 1' 98.6, CH 4.53 (d, 9.6) 2' 1, 3' 1, 3', 5' 2' 73.3, CH 2.98 (dd, 10.2, 9.6) 1', 3' 1', 3' 3' 76.7, CH 3.16 (dd, 10.2, 10.2) 2' 2', 4' 1' 4' 70.0, CH 3.04 (dd, 12.0, 10.2) 5', 6' 6' 5' 77.2, CH 3.11 (brddd, 12.0, 7.2, 6' 4', 6'w 1', 6' 1.8) 6' 61.0, CH2 3.65 (brd, 13.8) 5' 4', 5' 4', 5' 3.41 (dd, 13.8, 7.2) a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

118

Geniposidic acid (155) has been studied for a wide range of biological activities, including anti-obesity,295 purgative,296 anti-inflammatory,297,298 antioxidant,299 anti- tumor,300,301 antimutagenic,302 and cardioactivity.175 Geniposidic acid was also found to inhibit plant cell growth.303 This NP has been found in at least three Eremophila spp..169,172,175

The caffeic acid ester glycoside verbascoside (110) was also found to be a major metabolite of the E. eriocalyx sample. The 1D/2D NMR (Table 18), MS, and [α]D values obtained for 110 correlated well with those reported in the literature.171,304

Verbascoside has been extensively studied (> 1000 references in SciFinder)289 and been found to possess biological activity in a range of screens including antioxidant,305,306 antifungal,305 antiinflammatory,306-309 antibacterial,310 anti- diabetic,311,312 antinociceptive,306,313 anticancer,314,315 antispasmodic,316 antihyperalgesic,317 analgesic,318 and cardioactivity.175

Dell et. al. demonstrated that lowering the amount of nutrients such as nitrogen, potassium or magnesium in the growth medium of Eremophila plant tissue cultures resulted in the increased production of verbascoside, indicating it was formed in response to nutrient stress.176

119

Table 18. NMR data for verbascoside (110)a b c Position δC δH COSY HMBC ROESY 1 125.6, C 2 114.7, CH 7.02 (s) 4, 6, 7 6''', 7, 8 3 145.8, C 4 148.5, C 5 115.9, CH 6.75 (d, 9.6) 6 1, 3, 4w 6 6 121.6, CH 6.96(dd, 9.6, 1.8) 5, 7w 2, 4, 5, 7 5, 6'''w, 7, 8 7 145.7, CH 7.45 (d, 19.2) 8, 6w 2, 6, 8, 9 2, 5'''w, 6'''w, 6 8 113.7, CH 6.19 (d, 19.2) 7 1, 9 2, 5''', 6''', 6 9 165.9, C 1' 129.3, C 2' 115.6, CH 6.62 (brs) 7'w 6', 7' 7', 8' 3' 145.1, C 4' 143.6, C 5' 116.4, CH 6.63 (d, 9.6) 6' 1', 3' 6' 119.7, CH 6.49 (dd, 9.6, 2.4) 5', 7'w 4', 5', 7' 7', 8' w w w 7' 35.1, CH2 2.69 (m) 2' , 6' , 8' 1', 2', 6', 8' 1'' , 2', 6', 8' 8' 70.4, CH2 3.59 (m) 7' 1', 1'', 7' 1'', 2', 6', 7' 3.87 (dt, 11.4, 8.4) 1'' 102.4, CH 4.34 (d, 9.6) 2'' 8' 2''w, 3'', 5'', 7'w, 8' 2'' 74.6, CH 3.21 (dd, 10.2, 9.6) 1'', 3'' 1'', 3'' 1''w, 4'' 3'' 79.3, CH 3.70 (dd, 11.4, 10.2) 2'', 4'' 2'', 4'' 1'' 4'' 69.3, CH 4.71 (dd, 11.4, 11.4) 3'', 5'' 3'', 5'', 6'', 9 2'', 6''w 5'' 74.5, CH 3.46 (m) 4'', 6'' 4'', 6''w 1'' w 6'' 60.9, CH2 3.40 (m) 5'' 5'' 4'' 3.31 (m) 1''' 101.3, CH 5.02 (brs) 2''' 3'', 2''' 2''' 2''' 70.6, CH 3.69 (m) 1''', 3''' 1''' 1''', 3''' 3''' 70.5, CH 3.29 (dd, 11.4, 3.6) 2''', 4''' 4''' 2''' 4''' 71.8, CH 3.10 (dd, 11.4, 11.4) 3''', 5''' 3''', 5''', 6''' 6''' 5''' 68.9, CH 3.34 (m) 4''', 6''' 4''' 6''' 6''' 18.3, CH3 0.95 (d, 7.2) 5''' 4''', 5''' 2, 4''', 5''', 6, 7, 8 a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

120

Fractions 29 and 46 from the C18-bonded silica HPLC purifications of the flash column fractions 3-5 were found to possess similar 1H NMR chemical shifts that were indicative of the iridoid class of NPs.319 These fractions were further purified by

Sephadex LH-20 size exclusion chromatography, resulting in two pure compounds that were identified as mussaenoside (181, Table 19)319-322 and ladroside (182, Table

20)322,323 by MS and 1D/2D NMR spectroscopic analysis and comparison with literature values.

Two studies have been undertaken describing the assignment of the absolute configuration of mussaenoside. Firstly, Takeda et. al. accomplished this by generating mussaenoside tetraacetate (183) then converting it to 10-deoxygeniposide tetraacetate

(184), which is of known absolute configuration (Scheme 20).320 In a subsequent publication Afifi-Yazar et. al. formed mussaenoside tetraacetate (183) from both mussaenoside (181) and 10-deoxygeniposide tetraacetate (184) via the epoxide 185, and compared the products generated (Scheme 20).322 The optical rotation and ROESY data obtained for mussaenoside and ladroside in this study were in agreement with the relative configuration assignment of these compounds in the literature.320,322 No reported biological activity was found for ladroside or mussaenoside.

121

Scheme 20. Synthetic methodology used to establish the absolute configuration of 181.320,322

Table 19. NMR data for mussaenoside (181)a b c Position δC δH COSY HMBC ROESY 1 93.5, CH 5.32 (d, 4.2) 9 1', 3, 5, 8 1', 10 2 3 150.2, CH 7.34 (s) 1, 4, 5, 11 4 111.4, C 5 30.3, CH 3.01 (dd, 7.2, 4.2) 6a, 6b, 9 1, 3, 4, 6, 7w, 8, 9 6a, 9 6a 29.3, CH2 2.13 (m) 5, 6b, 7a, 7b 4, 5, 7, 8, 9 5, 6b 6b 1.30 (dddd, 13.2, 5, 6a, 7a, 7b 4, 5, 7, 8, 9 6a 7.2, 7.2, 6.6) 7 39.2, CH2 1.59 (m) 6a, 6b, 7b 5, 8, 9, 10 1.54 (m) 6a, 6b, 7a 5, 8, 9, 10 8 78.1, C 9 50.5, CH 2.04 (dd, 9.0, 4.2) 1, 5 1, 4, 6, 7, 8, 10 5 10 24.3, CH3 1.18 (s) 7, 8, 9 1 11 166.7, C 12 50.8, CH3 3.61 (s) 11 1' 98.0, CH 4.48 (d, 7.8) 2' 1, 3' 1, 5' 2' 73.1, CH 2.95 (dd, 8.4, 7.8) 1', 3' 1', 3', 4' 3' 77.3, CH 3.14 (dd, 9.0, 8.4) 2', 4' 4', 5' 4' 70.0, CH 3.02 (dd, 9.0, 9.0) 3', 5' 5', 6' 5' 76.7, CH 3.12 (dd, 9.0, 6.6) 4', 6' 1', 3', 6' 6' 6' 61.1, CH2 3.68 (dd, 12.0, 1.2) 5' 4' 5' 3.42 (dd, 12.0, 6.6) 4', 5' a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). weak correlations.

122

Table 20. NMR data for ladroside (182)a b c Position δC δH COSY HMBC ROESY 1 94.1, CH 5.13 (d, 7.2) 9 1', 3, 5 1', 7, 10 2 3 150.3, CH 7.34 (s) 1, 4, 5, 11 4 111.3, C 5 31.1, CH 3.01 (m) 6a, 6b, 9 1, 3, 4, 6, 9, 11w 6a, 9 w w 6a 29.5, CH2 2.08 (dddd, 18.0, 16.2, 5, 6b, 7 7 , 9 5, 6b, 7 8.4, 3.0) 5, 6a, 7 4, 7 6a 6b 1.20 (brddd, 16.2, 15.6, 10.2) 7 39.0, CH2 1.52 (m) 6a, 6b 5, 8 1, 6a 8 78.5, C 8-OH 9 50.4, CH 1.96 (dd, 10.2, 7.2) 1, 5 1, 4w, 6, 7, 8 5 10 24.5, CH3 1.15 (s) 7, 8, 9 1 11 166.7, C 12 50.9, CH3 3.60 (s) 11 1' 98.2, CH 4.56 (d, 9.6) 2' 1 1, 3', 5' 2' 73.1, CH 3.02 (m) 1', 3' 1', 3' 3' 76.4, CH 3.19 (m) 2' 2', 4' 1', 5' 4' 70.1, CH 3.16 (m) 5' 3', 5' 5' 73.9, CH 3.43 (m) 4', 6' 3' 1', 3', 6' 6' 63.0, CH2 4.40 (dd, 13.8, 7.2) 5' 1'', 5' 5' 4.20 (dd, 13.8, 1.8) 1'' 166.3, C 2'' 113.7, CH 6.23 (d, 19.2) 3'' 1'', 4'' 5'', 9'' 3'' 145.2, CH 7.44 (d, 19.2) 2'' 1'', 5'', 9'' 5'', 9'' 4'' 125.4, C 5'' 114.7, CH 7.01 (s) 3'', 7'', 9'' 2'', 3'' 6'' 145.5, C 7'' 148.4, C 8'' 115.7, CH 6.74 (d, 10.2) 9'' 4'', 6'', 7'' 9'' 9'' 121.3,CH 6.95 (d, 10.2) 8'' 3'', 5'', 7'' 2'', 3'', 8'' a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). weak correlations.

123

5,19-Dihydroxy-3,14-viscidadien-20-oic acid (186) was isolated after C18- bonded silica semi-preparative HPLC followed by diol-bonded silica semi-preparative

HPLC. NP 186 was reported in the same publication in 1986 to have been isolated from

E. exotrachys, E. platythamnos, E. crenulata and E. sp. aff. gibsonii, and was characterised as its methyl ester.235 Given the small amount of 186 isolated in this project (0.6 mg), it was difficult to obtain complete NMR data. However, the 1D/2D

NMR data that was acquired (Table 21) supported the structural assignment of 186.

Ghisalberti et. al. found that NP 186 converted to 187 after prolonged storage.235

They also generated 187 by heating 186 under reduced pressure, and suggested that the formation of the ether proceeds by acid-catalysed protonation of the allylic hydroxyl group followed by substitution by the primary hydroxyl group.235 In this study, when

186 was analysed by LC/MS some months after its isolation, the (–)-MS spectrum indicated the presence of 186 (m/z 335 [M-H]-), while the (+)-MS spectrum indicated the sample contained 187 (m/z 319 [M+H]+). The 1H NMR spectrum still appeared to contain one compound, however, the signals were broadened, indicating that the two compounds had overlapping signals. A paucity of material (0.6 mg) prevented us from obtaining a well resolved 13C NMR spectrum. However, it would seem that 186 is slowly converting to 187. It is assumed that the relative configuration of 186 and 187 are the same as that reported in the literature.

124

Table 21. NMR data for 5,19-dihydroxy-3,14-viscidadien-20-oic acid (186)a b c Position δC δH COSY HMBC 1 46.3, C w 2a 30.4, CH2 1.97 (brdd, 18.0, 3.0) 2b, 3 1 , 3, 4, 7 2b 1.72 (dd, 18.0, 4.2) 2a, 3 1, 3, 4, 6, 7 3 128.7, CH 6.29 (brs) 2a, 2b 1, 2, 5, 4w, 20

4 139.0, C w w 5 65.6, CH 4.29 (brdd, 7.8, 6.6) 6 3 , 4 w 6 38.4, CH2 1.57 (m) 5 1, 2, 4, 5, 7, 10 7 47.8, CH 1.66 (m) 19 8 23.8, CH2 1.56 (m) 9a, 9b 9a 26.0, CH2 1.79 (m) 8, 9b 1, 10 9b 1.30 (m) 8, 9a, 10 10 51.1, CH 1.51 (ddd, 9.0, 8.4, 8.4) 9b 1, 2, 6, 8, 12, 18 11 33.2, CH 1.38 (m) 18 12 35.5, CH2 1.37 (m) 12b, 13a, 13b 1.01 (m) 12a, 13aw, 13b 11, 13, 14, 18 w 13a 24.8, CH2 1.98 (m) 12a, 12b , 13b, 14 13b 1.83 (m) 12a, 12b, 13a, 14 12, 14, 15 14 124.8, CH 5.07 (dd, 7.2, 7.2) 13a, 13b, 16, 17 16, 17 15 130.3, C 16 25.4, CH3 1.55 (s) 14 14, 15, 17 17 17.4, CH3 1.63 (s) 14 14, 15, 16 18 19.8, CH3 0.88 (d, 6.6) 11 10, 11, 12 w 19 61.2, CH2 3.53 (brd, 9.6) 7, 19-OH 8 3.09 (brddd, 10.2, 9.6, 4.8) 19-OH 4.21 (brdd, 4.8, 4.8) 19 20 170.7, C a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

3,15-Epoxy-19-hydroxycembra-7,11-dien-18-oic acid (188) was isolated as a minor component (0.6 mg) of the E. eriocalyx plant (Table 22). While 188 has been

240 synthesised by the conversion of the NP 189 using NaBH4 (Scheme 21), it has not previously been reported as a NP.

125

Scheme 21. Conversion of NP 189 to 188 by Ghisalberti et. al.240

Ghisalberti et. al. originally isolated NP 189 from E. granitica in 1983.240 The authors established the relative and absolute configuration of 189 through a combination of synthetic methods and comparison to the related structure 190, which had been

240 studied by X-ray crystallography. The []D value obtained for 188 [+ 8 (c 0.04,

CHCl3)] was in agreement with that obtained in the original report [+ 1.1 (c 2.0,

CHCl3)]. It is therefore assumed that the relative and absolute configurations of 188 are the same as that reported for the related cembranes 189 and 190.240 Given the small amount of 188 isolated in this project (0.6 mg), it was not possible to conclusively establish the relative or absolute configuration of the compound. The scarcity of material also prevented us from obtaining a well resolved 13C NMR spectrum, therefore the carbon chemical shifts for 188 (Table 22) have been assigned using the HSQC and

HMBC spectra.

126

Table 22. NMR data for 3,15-epoxy-19-hydroxycembra-7,11-dien-18-oic acid (188)a b c Position δC δH COSY HMBC ROESY 1 47.2, CH 1.58 (m) 2a, 2b, 14b 13, 15, 16, 17 2a, 3w, 16, 17 w 2a 33.3, CH2 2.21 (m) 1, 2b, 3 1, 3, 4, 15 1, 2b, 3 , 14a 2b 1.46 (brddd, 1, 2a, 3 1w, 3w, 4w, 14w 2a, 3, 14a, 14b, 17 11.4, 10.8, 10.8) 3 76.4, CH 4.34 (ddd, 9.0, 2a, 2b, 4 1w, 2aw, 2b, 4, 17 3.6, 3.0) 4 51.1, CH 2.13 (m) 3, 5a, 5b 3, 18 3, 5a 5a 30.0, CH2 1.65 (m) 4, 5b, 6a 4 4, 5b, 7 5b 1.29 (m) 4, 5a, 6a 4 5a, 6a, 7 w 6a 26.5, CH2 2.35 (dddd, 15.6, 5a, 5b, 6b, 7 5 , 7, 8 5b, 6b, 10a, 16 11.4, 10.2, 3.0) 6a, 7 5, 7, 8 6a 6b 1.62 (m) 7 126.3, CH 5.33 (dd, 10.2, 6a, 6b, 19w 5w, 6, 19 5a, 5b, 19 4.8) 8 138.1, C 9a 29.4, CH2 2.07 (m) 9b 7, 8, 19 9b, 19 9b 1.91 (m) 9a 7, 8, 10, 19 9a, 19 w w 10a 28.6, CH2 2.03 (m) 10b, 11 11 , 12 6a, 10b,11 10b 1.90 (m) 10a, 11 9, 12 10a, 11 11 125.9, CH 5.12 (brd, 7.8) 10a, 10b, 20 20w 10a, 10b, 20 12 133.5, C 13a 29.8, CH2 2.15 (d, 12.0) 13b, 14a, 14b 1, 11, 12, 14, 20 13b, 14b 13b 1.80 (dd, 12.0, 13a, 14a, 14b 1, 11, 12, 14, 20 13a, 14a, 14b 6.6)

w 14a 27.4, CH2 1.26 (m) 1, 13a, 13b 1, 2, 13, 12, 15 2a, 2b, 13b, 16, 17 14b 1.23 (m) 1, 13a, 13b 1w, 2, 12, 13 2b, 13a, 13b, 16, 17 15 79.4, C 16 26.2, CH3 1.06 (s) 17 1, 15, 17 1, 6a, 14a, 14b, 17 17 19.4, CH3 0.87 (s) 16 1, 15, 16 1, 2b, 3, 14a, 14b, 16 18 176.7, C w 19 64.6, CH2 3.85 (d, 13.2) 7 7, 8, 9 7, 9a, 9b 3.79 (d, 13.2) 20 23.4, CH3 1.66 (s) 11 11, 12, 13 11 a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

127

The MS and 1H NMR spectroscopic data acquired for a number of other fractions from the C18-bonded silica HPLC purifications indicated the presence of further serrulatane, viscidane, or cembrane molecules, however, the amounts obtained of these still impure compounds (< 0.5 mg) prevented further isolation and structure elucidation studies.

6.3. E. linsmithii

E. linsmithii, named after the botanist Lindsay Smith, grows as an erect shrub with white to pale lilac flowers (Figure 29).106 This species is only found in south- western Queensland.106

Figure 29. E. linsmithii, photo © M. Fagg, Australian National Botanic Gardens.111

128

The air-dried and ground plant E. linsmithii (10 g) was exhaustively extracted with sequential washes of CH2Cl2 and CH3OH (Scheme 22). The crude extracts were combined, loaded onto a C18-bonded silica flash column, and then fractionated using a

1 H2O/CH3OH gradient. Each fraction was evaluated by H NMR spectroscopy to ascertain which were worth pursuing and what conditions should be used for further purification.

E. linsmithii (10 g)

CH2Cl2/CH3OH

CH2Cl2/CH3OH (3.62 g)

C18 flash column, H2O/CH3OH

Fraction 3 Fraction 4 Fraction 5

a a a a a a a

mannitol Fraction 31 Fraction 28 (179, 41.2 mg, 0.412% dry wt) 3,15-epoxycembra-7,11-dien- verbascoside 18-oic acid (110, 130.6 mg, 1.31% dry wt) (191, 1.5 mg, 0.015% dry wt)

geniposidic acid-Na (180, 2.6 mg, 0.026% dry wt) 3-methylbut -3-enyl O-α-L-rhamnopyranosyl- (1→6)-O-β-D-glucopyranoside (193, 9.6 mg, 0.096% dry wt)

a C18-bonded silica HPLC (CH3OH/H2O)

b Sephadex LH -20 (CH3OH/CH2Cl2)

Scheme 22. Extraction and isolation procedure for E. linsmithii

129

By 1H NMR spectroscopy it was found that fractions 1-2 predominantly contained simple sugar molecules, and were not pursued. The 1H NMR spectra of fractions 3-5 also looked to contain a high proportion of glycosidic compounds, however, they also possessed interesting peaks in the aromatic region (6-8 ppm).

Fractions 3-5 were purified by C18-bonded silica semi-preparative HPLC (CH3OH/H2O) with the major NPs isolated being the known compounds mannitol (179), geniposidic acid-Na (180),291,292 and verbascoside (110).171

C18-bonded silica HPLC of fraction 5 led to the isolation of the previously reported cembrane 3,15-epoxycembra-7,11-dien-18-oic acid (191, 1.5 mg) as a minor metabolite. Ghisalberti et. al. originally isolated 191 from E. granitica in 1983, the same study that first identified the related cembrane 188, which was also isolated in this project from E. eriocalyx.240 The 1D/2D NMR (Table 23) and (+)-LRESIMS data (m/z

343 [M+Na]+) obtained in this study fully supported the assignment of structure 191.

130

Table 23. NMR data for 3,15-epoxycembra-7,11-dien-18-oic acid (191)a b c Position δC δH COSY HMBC ROESY 1 46.5, CH 1.62 (m) 14a, 14b 15, 16, 17 17 2a 33.4, CH2 2.13 (m) 2b, 3 4, 15 2b 2b 1.61 (m) 2a, 3 1, 4 2a, 3 3 75.7, CH 4.12 (dd, 7.2, 5.1) 2a, 2b, 4 2b, 4, 17 4 49.4, CH 2.35 (ddd, 8.4, 7.2, 3, 5a, 5b 2, 3, 5, 6, 18 2b, 3 7.2) w 5a 29.8, CH2 1.64 (m) 4, 5b, 6a, 6b 3, 4, 6, 18 5b 5b 1.31 (ddd, 13.2, 4, 5a, 6a, 6bw 5a 12.0, 6.6) 6a 26.0, CH2 2.14 (m) 5a, 5b, 6b, 7 5, 7, 8 6b, 7 6b 1.79 (m) 5aw, 5bw, 6a, 7 5, 7, 8 6a, 7 7 125.4, CH 5.17 (m) 6a, 6b, 19 9, 19 6a, 6b, 19 8 134.9, C

9a 33.0, CH2 2.15 (m) 9b, 10b 7, 8, 19 9b, 10b 9b 1.80 (m) 9a, 10a, 10b 7, 8, 10, 19 9a, 10a, 11 w w w 10a 27.4, CH2 2.09 (m) 9b, 10b , 11 9 , 11 9b, 10b, 11 10b 1.95 (brddd, 15.0, 9a, 9b, 10aw, 9w 9a, 10a 12.0, 11.4) 11 11 126.1, CH 5.15 (m) 10a, 10b, 20 9, 13w, 20 9b, 10a, 20 12 133.5, C 13a 29.7, CH2 2.18 (ddd, 15.0, 13b, 14a, 14b 1, 11, 12, 14, 20 13b 15.0, 14.2) 13b 1.88 (m) 13a, 14a, 14b 1, 11, 12, 14, 20 13a w w 14a 26.6, CH2 1.42 (brdd, 15.0, 1, 13a, 13b, 13 14b, 16 , 17 15.0) 14b 14b 1.23 (m) 1, 13a, 13b, 12w, 13, 15 14a, 17 14a 15 80.7, C 16 27.0, CH3 1.11 (s) 17 1, 15, 17 14a 17 20.7, CH3 0.90 (s) 16 1, 15, 16 1, 3, 14a, 14b 18 175.6, C 19 23.1, CH3 1.66 (s) 7 7, 8, 9 7 20 23.1, CH3 1.65 (s) 11 11, 12, 13 11 a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

131

As for NP 188, Ghisalberti et. al. established the relative and absolute

configuration of 191 through a combination of synthetic methods and comparison to

the related structure 190, which had been studied by X-ray crystallography.240 The

synthetic studies included the formation of the α,-unsaturated acid 192 from 191

using lithium diisopropylamide in THF, followed by ozonolysis to give (R)-

homoterpenyl methyl ketone, which was characterised as its crystalline

semicarbazone.240 This established that the absolute configuration at C-1 was R.240

27 The optical rotation for 191 ([α]D - 28 [c 0.015, CHCl3]) was found to be of the same sign but of somewhat larger magnitude as the literature value ([α]D - 1.7 [c 2.7,

240 CHCl3]), however, this data suggested that the relative and absolute configuration of

191 was the same as that reported.240 No biological activity has been reported for compound 191.

132

The new hemiterpene glycoside 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-

(1→6)-O-β-L-glucopyranoside (193) was isolated as an optically active dark brown gum. The 1H NMR spectrum of 193 (Figure 30, Table 24) contained signals for two methyl groups (δH 1.12 and 1.69) and a profusion of multiplets between δH 2.93 and

4.57 that suggested the presence of sugar moieties within the molecule.112 Four methylene groups (δH 2.24, 3.80/3.54, 3.80/3.40 and 4.71) were established by analysis of the 1H NMR, COSY, and HSQC spectra.

1 13 Figure 30. H and C NMR spectra for 193 in DMSO-d6

133

The chemical shifts of H-1' and C-1' (δH 4.12, δC 102.8) indicated that they were at the anomeric position of a sugar moiety.112 This sugar fragment was established as a glucose by using the observed COSY correlations (H-1'/H-2', H-2'/H-3', H-3'/H-4', H-

4'/H-5', H-5'/H-6') and was supported by key HMBC signals (H-5'/C-1', H-4'/C-6', H-

1 3'/C-1') (Figure 31). A resonance at δH 4.57 (H-1'') in the H NMR spectrum also possessed a chemical shift typical of an anomeric proton.112 Inspection of the 1D/2D

NMR data allowed for a second sugar fragment, rhamnose, to be elucidated. The COSY data allowed most of the rhamnose ring to be constructed (H-1''/H-2'', H-2''/H-3'', H-

5''/H-6''), however, overlapping proton resonances in the 1H NMR spectrum meant that the observed HMBC correlations were crucial in positioning the hydrogen signals (H-

1''/C-3'', H-1''/C-5'', H-2''/C-4'', H-4''/C-6'', H-6''/C-4''). Reciprocal HMBC signals between H-1''/C-6' and H-6'/C-1'' allowed the glucose and rhamnose rings to be joined through the anomeric oxygen of the rhamnose moiety (C-1'').

Figure 31. Key COSY, HMBC and ROESY correlations for 193

134

The methylene at H-4 (δH 4.71) showed HMBC correlations to a quaternary carbon at δC 142.4 (C-3) and a methyl group at δC 22.5 (C-5). It was proposed that H-4 was positioned at the terminus of an exocyclic double bond, with the methyl group (δH

1.69, C-5) attached at C-3. The methylene at δH 2.24 (H-2) showed HMBC correlations to C-4 (δC 111.5) and C-5 (δC 22.5). H-2 also possessed a strong COSY correlation to

H-1 (δH 3.80/3.54), which in turn showed a HMBC correlation to C-1' (δC 102.8), allowing for the positioning of the hemiterpene side chain at the anomeric oxygen of the glucose ring. The 1D/2D NMR spectroscopic data obtained for the hemiterpene side chain of 193 corresponded well with values obtained for similar structures in the literature.324,325 This confirmed the planar structure of 193, which corresponded to the molecular formula of C17H30O10 obtained by (+)-HRESIMS analysis.

The relative configuration of 193 was assigned by analysis of the ROESY spectrum and 1H-1H coupling constant data (Figure 31). ROESY correlations between

H-1'/H-3', H-3'/H-5', and H-5'/H'1' indicated the cis relationship of these protons. This assignment was supported by the 1H-1H coupling constants seen between the protons

113,114 around the glucose ring (J1',2' = 7.8 Hz, J2',3' = 9.0 Hz, J3',4' = 9.0 Hz, J4',5' = 9.6 Hz).

The assignment of the anomeric hydrogen as having a  configuration was corroborated

13 1 1 by the C NMR chemical shift of C-1' (δC 102.8) and the H- H coupling constant seen

324,326 between H-1'/H-2' (J1',2' = 7.8 Hz).

135

The overlapping nature of the rhamnose protons in the ROESY spectrum prevented it giving any useful information in regards to the relative configuration of this moiety. However, the relative configuration could be established as that shown in

1 1 structure 193 through analysis of the H- H coupling constant data (J1'',2'' = 1.2 Hz, J2'',3''

= 3.6 Hz, J3'',4'' = 9.6 Hz, J4'',5'' = 9.6 Hz). The α configuration of the anomeric proton was supported by the carbon chemical shifts of C-3'' (δC 70.6) and C-5'' (δC 68.3), which are typically around δC 74.1 and δC 73.4, respectively, when the anomeric proton has a  configuration.326 With the relative configuration determined, structure 193 was assigned to 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-D-glucopyranoside.

Table 24. NMR data for 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-D- glucopyranoside (193)a b c Position δC δH COSY HMBC ROESY w 1 67.0, CH2 3.80 (dt, 9.6, 7.2, 7.2) 2 1', 2, 3 1', 2 3.54 (dt, 9.6, 7.2, 7.2) w 2 37.3, CH2 2.24 (dd, 7.2, 7.2) 1, 5 1, 3, 4, 5 1 3 142.4, C w 4 111.5, CH2 4.71 (brd, 0.6) 5 2, 3, 5 w w 5 22.5, CH3 1.69 (s) 2 , 4 2, 3, 4 1' 102.8, CH 4.12 (d, 7.8) 2' 1, 3'w, 5'w 1, 2'w, 3', 5' 2' 73.3, CH 2.93 (dd, 9.0, 7.8) 1', 3' 1', 3', 4'w 1'w 3' 76.7, CH 3.12 (dd, 9.0, 9.0) 2', 4' 1'w, 2', 4', 5' 1', 5' 4' 70.2, CH 2.98 (dd, 9.6, 9.0) 3', 5' 2'w, 3', 5', 6' 5' 75.3, CH 3.25 (ddd, 9.6, 9.6, 1.8) 4', 6' 1', 3', 5', 6' 1', 3', 6' 6' 67.1, CH2 3.80 (m) 5' 1'', 4', 5' 1'', 3' 3.40 (m) 1'' 100.8, CH 4.57 (brd, 1.2) 2'' 3'', 5'', 6' 2'', 6' 2'' 70.5, CH 3.59 (brdd, 3.6, 1.2) 1'', 3'' 3'', 4'' 1'' 3'' 70.6, CH 3.42 (m) 2'', 4'' 1'', 4'' 4'' 71.9, CH 3.18 (dd, 9.6, 9.6) 3'', 5'' 3'', 5'', 6'' 6'' 5'' 68.3, CH 3.43 (m) 4'', 6'' 1'', 4'' 6'' 6'' 17.9, CH3 1.12 (d, 6.0) 5'' 4'', 5'' 4'', 5'' a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

136

A number of glycosides related to 193 have been reported in the literature (194-

197). Investigations of the tropical plant Morinda citrifolia (commonly known as 'noni') have led to the isolation of hemiterpene glycosides such as 194 and 195.325,327,328 Noni fruit juice has been extensively used as an alternative medicine for a wide range of diseases such as diabetes, cancer, arthritis, headaches, heart disease and gastric ulcers.

This has led to a number of studies on the isolation and identification of marker compounds for quality control purposes.325,327-329

Akihisa et. al. found that NPs 194 and 195 moderately inhibited Epstein-Barr virus early antigen activation induced by 12-O-tetradecanoylphorbol-13-acetate in a preliminary evaluation of the potential anti-tumour effects of these compounds.329 It has also been found that a number of hemiterpene glycosides including 194 and 195 inhibit melanogenesis, suggesting that these compounds may be used as skin whitening agents.328 NPs 196 and 197 were identified as potential molecular markers for grape varieties after they were isolated from Vitis vinifera.324,330 Non-volatile glycosidic conjugates such as these compounds are thought to be hydrolysed during the processing and maturation of the wine, releasing volatile flavorants that contribute to the distinctive taste produced by different grape varieties.324,330

137

6.4. Conclusion

The new NPs 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-D- glucopyranoside (193) and 3,15-epoxy-19-hydroxycembra-7,11-dien-18-oic acid (188) as well as the known compounds mannitol (179), geniposidic acid (155), geniposidic acid-Na (180), verbascoside (110), mussaenoside (181), ladroside (182), 5,19- dihydroxy-3,14-viscidadien-20-oic acid (186), and 3,15-epoxycembra-7,11-dien-18-oic acid (191) were isolated from E. linsmithii and E. eriocalyx.

138

Chapter 7. Serrulatanes from Eremophila microtheca

7.1. Introduction

E. microtheca, also known as heath-like Eremophila, grows as an erect shrub with lilac flowers (Figure 32).106 The name microtheca comes from the Greek language, with 'micro' meaning small, and 'theca' case or container, which refers to the small nature of the fruit of these plants.106 This species is relatively rare in the wild, being found in just a few localities in south-western Western Australia.106 However, this species is becoming a common addition in many plant nurseries within Australia due to its attractive flowers and drought tolerance. This species also possesses a distinctive aroma that some find offensive. Chinnock reports that the odour of the foliage can be discerned from some distance, and that one large population of E. microtheca was discovered after the searchers followed the smell to the source.106

Figure 32. E. microtheca, photos © M. Fagg, Australian National Botanic Gardens.111

139

On undertaking a search in both DNP65 and SciFinder289 for E. microtheca, only one reference could be found. Ndi et. al. included an extract of this plant in a study of the antimicrobial activity of 72 Eremophila plants.184 The E. microtheca extract exhibited MICs ranging from 31 to 125 μg/mL, and MBCs of 62 to 500 μg/mL, against the Gram-positive bacteria strains Staphylococcus aureus, Streptococcus pyogenes, and

Streptococcus pneumonia.184 No reports were found of NPs that had been isolated from

E. microtheca, indicating that this plant had high potential of yielding new and/or bioactive chemistry. The analytical HPLC chromatogram for E. microtheca can be seen in Figure 33.

Figure 33. Analytical HPLC UV chromatogram for E. microtheca extract

7.2. Extraction and purification of NPs from E. microtheca

The outer branches and leaves (10.4 g) of E. microtheca were air dried and ground before being extracted exhaustively with sequential washes of CH2Cl2 and

CH3OH (3.26 g, Scheme 23). This crude extract was purified by C18-bonded silica semi- preparative HPLC (H2O/CH3OH), and the fractions obtained analysed by (+)-

LRESIMS. The HPLC chromatogram and (+)-LRESIMS results were used to determine which fractions would be analysed by 1H NMR spectroscopy.

140

E. microtheca (10.4 g)

CH2Cl2/CH3OH

CH2Cl2/CH3OH (3.26 g)

C -bonded silica HPLC, CH OH/H O 18 3 2

3-acetoxy-7,8-dihydroxyserrulat- 3,19-diacetoxy-8- 14-en-19-oic acid (198) hydroxyserrulat-14-ene (200) (66.4 mg, 0.638% dry wt) (54 mg, 0.519 % dry wt)

3,7,8-trihydroxyserrulat- Fractions 41-42 verbascoside (110) 14-en-19-oic acid (199) (43.5 mg, 0.418% dry wt) LH-20 (452.9 mg, 4.355% dry wt) Sephadex CH3OH/CH2Cl2

jaceosidin (201) (9.2 mg, 0.088% dry wt)

Scheme 23. Extraction and isolation procedure for E. microtheca

Following 1H NMR spectroscopic data analysis, four pure NPs and one impure compound were identified. Further analysis by HRESIMS and 1D/2D NMR spectroscopy identified three of these compounds as the new serrulatanes, 3-acetoxy-

7,8-dihydroxyserrulat-14-en-19-oic acid (198, 66.4 mg, 0.638% dry wt), 3,7,8- trihydroxyserrulat-14-en-19-oic acid (199, 452.9 mg, 4.355% dry wt), and 3,19- diacetoxy-8-hydroxyserrulat-14-ene (200, 54 mg, 0.519 % dry wt). The previously reported compound verbascoside (110, 43.5 mg, 0.418% dry wt)171 was also identified, and further purification of the impure NP by size exclusion chromatography (Sephadex

LH-20) afforded the known flavonoid jaceosidin (201, 9.2 mg, 0.088% dry wt).331,332

141

3-Acetoxy-7,8-dihydroxyserrulat-14-en-19-oic acid (198) was isolated as an optically active brown gum and assigned a molecular formula of C22H30O6 (eight degrees of unsaturation) on the basis of NMR (Table 25) and HRESIMS data. The 1H

NMR spectrum of 198 (Figure 34) indicated the presence of five methyl groups (δH

0.47, 1.20, 1.56, 1.65 and 2.00), one –OCHR– moiety (δH 5.19), and two olefinic protons (δH 5.10 and 6.96) and also contained several upfield signals that integrated for eight protons. The 13C NMR (Figure 34) and edited HSQC spectra of 198 indicated a total of 22 carbons (Table 25), including five methyls (δC 17.5, 18.8, 21.0, 21.5, 25.5),

2 two carbonyls (δC 169.8 and 172.3), three methylenes (δC 25.4, 31.4, 37.4), six sp

2 quaternary carbons (δC 110.3, 126.6, 130.8, 134.7, 142.4, 147.7), two sp methines (δC

3 119.8 and 124.3), and four sp methines (δC 28.0, 31.2, 44.5, 69.6).

142

1 13 Figure 34. H and C NMR spectra for 198 in DMSO-d6

Figure 35. Key COSY, HMBC and ROESY correlations for 198 143

The extended spin system –(R)CH(CH3)CH2CH(OR)CH(R)CH(CH3)CH2CH2

CH=C(CH3)2 (Figure 35) was readily established following interpretation of the COSY data for 198 (Figure 36) and was supported by key HMBC correlations (Figure 37). For example, the methyl group at H-20 (δH 1.20) showed HMBC correlations to C-1 (δC

28.0), C-2 (δC 31.4), and C-9 (δC 134.7). HMBC correlations from δH 1.95 (H-13) to C-

14 (δC 124.3) and C-15 (δC 130.8) and from δH 5.10 (H-14) to C-16 (δC 25.5) and C-17

(δC 17.5) confirmed that the side chain terminated with a di-methylated olefin moiety.

Substructure searching of this spin system in conjunction with the taxonomic genus in

DNP65 indicated that 198 possessed a serrulatane skeleton.

Figure 36. COSY spectrum for 198 in DMSO-d6

144

Figure 37. HMBC spectrum for 198 in DMSO-d6

The remainder of structure 198 was assembled as follows. An acetoxy group was attached to C-3 on the basis of strong HMBC correlations from both the downfield proton at δH 5.19 (H-3) and the methyl group at δH 2.00 (H-22) to the carbonyl at δC

277 169.8 (C-21). H-4 showed HMBC correlations to carbons resonating at δC 119.8 (C-

5), δC 134.7 (C-9), and δC 126.6 (C-10), indicating that it was adjacent to an aromatic system. H-1 (δH 3.26) also showed HMBC correlations to C-9 and C-10, allowing a cyclohexene (ring A, Figure 35) to be formed. The proton at δH 6.96 was attached to a

2 carbon at δC 119.8 on the basis of HSQC data, and HMBC correlations from this sp methine to C-4 and C-9 of ring A allowed it to be assigned to H-5.

At this point a subunit of C4O4H3 remained to be elucidated. The four remaining carbons included one carbonyl (δC 172.3) and three aromatic quaternary (δC 110.3,

147.7 and 142.4) signals.

145

The proton at H-5 (δH 6.96) possessed a HMBC correlation to the carbon at δC

147.7, placing it at C-7, while H-1 (δH 3.26) demonstrated a strong three-bond HMBC correlation to δC 142.4, positioning it at C-8. This allowed the formation of the benzenoid system, ring B (Figure 35). The proton at H-5 also showed a HMBC correlation to the carbon at δC 172.3, hence a carbonyl group was placed at C-6 (δC

110.3). This left the equivalent of three -OH moieties to position within the structure, suggesting a carboxylic acid and two hydroxyl groups at C-6, C-7 and C-8, respectively.

While neither the phenolic or carboxylic acid protons were identified in the 1H NMR spectrum of 198, the NMR data supported the assigned substitution pattern of ring B after comparison with literature data.230 Furthermore, the IR spectrum of 198 showed strong absorptions at 3262, 1667, and 1731 cm-1, which confirmed the presence of phenolic, aromatic carboxylic acid and ester moieties, respectively.112 Thus the planar structure of 198 was assigned.

Of note is the shielding effect seen for the secondary methyl group at H-18. It has been reported that when a secondary hydroxyl group is located at C-3, the signal for

H-18 is shifted significantly upfield (δH ~ 0.60 in CDCl3, δH ~ 0.30 in DMSO-

108,228,232 d6) compared to when the hydroxyl group is not present (δH ~ 1.00 in CDCl3

108,228,232,277,278 and DMSO-d6). The acetyl group at C-3 in 198 mimics this effect, with the chemical shift of H-18 (δH 0.47) in agreement with the observed shielding effect.

The relative configuration of 198 was established through analysis of the

ROESY and 1H-1H coupling constant data (Table 25, Figure 35) and comparison to literature values. ROESY correlations between H-4/H-3, H-4/H-2β, H-3/H-2β, and H-

3/H-20 placed these protons on the same face of ring A. The small 1H-1H coupling constants between H-4/H-3 (J3,4 = 3.6 Hz) and H-3/H-2β (J3,2β = 4.2 Hz) supported the cis orientation of these protons.

146

Table 25. NMR data for 3-acetoxy-7,8-dihydroxyserrulat-14-en-19-oic acid (198)a b c Position δC δH COSY HMBC ROESY 1 28.0, CH 3.26 (m) 2α, 2β, 20 2, 3, 8, 9, 10, 20 2α, 2β, 20 2α 31.4, CH2 2.08 (ddd, 12.6, 1, 2β, 3 1, 3, 4, 20 1, 2β, 18 12.0, 6.6) 2β 1.58 (m) 1, 2α, 3 1, 3, 4, 20 1, 2α, 3, 4w, 20 3 69.6, CH 5.19 (ddd, 12.0, 2α, 2β, 4 1, 4, 11, 21 2α, 4, 12, 18, 20 4.2, 3.6) 4 44.5, CH 2.95 (brd, 3.6) 3, 5w, 11 2, 3, 5, 9, 10, 12, 18 2αw, 3, 5, 11, 12, 18 5 119.8, CH 6.96 (s) 4w 4, 7, 9, 19 4, 11, 12, 18 6 110.3, C 7 147.7, C 8 142.4, C 9 134.7, C 10 126.6, C 11 31.2, CH 1.96 (m) 4, 12, 18 3, 4, 10, 12, 13, 18 4, 5, 12, 18 12 37.4, CH2 1.36 (ddt, 13.8, 13 4, 11, 13, 14, 18 4, 5, 11, 14, 18 7.2, 6.6) 1.19 (m) 11, 13 4, 13, 14, 18 5, 11, 18 13 25.4, CH2 1.95 (m) 12, 14 11, 12, 14, 15 14, 17, 18 14 124.3, CH 5.10 (t, 6.6) 13, 16, 17 12, 16, 17 12, 13, 16 15 130.8, C 16 25.5, CH3 1.65 (s) 14 14, 15, 17 14 17 17.5, CH3 1.56 (s) 14 14, 15, 16 13 18 18.8, CH3 0.47 (d, 6.6) 11 4, 11, 12 2β, 3, 4, 5, 11, 12, 13 19 172.3, C 20 21.5, CH3 1.20 (d, 6.6) 1 1, 2, 9 1, 2β, 3 21 169.8, C 22 21.0, CH3 2.00 (s) 21 a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

The relative configuration of the cyclohexene system present in 198 is consistent with that reported for the majority of serrulatanes described in the literature. Only one serrulatane, compound 202, isolated from E. phyllopoda, has been found to have the opposite configuration at C-1.228

147

The relative configuration of the stereocentre at C-11 cannot be ascertained using NMR data alone. The relative configuration of the related natural products 203 and 204 was established by X-ray crystallographic analysis,224,226 which allowed for the assignment of the stereocentre at C-11. The majority of previous studies on serrulatanes have assigned the C-11 relative configuration to be identical to that of 203 and 204, on the basis of similar NMR spectroscopic data and biosynthetic grounds.108,277,278

Comparison of the NMR data of 198 with related serrulatanes showed a high degree of similarity for 1H and 13C chemical shifts about C-11, thus the relative configuration was determined to be the same as that of previously reported metabolites.108,228,232

Consequently the chemical structure of 198 was assigned as 3-acetoxy-7,8- dihydroxyserrulat-14-en-19-oic acid. Attempts to crystallise the new serrulatanes (198-

200) in order to undertake X-ray crystallography studies to confirm the relative configuration of these NPs are discussed below.

148

3,7,8-Trihydroxyserrulat-14-en-19-oic acid (199) was determined to have a molecular formula of C20H28O5 based on the sodiated pseudomolecular ion at m/z

371.1844 in the (+)-HRESIMS (calcd. 371.1829), which equated to seven degrees of unsaturation. The molecular weight difference of 42 Da between 198 and 199 suggested that 199 contained a hydroxyl rather than an acetoxy moiety.

1 13 Figure 38. H and C NMR spectra for 199 in DMSO-d6

149

Comparison of the NMR data of compounds 198 and 199 showed they had almost identical 1H and 13C chemical shifts for both rings and the alkene side chain. The only major differences were that 199 lacked the 1H and 13C signals of an acetoxy group, and that the H-3 and C-3 resonances of 199 (δH 4.05 and δC 64.9) resonated further upfield than that of 198 (δH 5.19 and δC 69.6) (Figure 38). These data indicated that 199 was the de-acetyl derivative of 198. Furthermore, the chemical shift of H-18 (δH 0.36) was in agreement with literature values reported when a hydroxyl group is located at C-

108,228,232,277,278 3 in the pseudo-equatorial position (δH ~ 0.30 in DMSO-d6).

While the phenolic protons were not observed in the 1H NMR spectrum of 198, a downfield hydroxyl signal (δH 8.60) was detected for 199 and attached to C-8 based on

HMBC correlations from this resonance to C-7, C-8, and C-9. The ROESY and 1H-1H coupling constant data for 199 were essentially identical to that of 198, hence the chemical structure of 199 was assigned to 3,7,8-trihydroxyserrulat-14-en-19-oic acid.

150

Table 26. NMR data for 3,7,8-trihydroxyserrulat-14-en-19-oic acid (199)a b c Position δC δH COSY HMBC ROESY 1 28.5, CH 3.19 (brdq, 7.2, 7.2) 2α, 2β, 20 2, 3, 8, 9, 10, 20 2α, 2β, 18 2α 34.6, CH2 1.87 (ddd, 12.5, 12.0, 7.2) 1, 2β, 3 1, 3, 4, 20 1, 2β, 4 2β 1.45 (m) 1, 2α, 3 1, 3, 4, 9, 20 1, 2α, 11, 20 3 64.9, CH 4.05 (brdd, 12.5, 4.2) 2α, 2β, 3- 1, 11 4, 2 β, 11w, OH, 4 18, 20 3-OH 4.62 (brs) 3 2w, 3w, 4w 4 48.2, CH 2.69 (brd, 4.2) 3, 5w, 11 2, 3, 5, 9, 10, 2β, 3, 11, 11, 12, 18 12, 13, 18 5 120.0, CH 6.94 (s) 4w 4, 7, 9, 19 4, 11, 12, 13, 18 6 110.1, C 7 147.4, C 8 142.3, C 8-OH 8.60 (brs) 7w, 8w, 9 9 135.3, C 10 128.3, C 11 29.9, CH 2.10 (brdq, 13.8, 6.6) 4, 12, 18 3, 4, 10, 12, 13, 2β, 3w, 4, 5, 18 12, 14w 12 38.5, CH2 1.47 (m) 11, 13 4, 11, 13, 14, 18 4, 11, 14, 18 1.25 (brddt, 13.8, 7.8, 7.8) 11, 13 4, 11, 13, 14, 18 4, 11, 13, 14, 18 13 25.7, CH2 1.97 (ddd, 7.8, 7.2, 7.2) 12, 14 11, 12, 14, 15 4, 5, 18 14 124.8, CH 5.13 (t, 7.2) 13, 16, 17 12, 16, 17 11w, 12, 16, 18w 15 130.2, C 16 25.5, CH3 1.64 (s) 14 14, 15, 17 14 17 17.5, CH3 1.57 (s) 14 14, 15, 16 18 18.8, CH3 0.36 (d, 6.6) 11 4, 11, 12 1, 2β, 3, 4, 5, 12, 13, 14w 19 172.4, C 20 21.9, CH3 1.17 (d, 7.2) 1 1, 2, 9 2β, 3 a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

151

The third new serrulatane to be identified, compound 200, also displayed a very similar 1H NMR spectrum to that of NP 198 (Figure 39, Table 27). The only observed differences were that 200 possessed an additional aromatic proton resonance (δH 6.64),

13 an extra methyl group (δH 2.03), and an –OCH2– moiety (δH 4.92). The C NMR spectrum of 200 (Figure 40) was similar to that of 198 in that it contained two carbonyl signals (δC 170.1 and 169.8), however, it possessed an additional oxygenated carbon signal (δC 65.4) and lacked one of the phenolic carbons of 198. The 2D NMR data of

200 indicated that it differed from 198 in the substitution of the benzene ring (ring B). It was determined that the hydroxyl at C-7 in 198 had been replaced by a hydrogen (δH

6.64) in 200 as indicated by HMBC correlations from this proton to C-5 (δC 119.9) and

C-9 (δC 127.4). H-5 and H-7 shared a HMBC correlation to δC 65.4 (C-19), which allowed the –OCH2– moiety to be substituted at C-6. The methylene protons at H-19 (δH

4.92) showed a HMBC correlation to the carbonyl carbon at δC 170.1, as did the methyl group at δH 2.03, which established an acetoxy moiety at C-19.

1 Figure 39. H NMR spectrum for 200 in DMSO-d6

152

Table 27. NMR data for 3,19-diacetoxy-8-hydroxyserrulat-14-ene (200)a b c Position δC δH COSY HMBC ROESY 1 27.4, CH 3.17 (brqdd, 7.2, 2α, 2β, 20 2, 3, 8w, 9, 10, 2α, 2β 6.6, 1.8) 20 2α 31.5, CH2 2.08 (ddd, 12.6, 1, 2β, 3 1, 3, 4, 20 1, 2β, 18 12.6, 6.6) 2β 1.57 (m) 1, 2α, 3 3, 4, 9, 20 1, 2α, 3, 4, 20 3 69.7, CH 5.18 (ddd, 12.6, 2α, 2β, 4 1, 11, 21 2β, 4, 18, 20 4.8, 3.6) 4 45.1, CH 2.97 (brd, 4.8) 3, 5w, 11 3, 5, 9, 10, 11, 2β, 3, 11, 12, 18 12, 18 5 119.9, CH 6.46 (brd, 1.2) 4w 4, 7, 9, 19 4, 12, 18, 19 6 133.7, C 7 112.6, CH 6.64 (brd, 1.2) 5, 8, 9, 19 19 8 154.7, C 9 127.4, C 10 137.0, C 11 31.1, CH 1.96 (m) 4, 18 3, 4, 10w, 12, 4, 12, 18 13w, 18w 12 37.9, CH2 1.38 (ddt, 13.8, 13 4, 11, 13, 14, 18 4, 5, 11, 13, 18 7.8, 6.6) 1.25 (m) 13 4, 11, 13, 14, 18 4, 5, 11, 13, 18 13 25.5, CH2 1.94 (m) 12, 14 12, 14, 15 12, 18 14 124.3, CH 5.10 (t, 7.2) 13, 16, 17 12, 16, 17 12, 16 15 130.7, C 16 25.5, CH3 1.65 (s) 14 14, 15, 17 14 17 17.4, CH3 1.56 (s) 14 14, 15, 16 18 18.8, CH3 0.46 (d, 6.6) 11 4, 11, 12 2α, 3, 4, 5, 11, 12, 13 19 65.4, CH2 4.92 (s) 5, 6, 7, 23 5, 7 20 21.7, CH3 1.18 (d, 7.2) 1 1, 2, 9 1, 2β, 3 21 169.8, C 22 21.0, CH3 2.01 (s) 21 23 170.1, C 24 20.7, CH3 2.03 (s) 23 a b 13 c 1 w Spectra were recorded in DMSO-d6 at 30 °C. C, mult. H (mult., J in Hz). Weak correlations.

153

13 Figure 40. C NMR spectrum for 200 in DMSO-d6

The molecular formula obtained for 200 of C24H34O5 (eight degrees of unsaturation) from the HRESIMS data was in agreement with the NMR data. After analyses of the ROESY spectrum and 1H-1H coupling constants the relative configuration of 200 was found to be the same as that of 198 and 199. The chemical structure of 200 was therefore determined to be 3,19-diacetoxy-8-hydroxyserrulat-14- ene.

NP 200 is a positional isomer of serrulatane 160 which was isolated from E.

277 neglecta and whose NMR data was recorded in CD3OD. In order to directly compare

1 1 the H NMR data of 200 with that of 160, a H NMR spectrum was obtained in CD3OD.

The 1H NMR values for these two NPs were found to be clearly different, particularly in regards to the chemical shifts of H-18 (δH 0.52 for 200 and δH 1.05 for 160).

154

The known NPs verbascoside171 (Chapter 6, Table 18) and jaceosidin331,332

(Table 28) were assigned as structures 110 and 201, respectively, after MS and 1D/2D

NMR spectroscopic data analyses and comparison with literature values.

Jaceosidin (201) has been found to possess a range of biological activities including anti-oxidant,333,334 anti-cancer,332,335-338 anti-inflammatory,339-342 and immunosuppressive,343 and its mechanism of action against many of these disease states has been explored.

Table 28. NMR data for jaceosidin (201)a b c Position δC δH COSY HMBC 2 163.6, C 3 102.4, CH 6.88 (s) 1', 2, 4, 10 4 182.1, C 5 152.7, C 6 131.3, C 7 157.4, C 8 94.1, CH 6.61 (s) 4, 6, 7, 9, 10 9 152.4, C 10 104.0, C 1' 121.4, C w 2' 109.9, CH 7.55 (m) 4'-OCH3 2, 3', 4', 6' 3' 148.1, C 4' 150.7, C 5' 115.6, CH 6.93 (d, 9.0) 6' 1', 3', 4' 6' 120.2, CH 7.56 (m) 5' 2, 2', 4' 5-OH 13.07 (brs) 5, 6, 10 6-OCH3 59.6, CH3 3.75 (s) 7-OH 3'-OH w 4'-OCH3 55.7, CH3 3.88 (s) 2' 3' a b 13 c 1 Spectra were recorded in DMSO-d6 at 30 °C; C, mult.; H (mult., J in Hz); w Weak correlations.

155

7.3. Semi-synthetic analogues of serrulatane 199

As serrulatane 199 had been isolated in such large amounts (453 mg), it was possible to generate simple analogues for crystallisation and bioactivity studies. In

344 parallel, 199 was acetylated using Ac2O and dry pyridine (Scheme 24) and

166 methylated using TMS-diazomethane. Both reaction crudes were purified using C18- bonded silica semi-preparative HPLC (CH3OH/H2O). The acetylation reaction afforded compound 205 (15.7 mg, 53% yield) while the methylation procedure resulted in two products, 206 (8.8 mg, 35% yield, Figure 41) and 207 (2.4 mg, 9% yield). The

HRESIMS and 1D/2D NMR spectroscopic data obtained for each of these compounds confirmed their structures as shown.

Scheme 24. Acetylation (i) and methylation (ii) of NP 199. (i) Ac2O:py (1:1), rt, 16 h.

(ii) TMS-CH2N2, CH3OH:CH2Cl2 (1:1), rt, 20 min

156

1 13 Figure 41. H and C NMR spectra for 206 in DMSO-d6

NPs 198-200 and the semi-synthetics 205-207 were divided into portions and dissolved in a range of different solvents such as CH2Cl2, diethyl ether, CH3OH, and mixtures of these solvents with n-hexanes. Crystals of NP 199 were obtained from

CH2Cl2, however, they were not of suitable quality for X-ray crystallography studies.

Crystallisation studies are on-going.

157

7.4. Bioactivity of isolated NPs and semi-synthetic analogues

Literature reports have identified a number of serrulatanes (63, 64, 157-162,

164-167) that demonstrate selective antimicrobial activity towards Gram-positive bacteria, and just one (161) has been shown to inhibit a Gram-negative bacterial strain.108,277-279 For example, NPs 63 and 64 from E. sturtii exhibited bactericidal activity against S. aureus with MBCs of 200 μg/mL and 15 μg/mL, respectively.108

Serrulatane 63 also strongly inhibited the inflammation pathway enzymes COX-1 (IC50

108 27 μg/mL) and COX-2 (IC50 73 μg/mL).

E. serrulata provided serrulatanes 157-159, which possessed MICs ranging from

125-250 μg/mL against S. aureus.278 Compounds 161 and 162 from E. neglecta showed

MIC values ranging from 12.9-49.3 μM against strains of S. aureus, S. pyogenes, and S. pneumonia.277 The isomer of NP 200, compound 160, was found to be inactive in this same study. NPs 161 and 162 were examined further and were found to inhibit 24

Gram-positive bacterial strains with MIC values ranging from 3.1-100 μg/mL.281

Compound 161 also demonstrated activity against the Gram-negative strain Moraxella catarrhalis (ATCC 49143, MIC 3.1 μg/mL). Serrulatanes 161 and 162 were also screened against Vero mammalian cells and found to possess moderate cytotoxicity

281 (CC50 9.2 μg/mL and 20 μg/mL, respectively).

Serrulatanes 164 and 165 from E. duttonii226,232,279 were found to possess MIC values ranging from 23–1500 μg/mL against strains of S. aureus, S. epidermidis, and S. pneumonia.279 In a separate study Ndi et. al. found that acetylation of the hydroxyl groups of 166 and 167 resulted in a loss of antibacterial activity against S. aureus

(ATCC 29213).282

158

Given the known antibacterial activity of the serrulatane structure class and the fact that an extract of E. microtheca has been found to be active against S. aureus, S. pyogenes, and S. pneumonia (MICs 31-125 μg/mL),184 NPs 198-199 along with the semi-synthetic analogues 205 and 206 were analysed against a panel of nine Gram- positive and one Gram-negative bacterial strains (Table 29). The known compounds verbascoside (110) and jaceosidin (201) were also tested. Compound 207 had only been obtained at 80% purity, so was not included in the screen.

159

Most of the compounds were found to be inactive at the highest concentration tested (128 μg/ mL) against most bacteria, though all the serrulatane compounds were found to be moderately active (64-128 μg/ mL) against S. pyogenes. Verbascoside (110) was inactive against all bacterial strains.

NP 198 demonstrated activity at 64 or 128 μg/mL against all bacterial strains except for E. faecium and E. faecalis. Jaceosidin (201) had the greatest potency (16-32

μg/mL) against most S. aureus strains.

As previously mentioned, Ndi et. al. had shown that acetylation of serrulatanes

164 and 165 reduced their activity.282 In our study, the acetylated compound 205 was active against S. aureus (ATCC 43300) and S. pyogenes (128 μg/mL and 64 μg/mL, respectively) while NP 199 was only active against S. pyogenes (MIC 64 μg/mL). It has been suggested that the antibacterial activity of the serrulatanes is increased by the presence of phenolic hydroxyl groups, but that they may also cause instability of these compounds as diphenols can be easily oxidised to form quinones.282 As only partial acetylation was achieved for 205, it may be that this has stabilised the compound while leaving one of the phenolic hydroxyl groups free for interaction with the target.

Ndi et. al. also found that serrulatane 160, which is an isomer of NP 200, was inactive against strains of S. aureus, S. pyogenes, and S. pneumonia,277 however in this investigation 200 was active against S. pyogenes (MIC 128 μg/mL).

160

Table 29. Antibacterial activity of 198-201, 205 and 206‡

MIC (MBC) μg/mL Bacterial strain 198 199 200 201 205 206 Staphylococcus aureus 128 (128) a a 16 (16) a a MRSA, clinical isolate Staphylococcus aureus 128 (128) a a a 128 (128) a MRSA, ATCC 43300 Staphylococcus aureus 128 (128) a a 32 (32) a a GISA, NRS 17 Staphylococcus aureus 128 (128) a a 16 (16) a a MRSA, DapRSA Staphylococcus aureus 128 (128) a a 32 (32) a a GISA, NRS 1 Streptococcus pneumonia 128 (128) a a a a a MDR, ATCC 700677 Streptococcus pyogenes 64 (64) 128 (128) 128 (128) a 64 (64) 128 (128) Group A, ATCC 12344 Enterococcus faecium MDR-Van A, ATCC a a a a a a 51559 Enterococcus faecalis a a a a a a VanA, clinical isolate Escherichia coli a a a a a a ATCC 25922 a not active at maximum concentration (128 μg/mL); Positive controls included vancomycin, colistin and daptomycin, control MIC values given in Chapter 8 (Experimental).

7.5. Conclusion

Three new members of the serrulatane structure class (198-200) and the known compounds verbascoside (110) and jaceosidin (201) were isolated from the leaves and outer branches of E. microtheca. The methylated and acetylated derivatives 205-207 were generated from 199 and crystallisation studies undertaken. The antibacterial activity of the NPs (110, 198-201) and the semi-synthetic analogues 205 and 206 was also assessed against a panel of Gram-positive and Gram-negative bacterial strains.

‡Antibacterial assay was undertaken in the laboratory of Professor Matthew Cooper of the Institute for Molecular Bioscience, University of Queensland. 161

Chapter 8: Experimental

8.1. General experimental

Optical rotations were recorded on a JASCO P-1020 polarimeter. IR and UV spectra were recorded on a Bruker Tensor 27 spectrophotometer and a Jasco V-650

UV/Vis spectrophotometer, respectively. NMR spectra were recorded at 30 °C on either a Varian 400 MHz, 500 MHz or 600 MHz Unity INOVA spectrometer. The latter spectrometer was equipped with a triple resonance cold probe. The 1H and 13C chemical shifts were referenced to the solvent peaks for DMSO-d6 at δH 2.49 and δC 39.5, for

CDCl3 at δH 7.27 and δC 77.2, and for CD3OD at δH 3.31 and δC 49.1. J values are given in Hz. LRESIMS were recorded on a Mariner Time-of-Flight spectrometer equipped with a Gilson 215 eight probe injector or a Waters ZQ mass spectrometer. HRESIMS were recorded on a Bruker Apex III 4.7 Tesla Fourier transform ion cyclotron resonance mass spectrometer or a Bruker micrOTOF-Q instrument with a Bruker ESI source. A

Fritsch pulverisette 19 universal cutting mill was utilised to grind the biota samples. An

Edwards Instrument company Bio-line orbital shaker was used for plant extractions.

Fluka silica gel TLC aluminium cards were used for TLC. Phenomenex solid phase extraction (SPE) polypropylene cartridges were used for reaction purifications. For analytical HPLC purity analysis a Waters 600 pump equipped with a Waters 966 PDA detector and Gilson 715 liquid handler (5 mL syringe, 200 µL Rheodyne sample loop) was used. A Waters 600 pump equipped with a Waters 966 PDA detector and a Waters

717 Plus Autosampler connected to a Gilson FC204 fraction collector were used for semi-preparative separations. A Phenomenex C18 Onyx monolithic (4.6  100 mm) column was used for analytical HPLC purity analysis. Alltech Davisil diol-bonded silica, 30-40 μm, 60 Å, or Alltech C18 bonded silica, 35-75 μm, 150 Å, or Merck silica gel 60 (0.040-0.063 mm) were used for pre-adsorption work. Either a YMC diol 5 μm

162

120 Å (20 × 150 mm) column or a ThermoElectron C18 Betasil 5 μm 143 Å (21.2 × 150 mm) column were used for semi-preparative HPLC separations. All solvents used for chromatography, [α]D, UV, and MS were Lab Scan HPLC grade, and the H2O was

Millipore Milli-Q PF filtered. All synthetic reagents were obtained from Sigma–Aldrich and used without further purification.

8.2. Chapter 2. Experimental

Plant material. The leaves of E. sturtii (QID005836) were collected from

Currawinya National Park, QLD, Australia, in March, 1996. Collection and identification were undertaken by P. Forster and G. Guymer from the Queensland

Herbarium. A voucher specimen (AQ603041) has been deposited at the Queensland

Herbarium, Brisbane, Australia.

Extraction and isolation of 14-Hydroxy-6,12-muuroloadien-15-oic acid (62).

The air-dried and ground leaves of E. sturtii (27.5 g) were poured into a conical flask (1

L), n-hexane (250 mL) was added and the flask was shaken at 200 rpm for 2.5 h. The n- hexane extract was filtered under gravity then discarded. CH2Cl2 (250 mL) was added to the de-fatted sample in the conical flask and shaken at 200 rpm for 2.5 h. The resulting extract was filtered under gravity, and set aside. Another volume of CH2Cl2 (250 mL) was added to the plant sample, and the flask shaken at 200 rpm for 16 h before filtration. CH3OH (250 mL) was then added and the CH3OH/plant mixture shaken for a further 2.5 h at 200 rpm. This step was repeated for another 2.5 h. Following gravity filtration, the plant material was extracted with another volume of CH3OH (250 mL), while being shaken at 200 rpm for 16 h. All CH2Cl2/CH3OH extractions were combined and dried under reduced pressure to yield a dark brown gum (5.76 g).

163

Two portions of this crude extract were taken for method development. The first portion (0.79 g) was pre-adsorbed to diol-bonded silica, loaded onto a diol-bonded silica flash column (35 × 130 mm), and a 20% stepwise gradient from n-hexane to EtOAc followed by CH3OH was performed (100 mL washes). Fraction 5 (20% n-hexane/80%

EtOAc, 160.3 mg) was found to contain 62 after 1H NMR spectroscopic and (+)-

LRESIMS analysis. This fraction was pre-adsorbed to diol-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a diol-bonded silica column (20 × 150 mm). Isocratic conditions of n-hexane were held for 10 min, followed by a linear gradient to 20% i-PrOH/80% n-hexane over 40 min, then isocratic conditions of 20% i-PrOH/80% n-hexane for 10 min, all at a flow rate of

9 mL/min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy. Fractions 32 and 33 (53.0 mg, 90% pure) were found to contain 62. Attempts to obtain 62 in higher purity were performed as follows. Fractions

32 and 33 were pre-adsorbed to C18-bonded silica, packed into a stainless steel cartridge

(10 × 30 mm), and subjected to semi-preparative HPLC using a C18-bonded silica column (21.2 × 150 mm) at a flow rate of 9 mL/min. Isocratic conditions of 10%

CH3OH/90% H2O were held for 10 min, followed by a linear gradient to CH3OH over

40 min, then isocratic conditions of CH3OH for 10 min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy. Fractions

46-48 were found to contain still impure 62 (40.5 mg, > 90% pure). A portion of these fractions (7 mg) was re-injected onto the same C18-bonded silica column, however, this time conditions of 50% CH3OH/50% H2O for 5 min, followed by a linear gradient to

CH3OH over 20 min, then isocratic conditions of CH3OH for 10 min were used. Thirty- five fractions (35 × 1 min) were collected.

164

Another quantity of impure 62 (7 mg) was injected onto the C18-bonded silica column using the same conditions as just listed, except 70 fractions (70 × 0.5 min) were collected, resulting in the isolation of 0.7 mg of compound 62 (> 95% pure).

The second portion of the E. sturtii crude extract (0.82 g) was pre-adsorbed to

C18-bonded silica, loaded onto a C18-bonded silica flash column (35 × 130 mm), and a

20% stepwise gradient from 20% CH3OH/80% H2O to CH3OH was performed (100 mL washes). Fraction 4 (80% CH3OH/20% H2O, 154.2 mg) was found to contain 62 after

(+)-LRESIMS and 1H NMR spectroscopic analysis. This fraction was pre-adsorbed to

C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a C18-bonded silica column (21.2 × 150 mm) at a flow rate of 9 mL/min and isocratic conditions of 10% CH3OH/90% H2O for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of

CH3OH for 10 min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-

LRESIMS and 1H NMR spectroscopy. Fractions 46 and 47 (27.0 mg, 95% pure) were found to contain 62, however, these fractions were not pure by 1H NMR spectroscopy.

Fraction 46 was pre-adsorbed to diol-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a diol-bonded silica column (20 × 150 mm) at a flow rate of 9 mL/min and isocratic conditions of n- hexane for 10 min, followed by a linear gradient to 20% i-PrOH/80% n-hexane over 40 min, then isocratic conditions of 20% i-PrOH/80% n-hexane for 10 min. Sixty fractions

(60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy.

This resulted in 13 mg of pure 62 (fraction 33) and 5.8 mg of impure 62 (fraction 32,

95% pure).

165

Fraction 32 was dissolved in CH3OH and injected in one shot onto a semi- preparative diol-bonded silica column (20 × 150 mm). Isocratic conditions of n-hexane were used for 5 min, followed by a linear gradient to 15% i-PrOH/85% n-hexane over

20 min, then isocratic conditions of 15% i-PrOH/85% n-hexane for 10 min all at a flow rate of 9 mL/min. Thirty-five fractions (35 × 1 min) were collected. This resulted in the isolation of 3.4 mg of impure compound 62 (fraction 24, 95% pure). Fraction 24 was re- purified using the same conditions as just described but this time seventy fractions (70 ×

0.5 min) were collected. Fraction 47 contained 62 along with a minor impurity (0.6 mg) and fraction 48 contained pure 62 (0.8 mg).

After considering the yields and purities obtained, the remaining crude from the extraction of E. sturtii was divided into ~ 500 mg portions and purified by diol-bonded silica flash column and diol-bonded silica HPLC, as outlined above. The mass of natural product 62 obtained was 201.7 mg (0.73 % dry wt, ≥ 90% purity).

14-Hydroxy-6,12-muuroloadien-15-oic acid (62): Light

26 green gum; []D - 154 (c 0.120, CHCl3); UV (CH3OH) λmax

nm (log ε) 203 (4.68), 220 (4.75); IR νmax (KBr) 3340, 3074,

2628, 1685, 1644, 1531, 1453, 1432, 1268, 1219, 1086, 1051, 903 cm-1; 1H and 13C

+ NMR data (DMSO-d6), see Table 1; (+)-LRESIMS m/z (rel. int.) 273 (100) [M + Na] ;

+ (+)-HRESIMS m/z 273.1464 (C15H22O3Na [M + Na] requires 273.1461).

166

Extraction and isolation of NPs 68-74. The remaining fractions from the diol- bonded silica flash column undertaken on the E. sturtii leaf extract were further purified as follows. The 60% n-hexane/40% EtOAc fraction was purified by diol-bonded silica semi-preparative HPLC. Isocratic conditions of n-hexane were held for the first 10 min, followed by a linear gradient to 20% i-PrOH/80% n-hexane in 40 min, then isocratic conditions of 20% i-PrOH/80% n-hexane for 10 min, all at a flow rate of 9 mL/min.

Sixty × one min fractions were collected and analysed by (+)-LRESIMS and 1H NMR spectroscopy. Fractions 25-26 contained impure mitchellene B (69, 56.5 mg) and were further purified by semi-preparative diol-bonded silica HPLC. Isocratic conditions of n- hexane were held for the first 5 min, followed by a linear gradient to 15% i-PrOH/75% n-hexane over 40 min, then a linear gradient to 20% i-PrOH/80% n-hexane over 5 min, followed by isocratic conditions of 20% i-PrOH/80% n-hexane for 10 min all at a flow rate of 9 mL/min. Sixty × one min fractions were collected. Mitchellene B (69) eluted in fractions 41-42 (28.3 mg, 0.103% dry wt).

The 40% n-hexane/60% EtOAc fraction from the diol-bonded silica flash column was further purified by C18-bonded silica semi-preparative HPLC at a flow rate of 9 mL/min and isocratic conditions of 10% CH3OH/90% H2O for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of CH3OH for 10 min. Sixty × one min fractions were collected and analysed by (+)-LRESIMS and 1H

NMR spectroscopy. Fraction 45 contained semi-pure mitchellene C (70, 5.4 mg) and was further purified by semi-preparative diol-bonded silica HPLC, using isocratic conditions of n-hexane for 10 min, followed by a linear gradient to 20% i-PrOH/80% n- hexane over 40 min, then isocratic conditions of 20% i-PrOH/80% n-hexane for 10 min all at a flow rate of 9 mL/min. Sixty × one min fractions were collected. Fraction 29 contained pure 70 (2.2 mg, 0.008% dry wt).

167

The 20% n-hexane/80% EtOAc and EtOAc fractions from the initial diol flash column were combined and further purified using the same diol-bonded silica column and conditions as for the 40% EtOAc/60% n-hexane fraction, above. This yielded mitchellene D in fraction 31 (71, 9.6 mg, 0.035% dry wt), mitchellene E in fraction 36

(72, 8.4 mg, 0.031% dry wt), mitchellene A in fractions 37-38 (68, 10.5 mg, 0.038% dry wt), casticin in fractions 47-49 (73, 9 mg, 0.033% dry wt), and centaureidin in fractions

53-54 (74, 9.7 mg, 0.035% dry wt).

25 Mitchellene A (68): brown gum; []D - 45 (c 0.060, CHCl3); UV

(CH3OH) λmax nm (log ε) 205 (4.20), 214 (4.22), 289 (3.10); IR νmax

(KBr) 3420, 1744, 1457, 1375, 1278, 1202, 1059, 983, 956 cm-1; 1H

13 and C NMR data (DMSO-d6) see Table 2; (+)-LRESIMS m/z (rel. int.) 271 (100) [M + Na]+, 249 (100) [M + H]+; (+)-HRESIMS m/z 271.1300

+ (C15H20O3Na [M + Na] requires 271.1305).

25 Mitchellene B (69): opaque gum; []D - 58 (c 0.080, CHCl3);

UV (CH3OH) λmax nm (log ε) 204 (4.54), 216 (4.61); IR νmax (KBr)

2920, 1761, 1667, 1454, 1379, 1347, 1287, 1197, 1166, 1141,

1124, 1062, 1012, 981, 962, 952 cm-1; 1H and 13C NMR data

+ (DMSO-d6) see Table 3; (+)-LRESIMS m/z (rel. int.) 255 (100) [M + Na] , 233 (100)

+ + [M + H] ; (+)-HRESIMS m/z 255.1355 (C15H20O2Na [M + Na] requires 255.1356).

168

25 Mitchellene C (70): opaque gum; []D - 58 (c 0.107, CHCl3); UV

(CH3OH) λmax nm (log ε) 205 (4.36), 232 (3.74); IR νmax (KBr) 3382,

1764, 1594, 1455, 1378, 1357, 1289, 1199, 1094, 1033, 989 cm-1; 1H

13 and C NMR data (DMSO-d6) see Table 4; (+)-LRESIMS m/z (rel.

+ + int.) 271 (100) [M + Na] ; (+)-HRESIMS m/z 271.1311 (C15H20O3Na [M + Na] requires 271.1305).

26 Mitchellene D (71): light brown gum; []D - 34 (c 0.120,

CHCl3); UV (CH3OH) λmax nm (log ε) 203 (3.92), 219 (4.04); IR

νmax (KBr) 3390, 1694, 1644, 1454, 1416, 1380, 1269, 1224,

-1 1 13 1180 cm ; H and C NMR data (DMSO-d6) see Table 5; (+)-

LRESIMS m/z (rel. int.) 289 (100) [M + Na]+; (+)-HRESIMS m/z 289.1402

+ (C15H22O4Na [M + Na] requires 289.1410).

26 Mitchellene E (72): light brown gum; []D - 30 (c 0.093,

CHCl3); UV (CH3OH) λmax nm (log ε) 203 (4.11), 225 (3.70); IR

νmax (KBr) 3409, 1722, 1709, 1693, 1679, 1513, 1380, 1238,

-1 1 13 1224, 994 cm ; H and C NMR data (DMSO-d6) see Table 6;

(+)-LRESIMS m/z (rel. int.) 275 (100) [M + Na]+; (+)-HRESIMS m/z 275.1629

+ (C15H24O3Na [M + Na] requires 275.1618).

Casticin (73): yellow gum; 1H and 13C NMR data

(DMSO-d6) see Table 7; (+)-LRESIMS m/z (rel.

int.) 375 (100) [M + H]+.

169

Centaureidin (74): yellow gum; 1H and 13C NMR

data (DMSO-d6), see Table 8; (+)-LRESIMS m/z

(rel. int.) 360 (100) [M + H]+.

Procedure for oxidation of mitchellene B (69).§ Dimethyldioxirane (DMDO) in acetone (0.08 M) was generated following a literature procedure from acetone and potassium peroxymonosulfate.149 Mitchellene B (69, 5 mg, 0.022 mmol) was dissolved in acetone, cooled to – 78 °C and DMDO (270 μL) added. The solution was left to stir for 2 h, then another 10 eq of DMDO (2.7 mL) was added. The solution was allowed to warm to rt and left stirring overnight. The solution was evaporated under N2 and the crude obtained (5 mg) analysed by (+)-LRESIMS and 1H NMR spectroscopy.

1 Epoxide (81): white gum; H NMR (600 MHz, DMSO-d6) δ 0.81

(3H, d, J = 6.6 Hz, H-11), 0.93 (3H, d, J = 6.6 Hz, H-13), 0.97

(1H, m, H-4), 1.44 (1H, m, H-5), 1.60 (1H, m, H-12), 1.67 (1H,

m, H-1), 1.70 (1H, m, H-10), 2.86 (1H, dd, J = 7.8, 7.2 Hz, H-6),

3.38 (1H, brd, J = 2.4 Hz, H-8), 5.01 (1H, dd, J = 7.2, 7.2 Hz, H-14); (+)-LRESIMS m/z

(rel. int.) 249 (100) [M + H]+.

§ DMDO reaction was undertaken by Dr Brett Schwartz of the Eskitis Institute, Griffith University.

170

8.3. Chapter 3. Experimental

General procedure for amide formation using DMTMM.101 DMTMM was generated following a literature procedure from its precursor 2-chloro-4,6-dimethoxy-

1,3,5-triazine (CDMT) and N-methylmorpholine.101 Scaffold 62 (~ 10 mg) and the relevant commercially available amine (10 mmol excess) were stirred in dry THF (4 mL) for 15 min at rt under Ar. DMTMM (10 mmol excess) was added and the solution stirred at rt for 4 h. The solution was poured into H2O (30 mL) and extracted with

CH2Cl2 (30 mL). The organic phase was further washed with NaHCO3 (30 mL), 3.5 %

HCl (30 mL), and H2O (30 mL) before being dried over MgSO4.

Purification of DMTMM amine coupling products. After coupling 62 with phenethylamine, the crude product was purified using a SPE cartridge packed with diol- bonded silica (10 × 40 mm) and 10 mL eluent flushes of n-hexane, 5% i-PrOH/n- hexane, 10% i-PrOH/n-hexane, 20% i-PrOH/n-hexane, then CH3OH. The 10% i-

PrOH/90% n-hexane fraction was found to contain impure product following TLC and

1H NMR spectroscopic analysis. This fraction was pre-adsorbed to diol-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative

HPLC using a diol-bonded silica column at a flow rate of 9 mL/min and isocratic conditions of n-hexane for 10 min, followed by a linear gradient to 15% i-PrOH/85% n- hexane over 40 min, then isocratic conditions of 15% i-PrOH/85% n-hexane for 10 min.

Sixty × one min fractions were collected. Fraction 46 was found to contain 82 (2.3 mg,

19%).

171

Reaction of 62 with ammonia gave impure compound 83 which was purified in a similar manner to 82, except that a linear gradient to 20% i-PrOH/80% n-hexane over

40 min was utilised in the diol-bonded silica HPLC step to give 83 in fractions 46-48

(3.7 mg, 27%). Impure compound 84 was generated by coupling 62 with N-(2- aminoethyl)acetamide, and was first purified using a SPE cartridge packed with diol- bonded silica (10 × 40 mm) using 10 mL washes from n-hexane to EtOAc in 20% steps followed by CH3OH. Impure product was found in the CH3OH fraction, and was further purified by C18-bonded silica semi-preparative HPLC at a flow rate of 9 mL/min and isocratic conditions of 10% CH3OH/90% H2O for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of CH3OH for 10 min. Sixty fractions

(one min each) were collected and fraction 37 found to contain 84 (3.2 mg, 31%).

3-Chloro-4-hydroxyphenylacetic acid (85) was coupled with histamine using

DMTMM and the outlined procedure, above, to give impure 86. Compound 86 was extracted into the H2O fraction during the work up. This fraction was dried down, pre- adsorbed to C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a C18-bonded silica column (21.2 × 150 mm) at a flow rate of 9 mL/min and isocratic conditions of 10% CH3OH/90% H2O for

10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of CH3OH for 10 min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-

LRESIMS and 1H NMR spectroscopy. Fraction 11 contained pure 86 (2.4 mg, 15% yield).

172

25 Compound 82: Yield 19%, white gum; []D –

85 (c 0.073, CHCl3); UV (CH3OH) λmax nm (log

1 ε) 204 (4.13); H NMR (500 MHz, DMSO-d6) δ

0.91 (3H, d, J = 6.6 Hz, H-11), 1.11 (1H, dddd, J = 13.2, 12.6, 12.6, 3.0 Hz, H-2β), 1.31

(1H, ddd, J = 12.6, 12.6, 3.0 Hz, H-3α), 1.37 (1H, m, H-2α), 1.38 (1H, m, H-9β), 1.59

(1H, brdd, J = 12.5, 5.4 Hz, H-9α), 1.61 (1H, m, H-3β), 1.62 (1H, m, H-10), 1.72 (1H, m, H-1), 1.81 (1H, ddd, J = 12.6, 11.4, 2.4 Hz, H-4), 1.95 (1H, m, H-8α), 2.22 (1H, ddd,

J = 11.4, 6.0, 4.2 Hz, H-5), 2.40 (1H, brdd, J = 18.0, 5.4 Hz, H-8β), 2.71 (2H, t, J = 7.8

Hz, H-18), 3.27 (2H, dt, J = 7.8, 5.4 Hz, H-17), 3.82 (1H, d, J = 15.0 Hz, H-14), 3.87

(1H, d, J = 15.0 Hz, H-14), 4.79 (1H, brs, 14-OH), 4.90 (1H, s, H-13), 5.07 (1H, s, H-

13), 6.45 (1H, brd, J = 4.2 Hz, H-6), 7.17 (2H, d, J = 7.8 Hz, H-20), 7.18 (1H, t, J = 7.2

Hz, H-22), 7.27 (2H, dd, J = 7.8, 7.2 Hz, H-21), 7.66 (1H, t, J = 5.4 Hz, 16-NH); 13C

NMR (125 MHz, DMSO-d6) δ 16.0 (C-9), 19.3 (C-11), 25.0 (C-8), 28.9 (C-2), 33.9 (C-

3), 34.2 (C-1), 35.1 (C-18), 38.7 (C-10), 40.5 (C-17), 40.8 (C-5), 42.3 (C-4), 63.3 (C-

14), 107.3 (C-13), 125.9 (C-22), 128.2 (C-21), 128.5 (C-20), 132.8 (C-7), 135.1 (C-6),

139.6 (C-19), 152.6 (C-12), 167.4 (C-15); (+)-LRESIMS m/z (rel. int.) 354 (100) [M +

+ + H] ; (+)-HRESIMS m/z 376.2250 (C23H31NO2Na [M + Na] requires 376.2247).

25 Compound 83: Yield 27%, white gum; []D – 113 (c

0.107, CHCl3); UV (CH3OH) λmax nm (log ε) 202 (3.84), 219

1 (3.71); H NMR (500 MHz, DMSO-d6) δ 0.91 (3H, d, J = 7.2

Hz, H-11), 1.11 (1H, dddd, J = 12.6, 12.6, 12.6, 3.0 Hz, H-2β), 1.31 (1H, ddd, J = 13.2,

12.6, 2.4 Hz, H-3α), 1.35 (1H, m, H-9β), 1.37 (1H, m, H-2α), 1.55 (1H, brdd, J = 12.6,

5.4 Hz, H-9α), 1.61 (1H, m, H-3β), 1.62 (1H, m, H-10), 1.73 (1H, m, H-1), 1.80 (1H, ddd, J = 12.0, 11.4, 3.0 Hz, H-4), 1.96 (1H, m, H-8α), 2.21 (1H, ddd, J = 11.4, 6.0, 4.2

173

Hz, H-5), 2.38 (1H, brdd, J = 18.0, 5.4 Hz, H-8β), 3.81 (1H, d, J = 15.0 Hz, H-14), 3.86

(1H, d, J = 15.0 Hz, H-14), 4.75 (1H, brs, 14-OH), 4.90 (1H, s, H-13), 5.05 (1H, s, H-

13 13), 6.49 (1H, brd, J = 4.2 Hz, H-6), 7.07 (2H, s, 16-NH2); C NMR (125 MHz,

DMSO-d6) δ 16.1 (C-9), 19.3 (C-11), 25.1 (C-8), 29.0 (C-2), 33.8 (C-3), 34.2 (C-1),

38.7 (C-10), 41.0 (C-5), 42.3 (C-4), 63.3 (C-14), 107.4 (C-13), 132.5 (C-7), 135.9 (C-6),

152.6 (C-12), 169.3 (C-15); (+)-LRESIMS m/z (rel. int.) 250 (100) [M + H]+; (+)-

+ HRESIMS m/z 272.1620 (C15H23NO2Na [M + Na] requires 272.1621).

26 Compound 84: Yield 31%, orange gum; []D –

53 (c 0.04, CHCl3); UV (CH3OH) λmax nm (log ε)

202 (4.00), 221 (3.82); 1H NMR (500 MHz,

DMSO-d6) δ 0.91 (3H, d, J = 7.2 Hz, H-11), 1.11 (1H, dddd, J = 13.2, 12.6, 12.6, 3.6

Hz, H-2β), 1.29 (1H, ddd, J = 12.6, 12.6, 3.0 Hz, H-3α), 1.34 (1H, m, H-9β), 1.36 (1H, m, H-2α), 1.56 (1H, m, H-9α), 1.61 (1H, m, H-3β), 1.62 (1H, m, H-10), 1.72 (1H, m, H-

1), 1.78 (3H, s, H-21), 1.80 (1H, dd, J = 12.6, 4.2 Hz, H-4), 1.98 (1H, m, H-8α), 2.21

(1H, ddd, J = 11.4, 6.0, 5.4 Hz, H-5), 2.38 (1H, brdd, J = 18.6, 4.2 Hz, H-8β), 3.08 (2H, m, H-17), 3.13 (2H, m, H-18), 3.81 (1H, d, J = 15.0 Hz, H-14), 3.86 (1H, d, J = 15.0

Hz, H-14), 4.83 (1H, brs, 14-OH), 4.90 (1H, s, H-13), 5.06 (1H, s, H-13), 6.47 (1H, brd,

J = 4.8 Hz, H-6), 7.64 (1H, brs, 16-NH), 7.92 (1H, brs, 19-NH); 13C NMR (125 MHz,

DMSO-d6) δ 16.0 (C-9), 19.3 (C-11), 22.5 (C-21), 25.0 (C-8), 28.9 (C-2), 33.8 (C-3),

34.2 (C-1), 38.3 (C-17), 38.7 (C-10), 39.0 (C-18), 40.9 (C-5), 42.2 (C-4), 63.3 (C-14),

107.3 (C-13), 132.5 (C-7), 135.6 (C-6), 152.6 (C-12), 167.5 (C-15), 169.5 (C-20); (+)-

LRESIMS m/z (rel. int.) 357 (100) [M + Na]+, 335 (100) [M + H]+ ; (+)-HRESIMS m/z

+ 335.2318 (C19H31N2O3 [M + H] requires 335.2329).

174

Compound 86: Yield 23%, opaque gum; 1H NMR

(600 MHz, DMSO-d6) δ 2.75 (2H, t, J = 7.2 Hz, H-

11), 3.26 (2H, s, H-7), 3.32 (2H, dt, J = 7.2, 4.8 Hz, H-10), 6.88 (1H, d, J = 8.4 Hz, H-

5), 6.94 (1H, m, H-15), 6.95 (1H, d, J = 8.4 Hz, H-6), 7.15 (1H, s, H-2), 7.32 (1H, s, H-

13), 8.16 (1H, t, J = 4.8 Hz, 9-NH), 8.83 (1H, brs, 14-NH); 13C NMR (125 MHz,

DMSO-d6) δ 24.6 (C-11), 37.6 (C-10), 40.9 (C-7), 116.1 (C-13), 116.3 (C-5), 119.1 (C-

3), 127.9 (C-1), 128.4 (C-6), 130.0 (C-2), 131.4 (C-12), 133.6 (C-15), 151.5 (C-4),

170.3 (C-8); (+)-LRESIMS m/z (rel. int.) 280 (100) [M + H]+.

General procedure for amide formation using EDCI.95 Scaffold 62 (~ 10 mg), EDCI (1.5 mmol excess), and DMAP (0.5 mg) were stirred in CH3CN (3 mL) for

1 h at rt under Ar. The relevant commercially available amine (10 mmol excess) was added and the mixture stirred for 16 h. The solution was poured into CH2Cl2 (20 mL), before being extracted with H2O (20 mL) then 2 N HCl (20 mL).

Purification of EDCI amine coupling products. After coupling 62 with phenethylamine and N-(2-aminoethyl)acetamide, the crude products obtained were purified as above for the DMTMM coupling reactions to give 82 (1.6 mg, 12%) and 84

(3.2 mg, 22%). 3-Chloro-4-hydroxyphenylacetic acid (85) was coupled with histamine and phenethylamine using EDCI and the outlined procedure, above. After coupling with histamine, compound 86 was purified using the same method as outlined for the

DMTMM coupling, above, to give a yield of 23% (3.4 mg). Impure compound 87 from coupling with phenethylamine was purified using a silica SPE cartridge (10 × 40 mm) and a 20% stepwise gradient from n-hexane to EtOAc (10 mL flushes). The 40% n- hexane/60% EtOAc and 60% n-hexane/40% EtOAc fractions were found to contain pure 87 following 1H NMR spectroscopic analysis (14.7 mg, 92% yield).

175

Compound 87: Yield 92%, light brown gum; 1H

NMR (600 MHz, DMSO-d6) δ 2.68 (2H, t, J = 7.2

Hz, H-11), 3.25 (2H, m, H-10), 3.25 (2H, s, H-7), 6.86 (1H, d, J = 8.4 Hz, H-5), 6.95

(1H, d, J = 8.4 Hz, H-6), 7.14 (2H, d, J = 7.2 Hz, H-13), 7.17 (1H, s, H-2), 7.18 (1H, t, J

= 7.8 Hz, H-15), 7.25 (2H, dd, J = 7.8, 7.2 Hz, H-14), 7.99 (1H, t, J = 4.8 Hz, 9-NH);

13 C NMR (125 MHz, DMSO-d6) δ 35.0 (C-11), 40.2 (C-10), 41.0 (C-7), 116.3 (C-5),

119.2 (C-3), 127.9 (C-1), 126.0 (C-15), 128.2 (C-14), 128.5 (C-6), 128.6 (C-13), 130.0

(C-2), 139.4 (C-12), 151.6 (C-4), 170.0 (C-8); (+)-LRESIMS m/z (rel. int.) 290 (100)

[M + H]+.

Coupling of 3-chlorophenylacetic acid (89) with phenethylamine using

155 (COCl)2. 3-Chlorophenylacetic acid (89) (43 mg, 0.250 mmol) was dissolved in dry

CH2Cl2 (3 mL) under Ar. (COCl)2 (45μL, 0.500 mmol) was initially added, then anhydrous DMF (10 μL, 0.1293 mmol) was added dropwise. Gas evolution was observed on addition of DMF. In a second vial, phenethylamine (160 μL, 1.250 mmol) and anhydrous pyridine (100 μL) were stirred in dry CH2Cl2 (2 mL) under Ar. This vial was cooled to 0 °C and the acid chloride generated added dropwise. The solution was stirred for 30 min before being poured into CH2Cl2 (10 mL) and extracted with 2 N HCl

(30 mL). The crude obtained was purified using a silica SPE cartridge (10 × 40 mm) and a 20% stepwise gradient from n-hexane to EtOAc (10 mL flushes). The 40% n- hexane/60% EtOAc and 60% n-hexane/40% EtOAc fractions were found to contain product 90 following 1H NMR spectroscopic analysis (56.8 mg, 83% yield).

176

Compound 90: Yield 83%, light brown crystalline

1 solid; H NMR (500 MHz, DMSO-d6) δ 2.70 (2H, t, J = 7.2 Hz, H-11), 3.28 (2H, dt, J = 7.2, 4.8 Hz, H-10), 3.40 (2H, s, H-7), 7.16 (2H, d,

J = 7.2 Hz, H-13), 7.17 (1H, d, J = 7.8 Hz, H-6), 7.18 (1H, t, J = 7.8 Hz, H-15), 7.25

(2H, dd, J = 7.8, 7.2 Hz, H-14), 7.28 (1H, m, H-5), 7.29 (1H, m, H-4), 7.30 (1H, m, H-

13 2), 8.12 (1H, t, J = 4.8 Hz, 9-NH); C NMR (125 MHz, DMSO-d6) δ 35.0 (C-11), 40.2

(C-10), 41.8 (C-7), 125.9 (C-15), 126.2 (C-2), 127.6 (C-6), 128.2 (C-14), 128.6 (C-13),

128.8 (C-4), 129.8 (C-5), 132.7 (C-3), 138.8 (C-1), 139.3 (C-12), 169.3 (C-8); (+)-

LRESIMS m/z (rel. int.) 273 (100) [M + H]+.

155 General procedure for amide formation using scaffold 62 and (COCl)2.

Scaffold 62 (~ 10 mg) was dissolved in dry CH2Cl2 (2 mL) under Ar. (COCl)2 (3 mmol excess) was initially added, then anhydrous DMF (10 μL, 0.1293 mmol) was added dropwise. Gas evolution was observed on addition of DMF. In a second vial, the relevant commercially available amine (5 mmol excess) and anhydrous pyridine (100

μL) were stirred in dry CH2Cl2 (2 mL) under Ar. This vial was cooled to 0 °C and the acid chloride generated added dropwise. The solution was stirred for 20 min before being poured into CH2Cl2 (10 mL) and extracted with 2 N HCl (30 mL).

Purification of (COCl)2 amine coupling products. After coupling 62 with phenethylamine and N-(2-aminoethyl)acetamide the crude products obtained were purified as above for the DMTMM coupling reactions to give 82 (11.6 mg, 77%) and 84

(0.9 mg, 5%), respectively. Coupling of 62 with 2-(4-chlorophenyl)ethylamine gave impure 91 which was purified off a silica SPE cartridge (10 × 40 mm) by using a 10% stepwise gradient from n-hexane to EtOAc followed by CH3OH (10 mL elutions). Pure product was found in fractions 5-6 (8.6 mg, 48%).

177

Reaction of 62 with 4-methoxyphenethylamine gave 92 which was first purified by silica SPE cartridge (10 × 40 mm) using a 10% stepwise gradient from n-hexane to

EtOAc followed by CH3OH (10 mL elutions). Fractions 5-7 were then subjected to semi-preparative HPLC using a diol-bonded silica column at a flow rate of 9 mL/min and isocratic conditions of n-hexane for 10 min, followed by a linear gradient to 20% i-

PrOH/80% n-hexane over 40 min, then isocratic conditions of 20% i-PrOH/80% n- hexane for 10 min. Sixty fractions (60 × 1 min) were collected. Fraction 41 was found to contain 92 (2.5 mg, 5%). Compounds 94 and 95 (0.4 mg) eluted in fractions 44 and

46 of this HPLC run, respectively. Coupling of 62 with tyramine gave impure 93 which was purified in the same manner as 92, and eluted in fractions 53-54 (1.7 mg, 8%) after diol-bonded silica HPLC.

25 Compound 91: Yield 48%, white gum; []D

– 55 (c 0.07, CHCl3); UV (CH3OH) λmax nm

(log ε) 202 (4.23), 221 (4.21); 1H NMR (500

MHz, DMSO-d6) δ 0.91 (3H, d, J = 7.2 Hz, H-11), 1.11 (1H, dddd, J = 13.2, 13.2, 12.6,

2.4 Hz, H-2β), 1.31 (1H, m, H-3α), 1.34 (1H, m, H-9β), 1.36 (1H, m, H-2α), 1.56 (1H, brdd, J = 12.6, 6.0 Hz, H-9α), 1.61 (1H, m, H-3β), 1.62 (1H, m, H-10), 1.72 (1H, m, H-

1), 1.80 (1H, ddd, J = 12.6, 11.4, 3.0 Hz, H-4), 1.93 (1H, m, H-8α), 2.21 (1H, ddd, J =

11.4, 6.0, 4.2 Hz, H-5), 2.40 (1H, brdd, J = 17.4, 4.8 Hz, H-8β), 2.70 (2H, t, J = 7.2 Hz,

H-18), 3.24 (2H, m, H-17), 3.81 (1H, d, J = 15.0 Hz, H-14), 3.87 (1H, d, J = 15.0 Hz,

H-14), 4.80 (1H, brs, 14-OH), 4.89 (1H, s, H-13), 5.07 (1H, s, H-13), 6.45 (1H, brd, J =

4.2 Hz, H-6), 7.18 (2H, d, J = 7.8 Hz, H-20), 7.32 (2H, d, J = 7.8 Hz, H-21), 7.66 (1H, t,

13 J = 5.4 Hz, 16-NH); C NMR (125 MHz, DMSO-d6) δ 16.0 (C-9), 19.3 (C-11), 25.0

(C-8), 29.0 (C-2), 33.9 (C-3), 34.2 (C-1), 34.3 (C-18), 38.7 (C-10), 40.2 (C-17), 40.9

(C-5), 42.3 (C-4), 63.3 (C-14), 107.3 (C-13), 128.1 (C-21), 130.5 (C-20), 130.6 (C-22),

178

132.8 (C-7), 135.1 (C-6), 138.6 (C-19), 152.6 (C-12), 167.5 (C-15); (+)-LRESIMS m/z

+ + (rel. int.) 388 (100) [M + H] ; (+)-HRESIMS m/z 410.1865 (C23H30NClO2Na [M + Na] requires 410.1857).

Compound 92: Yield 5%, white gum;

25 []D – 78 (c 0.153, CHCl3); UV (CH3OH)

λmax nm (log ε) 203 (4.62), 225 (4.60), 278

1 (3.54), 284 (3.48); H NMR (500 MHz, DMSO-d6) δ 0.91 (3H, d, J = 7.2 Hz, H-11),

1.11 (1H, dddd, J = 13.2, 12.6, 12.6, 3.0 Hz, H-2β), 1.31 (1H, ddd, J = 13.2, 12.0, 2.4

Hz, H-3α), 1.35 (1H, m, H-9β), 1.37 (1H, m, H-2α), 1.56 (1H, brdd, J = 12.6, 6.0 Hz, H-

9α), 1.61 (1H, m, H-3β), 1.62 (1H, m, H-10), 1.72 (1H, m, H-1), 1.81 (1H, ddd, J =

12.0, 11.4, 2.4 Hz, H-4), 1.95 (1H, m, H-8α), 2.22 (1H, ddd, J = 11.4, 5.4, 4.8 Hz, H-5),

2.40 (1H, brdd, J = 17.4, 4.8 Hz, H-8β), 2.71 (2H, t, J = 7.8 Hz, H-18), 3.22 (2H, m, H-

17), 3.70 (3H, s, H-23), 3.82 (1H, d, J = 14.4 Hz, H-14), 3.87 (1H, d, J = 14.4 Hz, H-

14), 4.79 (1H, brt, J = 4.2 Hz, 14-OH), 4.90 (1H, s, H-13), 5.07 (1H, s, H-13), 6.45 (1H, brd, J = 4.8 Hz, H-6), 6.84 (2H, d, J = 7.8 Hz, H-21), 7.07 (2H, d, J = 7.8 Hz, H-20),

13 7.62 (1H, t, J = 5.4 Hz, 16-NH); C NMR (125 MHz, DMSO-d6) δ 16.1 (C-9), 19.3 (C-

11), 25.0 (C-8), 28.9 (C-2), 33.9 (C-3), 34.0 (C-1), 34.2 (C-18), 38.7 (C-10), 40.7 (C-

17), 40.8 (C-5), 42.3 (C-4), 54.9 (C-23), 63.3 (C-14), 107.3 (C-13), 113.7 (C-21), 129.5

(C-20), 131.4 (C-19), 132.8 (C-7), 135.1 (C-6), 152.6 (C-12), 157.6 (C-22), 167.4 (C-

15); (+)-LRESIMS m/z (rel. int.) 384 (100) [M + H]+; (+)-HRESIMS m/z 406.2339

+ (C24H33NO3Na [M + Na] requires 406.2353).

179

25 Compound 93: Yield 8%, white gum; []D

– 48 (c 0.087, CHCl3); UV (CH3OH) λmax nm

(log ε) 203 (4.06), 224 (4.00), 278 (3.02), 287

1 (2.88); H NMR (500 MHz, DMSO-d6) δ 0.91 (3H, d, J = 6.6 Hz, H-11), 1.11 (1H, dddd, J = 13.2, 13.2, 12.6, 2.4 Hz, H-2β), 1.30 (1H, m, H-3α), 1.35 (1H, m, H-9β), 1.37

(1H, m, H-2α), 1.56 (1H, m, H-9α), 1.61 (1H, m, H-3β), 1.62 (1H, m, H-10), 1.72 (1H, m, H-1), 1.81 (1H, ddd, J = 12.6, 11.4, 1.8 Hz, H-4), 1.95 (1H, m, H-8α), 2.21 (1H, ddd,

J = 11.4, 6.0, 4.8 Hz, H-5), 2.40 (1H, brdd, J = 17.4, 4.8 Hz, H-8β), 2.58 (2H, t, J = 7.8

Hz, H-18), 3.20 (2H, m, H-17), 3.82 (1H, d, J = 15.0 Hz, H-14), 3.87 (1H, d, J = 15.0

Hz, H-14), 4.90 (1H, s, H-13), 5.07 (1H, s, H-13), 6.43 (1H, brd, J = 4.8 Hz, H-6), 6.65

(2H, d, J = 7.8 Hz, H-21), 6.94 (2H, d, J = 7.8 Hz, H-20), 7.60 (1H, t, J = 5.4 Hz, 16-

13 NH); C NMR (125 MHz, DMSO-d6) δ 16.1 (C-9), 19.3 (C-11), 25.0 (C-8), 28.9 (C-2),

33.9 (C-3), 34.2 (C-1), 34.3 (C-18), 38.7 (C-10), 40.1 (C-17), 40.8 (C-5), 42.3 (C-4),

63.3 (C-14), 107.3 (C-13), 115.0 (C-21), 129.3 (C-20), 129.5 (C-19), 132.8 (C-7), 135.1

(C-6), 152.6 (C-12), 155.6 (C-22), 167.4 (C-15); (+)-LRESIMS m/z (rel. int.) 370 (100)

+ + [M + H] ; (+)-HRESIMS m/z 392.2199 (C23H31NO3Na [M + Na] requires 392.2196).

Compound 95: White gum, 0.4 mg; 1H

NMR (500 MHz, DMSO-d6) δ 0.91 (1H, d,

J = 8.4 Hz, H-11), 1.11 (1H, dddd, J = 12.6,

12.6, 10.2, 2.4 Hz, H-2β), 1.35 (1H, m, H-

3α), 1.37 (1H, m, H-2α), 1.39 (1H, m, 9β),

1.55 (1H, m, H-9α), 1.61 (1H, m, H-3β), 1.63 (1H, m, H-10), 1.72 (1H, m, H-1), 1.81

(1H, ddd, J = 12.0, 11.4, 2.4 Hz, H-4), 1.96 (1H, m, H-8α), 2.21 (1H, ddd, J = 11.4, 5.4,

3.6 Hz H-5), 2.41 (1H, dd, J = 17.4, 5.4 Hz, H-8β), 2.63 (2H, t, J = 7.2, H-18), 2.70

180

(2H, t, J = 7.2 Hz, H-28), 3.22 (2H, m, H-17), 3.31 (2H, m, H-27), 3.70 (3H, s, H-23),

3.75 (3H, s, H-33), 3.82 (1H, dd, J = 15.0, 4.8 Hz, H-14), 3.87 (1H, dd, J = 15.0, 4.8 Hz,

H-14), 4.90 (1H, s, H-13), 5.07 (1H, s, H-13), 6.43 (1H, brd, J = 3.6 Hz, H-6), 6.84 (4H, m, H-21, H-31), 7.08 (4H, m, H-20, H-30), 7.62 (1H, t, J = 5.4 Hz, 16-NH), 8.93 (1H,

13 brt, J = 4.8 Hz, 27-NH); C NMR (125 MHz, DMSO-d6) δ 16.1 (C-9), 19.3 (C-11),

25.0 (C-8), 29.0 (C-2), 33.5 (C-28), 33.9 (C-3), 34.2 (C-18), 34.3 (C-1), 38.8 (C-10),

40.6 (C-17), 40.7 (C-27), 40.9 (C-5), 42.3 (C-4), 52.7 (C-33), 54.9 (C-23), 63.3 (C-14),

107.4 (C-13), 113.7 (C-21), 113.7 (C-31), 129.5 (C-20), 129.5 (C-30), 130.8 (C-29),

131.5 (C-19), 132.8 (C-7), 135.1 (C-6), 152.6 (C-12), 156.7 (C-25), 157.6 (C-22), 157.7

(C-24), 161.1 (C-32), 167.4 (C-15); (+)-LRESIMS m/z (rel. int.) 589 (100) [M + H]+,

611 (100) [M + Na]+.

TBDPS protection of scaffold 62 and subsequent coupling with 4- methoxyphenethylamine. Compound 62 (163 mg, 0.6520 mmol) was dissolved in dry

CH2Cl2 (10 mL) before t-butyl(chloro)diphenylsilane (500 μL, 1.9600 mmol), Et3N

(275 μL, 1.9767 mmol), and DMAP (cat.) were added. The solution was stirred at rt under argon for 3.5 h before being filtered through paper. The solution was dried under reduced pressure to give a light brown crude (1.13 g). This crude was redissolved in

THF:H2O (18 mL, 2:1) and potassium carbonate (930 mg, 6.7391 mmol) added. The solution was stirred for 1 h at rt, then the THF removed under vacuum. The solution was cooled to 0 °C and HCl added until a pH of 4 was reached. More H2O was added

(10 mL) and the solution extracted with CH2Cl2 (2 × 30 mL). The organic phase was washed with brine, dried over MgSO4, then dried under vacuum to give a light brown oil (593.4 mg). This crude was purified by silica flash column chromatography and a

CH3OH/CH2Cl2 gradient.

181

After TLC and 1H NMR spectroscopic analysis of the fractions, product 96 was found in the 10% CH3OH/90% CH2Cl2 fractions (121.8 mg, 25%). A portion of 96 (9.8 mg) was coupled with 4-methoxyphenethylamine using (COCl)2 as outlined for compounds 82-84 and 91-93, above. Purification by SPE silica column chromatography and CH3OH/CH2Cl2 gave compound 97 (4.7 mg, 56%).

TBAF de-protection of compound 97. To a solution of 97 (4.7 mg, 0.0076 mmol) in dry THF (2 mL) was added a solution of tetrabutylammonium fluoride

(TBAF, 10 μL, 0.0010 mmol) in THF. The reaction was stirred for 24 h at rt under argon. The reaction mixture was poured into NaHCO3 (30 mL) and extracted with

CH2Cl2 (30 mL × 2). The organic phase was pre-adsorbed to silica and loaded onto a silica SPE cartridge (10 × 35 mm) that was flushed with 10 mL washes of CH2Cl2, 5%

CH2Cl2/95% CH3OH, 10% CH2Cl2/90% CH3OH, 20% CH2Cl2/80% CH3OH, 50%

CH2Cl2/50% CH3OH, then CH3OH. Product 92 was obtained impure in the 5%

CH2Cl2/95% CH3OH fraction. This mixture was re-purified off another silica SPE cartridge (10 × 35 mm) this time using flushes of n-hexane, 10% EtOAc/90% n-hexane, then a stepwise gradient from 20% EtOAc/80% n-hexane to EtOAc, then CH3OH (10 mL elutions). Compound 92 was obtained pure in the 60% n-hexane/40% EtOAc and

40% n-hexane/60% EtOAc washes (1 mg, 34%).

182

Second attempt at amine coupling and de-protection of 96 to give 92. A second portion of 96 (17 mg, 0.0234 mmol) was coupled with 4- methoxyphenethylamine and de-protected as outlined above, without purification steps until after de-protection. The crude obtained (28.6 mg) was purified by silica SPE column (10 × 35 mm) and flushes of n-hexane, 5% EtOAc/95% n-hexane, 10%

EtOAc/90% n-hexane, then a stepwise gradient from 20% EtOAc/80% n-hexane to

EtOAc, then CH3OH (10 mL elutions). Fractions 7-8 were found to contain 3.9 mg of the wanted product 92 (44% over both steps).

Procedure for carbamate formation using cinnamyl alcohol (98). Compound

98 (10 μL, 0.0746 mmol) was dissolved in dry CH2Cl2 (0.5 mL), and benzyl isocyanate

(15 μL, 0.1190 mmol) added followed by Et3N (cat. 10 μL). The solution was stirred at rt under Ar for 4 h then was quenched with CH3OH. The solution was dried under N2 and the crude loaded onto a SPE silica column (10 × 45 mm) that was washed with n- hexane, 5% EtOAc/95% n-hexane, 10% EtOAc/90% n-hexane, 20% EtOAc/80% n- hexane, EtOAc, then CH3OH in 10 mL elutions. Product 99 eluted in fractions 5-7 (16.5 mg, 83% yield). The reaction was repeated at 40 °C to give a yield of 68% of 99 (13.5 mg). The side product 101 was obtained in fraction 8 after purification (1.9 mg at rt, 6.3 mg at 40 °C). The same reaction and purification procedures were used to couple butyl isocyanate (14 μL, 0.1190 mmol) to cinnamyl alcohol to give 100 at a yield of 36% at rt and 20% at 40°C.

183

Compound 99: Yield 83%, white crystals; 1H

NMR (500 MHz, DMSO-d6) δ 4.21 (2H, d, J = 6.0

Hz, H-10), 4.66 (2H, d, J = 6.0 Hz, H-1), 6.37 (1H, dt, J = 15.6, 6.0 Hz, H-2), 6.65 (1H, d, J = 15.6 Hz, H-3), 7.23 (1H, t, J = 7.8 Hz, H-7), 7.24 (1H, t, J = 7.8 Hz, H-14), 7.25

(2H, d, J = 7.2 Hz, H-12), 7.31 (2H, dd, J = 7.8, 7.8 Hz, H-13), 7.33 (2H, dd, J = 7.8,

7.2 Hz, H-6), 7.43 (2H, d, J = 7.2 Hz, H-5), 7.77 (1H, t, J = 6.0 Hz, 9-NH); 13C NMR

(125 MHz, DMSO-d6) δ 43.8 (C-10), 64.2 (C-1), 124.8 (C-2), 126.3 (C-5), 126.7 (C-

12), 126.9 (C-14), 127.8 (C-7), 128.1 (C-13), 128.6 (C-6), 132.2 (C-3), 136.1 (C-4),

139.7 (C-11), 156.2 (C-8); (+)-LRESIMS m/z (rel. int.) 290 (100) [M + Na]+, 557 (100)

[2M + H]+.

Compound 100: Yield 36%, white crystals; δ

0.85 (3H, t, J = 7.8 Hz, H-13), 4.66 (2H, d, J =

6.0 Hz, H-1), 1.26 (2H, tq, J = 14.4, 7.8 Hz, H-12), 1.38 (2H, tt, J = 14.4, 12.6 Hz, H-

11), 2.98 (2H, td, J = 12.6, 6.0 Hz, H-10), 6.34 (1H, dt, J = 16.2, 6.0 Hz, H-2), 6.63 (1H, d, J = 16.2 Hz, H-3), 7.16 (1H, t, J = 6.0 Hz, 9-NH), 7.25 (1H, t, J = 7.2 Hz, H-7), 7.33

(2H, dd, J = 7.2, 7.2 Hz, H-6), 7.43 (2H, d, J = 7.2 Hz, H-5); 13C NMR (125 MHz,

DMSO-d6) δ 13.6 (C-13), 19.3 (C-12), 31.5 (C-11), 40.0 (C-10), 63.9 (C-1), 125.0 (C-

2), 126.3 (C-5), 127.7 (C-7), 128.6 (C-6), 132.0 (C-3), 136.1 (C-4), 155.9 (C-8); (+)-

LRESIMS m/z (rel. int.) 256 (100) [M + Na]+.

184

Compound 101: opaque gum; 1H NMR (500 MHz,

DMSO-d6) δ 4.30 (4H, brs, H-4), 5.04 (2H, s, H-9),

7.12 (4H, d, J = 7.8 Hz, H-6), 7.19 (2H, m, H-8),

7.19 (2H, m, H-11), 7.24 (4H, m, H-7), 7.24 (1H, m, H-13), 7.33 (2H, brt, J = 7.8, 7.2

13 Hz, H-12), 8.77 (2H, brs, 3-NH); C NMR (125 MHz, DMSO-d6) δ 43.5 (C-4), 44.9

(C-9), 126.1 (C-8), 126.1 (C-11), 126.8 (C-6), 128.1 (C-7), 128.1 (C-13), 128.3 (C-12),

138.1 (C-10), 139.3 (C-5), 156.0 (C-2); (+)-LRESIMS m/z (rel. int.) 396 (100) [M +

Na]+, 374 (100) [M + H]+.

Methylation of 62 using TMS-diazomethane.166 Scaffold 62 (12.8 mg, 0.0512 mmol) was dissolved in CH3OH:CH2Cl2 (1:1, 1 mL) before TMS-diazomethane (2.0 M in diethyl ether, 77 μL, 0.1536 mmol) was added dropwise. The reaction was stirred for

20 min at rt then quenched with AcOH (50 μL). The solvent was removed under reduced pressure to give 102 (13.7 mg, 99%).

25 Compound 102: White gum; []D – 134 (c 0.087, CHCl3);

1 UV (CH3OH) λmax nm (log ε) 202 (3.87), 222 (3.93); H

NMR (500 MHz, DMSO-d6) δ 0.89 (3H, d, J = 6.5 Hz, H-

11), 1.11 (1H, dddd, J = 12.5, 12.5, 12.0, 2.0 Hz, H-2β), 1.32 (1H, ddd, J = 13.0, 11.5,

2.5 Hz, H-3α), 1.35 (1H, m, H-2α), 1.37 (1H, m, H-9β), 1.56 (1H, m, H-9α), 1.62 (1H, m, H-3β), 1.63 (1H, m, H-10), 1.71 (1H, m, H-1), 1.83 (1H, ddd, J = 11.5, 11.5, 2.5 Hz,

H-4), 2.03 (1H, m, H-8α), 2.32 (1H, m, H-5), 2.37 (1H, dd, J = 18.5, 6.0 Hz, H-8β),

3.60 (3H, s, H-16), 3.79 (1H, d, J = 15.0 Hz, H-14), 3.86 (1H, d, J = 15.0 Hz, H-14),

4.90 (1H, s, H-13), 5.04 (1H, s, H-13), 6.83 (1H, m, H-6); 13C NMR (125 MHz, DMSO- d6) δ 15.9 (C-9), 19.2 (C-11), 25.0 (C-8), 28.9 (C-2), 33.7 (C-3), 34.0 (C-1), 38.4 (C-

185

10), 40.9 (C-5), 42.2 (C-4), 51.3 (C-16), 63.0 (C-14), 107.6 (C-13), 128.9 (C-7), 142.8

(C-6), 152.2 (C-12), 167.1 (C-15); (+)-LRESIMS m/z (rel. int.) 287 (100) [M + Na]+;

+ (+)-HRESIMS m/z 287.1606 (C16H24O3Na [M + Na] requires 287.1618).

General procedure for carbamate formation using scaffold 102.102 Scaffold

102 (~ 10 mg) was dissolved in dry CH2Cl2, and the relevant commercially available isocyanate (3 mmol excess) was added followed by Et3N (cat. 10 μL). The solution was stirred at rt under Ar for 16 h before being quenched with CH3OH then dried under N2.

Purification of carbamate products. After reaction with 3,4-(methylenedioxy)- phenyl isocyanate the crude obtained was loaded onto a SPE silica column (10 × 40 mm) that was washed with n-hexane, 10% EtOAc/90% n-hexane, 20% EtOAc/80% n- hexane, 50% EtOAc/50% n-hexane, 80% EtOAc/20% n-hexane, EtOAc, then CH3OH in 10 mL elutions. Fractions containing impure product by 1H NMR spectroscopy were further purified by semi-preparative diol HPLC. These fractions were injected onto a semi-preparative diol-bonded silica column at a flow rate of 9 mL/min and isocratic conditions of n-hexane for 10 min, followed by a linear gradient to 20% i-PrOH/80% n- hexane over 40 min, then isocratic conditions of 20% i-PrOH/80% n-hexane for 10 min.

Sixty fractions (60 × 1 min) were collected. Fractions 28-29 were found to contain 103

(6.9 mg, 49%). After coupling 102 with benzyl isocyanate, compound 104 was found to elute in the 20% EtOAc/80% n-hexane fraction when purified off a SPE silica column in the same manner as compound 103 (8.2 mg, 64%). Coupling of 102 with butyl isocyanate gave impure 105 which was purified in an identical way to compound 103, and eluted in fractions 19-20 off the diol-bonded silica HPLC column (9.7 mg, 76%).

186

The crude obtained from reaction of 102 with cyclohexyl isocyanate was first purified using a silica packed SPE column (10 × 40 mm) and elutions of n-hexane, 10%

EtOAc/90% n-hexane, 20% EtOAc/80% n-hexane, then CH3OH. Impure fractions were further purified by C18-bonded silica semi-preparative HPLC at a flow rate of 9 mL/min

(120 × 0.5 min fractions) and isocratic conditions of 10% CH3OH/90% H2O for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of

CH3OH for 10 min. Fractions 95-97 contained 106 (4.5 mg, 32%). Reaction of 102 with

4-methoxybenzyl isocyanate gave impure compound 107 which was pre-adsorbed to silica and loaded onto a silica flash column (15 × 40 mm) that was washed with 20%

EtOAc/80% n-hexane and collected in 1 mL fractions. Compound 107 eluted in fractions 20-26 (2.9 mg, 31%). Coupling with 3,4-dimethoxybenzyl isocyanate gave compound 108 which was purified in the same manner as 107, except that 25%

EtOAc/75% n-hexane was utilised. Compound 108 eluted into fractions 20-26 (8 mg,

69%).

Compound 103: Yield 49%, light yellow

27 gum; []D – 152 (c 0.067, CHCl3); UV

(CH3OH) λmax (log ε) 207 (3.60), 248

(3.03), 297 (2.58) nm; 1H NMR (500

MHz, DMSO-d6) δ 0.91 (3H, d, J = 6.5 Hz, H-11), 1.15 (1H, dddd, J = 12.5, 12.0, 12.0,

2.5 Hz, H-2β), 1.35 (1H, m, H-3α), 1.38 (1H, m, H-9β), 1.40 (1H, m, H-2α), 1.58 (1H, brdd, J = 12.0, 6.0 Hz, H-9α), 1.67 (1H, m, H-10), 1.68 (1H, m, H-3β), 1.74 (1H, m, H-

1), 1.99 (1H, ddd, J = 12.0, 11.0, 3.0 Hz, H-4), 2.04 (1H, m, H-8α), 2.35 (1H, m, H-5),

2.38 (1H, brdd, J = 18.5, 5.0 Hz, H-8β), 3.59 (3H, s, H-16), 4.49 (1H, d, J = 14.5 Hz, H-

14), 4.58 (1H, d, J = 14.5 Hz, H-14), 5.06 (1H, s, H-13), 5.11 (1H, s, H-13), 5.95 (2H, s,

H-22), 6.82 (1H, s, H-20), 6.83 (1H, brd, J = 7.0 Hz, H-24), 6.84 (1H, brd, J = 5.0 Hz,

187

13 H-6), 7.10 (1H, brs, H-25); C NMR (125 MHz, DMSO-d6) δ 15.9 (C-9), 19.2 (C-11),

25.0 (C-8), 28.7 (C-2), 33.2 (C-3), 33.9 (C-1), 38.3 (C-10), 40.5 (C-5), 42.9 (C-4), 51.3

(C-16), 65.0 (C-14), 100.7 (C-25), 100.8 (C-22), 108.0 (C-24), 111.0 (C-13), 111.1

(C-20), 129.2 (C-7), 133.4 (C-19), 142.3 (C-6), 142.4 (C-23), 147.0 (C-12), 147.1 (C-

21), 153.1 (C-17), 167.0 (C-15); (+)-LRESIMS m/z (rel. int.) 450 (100) [M + Na]+,

+ + 438 (100) [M + H] ; (+)-HRESIMS m/z 450.1880 (C24H29NO6Na [M + Na] requires

450.1887).

Compound 104: Yield 64%, opaque gum;

24 []D – 94 (c 0.08, CHCl3); UV (CH3OH)

λmax (log ε) 203 (4.31), 210 (4.30), 217

(4.22), 229 (4.04) nm; 1H NMR (500 MHz,

DMSO-d6) δ 0.90 (3H, d, J = 6.6 Hz, H-11), 1.11 (1H, dddd, J = 12.6, 12.6, 12.6, 2.4

Hz, H-2β), 1.34 (1H, m, H-3α), 1.37 (1H, m, H-2α), 1.38 (1H, m, H-9β), 1.58 (1H, brdd,

J = 12.6, 5.4 Hz, H-9α), 1.64 (1H, m, H-3β), 1.65 (1H, m, H-10), 1.71 (1H, m, H-1),

1.95 (1H, ddd, J = 12.0, 11.4, 3.0 Hz, H-4), 2.04 (1H, m, H-8α), 2.35 (1H, dd, J = 11.4,

5.4 Hz, H-5), 2.38 (1H, brdd, J = 18.0, 5.4 Hz, H-8β), 3.60 (3H, s, H-16), 4.17 (2H, d, J

= 6.0, H-19), 4.42 (1H, d, J = 13.8 Hz, H-14), 4.50 (1H, d, J = 13.8 Hz, H-14), 5.00

(1H, s, H-13), 5.05 (1H, s, H-13), 6.84 (1H, brd, J = 5.4 Hz, H-6), 7.22 (1H, t, J = 7.2

Hz, H-23), 7.23 (2H, d, J = 7.8 Hz, H-21), 7.30 (2H, dd, J = 7.2, 7.8 Hz, H-22), 7.78

13 (1H, t, J = 6.0 Hz, 18-NH); C NMR (125 MHz, DMSO-d6) δ 15.8 (C-9), 19.2 (C-11),

25.0 (C-8), 28.7 (C-2), 33.2 (C-3), 34.0 (C-1), 38.3 (C-10), 40.4 (C-5), 42.9 (C-4), 43.7

(C-19), 51.3 (C-16), 64.9 (C-14), 110.9 (C-13), 126.7 (C-23), 126.9 (C-21), 128.2 (C-

22), 129.1 (C-7), 139.7 (C-20), 142.5 (C-6), 147.3 (C-12), 156.1 (C-17), 167.1 (C-15);

(+)-LRESIMS m/z (rel. int.) 420 (100) [M + Na]+, 398 (100) [M + H]+ ; (+)-HRESIMS

+ m/z 398.2308 (C24H32NO4 [M + H] requires 398.2326). 188

Compound 105: Yield 76%, light yellow

29 gum; []D – 52 (c 0.073, CHCl3); UV

(CH3OH) λmax (log ε) 202 (4.41), 221 (4.43)

1 nm; H NMR (500 MHz, DMSO-d6) δ 0.85

(3H, t, J = 7.5 Hz, H-22), 0.91 (3H, d, J = 7.0 Hz, H-11), 1.12 (1H, dddd, J = 12.5,

12.5, 12.5, 2.5 Hz, H-2β), 1.25 (2H, m, H-21), 1.34 (1H, m, H-3α), 1.36 (2H, m, H-

20), 1.37 (1H, m, H-9β), 1.38 (1H, m, H-2α), 1.59 (1H, m, H-9α), 1.64 (1H, m, H-3β),

1.65 (1H, m, H-10), 1.73 (1H, m, H-1), 1.94 (1H, brdd, J = 11.5, 11.5 Hz, H-4), 2.04

(1H, m, H-8α), 2.34 (1H, m, H-5), 2.38 (1H, brdd, J = 18.5, 5.0 Hz, H-8β), 2.96 (2H, m,

H-19), 3.61 (3H, s, H-16), 4.37 (1H, d, J = 14.5 Hz, H-14), 4.45 (1H, d, J = 14.5 Hz, H-

14), 4.99 (1H, s, H-13), 5.03 (1H, s, H-13), 6.83 (1H, brd, J = 3.5 Hz, H-6), 7.12 (1H, t,

13 J = 5.5 Hz, 18-NH); C NMR (125 MHz, DMSO-d6) δ 13.6 (C-22), 15.9 (C-9), 19.2

(C-11), 19.3 (C-21), 25.0 (C-8), 28.7 (C-2), 31.4 (C-20), 33.2 (C-3), 34.0 (C-1), 38.3

(C-10), 39.0 (C-19), 40.4 (C-5), 42.9 (C-4), 51.3 (C-16), 64.5 (C-14), 110.6 (C-13),

129.1 (C-7), 142.5 (C-6), 147.4 (C-12), 155.8 (C-17), 167.0 (C-15); (+)-LRESIMS m/z

(rel. int.) 386 (100) [M + Na]+, 364 (100) [M + H]+ ; (+)-HRESIMS m/z 364.2498

+ (C21H34NO4 [M + H] requires 364.2482).

Compound 106: Yield 32%, yellow gum;

27 []D – 101 (c 0.120, CHCl3); UV (CH3OH)

1 λmax (log ε) 197 (3.13), 220 (3.41) nm; H

NMR (500 MHz, DMSO-d6) δ 0.91 (3H, d, J =

7.0 Hz, H-11), 1.04 (2H, m, H-22), 1.10 (1H, m, H-2β), 1.11 (2H, m, H-20), 1.18 (2H, m, H-21), 1.33 (1H, m, H-3α), 1.36 (1H, m, H-2α), 1.38 (1H, m, H-9β), 1.52 (2H, brd,

J = 12.0 Hz, H-22), 1.58 (1H, brdd, J = 12.5, 5.5 Hz, H-9α), 1.63 (1H, m, H-3β), 1.64

189

(2H, m, H-21), 1.66 (1H, m, H-10), 1.71 (2H, m, H-20), 1.72 (1H, m, H-1), 1.94 (1H, brdd, J = 11.5, 10.5 Hz, H-4), 2.04 (1H, m, H-8α), 2.34 (1H, m, H-5), 2.38 (1H, brdd, J

= 18.0, 5.0 Hz, H-8β), 3.22 (1H, m, H-19), 3.59 (3H, s, H-16), 4.36 (1H, d, J = 14.0 Hz,

H-14), 4.44 (1H, d, J = 14.0 Hz, H-14), 4.98 (1H, s, H-13), 5.02 (1H, s, H-13), 6.83

13 (1H, brs, H-6), 7.05 (1H, d, J = 8.0 Hz, 18-NH); C NMR (125 MHz, DMSO-d6) δ

15.9 (C-9), 19.2 (C-11), 24.5 (C-21), 25.0 (C-8), 25.1 (C-22), 28.7 (C-2), 32.6 (C-20),

33.2 (C-3), 34.0 (C-1), 38.3 (C-10), 40.4 (C-5), 42.9 (C-4), 49.4 (C-19), 51.3 (C-16),

64.4 (C-14), 110.5 (C-13), 129.1 (C-7), 142.5 (C-6), 147.4 (C-12), 155.0 (C-17), 167.0

(C-15); (+)-LRESIMS m/z (rel. int.) 413 (100) [M + Na]+, 390 (100) [M + H]+ ; (+)-

+ HRESIMS m/z 390.2634 (C23H36NO4 [M + H] requires 390.2639).

Compound 107: Yield 31%, opaque

27 gum; []D – 67 (c 0.093, CHCl3); UV

(CH3OH) λmax (log ε) 202 (4.07), 225

(4.00), 276 (2.88), 282 (2.82) nm; 1H

NMR (500 MHz, DMSO-d6) δ 0.91 (3H, d, J = 7.0 Hz, H-11), 1.11 (1H, brddd, J =

15.0, 14.0, 13.0 Hz, H-2β), 1.33 (1H, m, H-3α), 1.36 (1H, m, H-2α), 1.38 (1H, m, H-

9β), 1.58 (1H, m, H-9α), 1.63 (1H, m, H-3β), 1.65 (1H, m, H-10), 1.71 (1H, m, H-1),

1.95 (1H, brdd, J = 12.0, 10.5 Hz, H-4), 2.04 (1H, m, H-8α), 2.34 (1H, m, H-5), 2.38

(1H, brdd, J = 18.5, 5.0 Hz, H-8β), 3.61 (3H, s, H-16), 3.71 (3H, s, H-24), 4.10 (2H, d,

J = 6.0, H-19), 4.41 (1H, d, J = 14.0 Hz, H-14), 4.49 (1H, d, J = 14.0 Hz, H-14), 4.99

(1H, s, H-13), 5.04 (1H, s, H-13), 6.83 (1H, brs, H-6), 6.85 (2H, d, J = 8.5 Hz, H-22),

7.16 (2H, d, J = 8.5 Hz, H-21), 7.66 (1H, t, J = 6.0 Hz, 18-NH); 13C NMR (125 MHz,

DMSO-d6) δ 15.8 (C-9), 19.2 (C-11), 25.0 (C-8), 28.7 (C-2), 33.2 (C-3), 33.9 (C-1),

38.3 (C-10), 40.4 (C-5), 42.9 (C-4), 43.2 (C-19), 51.3 (C-16), 55.0 (C-24), 64.8 (C-14),

190

110.8 (C-13), 113.6 (C-22), 128.3 (C-21), 129.1 (C-7), 131.7 (C-20), 142.4 (C-6),

147.2 (C-12), 156.0 (C-17), 158.1 (C-23), 167.0 (C-15); (+)-LRESIMS m/z (rel. int.)

+ + 450 (100) [M + Na] , 428 (100) [M + H] ; (+)-HRESIMS m/z 428.2448 (C25H34NO5

[M + H]+ requires 428.2432).

Compound 108: Yield 69%, opaque

26 gum; []D – 79 (c 0.08, CHCl3); UV

(CH3OH) λmax (log ε) 202 (4.16), 226

(3.65), 278 (2.88) nm; 1H NMR (500

MHz, DMSO-d6) δ 0.90 (3H, d, J = 7.0 Hz, H-11), 1.10 (1H, brddd, J = 15.0, 13.5, 13.0

Hz, H-2β), 1.32 (1H, m, H-3α), 1.36 (1H, m, H-2α), 1.37 (1H, m, H-9β), 1.58 (1H, brdd,

J = 13.0, 5.0 Hz, H-9α), 1.64 (1H, m, H-3β), 1.65 (1H, m, H-10), 1.70 (1H, m, H-1),

1.95 (1H, brddd, J = 12.0, 11.5, 2.5 Hz, H-4), 2.04 (1H, m, H-8α), 2.33 (1H, m, H-5),

2.38 (1H, brdd, J = 17.5, 4.5 Hz, H-8β), 3.60 (3H, s, H-16), 3.71 (6H, s, H-26, H-27),

4.10 (2H, d, J = 6.0, H-19), 4.41 (1H, d, J = 14.5 Hz, H-14), 4.49 (1H, d, J = 14.5 Hz,

H-14), 4.99 (1H, s, H-13), 5.04 (1H, s, H-13), 6.83 (1H, brd, J = 4.5 Hz, H-6), 6.75 (1H, brd, J = 8.0 Hz, H-21), 6.85 (1H, s, H-25), 6.86 (1H, brd, J = 8.0 Hz, H-22), 7.67 (1H,

13 t, J = 6.0 Hz, 18-NH); C NMR (125 MHz, DMSO-d6) δ 15.8 (C-9), 19.2 (C-11), 25.0

(C-8), 28.7 (C-2), 33.2 (C-3), 33.9 (C-1), 38.3 (C-10), 40.4 (C-5), 42.9 (C-4), 43.5 (C-

19), 51.3 (C-16), 55.3 (C-27), 55.5 (C-26), 64.8 (C-14), 110.7 (C-13), 111.1 (C-25),

111.7 (C-22), 119.0 (C-21), 129.1 (C-7), 132.2 (C-20), 142.5 (C-6), 147.3 (C-23),

147.7 (C-24), 148.6 (C-12), 156.0 (C-17), 167.0 (C-15); (+)-LRESIMS m/z (rel. int.)

+ + 480 (100) [M + Na] , 458 (100) [M + H] ; (+)-HRESIMS m/z 458.2551 (C26H36NO6

[M + H]+ requires 458.2537).

191

Formation of di-substituted analogue of 62 (109). Scaffold 62 (36.2 mg,

0.1448 mmol) was first coupled with phenethylamine (187.0 μL, 1.4480 mmol) using the procedure outlined in the General procedure for amide formation using

155 (COCl)2, above. The crude product was purified using a silica flash column (15 × 50 mm) and a 20% stepwise gradient from n-hexane to EtOAc followed by CH3OH (40 mL per wash, collected as 10 mL fractions). After (+)-LRESIMS and 1H NMR spectroscopic analysis, fractions 10-14 were found to contain impure product. These fractions were pre-adsorbed to silica and packed onto five separate SPE silica columns

(10 × 40 mm) which were subjected to a 10% stepwise gradient from n-hexane to 40% n-hexane/60% EtOAc, followed by 20% n-hexane/80% EtOAc, EtOAc, then CH3OH

(10 mL elutions). Fractions 6-7 from each column were found to contain product 82 after 1H NMR spectroscopy (18.3 mg, 36%), so were recombined for the next reaction step. Compound 82 (18.3 mg, 0.0518 mmol) was then reacted with cyclohexyl isocyanate (132.2 μL, 1.0368 mmol) as outlined in the General procedure for carbamate formation,102 above. The crude obtained was wet loaded onto a silica flash column (15 × 60 mm) and flushed through using a 20% stepwise gradient from n-

1 hexane to EtOAc followed by CH3OH (40 mL per wash). After (+)-LRESIMS and H

NMR spectroscopic analysis, fractions 7-8 were found to contain product 109 (7.1 mg,

29%). The yield over both reaction steps was 10%.

Compound 109: Yield 10%, opaque

25 gum; []D – 95 (c 0.04, CHCl3); UV

(CH3OH) λmax nm (log ε) 204 (4.26),

219 (4.10), 233 (3.84); 1H NMR (600

MHz, DMSO-d6) δ 0.91 (3H, d, J = 6.6 Hz, H-11), 1.05 (1H, m, H-22'), 1.10 (1H, m, H-

2β), 1.14 (1H, m, H-20'), 1.19 (1H, m, H-21'), 1.32 (1H, m, H-9β), 1.36 (1H, m, H-3α),

192

1.37 (1H, m, H-2α), 1.52 (1H, m, H-22'), 1.57 (1H, m, H-9α), 1.62 (1H, m, H-21'), 1.64

(1H, m, H-3β), 1.66 (1H, m, H-10), 1.74 (1H, m, H-1), 1.75 (1H, m, H-20'), 1.83 (1H, dd, J = 11.4, 10.8 Hz, H-4), 2.00 (1H, m, H-8α), 2.19 (1H, brdd, J = 10.8, 4.8, 3.0 Hz,

H-5), 2.35 (1H, brdd, J = 18.0, 4.8 Hz, H-8β), 2.72 (2H, t, J = 7.8 Hz, H-18), 3.23 (1H, m, H-19'), 3.25 (2H, m, H-17), 4.33 (1H, d, J = 14.4 Hz, H-14), 4.42 (1H, d, J = 14.4

Hz, H-14), 4.99 (1H, s, H-13), 5.04 (1H, s, H-13), 6.43 (1H, brd, J = 3.0 Hz, H-6), 7.15

(2H, d, J = 7.2 Hz, H-20), 7.17 (1H, t, J = 7.8, H-22), 7.26 (2H, dd, J = 7.8, 7.2 Hz, H-

13 21), 7.68 (1H, t, J = 4.8 Hz, 16-NH); C NMR (125 MHz, DMSO-d6) δ 16.0 (C-9),

19.3 (C-11), 24.6 (C-21), 25.0 (C-22'), 25.1 (C-8), 28.9 (C-2), 32.6 (C-20), 33.3 (C-3),

34.1 (C-1), 35.1 (C-18), 38.7 (C-10), 40.5 (C-17), 41.3 (C-5), 42.5 (C-4), 49.5 (C-19),

65.0 (C-14), 109.4 (C-13), 125.9 (C-22), 128.2 (C-21), 128.5 (C-20), 133.1 (C-7), 134.6

(C-6), 139.5 (C-19), 148.3 (C-12), 155.1 (C-17), 167.7 (C-15); (+)-LRESIMS m/z (rel. int.) 501 (100) [M + Na]+, 479 (100) [M + H]+ ; (+)-HRESIMS m/z 501.306401

+ (C30H42N2O3Na [M + Na] requires 501.308764).

Compound purity analysis. Compounds 62, 68-72, 82-84, 91-93, and 102-108 were prepared at concentrations of 0.1 mg/100 μL in DMSO and injected (100 μL) onto a Phenomenex C18 Onyx monolithic column. HPLC fractionation conditions consisted of a linear gradient (curve #6) from 90% H2O (0.1% TFA)/10% CH3OH (0.1% TFA) to

50% H2O (0.1% TFA)/50% CH3OH (0.1% TFA) in 3 min at a flow rate of 4 mL/min, a convex gradient (curve #5) to CH3OH (0.1% TFA) in 3.50 min at a flow rate of 3 mL/min, held at CH3OH (0.1% TFA) for 0.50 min at a flow rate of 3 mL/min, held at

CH3OH (0.1% TFA) for a further 1.0 min at a flow rate of 4 mL/min, then a linear gradient (curve #6) back to 90% H2O (0.1% TFA)/10% CH3OH (0.1% TFA) in 1 min at a flow rate of 4 mL/min, then held at 90% H2O (0.1% TFA)/10% CH3OH (0.1% TFA) for 2 min at a flow rate of 4 mL/min.

193

Total run time for each analytical injection was 11 min. Compound purity was determined by extracting each chromatogram at 210 nm and integrating all UV peaks.

The percent purity and retention time for the compounds were as follows: 62 (90%, 6.22 min), 82 (95%, 6.52 min), 83 (90%, 5.89 min), 84 (100%, 5.88 min), 91 (93%, 6.72 min), 92 (96%, 6.51 min), 93 (97%, 6.17 min), 102 (90%, 6.53 min), 103 (100%, 6.92 min), 104 (95%, 6.90 min), 105 (96%, 6.89 min), 106 (97%, 7.03 min), 107 (99%, 6.85 min), 108 (98%, 6.73 min), 68 (90%, 4.46 min), 69 (92%, 6.29 min), 70 (93%, 5.96 min), 71 (97%, 6.19 min) and 72 (90%, 6.16 min).

Biological Experiments

P. falciparum growth inhibition assay.** As previously described, P. falciparum growth inhibition assays were carried out using an isotopic microtest.345

Briefly, ring-stage infected erythrocytes (0.5% parasitemia and 2.5% hematocrit) were seeded into triplicate wells of 96 well tissue culture plates containing serial dilutions of control (chloroquine) or test compounds and incubated under standard P. falciparum culture conditions. After 48 h, 0.5 μCi [3H]-hypoxanthine was added to each well after which the plates were cultured for a further 24 h. Cells were harvested onto 1450

MicroBeta filter mats (Wallac) and [3H] incorporation determined using a 1450

MicroBeta liquid scintillation counter. Percentage inhibition of growth compared to matched DMSO controls (0.5%) was determined and IC50 values were calculated using linear interpolation of inhibition curves.346 Chloroquine (Sigma Aldrich, C6628) was used as a positive control. The mean IC50 (± SD) was calculated over three independent experiments, each carried out in triplicate.

**Antimalarial and NFF cytotoxicity assays were undertaken by Vanida Choomuenwai and Dr. Katherine Andrews of the Eskitis Institute, Griffith University.

194

Cytotoxicity assays.347 Neonatal foreskin fibroblast (NFF) cells were cultured at

37 °C and 5% CO2 in RPMI 1640 media (Life Technologies, Inc., Rockville, MD) supplemented with 10% FCS (CSL Biosciences, Parkville, Victoria, Australia) and 1% streptomycin (Life Technologies, Inc., Rockville, MD; complete medium). Cells were maintained in log phase growth before being seeded (3000/well) into 96 well tissue culture plates (Corning, USA). Treatment was undertaken after 24 h of growth.

Compounds were dissolved in DMSO and diluted in complete medium; the DMSO concentration in the medium did not exceed 1%. An equivalent dose of DMSO was used to treat the control cells. The cells were washed with PBS and fixed in methylated spirits after three days of treatment. Total protein was determined using sulforhodamine

B as described previously.347 Compounds were tested in triplicate in three independent experiments.

LNCaP cytotoxicity assay.†† The library members generated were tested against the prostate cancer cell line LNCaP using a real time cell analyser

(xCELLigence, Roche Applied Systems). LNCaP cells were routinely grown in RPMI medium supplemented with 5% (v/v) FBS (Invitrogen, USA) at 37ºC in the presence of

3 5% CO2. These cells were seeded in a 96-well E-plate at a density of 10 × 10 cells in a final volume of 150 µL. The attachment of the cells was monitored for 4 h every 2 min.

After this period, the conditions of the cells were analysed every 1 h. After 24 h, the cells were treated with the compounds at a final concentration of 10 µM. The biological status of the cells were monitored for 2 h every minute and then for 70 h every hour.

Doxorubicin (5 µM) and DMSO (0.1%) were used as controls. The output curves were normalized at the last measurement time point prior to the addition of the drug.

†† LNCap cytotoxicity assay was undertaken by Michelle Liberio of the Eskitis Institute, Griffith University.

195

8.4. Chapter 4. Experimental

Analytical HPLC of E. mitchellii samples. Seven air dried and ground biota samples of E. mitchellii (200 mg) were packed separately into SPE cartridges before being washed with CH2Cl2 (8 mL) then CH3OH (8 mL). These fractions were combined, dried down, and resuspended in DMSO (800 μL). A sample of pure compound 62 (in 800 μL DMSO), the crude obtained from the original biota extraction

(QID005836, in 800 μL DMSO), and a standard (methyl 4-hydroxy benzoate, ethyl 4- hydroxy benzoate, benzophenone, and uracil all at 0.125 mg/mL in DMSO) were also prepared. All samples were consecutively injected (100 μL) onto a Phenomenex C18

Onyx monolithic column. HPLC fractionation conditions consisted of a linear gradient

(curve #6) from 90% H2O (0.1% TFA)/10% CH3OH (0.1% TFA) to 50% H2O (0.1%

TFA)/50% CH3OH (0.1% TFA) in 3 min at a flow rate of 4 mL/min, a convex gradient

(curve #5) to CH3OH (0.1% TFA) in 3.50 min at a flow rate of 3 mL/min, held at

CH3OH (0.1% TFA) for 0.50 min at a flow rate of 3 mL/min, held at CH3OH (0.1%

TFA) for a further 1.0 min at a flow rate of 4 mL/min, then a linear gradient (curve #6) back to 90% H2O (0.1% TFA)/10% CH3OH (0.1% TFA) in 1 min at a flow rate of 4 mL/min, then held at 90% H2O (0.1% TFA)/10% CH3OH (0.1% TFA) for 2 min at a flow rate of 4 mL/min. Total run time for each analytical injection was 11 min.

Compound 62 eluted in fractions 9-11, so these fractions from each run were analysed by (+)-LRESIMS.

196

Plant material (QID004133). The leaves of E. mitchellii Benth. (QID004133) were collected 5 km from Inglewood along Tobacco Road, QLD, Australia, in October,

1993. Collection and identification was undertaken by P. Forster and G. Guymer from the Queensland Herbarium. A voucher specimen (AQ600063) has been deposited at the

Queensland Herbarium, Brisbane, Australia.

Extraction and isolation (QID004133). The air-dried and ground leaves of E. mitchellii (QID004133, 5 g) were poured into a conical flask (1 L) and extracted with n- hexane (250 mL) for 2.5 h while being shaken at 200 rpm. The n-hexane extract was filtered under gravity then discarded. CH2Cl (250 mL) was added to the de-fatted sample and the mixture shaken at 200 rpm for 2.5 h before filtration. CH3OH (250 mL) was then added and the CH3OH/plant mixture shaken for a further 16 h at 200 rpm. This step was repeated for another 2.5 h. All CH2Cl/CH3OH extractions were combined and dried under reduced pressure to yield a dark brown gum (2 g). The crude extract was divided into ~ 500 mg portions, each of which was pre-adsorbed to diol-bonded silica, loaded onto a diol-bonded silica flash column (35 × 130 mm), and fractionated using a

20% stepwise gradient from n-hexane to EtOAc followed by CH3OH (100 mL washes).

Fractions 5 and 6 were pre-adsorbed to diol-bonded silica, packed into stainless steel cartridges (10 × 30 mm), and subjected to semi-preparative HPLC using a diol-bonded silica column (20 × 150 mm). Isocratic conditions of n-hexane were held for 10 min, then a linear gradient to 20% i-PrOH/80% n-hexane over 40 min was employed, followed by isocratic conditions of 20% i-PrOH/80% for 10 min, all at a flow rate of 9 mL/min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy.

197

Analytical HPLC of 40 Eremophila samples. A portion (200 mg) of each of the 40 air dried and ground Eremophila samples within the Nature Bank biota library that were labelled as either 'leaf' or 'aerial parts' were packed separately into SPE cartridges which were subsequently washed with CH2Cl2 (8 mL) then CH3OH (8 mL).

These fractions were combined, dried down and weighed and the resulting crudes made up into 2 mg/100 μL solutions in DMSO. All samples and a standard were successively injected (100 μL) onto a Phenomenex C18 Onyx monolithic column as detailed for the

E. mitchellii samples, above. Fractions 9-11 from all the runs were analysed by (+)-

LRESIMS.

Plant material (QID032241 and QID027518). E. drummondii F.Muell.

(QID032241) and the unidentified Eremophila sp. (QID027518) were both cultivated by

P. & A. Vaughn at Mt Cassel Plant Nursery, Pomonal, Victoria, Australia, and collected in October, 1998. Collection and identification were undertaken by P. Forster and R.

Booth from the Queensland Herbarium. Voucher specimens (AQ674466 and

AQ606490, respectively) are being held at the Queensland Herbarium, Brisbane,

Australia.

Extraction and isolation of QID032241 and QID027518. A portion (1 g each) of the air dried and ground E. drummondii and Eremophila sp. samples were extracted with successive washes of n-hexanes (100 mL, 2 h), CH2Cl2 (200 mL, 2 h, then 200 mL,

16 h), and CH3OH (200 mL, 2 h, then 200 mL, 64 h). The CH2Cl2 and CH3OH extracts were combined and dried down under vacuum. The crudes obtained (340 mg and 210 mg, respectively) were pre-adsorbed to C18-bonded silica, packed into stainless steel cartridges (10 × 30 mm), and subjected to semi-preparative HPLC using a C18-bonded silica column (21.2 × 150 mm) and a flow rate of 9 mL/min.

198

Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of CH3OH for 10 min. Sixty fractions (60 × 1 min) were collected. Fractions 40-53 from both HPLC runs were analysed by (+)-LRESIMS, then fractions 44 to 48 by 1H NMR spectroscopy. No scaffold 62 was isolated from QID032241, and < 0.5 mg of impure 62 was obtained from QID027518.

Plant material (Kalbar samples 1-3). The leaves and outer branches of the three E. mitchellii Benth. samples were collected from Kalbar, QLD, Australia, in

August, 2011. Collection and identification was undertaken by Barry Jahnke, and a voucher specimen (RAD042) is being held at the Eskitis Institute, Griffith University,

Brisbane, Australia.

Analytical HPLC of E. mitchellii samples (Kalbar samples 1-3). A portion

(300 mg) of the three air dried and ground E. mitchellii samples were packed separately into SPE cartridges which were subsequently washed with CH2Cl2 (8 mL) then CH3OH

(8 mL). These fractions were combined, dried down and weighed and the resulting crudes made up into 2 mg/100 μL solutions in DMSO. All samples as well as a standard were consecutively injected (100 μL) onto a Phenomenex C18 Onyx monolithic column as detailed for the E. mitchellii samples, above. Fractions 9-11 from all the runs were analysed by (+)-LRESIMS.

199

Extraction and isolation of Kalbar sample 1. A portion (5 g) of the air dried and ground leaves and terminal branches of E. mitchellii (Kalbar sample 1) was extracted with successive washes of CH2Cl2 (250 mL, 1.5 h, then 250 mL, 16 h, then

250 mL, 2.5h) and CH3OH (250 mL, 2.5 h, then 250 mL, 1.5 h, then 250 mL, 16 h) while being shaken at 150 rpm. The CH2Cl2 and CH3OH fractions were combined and dried down under vacuum to give a dark brown crude extract (2.05 g). This extract was pre-adsorbed to diol-bonded silica, loaded onto a diol-bonded silica flash column (45 ×

90 mm) which was flushed with successive washes of n-hexane, 50:50 n- hexane:EtOAc, EtOAc, then CH3OH (750 mL washes). The EtOAc fraction was pre- adsorbed to diol-bonded silica and packed into a stainless steel cartridge (10 × 30 mm) before being subjected to semi-preparative HPLC using a diol-bonded silica column (20

× 150 mm). Isocratic conditions of n-hexane were held for 10 min, followed by a linear gradient to 20% i-PrOH/80% n-hexane over 40 min, followed by isocratic conditions of

20% i-PrOH/80% for 10 min, all at a flow rate of 9 mL/min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy.

Compound 62 was found impure in fractions 36-37 (7.5 mg, ~ 60% pure, 0.15% dry weight).

Plant material (QID009029, QID009030, QID009031). Three samples of the same E. sturtii plant (roots [QID009029], mixed [QID009030], and wood

[QID009031]) were collected 1 km north of the Currawinya National Park boundary on the road to Eulo, QLD, Australia, in August, 1995. Collection and identification were undertaken by D.A. Halford from the Queensland Herbarium. A voucher specimen

(AQ602014) has been deposited at the Queensland Herbarium, Brisbane, Australia.

200

Extraction and isolation of QID009029, QID009030, and QID009031. A portion (1 g) of each of the three air dried and ground E. sturtii samples were extracted with successive washes of CH2Cl2 (250 mL, 1.5 h, then 250 mL, 1.5 h) and CH3OH

(250 mL, 16 h, then 250 mL, 4 h) while being shaken at 150 rpm. The CH2Cl2 and

CH3OH fractions were combined and dried down under vacuum. A portion of each crude was set aside for analytical HPLC. The remaining extracts were pre-adsorbed to diol-bonded silica, packed into stainless steel cartridges (10 × 30 mm), and subjected to semi-preparative HPLC using a diol-bonded silica column (20 × 150 mm) at a flow rate of 9 mL/min and isocratic conditions of n-hexane for 10 min, followed by a linear gradient to 20% i-PrOH/80% n-hexane over 40 min, then isocratic conditions of 20% i-

PrOH/80% n-hexane for 10 min. Sixty fractions (60 × 1 min) were collected, then fractions 25-60 from each run analysed by (+)-LRESIMS and 1H NMR spectroscopy.

The mixed (QID009030) and root (QID009029) samples were found to contain 62 in fractions 36-38 at 6.5 mg (0.63% dry wt) and 1 mg (0.10% dry wt), respectively.

Mitchellenes B-E (68-72) were found in fractions 25 (69, 7.9 mg), 29 (70, impure, 2.5 mg), 34 (71, 2.3 mg), and 35 (72, impure, 2.1 mg). Centaureidin (74, impure, 2.6 mg) was found in fractions 58 and 59 and verbascoside in fraction 30 (110, 15 mg).

Analytical HPLC of E. sturtii samples (QID009029, QID009030,

QID009031). A portion of the crude extracts of the three E. sturtii samples were made up into 2 mg/100 μL solutions in DMSO. All samples as well as a standard were successively injected (100 μL) onto a Phenomenex C18 Onyx monolithic column, as detailed above for the E. mitchellii samples. Fractions 9-11 from all the runs were analysed by (+)-LRESIMS.

201

8.5. Chapter 6. Experimental

8.5.1. E. eriocalyx

Plant material. E. eriocalyx F.Muell. was cultivated by P. & A. Vaughn at Mt

Cassel Plant Nursery, Pomonal, Victoria, Australia and collected from there in October,

1998, by P.I. Forster and R. Booth of the Queensland Herbarium. A voucher specimen

(AQ606401) has been deposited at the Queensland Herbarium, Brisbane, Australia.

Extraction and isolation. The air-dried and ground plant E. eriocalyx (10 g) was poured into a conical flask (1 L), CH2Cl2 (250 mL) was added and the flask was shaken at 150 rpm for 3 h. The resulting extract was filtered under gravity, and set aside. Another volume of CH2Cl2 (250 mL) was added to the plant sample, and the flask shaken at 150 rpm for 1.5 h before filtration. CH3OH (250 mL) was then added and the

CH3OH/plant mixture shaken for 16 h at 150 rpm. Following gravity filtration, the plant material was extracted with another volume of CH3OH (250 mL), while being shaken at

150 rpm for 3 h. All CH2Cl2/CH3OH extractions were combined and dried under reduced pressure to yield a brown gum (2.51 g). This crude extract was divided into three portions, each of which was pre-adsorbed to C18-bonded silica (~ 7 g) and then loaded onto a C18-bonded silica flash column (30 × 150 mm). The extract was first fractionated off the columns using 10% CH3OH/90% H2O, followed by a 20% stepwise gradient from 20% CH3OH/80% H2O to CH3OH (100 mL washes). Like fractions from each column were combined, dried under vacuum, and then analysed by 1H NMR spectroscopy. Fractions 1-2 (20 mg and 70 mg, respectively) were found to predominantly contain mannitol (179) and geniposidic acid (155) by (+)-LRESIMS and

1H NMR spectroscopy. These fractions were not pursued further.

202

Fraction 3 (660 mg) was pre-adsorbed to C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a

C18-bonded silica column (21.2 mm × 150 mm) at a flow rate of 9 mL/min. Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to 60% CH3OH/40% H2O over 40 min, then isocratic conditions of 60%

CH3OH/40% H2O for 10 min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy. Fraction 4 contained mannitol (179), fractions 5-7 geniposidic acid-Na (180), fractions 20-21 geniposidic acid (155), and fractions 38-40 verbascoside (110). The masses and % dry weights of these compounds are given below, after they were combined with like fractions from other HPLC runs.

Fraction 36 contained impure mussaenoside (181, 7 mg), whose purification is described below. Fraction 46 contained impure ladroside (182, 10.9 mg), which was loaded onto a LH-20 Sephadex size exclusion column (35 × 310 mm) and flushed through with 1:1 CH2Cl2:CH3OH. Initially 100 mL was collected into a conical flask and discarded. Twenty fractions were then collected at 10 mL each. Fractions 8-10 were found to contain pure ladroside (182, 4.1 mg, 0.041% dry wt) after (+)-LRESIMS and 1H NMR spectroscopic analyses.

Ladroside (182): brown gum;

25 []D - 54 (c 0.055, CH3OH);

1H and 13C NMR data (DMSO-

d6) see Table 20; (+)-LRESIMS m/z (rel. int.) 575 (100) [M + Na]+.

203

Fraction 4 from the C18-bonded silica flash column (790 mg) was pre-adsorbed to C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a C18-bonded silica column (21.2 mm × 150 mm) at a flow rate of 9 mL/min. Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to 80% CH3OH/20% H2O over 40 min, then isocratic conditions of 80% CH3OH/20% H2O for 10 min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy. Fraction 4 contained mannitol (179), fractions 5-7 geniposidic acid-Na (180), fractions 19-20 geniposidic acid (155), and fraction 34 verbascoside (110). Fraction 32 was found to contain impure mussaenoside (181), whose purification is described below.

Fraction 5 from the C18-bonded silica flash column (350 mg) was pre-adsorbed to C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a C18-bonded silica column (21.2 mm × 150 mm) at a flow rate of 9 mL/min. Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to CH3OH in 40 min, then isocratic conditions of

CH3OH for 10 min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-

LRESIMS and 1H NMR spectroscopy. Fraction 29 was found to contain impure mussaenoside (181, 5.4 mg). The fractions of impure mussaenoside obtained from each of the flash column fraction purifications were recombined and loaded onto a LH-20

Sephadex size exclusion column (35 × 320 mm) and flushed through with CH3OH.

Initially 100 mL was collected into a conical flask and discarded. Twenty fractions were then collected at 10 mL each. Fractions 13-14 were found to contain pure mussaenoside

(181, 7.6 mg, 0.076% dry wt) after (+)-LRESIMS and 1D/2D NMR spectroscopic analysis. HPLC fractions 45-48 were separately injected onto a diol-bonded silica column (20 × 150 mm) and subjected to semi-preparative HPLC.

204

Isocratic conditions of n-hexane were held for 10 min, followed by a linear gradient to 20% i-PrOH/80% n-hexane over 40 min, then isocratic conditions of 20% i-

PrOH/80% n-hexane for 10 min, all at a flow rate of 9 mL/min. Sixty fractions (60 × 1 min) were collected for each HPLC run, dried down, then analysed by (+)-LRESIMS and 1H NMR spectroscopy. Like fractions from the different HPLC runs were recombined to give 3,15-epoxy-19-hydroxycembra-7,11-dien-18-oic acid (188, 0.6 mg,

0.006% dry wt) and 5,19-dihydroxy-3,14-viscidadien-20-oic acid (186, 0.6 mg, 0.006% dry wt), which converted to the ether acid 187 after prolonged storage.

25 Mussaenoside (181): brown gum; []D - 31 (c

1 13 0.07, CH3OH); H and C NMR data (DMSO-d6)

see Table 19; (+)-LRESIMS m/z (rel. int.) 413

(100) [M + Na]+, 803 (100) [2M + H]+.

3,15-Epoxy-19-hydroxycembra-7,11-dien-18-oic acid (188):

25 1 13 brown gum; []D + 8 (c 0.04, CHCl3); H and C NMR data

(DMSO-d6) see Table 22; (+)-LRESIMS m/z (rel. int.) 359 (100)

[M + Na]+, 695 (100) [2M + H]+.

205

5,19-Dihydroxy-3,14-viscidadien-20-oic acid (186): opaque

1 13 gum; H and C NMR data (DMSO-d6) see Table 21; (+)-

LRESIMS m/z (rel. int.) 359 (100) [M + Na]+, 695 (100) [2M +

H]+; (-)-LRESIMS m/z (rel. int.) 335 (100) [M - H]-.

Ether acid (187): opaque gum; (+)-LRESIMS m/z (rel. int.) 319

(100) [M + H]+.

The total amount and % dry weights of the known compounds that eluted across all the flash column purifications were as follows: mannitol (179, 51.5 mg, 0.515% dry weight), geniposidic acid-Na (180, 15.9 mg, 0.159% dry weight), geniposidic acid (155,

37.2 mg, 0.372% dry weight), and verbascoside (110, 36.3 mg, 0.363% dry wt).

Mannitol (179): Brown gum; (+)-LRESIMS m/z (rel. int.) 205

(100) [M + Na]+, 387 (100) [2M + H]+.

25 Geniposidic acid (155): brown gum; []D + 8 (c

1 13 0.04, CH3OH); H and C NMR data (DMSO-d6)

see Table 17; (+)-LRESIMS m/z (rel. int.) 397 (100)

[M + Na]+, 771 (100) [2M + H]+.

206

27 Geniposidic acid-Na (180): brown gum; []D +

10 (c 0.03, CH3OH); (+)-LRESIMS m/z (rel. int.)

397 (100) [M + Na]+, 771 (100) [2M + H]+.

Verbascoside (110): brown gum;

27 1 []D - 68 (c 0.05, CH3OH); H

13 and C NMR data (DMSO-d6) see Table 18; (+)-LRESIMS m/z (rel. int.) 647 (100) [M + Na]+.

8.5.2. E. linsmithii

Plant material. E. linsmithii R. J. F. Hend. was cultivated by P. & A. Vaughn at

Mt Cassel Plant Nursery, Pomonal, Victoria, Australia and collected from there in

October, 1998, by P.I. Forster and R. Booth of the Queensland Herbarium. A voucher specimen (AQ606028) has been deposited at the Queensland Herbarium, Brisbane,

Australia.

Extraction and isolation. The air-dried and ground plant E. linsmithii (10 g) was poured into a conical flask (1 L), CH2Cl2 (250 mL) was added and the flask was shaken for 3 h. The resulting extract was filtered under gravity, and set aside. Another volume of CH2Cl2 (250 mL) was added to the plant sample, and the flask shaken for 1.5 h before filtration. CH3OH (250 mL) was then added and the CH3OH/plant mixture shaken for 16 h. Following gravity filtration, the plant material was extracted with another volume of CH3OH (250 mL) for 3 h. All the extractions were shaken at 150 rpm.

207

All CH2Cl2/CH3OH extractions were combined and dried under reduced pressure to yield a brown gum (3.62 g). This crude extract was divided into four portions, each of which was pre-adsorbed to C18-bonded silica (~ 7 g) and then loaded onto a C18-bonded silica flash column (30 × 150 mm). The extract was first fractionated off the columns using 10% CH3OH/90% H2O, followed by a 20% stepwise gradient from 20% CH3OH/80% H2O to CH3OH (100 mL washes). Like fractions from each column were combined, dried under vacuum, and then analysed by 1H NMR spectroscopy. Fractions 1-2 (80 mg and 150 mg, respectively) were found to contain simple sugar molecules, and were not pursued further.

Fraction 3 (250 mg) was pre-adsorbed to C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a

C18-bonded silica column (21.2 mm × 150 mm) and a flow rate of 9 mL/min. Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to 60% CH3OH/40% H2O over 40 min, then isocratic conditions of 60%

CH3OH/40% H2O for 10 min. Sixty fractions (60 × 1 min) were collected then analysed by (+)-LRESIMS and 1H NMR spectroscopy. Fraction 4 contained mannitol (179) and fraction 6 geniposidic acid-Na (180, 2.6 mg, 0.026% dry weight). The mass and % dry weight of mannitol is given below, after it was combined with like fractions from other

HPLC runs. Fraction 36 contained impure mussaenoside (1.5 mg), but was not purified further.

Fraction 4 from the C18-bonded silica flash column (1.38 g) was divided in two, each portion of which was pre-adsorbed to C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a C18- bonded silica column (21.2 mm × 150 mm) and a flow rate of 9 mL/min.

208

Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to 80% CH3OH/20% H2O over 40 min, then isocratic conditions of

80% CH3OH/20% H2O for 10 min. Sixty fractions (60 × 1 min) were collected for each run, and like fractions were combined, dried under vacuum, then analysed by (+)-

LRESIMS and 1H NMR spectroscopy. Fraction 5 contained mannitol (179) and fraction

33 verbascoside (110). Impure 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-

1 D-glucopyranoside (193) could be seen in fraction 31 by H NMR spectroscopy. The purification of 193 is described below.

Fraction 5 from the C18-bonded silica flash column (1.20 g) was divided in two, and each portion pre-adsorbed to C18-bonded silica, packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a C18-bonded silica column (21.2 mm × 150 mm) and a flow rate of 9 mL/min. Isocratic conditions of

10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of CH3OH for 10 min. Sixty fractions (60 × 1 min) were collected for each HPLC run, with like fractions being combined and dried under vacuum before being analysed by (+)-LRESIMS and 1H NMR spectroscopy.

Fraction 4 contained mannitol (179) and fraction 30 verbascoside (110). Fraction 52 was found to contain 3,15-epoxycembra-7,11-dien-18-oic acid (191, 1.5 mg, 0.015% dry wt).

209

Impure 3-methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-D- glucopyranoside was found in fraction 28, and was combined with the fraction obtained from the purification of flash column fraction 4, above. The impure material was loaded onto a LH-20 Sephadex size exclusion column (35 × 320 mm) and flushed through with

CH3OH. Initially 100 mL was collected into a conical flask and discarded. Twenty-nine fractions were then collected at 10 mL each. Fractions 8-9 were found to contain pure 3- methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-D-glucopyranoside (193, 9.6 mg,

0.096% dry wt) after (+)-LRESIMS and 1D/2D NMR spectroscopic analysis.

3,15-Epoxycembra-7,11-dien-18-oic acid (191): brown gum;

27 1 13 []D - 28 (c 0.015, CHCl3); H and C NMR data (DMSO-d6)

see Table 23; (+)-LRESIMS m/z (rel. int.) 343 (100) [M + Na]+,

663 (100) [2M + H]+.

3-Methylbut-3-enyl-O-α-L-rhamno-

pyranosyl-(1→6)-O-β-D-glucopyranoside

25 (193): opaque gum; []D - 67 (c 0.04,

CH3OH); UV (CH3OH) λmax nm (log ε) 231 (2.74), 244 (2.58); IR νmax (KBr) 3407,

-1 1 13 2094, 1653, 1378, 1069, 1045, 1025, 985 cm ; H and C NMR data (DMSO-d6) see

Table 24; (+)-LRESIMS m/z (rel. int.) 417 (100) [M + Na]+, 811 (100) [2M + Na]+; (+)-

+ HRESIMS m/z 417.174828 (C17H30O10Na [M + Na] requires 417.173118).

The total amount and % dry weights of the known compounds that eluted across all the flash column purifications were as follows: mannitol (179, 41.2 mg, 0.412% dry weight) and verbascoside (110, 130.6 mg, 1.31% dry wt).

210

8.6. Chapter 7. Experimental

Plant material. E. microtheca was cultivated and identified by Jan Glazebrook at Logan Village, Queensland, Australia and the leaves and outer branches harvested in

March 2011. A voucher specimen (RAD039) has been deposited at the Eskitis Institute,

Griffith University, Brisbane, Australia.

Extraction and isolation. The air-dried and ground leaves and outer branches of E. microtheca (10.4 g) were poured into a conical flask (1 L), then extracted with four successive washes of CH2Cl2 (250 mL each), the first three washes for 2.5 h each and the last for 16 h, while being shaken at 150 rpm. After each wash the resulting extract was filtered under gravity, and set aside. Four CH3OH extractions were then undertaken (250 mL each), the first two washes for 2.5 h, the third for 24 h, and the fourth for 40 h, all while being shaken at 150 rpm. After each wash the resulting extract was filtered under gravity, and set aside. All CH2Cl2/CH3OH extractions were combined and dried under reduced pressure to yield a brown gum (3.26 g). This crude extract was divided into quantities of ~ 300 mg, and each portion pre-adsorbed to C18-bonded silica

(~ 1 g), packed into stainless steel cartridges (10 × 30 mm), and subjected to semi- preparative HPLC using a C18-bonded silica column (21.2 × 150 mm). Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of CH3OH for 10 min, all at a flow rate of 9 mL/min. Sixty fractions (60 × 1 min) were collected. Ten HPLC runs were completed. Like fractions were then combined and dried under reduced pressure.

Fractions 25-60 were analysed by (+)-LRESIMS and 1H NMR spectroscopy.

211

Fraction 30 contained verbascoside (110, 43.5 mg, 0.418% dry wt), fractions 41 and 42 impure jaceosidin (201), fractions 46-47 3,7,8-trihydroxyserrulat-14-en-19-oic acid (199, 54 mg, 0.519 % dry wt), fraction 51 3-acetoxy-7,8-dihydroxyserrulat-14-en-

19-oic acid (198, 66.4 mg, 0.638% dry wt), and fraction 52 3,19-diacetoxy-8- hydroxyserrulat-14-ene (200, 452.9 mg, 4.355% dry wt). Fractions 41 and 42 were loaded onto a LH-20 Sephadex size exclusion column (35 × 310 mm) and flushed through with CH2Cl2:CH3OH (1:1). Initially 100 mL was collected into a conical flask and discarded. Twenty-five fractions were then collected at 10 mL each. Fractions 9 and

10 were found to contain pure jaceosidin (201, 9.2 mg, 0.088% dry wt) after (+)-

LRESIMS and 1H NMR spectroscopic analysis.

3-Acetoxy-7,8-dihydroxyserrulat-14-en-19-oic acid

25 (198): brown gum; [α]D + 6 (c 0.033, CH3OH); UV

(CH3OH) λmax nm (log ε) 215 (4.28), 257 (3.75), 320

(3.27); IR νmax (KBr) 3262 (br), 2961, 2928, 2874, 2565,

1731, 1667, 1618, 1481, 1434, 1371, 1291, 1245, 1026,

-1 1 13 999 cm ; H and C NMR data (DMSO-d6) see Table 25; (-)-LRESIMS m/z (rel. int.)

389 (100) [M - H]-; (+)-LRESIMS m/z (rel. int.) 413 (100) [M + Na]+; (+)-HRESIMS

+ m/z 413.1955 (C22H30O6Na [M + Na] requires 413.1935).

212

3,7,8-Trihydroxyserrulat-14-en-19-oic acid (199): brown

25 gum; [α]D + 9.1 (c 0.044, CH3OH); UV (CH3OH) λmax nm

(log ε) 216 (4.35), 225 (4.26), 259 (3.91), 325 (3.38); IR νmax

(KBr) 3267 (br), 2960, 2929, 2558, 1664, 1614, 1512, 1433,

1377, 1292, 1219, 1025, 998 cm-1; 1H and 13C NMR data

- (DMSO-d6) see Table 26; (-)-LRESIMS m/z (rel. int.) 347 (100) [M - H] ; (+)-

+ HRESIMS m/z 371.1844 (C20H28O5Na [M + Na] requires 371.1829).

3,19-Diacetoxy-8-hydroxyserrulat-14-ene (200):

25 brown gum; [α]D + 5.5 (c 0.037, CH3OH); UV

(CH3OH) λmax nm (log ε) 207 (3.83), 227 (3.42), 260

(2.94), 287 (2.62), 324 (2.45); IR νmax (KBr) 3262 (br),

2960, 2931, 2360, 2342, 1733, 1668, 1618, 1586,

-1 1 13 1456, 1433, 1374, 1292, 1243, 1026, 1003 cm ; H and C NMR data (DMSO-d6) see

Table 27; (-)-LRESIMS m/z (rel. int.) 401 (100) [M - H]-; (+)-LRESIMS m/z (rel. int.)

+ + 425 (100) [M + Na] ; (+)-HRESIMS m/z 425.2319 (C24H34O5Na [M + Na] requires

425.2298).

Verbascoside (110): brown gum;

27 1 []D - 68 (c 0.05, CH3OH); H

13 and C NMR data (DMSO-d6) see Table 18; (+)-LRESIMS m/z (rel. int.) 647 (100) [M + Na]+.

213

Jaceosidin (201): yellow gum; 1H and 13C NMR data

(DMSO-d6) see Table 28; (+)-LRESIMS m/z (rel. int.)

331 (100) [M + H]+.

Acetylation of 199.344 Serrulatane 199 (23.9 mg, 0.0687 mmol) was dissolved in

Ac2O (1 mL) and dry pyridine (1 mL) and stirred at rt overnight. The solution was dried under N2 before being pre-adsorbed to C18-bonded silica (~ 1 g), packed into a stainless steel cartridge (10 × 30 mm), and subjected to semi-preparative HPLC using a C18- bonded silica column (21.2 × 150 mm). Isocratic conditions of 10% CH3OH/90% H2O were held for 10 min, followed by a linear gradient to CH3OH over 40 min, then isocratic conditions of CH3OH for 10 min all at a flow rate of 9 mL/min. Sixty fractions

(60 × 1 min) were collected. After 1H NMR spectroscopic analysis of the main UV peak in the HPLC chromatogram, compound 205 was found to have eluted in fractions 40-43

(15.7 mg, 53%).

214

3,8-Diacetoxy-7-hydroxyserrulat-14-en-19-oic acid

26 (205): opaque gum; [α]D + 17.2 (c 0.07, CH3OH); UV

(CH3OH) λmax nm (log ε) 211 (3.64), 246 (2.94), 309

1 (2.59); H NMR (600 MHz, DMSO-d6) δH 0.44 (3H, d, J

= 6.6 Hz, H-18), 1.14 (3H, d, J = 6.6 Hz, H-20), 1.23 (1H,

ddt, J = 13.2, 6.6, 6.0 Hz, H-12), 1.39 (1H, ddt, J = 13.2,

7.2, 6.0 Hz, H-12), 1.56 (1H, m, H-2β), 1.57 (3H, s, H-17), 1.65 (3H, s, H-16), 1.93

(1H, m, H-11), 1.96 (2H, brd, J = 6.6 Hz, H-13), 2.00 (3H, s, H-22), 2.03 (1H, m, H-2α),

2.22 (3H, s, H-24), 2.94 (1H, d, brd, J = 3.0 Hz, H-4), 3.01 (1H, qd, J = 6.6, 6.0 Hz, H-

1), 5.11 (1H, t, J = 7.8 Hz, H-14), 5.18 (1H, ddd, J = 12.0, 4.8, 3.0 Hz, H-3), 7.24 (1H,

13 s, H-5); C NMR (125 MHz, DMSO-d6) δC 17.5 (C-17), 18.9 (C-18), 20.4 (C-24), 21.0

(C-22), 22.5 (C-20), 25.4 (C-13), 25.5 (C-16), 28.3 (C-1), 31.3 (C-2), 31.4 (C-11), 37.9

(C-12), 44.4 (C-4), 69.7 (C-3), 109.5 (C-6), 122.4 (C-10), 124.4 (C-14), 127.0 (C-5),

130.7 (C-15), 136.1 (C-9), 136.5 (C-8), 153.4 (C-7), 168.1 (C-23), 169.7 (C-21), 171.0

(C-19); (-)-LRESIMS m/z (rel. int.) 431 (100) [M - H]-; (-)-HRESIMS m/z 431.2077

- (C24H31O7 [M - H] requires 431.2075).

Methylation of 199.166 Serrulatane 199 (23.9 mg, 0.0687 mmol) was dissolved in CH3OH:CH2Cl2 (1:1, 2 mL) before TMS-diazomethane (2.0 M in diethyl ether, 172

μL, 0.3435 mmol) was added dropwise. The reaction was stirred for 20 min at rt then quenched with AcOH (50 μL). The solvent was removed under N2 and the reaction crude purified in the same manner as that of compound 205, above. After 1H NMR spectroscopic analysis of the main UV peaks in the HPLC chromatogram, compound

206 was found to have eluted in fraction 49 (2.4 mg, 9%) and compound 207 in fractions 50-51 (8.8 mg, 35%).

215

3,7,8-Trihydroxyserrulat-14-en-methyl-19-benzoate

25 (206): brown gum; [α]D + 3.2 (c 0.063, CH3OH); UV

(CH3OH) λmax nm (log ε) 217 (4.39), 227 (4.33), 262 (4.06),

1 328 (3.48); H NMR (600 MHz, DMSO-d6) δH 0.36 (3H, d, J

= 7.2 Hz, H-18), 1.17 (3H, d, J = 6.6 Hz, H-20), 1.25 (1H, m,

H-12), 1.45 (1H, brd, J = 12.6 Hz, H-2β), 1.46 (1H, m, H-12) 1.57 (3H, s, H-17), 1.65

(3H, s, H-16), 1.87 (1H, ddd, J = 12.6, 12.0, 6.0 Hz, H-2α), 1.97 (2H, ddd, J = 7.8, 7.2,

7.2 Hz, H-13), 2.11 (1H, brq, J = 7.2 Hz, H-11), 2.70 (1H, brd, J = 3.6 Hz, H-4), 3.20

(1H, qd, J = 6.6, 6.0 Hz, H-1), 3.87 (3H, s, H-21), 4.05 (1H, brdddd, J = 12.0, 4.2, 3.6,

3.0 Hz, H-3), 4.65 (1H, brd, J = 3.0 Hz, 3-OH), 5.13 (1H, t, J = 7.2 Hz, H-14), 6.95 (1H,

13 s, H-5); C NMR (125 MHz, DMSO-d6) δC 17.5 (C-17), 18.8 (C-18), 21.8 (C-20), 25.5

(C-13), 25.7 (C-16), 28.5 (C-1), 29.9 (C-11), 34.5 (C-2), 38.5 (C-12), 48.1 (C-4), 52.3

(C-21), 64.8 (C-3), 109.6 (C-6), 119.5 (C-5), 124.7 (C-14), 128.9 (C-10), 130.2 (C-15),

135.8 (C-9), 142.6 (C-8), 146.5 (C-7), 170.0 (C-19); (+)-LRESIMS m/z (rel. int.) 363

+ - (100) [M + H] ; (-)-HRESIMS m/z 361.2018 (C21H29O5 [M - H] requires 361.2020).

3,8-Dihydroxy-7-methoxyserrulat-14-en-methyl-19-

25 benzoate (207): brown gum; [α]D + 8 (c 0.04, CH3OH);

UV (CH3OH) λmax nm (log ε) 214 (4.01), 255 (3.48), 306

1 (3.04); H NMR (600 MHz, DMSO-d6) δ 0.36 (3H, d, J = 6.6

Hz, H-18), 1.18 (3H, d, J = 7.2 Hz, H-20), 1.27 (1H, m, H-

12), 1.46 (1H, m, H-2β), 1.47 (1H, m, H-12), 1.57 (3H, s, H-16), 1.65 (3H, s, H-17),

1.88 (1H, d, J = 12.6, 12.0, 6.0 Hz, H-2α), 1.97 (2H, ddd, J = 7.2, 7.2, 7.2 Hz, H-13),

2.12 (1H, brq, J = 6.6 Hz, H-11), 2.72 (1H, brd, J = 3.6 Hz, H-4), 3.18 (1H, brqd, J =

7.2, 6.6 Hz, H-1), 3.67 (3H, s, H-22), 3.79 (3H, s, H-21), 4.05 (1H, m, H-3), 4.65 (1H,

216 brd, J = 3.6 Hz, 3-OH), 5.12 (1H, t, J = 6.6 Hz, H-14), 6.87 (1H, s, H-5), 8.90 (1H, s, 8-

13 OH); C NMR (125 MHz, DMSO-d6) δ 17.5 (C-16), 18.8 (C-8), 21.9 (C-20), 25.5 (C-

13), 25.7 (C-17), 28.4 (C-1), 30.0 (C-11), 34.5 (C-2), 38.5 (C-12), 48.3 (C-4), 51.8 (C-

21), 61.4 (C-22), 64.7 (C-3), 120.7 (C-6), 120.8 (C-5), 124.7 (C-14), 130.2 (C-15),

133.6 (C-10), 134.4 (C-9), 144.9 (C-7), 147.7 (C-8), 165.9 (C-19); (+)-LRESIMS m/z

+ - (rel. int.) 377 (100) [M + H] ; (-)-HRESIMS m/z 375.2171 (C22H31O5 [M - H] requires

375.2177).

Antibacterial assay.‡‡ Compounds 198-201, 109, 205 and 206 were screened against one Gram-negative bacterial strain: Escherichia coli (ATCC 25922), and nine

Gram-positive bacterial strains: Enterococcus faecalis (VanA clinical isolate),

Enterococcus faecium (MDR Van A ATCC 51559), Streptococcus pyogenes (Group A

ATCC 12344), Streptococcus pneumoniae (MDR ATCC 700677), Staphylococcus aureus (mMRSA clinical isolate), Staphylococcus aureus (MRSA ATCC 43300),

Staphylococcus aureus (GISA NRS 17), Staphylococcus aureus (GISA NRS 1), and

Staphylococcus aureus (MRSA DapRSA clinical isolate). The experiments were all performed in duplicate (n = 2) with vancomycin, colistin, and daptomycin used as positive controls. Control MIC values for vancomycin were as follows: E. faecalis (> 64

μg/mL), E. faecium (> 64 μg/mL), S. pyogenes (0.5 μg/mL), S. pneumoniae (1 μg/mL), and S. aureus strains (1-4 μg/mL). The control MIC value for colistin was as follows: E. coli (≤ 0.03 μg/mL). Control MIC values for daptomycin were as follows: E. coli (> 64

μg/mL), E. faecalis (16 μg/mL), E. faecium (16 μg/mL), S. pyogenes (0.25 μg/mL), S.

‡‡Antibacterial assay was undertaken by Angela Kavanagh, Soumya Ramu, Dr Mark Blaskovich, and Professor Matthew Cooper of the Institute for Molecular Bioscience, University of Queensland.

217 pneumoniae (4 μg/mL), and S. aureus strains (2-16 μg/mL). Positive growth control rows of bacteria and DMSO + bacteria as well as a negative control row of only media were included for every plate tested.

MIC assay. MICs were determined by a two-fold serial broth microdilution according to the recommendation of CLSI standards with an inoculum of 5 × 105 cfu/mL. The compounds along with standard antibiotics were serially diluted twofold across the wells of 96-well non-binding surface plates (NBS, Corning). Standards ranged from 64—0.03 μg/mL, and the compounds from 128—0.06 μg/mL with final volumes of 50 μL per well. Gram-positive and Gram-negative bacteria were cultured in

Mueller Hinton broth (MHB) (Bacto laboratories, Cat. no. 211443) at 37 °C overnight.

A sample of each culture was then diluted 40-fold in fresh MHB broth and incubated at

37 °C for 2-3 h. The resultant mid-log phase cultures were diluted to the final concentration of 5 × 105 cfu/mL, then 50 μL was added to each well of the compound containing 96-well plates. All the plates were covered and incubated at 37 °C for 24 h.

MICs were the lowest concentration showing no visible growth.

MBC assay. Resazurin (30 μL, 0.01%) was added to each well of the 96-well plates after the MIC values were determined. The compounds were then incubated at 37

°C for a further 18 to 24 h. Wells with blue coloration indicated dead microorganisms, whereas wells with pink coloration indicated live microorganisms. The MBC value was determined by the lowest concentrations of the wells with blue coloration.

218

References

(1) Abel, U.; Koch, C.; Speitling, M.; Hansske, F. G. Curr. Op. Chem. Biol. 2002, 6, 453-458. (2) Butler, M. S. J. Nat. Prod. 2004, 67, 2141 -2153. (3) Rishton, G. M. Curr. Opin. Chem. Biol. 2008, 12, 340-351. (4) Gallop, M. A.; Barrett, R. W.; Dower, W. J.; Fodor, S. P. A.; Gordon, E. M. J. Med. Chem. 1994, 37, 1233-1251. (5) Bauer, R. A.; Wurst, J. M.; Tan, D. S. Curr. Op. Chem. Biol. 2010, 14, 308-314. (6) Koehn, F. E.; Carter, G. T. Nat. Rev. Drug Discov. 2005, 4, 206-220. (7) Lyttle, M. H. Drug Develop. Res. 1995, 35, 230-236. (8) Hong, J. Curr. Opin. Chem. Biol. 2011, 15, 350–354. (9) Kingston, D. G. I. J. Nat. Prod. 2011, 74, 496-511. (10) Kennedy, J. P.; Williams, L.; Bridges, T. M.; Daniels, R. N.; Weaver, D.; Lindsley, C. W. J. Comb. Chem. 2008, 10, 345-354. (11) Kodadek, T. Chem. Commun. 2011, 47, 9757-9763. (12) Breinbauer, R.; Vetter, I. R.; Waldmann, H. Angew. Chem. Int. Ed. 2002, 41, 2878-2890. (13) Ortholand, J.-Y.; Ganesan, A. Curr. Opin. Chem. Biol. 2004, 8, 271-280. (14) Nadin, A.; Hattotuwagama, C.; Churcher, I. Angew. Chem. Int. Ed. 2012, 51, 1114-1122. (15) Feher, M.; Schmidt, J. M. J. Chem. Inf. Comput. Sci. 2003, 43, 218-227. (16) Camp, D.; Davis, R. A.; Evans-Illidge, E. A.; Quinn, R. J. Future Med. Chem. 2012, 4, 1067-1084. (17) Gordon, E. M.; Barrett, R. W.; Dower, W. J.; Fodor, S. P. A.; Gallop, M. A. J. Med. Chem. 1994, 37, 1385-1401. (18) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Deliver. Rev. 1997, 23, 3-25. (19) Hann, M. M.; Oprea, T. I. Curr. Opin. Chem. Biol. 2004, 8, 255-263. (20) Martin, E. J.; Critchlow, R. E. J. Comb. Chem. 1999, 1, 32-45. (21) Martin, E. J.; Blaney, J. M.; Siani, M. A.; Spellmeyer, D. C.; Wong, A. K.; Moos, W. H. J. Med. Chem. 1995, 38, 1431-1436. (22) Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2012, 75, 311-335. (23) Nielsen, J. Curr. Opin. Chem. Biol. 2002, 6, 297-305. (24) Breinbauer, R.; Manger, M.; Scheck, M.; Waldmann, H. Curr. Med. Chem. 2002, 9, 2129-2145. (25) Oprea, T. I.; Gottfries, J. J. Comb. Chem. 2001, 3, 157-166. (26) Ertl, P.; Roggo, S.; Schuffenhauer, A. J. Chem. Inf. Model. 2008, 48, 68-74. (27) Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2007, 70, 461-477. (28) Newman, D. J. J. Med. Chem. 2008, 51, 2589-2599. (29) Sadowski, J.; Kubinyi, H. J. Med. Chem. 1998, 41, 3325-3329. (30) Balamurugan, R.; Dekker, F. J.; Waldmann, H. Mol. BioSyst. 2005, 1, 36-45. (31) Bon, R. S.; Waldmann, H. Accounts Chem. Res. 2010, 43, 1103-1114. (32) Grabowski, K.; Baringhaus, K.-H.; Schneider, G. Nat. Prod. Rep. 2008, 25, 892- 904. (33) Hegde, V.; Campitelli, M.; Quinn, R. J.; Camp, D. Org. Biomol. Chem. 2011, 9, 4570-4579. (34) Rosen, J.; Gottfries, J.; Muresan, S.; Backlund, A.; Oprea, T. I. J. Med. Chem. 2009, 52, 1953-1962.

219

(35) Hert, J.; Irwin, J. J.; Laggner, C.; Keiser, M. J.; Shoichet, B. K. Nat. Chem. Biol. 2009, 5, 479-483. (36) Ajay; Walters, W. P.; Murcko, M. A. J. Med. Chem. 1998, 41, 3314-3324. (37) Newman, D. J.; Cragg, G. M.; Snader, K. M. Nat. Prod. Rep. 2000, 17, 215-234. (38) Cragg, G. M.; Newman, D. J. J. Ethnopharmacol. 2005, 100, 72-79. (39) Mishra, B. B.; Tiwari, V. K. Eur. J. Med. Chem. 2011, 46, 4769-4807. (40) Quinn, R. J. Drug Develop. Res. 1999, 46, 250-254. (41) Harvey, A. L. Drug Discov. Today 2008, 13, 894-901. (42) Kumar, K.; Waldmann, H. Angew. Chem. Int. Ed. 2009, 48, 3224-3242. (43) Walsh, C. T.; Fischbach, M. A. J. Am. Chem. Soc. 2010, 132, 2469-2493. (44) Kellenberger, E.; Hofmann, A.; Quinn, R. J. Nat. Prod. Rep. 2011, 28, 1483- 1492. (45) McArdle, B. M.; Quinn, R. J. ChemBioChem 2007, 8, 788-798. (46) Koch, M. A.; Schuffenhauer, A.; Scheck, M.; Wetzel, S.; Casaulta, M.; Odermatt, A.; Ertl, P.; Waldmann, H. PNAS 2005, 102, 17272-17277. (47) Evans, B. E.; Rittle, K. E.; Bock, M. G.; DiPardo, R. M.; Freidinger, R. M.; Whitter, W. L.; Lundell, G. F.; Veber, D. F.; Anderson, P. S.; Chang, R. S. L.; Lotti, V. J.; Cerino, D. J.; Chen, T. B.; Kling, P. J.; Kunkel, K. A.; Springer, J. P.; Hirshfield, J. J. Med. Chem. 1988, 31, 2235-2246. (48) Tan, D. S.; Foley, M. A.; Stockwell, B. R.; Shair, M. D.; Schreiber, S. L. J. Am. Chem. Soc. 1999, 121, 9073-9087. (49) Nandy, J. P.; Prakesch, M.; Khadem, S.; Reddy, P. T.; Sharma, U.; Arya, P. Chem. Rev. 2009, 109, 1999-2060. (50) Reayi, A.; Arya, P. Curr. Opin. Chem. Biol. 2005, 9, 240-247. (51) Reddy, P. T.; Quevillon, S.; Gan, Z.; Forbes, N.; Leek, D. M.; Arya, P. J. Comb. Chem. 2006, 8, 856-871. (52) Arya, P.; Joseph, R.; Gan, Z.; Rakic, B. Chem. Biol. 2005, 12, 163-180. (53) Inoki, K.; Corradetti, M. N.; Guan, K.-L. Nat. Genet. 2004, 37, 19-24. (54) Brown, E. J.; Schreiber, S. L. Cell 1996, 86, 517-520. (55) Brown, E. J.; Albers, M. W.; Shin, T. B.; Ichikawa, K.; Keith, C. T.; Lane, W. S.; Schreiber, S. L. Nature 1994, 369, 756-758. (56) Shu, Y.-Z. J. Nat. Prod. 1998, 61, 1053-1071. (57) Li, J. W.-H.; Vederas, J. C. Science 2009, 325, 161-165. (58) Hall, D. G.; Manku, S.; Wang, F. J. Comb. Chem. 2001, 3, 125-150. (59) Bindseil, K. U.; Jakupovic, J.; Wolf, D.; Lavayre, J.; Leboul, J.; van der Pyl, D. Drug Discov. Today 2001, 6, 840-847. (60) Bugni, T. S.; Richards, B.; Bhoite, L.; Cimbora, D.; Harper, M. K.; Ireland, C. M. J. Nat. Prod. 2008, 71, 1095–1098. (61) Appleton, D. R.; Buss, A. D.; Butler, M. S. Chimia 2007, 61, 327–331. (62) Schroeder, F. C.; Gibson, D. M.; Churchill, A. C. L.; Sojikul, P.; Wursthorn, E. J.; Krasnoff, S. B.; Clardy, J. Angew. Chem. Int. Ed. 2007, 46, 901-904. (63) Clardy, J.; Walsh, C. Nature 2004, 432, 829-837. (64) Wolf, D.; Siems, K. Chimia 2007, 61, 339–345. (65) Dictionary of Natural Products (DVD); version 5.1; Taylor & Francis Group/CRC Press: London, UK, 2012. (66) Atuegbu, A.; Maclean, D.; Nguyen, C.; Gordon, E. M.; Jacobs, J. W. Bioorg. Med. Chem. 1996, 4, 1097-1106. (67) Davis, R. A.; Carroll, A. R.; Quinn, R. J. Aust. J. Chem. 2001, 54, 355-359. (68) Newman, D. J.; Cragg, G. M. Future Med. Chem. 2009, 1, 1415-1427.

220

(69) Cragg, G. M.; Grothaus, P. G.; Newman, D. J. Chem. Rev. 2009, 109, 3012- 3043. (70) Cragg, G. M.; Newman, D. J. Phytochem. Rev. 2009, 8, 313-331. (71) Lee, M.-L.; Schneider, G. J. Comb. Chem. 2001, 3, 284-289. (72) Wetzel, S.; Schuffenhauer, A.; Roggo, S.; Ertl, P.; Waldmann, H. Chimia 2007, 61, 355-360. (73) Pandey, R. C.; Toussaint, M. W.; Stroshane, R. M.; Kalita, C. C.; Aszalos, A. A.; Garretson, A. L.; Wei, T. T.; Byrne, K. M.; Geoghegan, R. F., Jr.; White, R. J. J. Antibiot. 1981, 34, 1389-1401. (74) Misra, R.; Pandey, R. C.; Silverton, J. V. J. Am. Chem. Soc. 1982, 104, 4478- 4479. (75) Misra, R.; Pandey, R. C.; Hilton, B. D.; Roller, P. P.; Silverton, J. V. J. Antibiot. 1987, 40, 786-802. (76) Warnick-Pickle, D. J.; Byrne, K. M.; Pandey, R. C.; White, R. J. J. Antibiot. 1981, 34, 1402-1407. (77) Latham, M. D.; King, C. K.; Gorycki, P.; Macdonald, T. L.; Ross, W. E. Cancer Chemother. Pharmacol. 1989, 24, 167-171. (78) Akai, S.; Tsujino, T.; Fukuda, N.; Iio, K.; Takeda, Y.; Kawaguchi, K.-I.; Naka, T.; Higuchi, K.; Akiyama, E.; Fujioka, H.; Kita, Y. Chem. Eur. J. 2005, 11, 6286-6297. (79) Braun, M.; Kergoet, G.; Kruska, F.; Frank, W. Synthesis 2010, 2023-2026. (80) Abel, U.; Simon, W.; Eckard, P.; Hansske, F. G. Bioorg. Med. Chem. Lett. 2006, 16, 3292-3297. (81) Sontag, B. World Patent WO2004024696, 2004. (82) Bhal, S. K.; Kassam, K.; Peirson, I. G.; Pearl, G. M. Mol. Pharm. 2007, 4, 556- 560. (83) Iwamoto, T.; Fujie, A.; Sakamoto, K.; Tsurumi, Y.; Shigematsu, N.; Yamashita, M.; Hashimoto, S.; Okuhara, M.; Kohsaka, M. J. Antibiot. 1994, 47, 1084-1091. (84) Iwamoto, T.; Fujie, A.; Nitta, K.; Hashimoto, S.; Okuhara, M.; Kohsaka, M. J. Antibiot. 1994, 47, 1092-1097. (85) Fujie, A.; Iwamoto, T.; Sato, B.; Muramatsu, H.; Kasahara, C.; Furuta, T.; Hori, Y.; Hino, M.; Hashimoto, S. Bioorg. Med. Chem. Lett. 2001, 11, 399-402. (86) Tomishima, M.; Ohki, H.; Yamada, A.; Maki, K.; Ikeda, F. Bioorg. Med. Chem. Lett. 2008, 18, 1474-1477. (87) Tomishima, M.; Ohki, H.; Yamada, A.; Takasugi, H.; Maki, K.; Tawara, S.; Tanaka, H. J. Antibiot. 1999, 52, 674-676. (88) Boger, D. L.; Desharnais, J.; Capps, K. Angew. Chem. Int. Ed. 2003, 42, 4138- 4176. (89) Nicolaou, K. C.; Pfefferkorn, J. A. Pept. Sci. 2001, 60, 171-193. (90) Dow, M.; Fisher, M.; James, T.; Marchetti, F.; Nelson, A. Org. Biomol. Chem. 2012, 10, 17-28. (91) Wender, P. A.; Jankowski, O. D.; Tabet, E. A.; Seto, H. Org. Lett. 2003, 5, 487- 490. (92) Lewis, C. A.; Longcore, K. E.; Miller, S. J.; Wender, P. A. J. Nat. Prod. 2009, 72, 1864-1869. (93) Gordeev, M. F.; Luehr, G. W.; Hui, H. C.; Gordon, E. M.; Patel, D. V. Tetrahedron 1998, 54, 15879-15890. (94) Boldi, A. M. Curr. Opin. Chem. Biol. 2004, 8, 281-286. (95) Davis, R. A.; Pierens, G. K.; Parsons, P. G. Magn. Reson. Chem. 2007, 45, 442- 445.

221

(96) Camp, D.; Davis, R. A.; Campitelli, M.; Ebdon, J.; Quinn, R. J. J. Nat. Prod. 2012, 75, 72-81. (97) Nature Bank, Eskitis Institute, Brisbane, QLD, Australia, 2011; viewed on 18th September 2012, www .nature-bank.com.au. (98) Convention on biological diversity, Secretariat of the Convention on Biological Diversity, Montreal, Canada, 2012; viewed 18th September 2012, www .cbd.int. (99) Ghisalberti, E. L.; Jefferies, P. R.; Skelton, B. W.; White, A. H.; Williams, R. S. F. Tetrahedron 1989, 45, 6297-6308. (100) Barnes, E. C.; Carroll, A. R.; Davis, R. A. J. Nat. Prod. 2011, 74, 1888-1893. (101) Kunishima, M.; Kawachi, C.; Morita, J.; Terao, K.; Iwasaki, F.; Tani, S. Tetrahedron 1999, 55, 13159-13170. (102) Van Hoof, S.; Lacey, C. J.; Rohrich, R. C.; Wiesner, J.; Jomaa, H.; Van Calenbergh, S. J. Org. Chem. 2008, 73, 1365-1370. (103) Tamaru, Y.; Kimura, M. Pure Appl. Chem. 2008, 80, 979-991. (104) Chalk, A. J.; Magennis, S. A. J. Org. Chem. 1976, 41, 1206-1209. (105) Ghisalberti, E. L.; Jefferies, P. R.; Vu, H. T. N. Phytochemistry 1990, 29, 2700- 2701. (106) Chinnock, R. J. Eremophila and Allied Genera; Rosenberg Publishing: Dural, NSW, 2007. (107) Jessup, L., Queensland Herbarium, Brisbane, QLD, Australia, Personal communication, 17th May 2012 (108) Liu, Q.; Harrington, D.; Kohen, J. L.; Vemulpad, S.; Jamie, J. F. Phytochemistry 2006, 67, 1256–1261. (109) Palombo, E. A.; Semple, S. J. J. Ethnopharmacol. 2001, 77, 151-157. (110) Semple, S. J.; Reynolds, G. D.; O'Leary, M. C.; Flower, R. L. P. J. Ethnopharmacol. 1998, 60, 163-172. (111) Fagg, M.; Australian Plant Image Index, Australian National Botanic Gardens, Canberra, Australia, 2012; viewed on 18th September 2012, www .anbg.gov.au. (112) Pretsch, E.; Buhlmann, P.; Badertscher, M. Structure Determination of Organic Compounds; Springer: Berlin, 2009. (113) Ghisalberti, E. L.; Jefferies, P. R.; Skelton, B. W.; White, A. H.; Williams, R. S. F. Tetrahedron 1989, 45, 6297-6308. (114) Davidson, B. S.; Plavcan, K. A.; Meinwald, J. J. Org. Chem. 1990, 55, 3912- 3917. (115) Borg-Karlson, A.-K.; Norin, T.; Talvitie, A. Tetrahedron 1981, 37, 425-430. (116) Belic, I.; Bergant-Dolar, J.; Morton, R. A. J. Chem. Soc. 1961, 2523-2525. (117) Lewin, G.; Maciuk, A.; Thoret, S.; Aubert, G.; Dubois, J.; Cresteil, T. J. Nat. Prod. 2010, 73, 702–706. (118) Chang, S.-L.; Chiang, Y.-M.; Chang, C. L.-T.; Yeh, H.-H.; Shyur, L.-F.; Kuo, Y.-H.; Wu, T.-K.; Yang, W.-C. J. Ethnopharmacol. 2007, 112, 232-236. (119) Farkas, L.; Horhammer, L.; Wagner, H.; Rosler, H.; Gurniak, R. Chem. Ber. 1964, 97, 1666. (120) Sy, L.-K.; Brown, G. D.; Haynes, R. Tetrahedron 1998, 54, 4345-4356. (121) Koh, D.-J.; Ahn, H.-S.; Chung, H.-S.; Lee, H.; Kim, Y.; Lee, J.-Y.; Kim, D.-G.; Hong, M.; Shin, M.; Bae, H. J. Ethnopharmacol. 2011, 136, 399-405. (122) Hajdú, Z.; Hohmann, J.; Forgo, P.; Martinek, T.; Dervarics, M.; Zupkó, I.; Falkay, G.; Cossuta, D.; Máthé, I. Phytother. Res. 2007, 21, 391-394. (123) Hu, Y.; Xin, H.-L.; Zhang, Q.-Y.; Zheng, H.-C.; Rahman, K.; Qin, L.-P. Phytomedicine 2007, 14, 668-674.

222

(124) Hu, Y.; Hou, T.-T.; Zhang, Q.-Y.; Xin, H.-L.; Zheng, H.-C.; Rahman, K.; Qin, L.-P. J. Pharm. Pharmacol. 2007, 59, 1307-1312. (125) Wang, Y.; Hamburger, M.; Gueho, J.; Hostettmann, K. Phytochemistry 1989, 28, 2323-2327. (126) Ye, Q.; Zhang, Q.-Y.; Zheng, C.-J.; Wang, Y.; Qin, L.-P. Acta Pharmacol. Sin. 2010, 31, 1564-1568. (127) Liu, K. C.-S. C.; Yang, S.-L.; Roberts, M. F.; Elford, B. C.; Phillipson, J. D. Plant Cell Rep. 1992, 11, 637-640. (128) Elford, B. C.; Roberts, M. F.; Phillipson, J. D.; Wilson, R. J. M. T. Roy. Soc. Trop. Med. H. 1987, 81, 434-436. (129) Bilia, A. R.; Lazari, D.; Messori, L.; Taglioli, V.; Temperini, C.; Vincieri, F. F. Life Sci. 2002, 70, 769-778. (130) Diaz, F.; Chavez, D.; Lee, D.; Mi, Q.; Chai, H.-B.; Tan, G. T.; Kardono, L. B. S.; Riswan, S.; Fairchild, C. R.; Wild, R.; Farnsworth, N. R.; Cordell, G. A.; Pezzuto, J. M.; Kinghorn, A. D. J. Nat. Prod. 2003, 66, 865-867. (131) Haidara, K.; Zamir, L.; Shi, Q.-W.; Batist, G. Cancer Lett. 2006, 242, 180-190. (132) Csupor-Löffler, B.; Hajdú, Z.; Zupkó, I.; Réthy, B.; Falkay, G.; Forgo, P.; Hohmann, J. Phytother. Res. 2009, 23, 672-676. (133) Ono, M.; Yanaka, T.; Yamamoto, M.; Ito, Y.; Nohara, T. J. Nat. Prod. 2002, 65, 537-541. (134) Kobayakawa, J.; Sato-Nishimori, F.; Moriyasu, M.; Matsukawa, Y. Cancer Lett. 2004, 208, 59-64. (135) Shen, J.-K.; Du, H.-P.; Yang, M.; Wang, Y.-G.; Jin, J. Ann. Hematol. 2009, 88, 743-752. (136) Trifunovic, S.; Vajs, V.; Juranic, Z.; Zizak, Z.; Tesevic, V.; Macura, S.; Milosavljevic, S. Phytochemistry 2006, 67, 887-893. (137) Pan, E.; Gorka, A. P.; Alumasa, J. N.; Slebodnick, C.; Harinantenaina, L.; Brodie, P. J.; Roepe, P. D.; Randrianaivo, R.; Birkinshaw, C.; Kingston, D. G. I. J. Nat. Prod. 2011, 74, 2174-2180. (138) Beutler, J. A.; Cardellina, J. H.; Lin, C. M.; Hamel, E.; Cragg, G. M.; Boyd, M. R. Bioorg. Med. Chem. Lett. 1993, 3, 581-584. (139) Beutler, J. A.; Hamel, E.; Vlietinck, A. J.; Haemers, A.; Rajan, P.; Roitman, J. N.; Cardellina, J. H.; Boyd, M. R. J. Med. Chem. 1998, 41, 2333-2338. (140) Abad, M. J.; Bermejo, P.; Villar, A.; Valverde, S. J. Nat. Prod. 1993, 56, 1164- 1167. (141) Guerra, J. A.; Molina, M. F.; Abad, M. J.; Villar, A. M.; Bermejo, P. Int. Immunopharmacol. 2006, 6, 1723-1728. (142) Aljancic, I.; Vajs, V.; Menkovic, N.; Karadzic, I.; Juranic, N.; Milosavljevic, S.; Macura, S. J. Nat. Prod. 1999, 62, 909-911. (143) Chang, Y.-J.; Song, S.-H.; Park, S.-H.; Kim, S.-U. Arch. Biochem. Biophys. 2000, 383, 178-184. (144) Cox, R. J. Annu. Rep. Prog. Chem. Sect. B 2002, 98, 223-251. (145) Hong, Y. J.; Tantillo, D. J. J. Am. Chem. Soc. 2009, 131, 7999-8015. (146) Hanson, J. R. Pure Appl. Chem. 1981, 53, 1155-1162. (147) Toyota, M.; Saito, T.; Asakawa, Y. Chem. Pharm. Bull. 1998, 46, 178-180. (148) Dong, J.-Q.; Wong, H. N. C. Angew. Chem. 2009, 121, 2387 –2390. (149) Murray, R. W.; Singh, M. Org. Syn. 1997, 74, 91-96. (150) Curci, R.; Dinoi, A.; Rubino, M. F. Pure Appl. Chem. 1995, 67, 811-822. (151) Reactor; ChemAxon: version 5.3.1, 2011; www .chemaxon.com/products/reactor.

223

(152) Instant JChem; ChemAxon: version 5.5, 2011; www .chemaxon.com/products/instant-jchem. (153) Camp, D.; Davis, R. A.; Campitelli, M.; Ebdon, J.; Quinn, R. J. J. Nat. Prod. 2011, 75, 72-81. (154) Sigma-Aldrich Castle Hill, NSW; www .sigmaaldrich.com/chemistry/chemistry. (155) Itoh, T.; Taguchi, T.; Kimberley, M. R.; Booker-Milburn, K. I.; Stephenson, G. R.; Ebizuka, Y.; Ichinose, K. Biochemistry 2007, 46, 8181-8188. (156) Martin, N. H.; Jefford, C. W. Helv. Chim. Acta 1982, 65, 762-774. (157) Nakamura, M.; Toganoh, M.; Ohara, H.; Nakamura, E. Org. Lett. 1999, 1, 7-9. (158) Matteson, D. S.; Soundararajan, R.; Ho, O. C.; Gatzweiler, W. Organometallics 1996, 15, 152-163. (159) Nahar, L.; Sarker, S. D.; Li, P. K.; Turner, A. B. Chem. Nat. Cmpd. 2004, 40, 50-53. (160) Zhao, Y.; Mao, C.; Li, Y.; Zhang, P.; Huang, Z.; Bi, F.; Huang, R.; Wang, Q. J. Agric. Food. Chem. 2008, 56, 7326-7332. (161) Yang, G.; Zhang, H.; Huang, Y.; Chen, Z. Synthetic Commun 2006, 36, 611– 619. (162) Loev, B.; Jones, H.; Brown, R. E.; Huang, F.; Khandwala, A.; Leibowitz, M. J.; Sonnino-Goldmans, P. J Med Chem 1985, 28, 24-27. (163) Craig, D.; Ford, M. J.; Gordon, R. S.; Stones, J. A.; White, A. J. P.; Williams, D. J. Tetrahedron 1999, 55, 15045-15066. (164) Chokki, Y.; Fujinami, K. Nippon Kagaku Kaishi 1974, 12, 2407-2413. (165) Blagbrough, I. S.; Mackenzie, N. E.; Ortiz, C.; Scott, A. I. Tetrahedron Lett. 1986, 27, 1251-1254. (166) Garfunkle, J.; Kimball, F. S.; Trzupek, J. D.; Takizawa, S.; Shimamura, H.; Tomishima, M.; Boger, D. L. J. Am. Chem. Soc. 2009, 131, 16036-16038. (167) Hofmann, A.; Wang, C. K.; Osman, A.; Camp, D. Struct. Chem. 2010, 21, 1117- 1129. (168) Townsend, K.; Eremophila Plant Gallery, Australian Native Plants Society, Australia, 2010; viewed on 18th September 2012, www. asgap.org.au. (169) Ghisalberti, E. L. Phytochemistry 1994, 35, 7-33. (170) Ghisalberti, E. L. J. Ethnopharmacol. 1994, 44, 1-9. (171) Andary, C.; Wylde, R.; Laffite, C.; Privat, G.; Winternitz, F. Phytochemistry 1982, 21, 1123-1127. (172) Ghisalberti, E. L.; Loh, J. S. C. Fitoterapia 1995, 66, 93-94. (173) Syah, Y. M.; Ghisalberti, E. L.; Skelton, B. W.; White, A. H. J. Nat. Prod. 1997, 60, 49-51. (174) Grice, I. D.; Garhnam, B.; Pierens, G.; Rogers, K.; Tindal, D.; Griffiths, L. R. J. Ethnopharmacol. 2003, 86, 123-125. (175) Pennacchio, M.; Syah, Y. M.; Ghisalberti, E. L.; Alexander, E. J. Ethnopharmacol. 1996, 53, 21-27. (176) Dell, B.; Elsegood, C. L.; Ghisalberti, E. L. Phytochemistry 1989, 28, 1871- 1872. (177) Ghisalberti, E. L.; Loh, J. S. C. Fitoterapia 1995, 66, 186-187. (178) Cribb, A. B.; Cribb, J. W. Wild Medicine in Australia; William Collins Pty Ltd: Sydney, 1981. (179) Barr, A. Traditional Bush Medicines: An Aboriginal Pharmacopoeia; Greenhouse Publications: Richmond, Australia 1988. (180) Latz, P. Bushfires and Bushtucker: Aboriginal Plant Use in Central Australia; IAD Press: Alice Springs, NT, 1995.

224

(181) Low, T. Bush Medicine: A Pharmacopoeia of Natural Remedies Collins/Angus & Robertson Publishers: North Ryde, Australia, 1990. (182) Richmond, G. S.; Ghisalberti, E. L. Econ. Bot. 1994, 48, 35-59. (183) Ghisalberti, E. L. In Studies in Natural Product Chemistry; Rahman, A.-U., Ed.; Elsevier Science: Amsterdam, 1995; Vol. 15. (184) Ndi, C. P.; Semple, S. J.; Griesser, H. J.; Barton, M. D. J. Basic Microbiol. 2007, 47, 158-164. (185) Cribb, A. B.; Cribb, J. W. Useful Wild Plants in Australia; William Collins Pty Ltd: Sydney, 1981. (186) Hegarty, B. F.; Kelly, J. R.; Park, R. J.; Sutherland, M. D. Aust. J. Chem. 1970, 23, 107-117. (187) Birch, A. J.; Massy-Westropp, R. A.; Wright, S. E.; Kubota, T.; Matsuura, T.; Sutherland, M. D. Chem. Ind. 1954, 902. (188) Seawright, A. A.; Hrdlicka, J.; Lee, J. A.; Ogunsan, E. A. J. Appl. Toxicol. 1982, 2, 75-82. (189) Finnemore, H.; Cox, C. B. J. Proc. Roy. Soc. NSW 1930, 63, 172-178. (190) Walters, B.; Eremophila Plant Gallery, Australian Native Plants Society, Australia, 2010; viewed on 18th September 2012, www. asgap.org.au. (191) Fagg, M.; Vallee, L.; Australian Plant Image Index, Australian National Botanic Gardens, Canberra, Australia, 2012; viewed on 18th September 2012, www .anbg.gov.au. (192) Clarke, P. A. Aboriginal People and Their Plants; Rosenberg Publishing Pty Ltd: Dural, NSW, 2007. (193) Dimitriadis, E.; Massy-Westropp, R. A. Aust. J. Chem. 1980, 33, 2729-2736. (194) Blackburne, I. D.; Park, R. J.; Sutherland, M. D. Aust. J. Chem. 1972, 25, 1787- 1796. (195) Russell, C. A.; Sutherland, M. D. Aust. J. Chem. 1982, 35, 1881-1894. (196) Massy-Westropp, R. A.; Reynolds, G. D.; Spotswood, T. M. Tetrahedron Lett. 1966, 7, 1939-1946. (197) Ingham, C. F.; Massy-Westropp, R. A.; Reynolds, G. D. Aust. J. Chem. 1974, 27, 1477-1489. (198) Begley, M. J.; Knight, D. W.; Pattenden, G. J. Chem. Soc. Perkin Trans. II 1975, 1863-1866. (199) Knight, D. W.; Pattenden, G. Tetrahedron Lett. 1975, 16, 1115-1116. (200) Dimitriadis, E.; Massy-Westropp, R. A. Aust. J. Chem. 1979, 32, 2003-2015. (201) Bradfield, A. E.; Penfold, A. R.; Simonsen, J. L. J. Chem. Soc. 1932, 2744-2759. (202) Bradfield, A. E.; Penfold, A. R.; Simonsen, J. L. J. Proc. Roy. Soc. NSW 1933, 66, 420-33. (203) Bradfield, A. E.; Hellstrom, N.; Penfold, A. R.; Simonsen, J. L. J. Chem. Soc. 1938, 767-774. (204) Beattie, K. D.; Waterman, P. G.; Forster, P. I.; Thompson, D. R.; Leach, D. N. Phytochemistry 2011, 72, 400-408. (205) Massy-Westropp, R. A.; Reynolds, G. D. Aust. J. Chem. 1966, 19, 303-307. (206) Chetty, G. L.; Zalkow, L. H.; Massy-Westropp, R. A. Tetrahedron Lett. 1969, 10, 307-309. (207) Abell, A. D.; Massy-Westropp, R. A. Aust. J. Chem. 1985, 38, 1263-1269. (208) Ruzicka, L. Proc. Chem. Soc. 1959, 341-360. (209) Zalkow, L. H.; Markley, F. X.; Djerassi, C. J. Am. Chem. Soc. 1959, 81, 2914- 2915.

225

(210) Zalkow, L. H.; Shaligram, A. M.; Hu, S.-E.; Djerassi, C. Tetrahedron 1966, 22, 337-350. (211) Zalkow, L. H.; Markley, F. X.; Djerassi, C. J. Am. Chem. Soc. 1960, 82, 6354- 6362. (212) Babidge, P. J.; Massy-Westropp, R. A. Aust. J. Chem. 1984, 37, 629-633. (213) Ghisalberti, E. L.; Jefferies, P. R.; Mori, T. A. Aust. J. Chem. 1984, 37, 635-647. (214) Ghisalberti, E. L.; Jefferies, P. R.; Vu, H. T. N. Phytochemistry 1990, 29, 316- 318. (215) Ghisalberti, E. L.; Jefferies, P. R.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1987, 40, 405-411. (216) Syah, Y. M.; Ghisalberti, E. L.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1996, 49, 707-710. (217) Lewis, D. E.; Massy-Westropp, R. A.; Ingham, C. F.; Wells, R. J. Aust. J. Chem. 1982, 35, 809-826. (218) Lewis, D. E.; Massy-Westropp, R. A.; Snow, M. R. Acta Crystallogr. Sect. B 1979, 35, 2253-2255. (219) Carrol, P. J.; Ghisalberti, E. L.; Ralph, D. E. Phytochemistry 1976, 15, 777-780. (220) Ghisalberti, E. L.; White, A. H.; Willis, A. C. J. Chem. Soc. Perkin Trans. II 1976, 12, 1300-1303. (221) Croft, K. D.; Ghisalberti, E. L.; Hocart, C. H.; Jefferies, P. R.; Raston, C. L.; White, A. H. J. Chem. Soc. Perkin Trans. I 1978, 1267-1270. (222) Bunko, J. D.; Ghisalberti, E. L.; Jefferies, P. R. Aust. J. Chem. 1981, 34, 2237- 2242. (223) Dastlik, K. A.; Forster, P. G.; Ghisalberti, E. L.; Jefferies, P. R. Phytochemistry 1989, 28, 1425-1426. (224) Croft, K. D.; Ghisalberti, E. L.; Jefferies, P. R.; Raston, C. L.; White, A. H.; Hall, S. R. Tetrahedron 1977, 33, 1475-1480. (225) Croft, K. D.; Ghisalberti, E. L.; Jefferies, P. R.; Stuart, A. D. Aust. J. Chem. 1979, 32, 2079-2083. (226) Croft, K. D.; Ghisalberti, E. L.; Jefferies, P. R.; Proudfoot, G. M. Aust. J. Chem. 1981, 34, 1951-1957. (227) Hall, S. R.; Raston, C. L.; Skelton, B. W.; White, A. H. J. Chem. Soc. Perkin Trans. II 1981, 1467-1472. (228) Syah, Y. M.; Ghisalberti, E. L. Phytochemistry 1997, 45, 1479-1482. (229) Abell, A. D.; Horn, E.; Jones, G. P.; Snow, M. R.; Massy-Westropp, R. A.; Riccio, R. Aust. J. Chem. 1985, 38, 1837-1845. (230) Forster, P. G.; Ghisalberti, E. L.; Jefferies, P. R.; Poletti, V. M.; Whiteside, N. J. Phytochemistry 1986, 25, 1377-1383. (231) Ghisalberti, E. L. Phytochemistry 1992, 31, 2168-2169. (232) Tippett, L. M.; Massy-Westropp, R. A. Phytochemistry 1993, 33, 417-421. (233) Ghisalberti, E. L.; Hocart, C. H.; Jefferies, P. R.; Proudfoot, G. M.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1983, 36, 993-1000. (234) Forster, P. G.; Ghisalberti, E. L.; Jefferies, P. R. Phytochemistry 1993, 32, 1225- 1228. (235) Forster, P. G.; Ghisalberti, E. L.; Jefferies, P. R. Aust. J. Chem. 1986, 39, 2111- 2120. (236) Syah, Y. M.; Ghisalberti, E. L.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1997, 50, 705-709. (237) Forster, P. G.; Ghisalberti, E. L.; Jefferies, P. R. Tetrahedron 1987, 43, 2999- 3007.

226

(238) Ghisalberti, E. L.; Jefferies, P. R.; Knox, J. R.; Sheppard, P. N. Tetrahedron 1977, 33, 3301-3303. (239) Coates, P.; Ghisalberti, E. L.; Jefferies, P. R. Aust. J. Chem. 1977, 30, 2717- 2721. (240) Ghisalberti, E. L.; Jefferies, P. R.; Mori, T. A.; Patrick, V. A.; White, A. H. Aust. J. Chem. 1983, 36, 1187-1196. (241) Ghisalberti, E. L.; Jefferies, P. R.; Mori, T. A. Aust. J. Chem. 1986, 39, 1703- 1710. (242) Ghisalberti, E. L.; Jefferies, P. R.; Mori, T. A.; Twiss, E. J. Nat. Prod. 1994, 57, 100-104. (243) Maslen, E. N.; Raston, C. L.; White, A. H. Tetrahedron 1977, 33, 3301-3303. (244) Ghisalberti, E. L.; Jefferies, P. R.; Proudfoot, G. M. Aust. J. Chem. 1981, 34, 1491-1499. (245) Forster, P. G.; Ghisalberti, E. L.; Jefferies, P. R.; Skelton, B. W.; White, A. H. Tetrahedron 1986, 42, 215-221. (246) Forster, P. G.; Ghisalberti, E. L.; Jefferies, P. R. J. Nat. Prod. 1993, 56, 147-152. (247) Syah, Y. M.; Ghisalberti, E. L. Phytochemistry 1996, 41, 859-861. (248) Birch, A. J.; Subba-Rao, G. S. R.; Turnbull, J. P. Tetrahedron Lett. 1966, 7, 4749-4751. (249) Ramage, R.; Owen, O. J. R.; Southwell, I. A. Tetrahedron Lett. 1983, 24, 4487- 4490. (250) Asaoka, M.; Ishibashi, K.; Yanagida, N.; Takei, H. Tetrahedron Lett. 1983, 24, 5127-5130. (251) Jefferies, P. R.; Knox, J. R.; Middleton, E. J. Aust. J. Chem. 1962, 15, 532-537. (252) Croft, K. D.; Ghisalberti, E. L.; Jefferies, P. R.; Mori, T. A.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1984, 37, 785-793. (253) Carroll, P. J.; Engelhardt, L. M.; Ghisalberti, E. L.; Jefferies, P. R.; Middleton, E. J.; Mori, T. A.; White, A. H. Aust. J. Chem. 1985, 38, 1351-1363. (254) Birch, A. J.; Grimshaw, J.; Turnbull, J. P. J. Chem. Soc. 1963, 2412-2417. (255) Ghisalberti, E. L.; Jefferies, P. R.; Sheppard, P. Tetrahedron Lett. 1975, 16, 1775-1778. (256) Ghisalberti, E. L.; Jefferies, P. R.; Sheppard, P. N. Tetrahedron 1980, 36, 3253- 3259. (257) Maslen, E. N.; Sheppard, P. N.; White, A. H.; Willis, A. C. J. Chem. Soc. Perkin Trans. II 1976, 263-266. (258) Croft, K. D.; Ghisalberti, E. L.; Jefferies, P. R.; Marshall, D. G.; Raston, C. L.; White, A. H. Aust. J. Chem. 1980, 33, 1529-1536. (259) Croft, K. D.; Ghisalberti, E. L.; Jefferies, P. R.; Stuart, A. D. Tetrahedron 1981, 37, 383-387. (260) Jefferies, P. R.; Knox, J. R. Aust. J. Chem. 1961, 14, 628-636. (261) Ghisalberti, E. L.; Jefferies, P. R.; Toia, R. F. Phytochemistry 1979, 18, 65-69. (262) Jefferies, P. R.; Knox, J. R.; White, D. E. Aust. J. Chem. 1961, 14, 175-177. (263) Bytheway, I. R.; Ghisalberti, E. L.; Gotsis, S.; Jefferies, P. R.; Skelton, B. W.; Sugars, K. E.; White, A. H. Aust. J. Chem. 1987, 40, 1913-1917. (264) Palombo, E. A.; Semple, S. J. J. Basic Microbiol. 2002, 42, 444-448. (265) Shah, A.; Cross, R. F.; Palombo, E. A. Phytother. Res. 2004, 18, 615-618. (266) Tomlinson, S.; Palombo, E. A. J. Basic Microbiol. 2005, 45, 363-370. (267) Kerr, R. W. Bulletin No. 261; CSIRO: Melbourne, Australia, 1951. (268) Wilkinson, J. M.; Cavanagh, H. M. A. Phytother. Res. 2005, 19, 643-646. (269) Owen, R. J.; Palombo, E. A. Food Control 2007, 18, 387-390.

227

(270) Sweeney, A. P.; Wyllie, S. G.; Shalliker, R. A.; Markham, J. L. J. Ethnopharmacol. 2001, 75, 273–277. (271) Rogers, K. L.; Grice, I. D.; Griffiths, L. R. Eur. J. Pharm. Sci. 2000, 9, 355-363. (272) Rogers, K. L.; Grice, I. D.; Griffiths, L. R. Life Sci. 2001, 69, 1817-1829. (273) Rogers, K. L.; Fong, W. F.; Redburn, J.; Griffiths, L. R. Eur. J. Pharm. Sci. 2002, 15, 321-330. (274) Pennacchio, M.; Alexander, E.; Ghisalberti, E. L.; Richmond, G. S. J. Ethnopharmacol. 1995, 47, 91-95. (275) Pennacchio, M.; Kemp, A. S.; Taylor, R. P.; Wickens, K. M.; Kienow, L. J. Ethnopharmacol. 2005, 96, 597-601. (276) Beattie, K., PhD Thesis, Southern Cross University, Lismore, 2009. (277) Ndi, C. P.; Semple, S. J.; Griesser, H. J.; Pyke, S. M.; Barton, M. D. J. Nat. Prod. 2007, 70, 1439-1443. (278) Ndi, C. P.; Semple, S. J.; Griesser, H. J.; Pyke, S. M.; Barton, M. D. Phytochemistry 2007, 68, 2684–2690. (279) Smith, J. E.; Tucker, D.; Watson, K.; Jones, G. L. J. Ethnopharmacol. 2007, 112, 386-393. (280) Comin, J.; de Lima, O. G.; Grant, H. N.; Jackman, L. M.; Keller-Schierlein, W.; Prelog, V. Helv. Chim. Acta 1963, 46, 409-415. (281) Anakok, O. F.; Ndi, C. P.; Barton, M. D.; Griesser, H. J.; Semple, S. J. J. Appl. Microbiol., 112, 197-204. (282) Ndi, C. P.; Semple, S. J.; Griesser, H. J. Aust. J. Chem. 2012, 65, 20-27. (283) Rodriguez, A. D. Tetrahedron 1995, 51, 4571-4618. (284) Look, S. A.; Fenical, W. Tetrahedron 1987, 43, 3363-3370. (285) Look, S. A.; Fenical, W.; Matsumoto, G. K.; Clardy, J. J. Org. Chem. 1986, 51, 5140-5145. (286) Look, S. A.; Fenical, W.; Jacobs, R. S.; Clardy, J. Proc. Natl. Acad. Sci. USA 1986, 83, 6238-6240. (287) Leach, D. N.; Spooner-Hart, R. N.; Eaton, G. F. World Patent WO2004/021784, 2004. (288) Pennacchio, M.; Syah, Y. M.; Ghisalberti, E. L.; Alexander, E. Phytomedicine 1997, 4, 325-330. (289) SciFinder; American Chemical Society, 2012; https: //scifinder.cas.org. (290) Jennings, C.; Eremophila Plant Gallery, Australian Native Plants Society, Australia, 2010; viewed on 18th September 2012, www. asgap.org.au. (291) El Bitar, H. E.; Nguyen, V. H.; Gramain, A.; Sevenet, T.; Bodo, B. Tetrahedron Lett. 2004, 45, 515–518. (292) Guarnaccia, R.; Madyastha, K. M.; Tegtmeyer, E.; Coscia, C. J. Tetrahedron Lett. 1972, 50, 5125-5127. (293) Konig, G.; Rimpler, H.; Hunkler, D. Phytochemistry 1987, 26, 423-427. (294) Rimpler, H.; Sauerbier, H. Biochem. Syst. Ecol. 1986, 14, 307-310. (295) Hirata, T.; Kobayashi, T.; Wada, A.; Ueda, T.; Fujikawa, T.; Miyashita, H.; Ikeda, T.; Tsukamoto, S.; Nohara, T. Bioorg. Med. Chem. Lett. 2011, 21, 1786-1791. (296) Inouye, H.; Takeda, Y.; Uobe, K.; Yamauchi, K.; Yabuuchi, N.; Kuwano, S. Planta Med. 1974, 25, 285-288. (297) Jin, X.; Sun, J.; Xie, W.; Wan, Z.; Jin, Y.; Zhu, J. Zhongguo Zhong Yao Za Zhi 2009, 34, 3082-3086. (298) Park, K. S.; Kim, B. H.; Chang, I.-M. Evidence-Based Complementary and Alternative Medicine, 7, 41-45.

228

(299) Toda, S.; Miyase, T.; Arichi, H.; Tanizawa, H.; Takino, Y. Chem. Pharm. Bull. 1985, 33, 1270-1273. (300) Hsu, H.-Y.; Yang, J.-J.; Lin, S.-Y.; Lin, C.-C. Cancer Lett. 1997, 113, 31-37. (301) Ueda, S.; Iwahashi, Y.; Tokuda, H. J. Nat. Prod. 1991, 54, 1677-1680. (302) Nakamura, T.; Nakazawa, Y.; Onizuka, S.; Satoh, S.; Chiba, A.; Sekihashi, K.; Miura, A.; Yasugahira, N.; Sasaki, Y. F. Mutat. Res.-Genet. Tox. 1997, 388, 7-20. (303) Cameron, D. W.; Feutrill, G. I.; Perlmutter, P.; Sasse, J. M. Phytochemistry 1984, 23, 533-535. (304) Scarpati, M. L.; Delle, M. F. Ann. Chim. 1963, 53, 356-367. (305) Casanova, E.; Garcia-Mina, J. M.; Calvo, M. I. Plant Foods Hum. Nutr. 2008, 63, 93-97. (306) Guevenc, A.; Okada, Y.; Akkol, E. K.; Duman, H.; Okuyama, T.; Calis, I. Food Chem. 2009, 118, 686-692. (307) Hausmann, M.; Obermeier, F.; Paper, D. H.; Balan, K.; Dunger, N.; Menzel, K.; Falk, W.; Schoelmerich, J.; Herfarth, H.; Rogler, G. Clin. Exp. Immunol. 2007, 148, 373-381. (308) Kim, K. H.; Kim, S.; Jung, M. Y.; Ham, I. H.; Whang, W. K. Arch. Pharm. Res. 2009, 32, 7-13. (309) Lee, J. Y.; Woo, E.-R.; Kang, K. W. J. Ethnopharmacol. 2005, 97, 561-566. (310) Ye, M.; Zhao, Y.; Norman, V. L.; Starks, C. M.; Rice, S. M.; Goering, M. G.; O'Neil-Johnson, M.; Eldridge, G. R.; Hu, J.-F. Phytother. Res. 2010, 24, 778-781. (311) Aparna, P.; Tiwari, A. K.; Srinivas, P. V.; Ali, A. Z.; Anuradha, V.; Rao, J. M. Phytother. Res. 2009, 23, 591-596. (312) Kim, J. K.; Lee, Y. S.; Kim, S. H.; Bae, Y. S.; Lim, S. S. Biol. Pharm. Bull. 2011, 34, 160-163. (313) Backhouse, N.; Delporte, C.; Apablaza, C.; Farias, M.; Goity, L.; Arrau, S.; Negrete, R.; Castro, C.; Miranda, H. J. Ethnopharmacol. 2008, 119, 160-165. (314) Zhang, F.; Jia, Z.; Deng, Z.; Wei, Y.; Zheng, R.; Yu, L. Planta Med. 2002, 68, 115–118. (315) Lee, K.-W.; Kim, H. J.; Lee, Y. S.; Park, H.-J.; Choi, J.-W.; Ha, J.; Lee, K.-T. Carcinogenesis 2007, 28, 1928-1936. (316) Fleer, H.; Verspohl, E. J. Phytomedicine 2007, 14, 409-415. (317) Isacchi, B.; Iacopi, R.; Bergonzi, M. C.; Ghelardini, C.; Galeotti, N.; Norcini, M.; Vivoli, E.; Vincieri, F. F.; Bilia, A. R. J. Pharm. Pharmacol. 2011, 63, 594-601. (318) Nakamura, T.; Okuyama, E.; Tsukada, A.; Yamazaki, M.; Satake, M.; Nishibe, S.; Deyama, T.; Moriya, A.; Maruno, M.; Nishimura, H. Chem. Pharm. Bull. 1997, 45, 499-504. (319) Boros, C. A.; Stermitz, F. R. J. Nat. Prod. 1990, 53, 1055-1147. (320) Takeda, Y.; Nishimura, H.; Inouye, H. Phytochemistry 1977, 16, 1401-1404. (321) Otsuka, H.; Watanabe, E.; Yuasa, K.; Ogimi, C.; Takushi, A.; Takeda, Y. Phytochemistry 1993, 32, 983-986. (322) Afifi-Yazar, F. U.; Sticher, O.; Uesato, S.; Nagajima, K.; Inouye, H. Helv. Chim. Acta 1981, 64, 16-24. (323) Chaudhuri, R. K.; Afifi-Yazar, F. U.; Sticher, O.; Winkler, T. Tetrahedron 1980, 36, 2317-2326. (324) Baltenweck-Guyot, R.; Trendel, J.-M.; Albrecht, P.; Schaeffer, A. J. Nat. Prod. 1997, 60, 1326-1327. (325) Samoylenko, V.; Zhao, J.; Dunbar, D. C.; Khan, I. A.; Rushing, J. W.; Muhammad, I. J. Agr. Food Chem. 2006, 54, 6398-6402. (326) Agrawal, P. K. Phytochemistry 1992, 31, 3307-3330.

229

(327) Wang, M.; Kikuzaki, H.; Jin, Y.; Nakatani, N.; Zhu, N.; Csiszar, K.; Boyd, C.; Rosen, R. T.; Ghai, G.; Ho, C.-T. J. Nat. Prod. 2000, 63, 1182-1183. (328) Akihisa, T.; Seino, K.-I.; Kaneko, E.; Watanabe, K.; Tochizawa, S.; Fukatsu, M.; Banno, N.; Metori, K.; Kimura, Y. J. Oleo Sci. 2010, 59, 49-57. (329) Akihisa, T.; Matsumoto, K.; Tokuda, H.; Yasukawa, K.; Seino, K.-I.; Nakamoto, K.; Kuninaga, H.; Suzuki, T.; Kimura, Y. J. Nat. Prod. 2007, 70, 754-757. (330) Baltenweck-Guyot, R.; Trendel, J.-M.; Albrecht, P.; Schaeffer, A. Phytochemistry 1996, 43, 621-624. (331) ApSimon, J. W.; Haynes, N. B.; Sim, K. Y.; Whalley, W. B. J. Chem. Soc. 1963, 3780-3782. (332) Nakasugi, T.; Nakashima, M.; Komai, K. J. Agric. Food Chem. 2000, 48, 3256- 3266. (333) Kim, M.-J.; Han, J.-M.; Jin, Y.-Y.; Baek, N.-I.; Bang, M.-H.; Chung, H.-G.; Choi, M.-S.; Lee, K.-T.; Sok, D.-E.; Jeong, T.-S. Arch. Pharm. Res. 2008, 31, 429-437. (334) Kim, A. R.; Zou, Y. N.; Park, T. H.; Shim, K. H.; Kim, M. S.; Kim, N. D.; Kim, J. D.; Bae, S. J.; Choi, J. S.; Chung, H. Y. Phytother. Res. 2004, 18, 1-7. (335) Kim, M.-J.; Kim, D.-H.; Lee, K. W.; Yoon, D.-Y.; Surh, Y.-J. Ann. N.Y. Acad. Sci. 2007, 1095, 483-495. (336) Lv, W.; Sheng, X.; Chen, T.; Xu, Q.; Xie, X. J. Biomed. Biotechnol. 2008, 2008, 1-6. (337) Jeong, M. A.; Lee, K. W.; Yoon, D.-Y.; Lee, H. J. Ann. N.Y. Acad. Sci. 2007, 1095, 458-466. (338) Lee, H.-G.; Yu, K.-A.; Oh, W.-K.; Baeg, T.-W.; Oh, H.-C.; Ahn, J.-S.; Jang, W.- C.; Kim, J.-W.; Lim, J.-S.; Choe, Y.-K.; Yoon, D.-Y. J. Ethnopharmacol. 2005, 98, 339-343. (339) Lee, S. H.; Bae, E.-A.; Park, E.-K.; Shin, Y.-W.; Baek, N.-I.; Han, E.-J.; Chung, H.-G.; Kim, D.-H. Int. Immunopharmacol. 2007, 7, 1678-1684. (340) Clavin, M.; Gorzalczany, S.; Macho, A.; Munoz, E.; Ferraro, G.; Acevedo, C.; Martino, V. J. Ethnopharmacol. 2007, 112, 585-589. (341) Min, S.-W.; Kim, N.-J.; Baek, N.-I.; Kim, D.-H. J. Ethnopharmacol. 2009, 125, 497-500. (342) Pelzer, L. E.; Guardia, T.; Juarez, A. O.; Guerreiro, E. Farmaco 1998, 53, 421- 424. (343) Yin, Y.; Sun, Y.; Gu, L.; Zheng, W.; Gong, F.; Wu, X.; Shen, Y.; Xu, Q. Eur. J. Pharmacol. 2011, 651, 205-211. (344) Davis, R. A.; Carroll, A. R.; Pierens, G. K.; Quinn, R. J. J. Nat. Prod. 1999, 62, 419-424. (345) Andrews, K. T.; Walduck, A.; Kelso, M. J.; Fairlie, D. P.; Saul, A.; Parsons, P. G. Int. J. Parasitol. 2000, 30, 761-768. (346) Huber, W.; Koella, J. C. Acta Trop. 1993, 55, 257-261. (347) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J.; Vistica, D.; Warren, J. T.; Bokesch, H.; Kenney, S.; Boyd, M. R. J. Natl. Cancer Inst. 1990, 82, 1107-1112.

230

Appendix I. Eremophila analytical HPLC chromatograms

E. goodwinii subsp. goodwinii E. longifolia

E. bignoniiflora E. latrobei subsp. glabra

E.longifolia E. maculata subsp. maculata

E. dalyana E. gilesii subsp. gilesii

E. bowmanii subsp. bowmanii UV wavelength key

E. bowmanii subsp. latifolia E. longifolia

E. latrobei subsp. latrobei E. longifolia

E. glabra subsp. glabra E. obovata

E. gilesii subsp. gilesii E. latrobei subsp. latrobei

E. gilesii subsp. gilesii UV wavelength key

Eremophila sp. E. glabra

E. pantonii E. resinosa

E. fraseri E. glabra

E. oldfieldii subsp. augustifolia E. eriocalyx

E.oldfieldii subsp. augustifolia UV wavelength key

E. linsmithii E. platycalyx

E. platycalyx E.spectabilis

E. longifolia E. rotundifolia

E. drummondii E. longifolia

Eremophila sp. UV wavelength key

E. calorhabdos E. eriocalyx

E. bignoniiflora E. microtheca

UV wavelength key

Appendix II. CD NMR data list for thesis compounds

NMR spectra were recorded at 30 °C on either a Varian 400 MHz, 500 MHz or

600 MHz Unity INOVA spectrometer. The latter spectrometer was equipped with a triple resonance cold probe. All the spectra given as supplementary material were

1 13 obtained in DMSO-d6, the H and C chemical shifts were referenced to the solvent peaks for DMSO-d6 at δH 2.49 and δC 39.5.

Each directory contains the relevant 1D/2D NMR spectra for each individual compound. The list below shows the directory code for each compound.

List of compound directories

Compound_62 [14-hydroxy-6,12-muuroloadien-15-oic acid] Compound_68 [Mitchellene A] Compound_69 [Mitchellene B] Compound_70 [Mitchellene C] Compound_71 [Mitchellene D] Compound_72 [Mitchellene E] Compound_73 [Casticin] Compound_74 [Centaureidin] Compound_82 Compound_83 Compound_84 Compound_86 Compound_87 Compound_90 Compound_91 Compound_92 Compound_93 Compound_95 Compound_99

Compound_100 Compound_101 Compound_102 Compound_103 Compound_104 Compound_105 Compound_106 Compound_107 Compound_108 Compound_109 Compound_110 [Verbascoside] Compound_155 [Geniposidic acid] Compound_181 [Mussaenoside] Compound_182 [Ladroside] Compound_186 [5,19-Dihydroxy-3,14-viscidadien-20-oic acid] Compound_188 [3,15-Epoxy-19-hydroxycembra-7,11-dien-18-oic acid] Compound_191 [3,15-Epoxycembra-7,11-dien-18-oic acid]

Compound_193 [3-Methylbut-3-enyl O-α-L-rhamnopyranosyl-(1→6)-O-β-D- glucopyranoside] Compound_198 [3-Acetoxy-7,8-dihydroxyserrulat-14-en-19-oic acid] Compound_199 [3,7,8-Trihydroxyserrulat-14-en-19-oic acid] Compound_200 [3,19-Diacetoxy-8-hydroxyserrulat-14-ene] Compound_201 [Jaceosidin] Compound_206 [3,8-Diacetoxy-7-hydroxyserrulat-14-en-19-oic acid] Compound_207 [3,7,8-Trihydroxyserrulat-14-en-methyl-19-benzoate] Compound_208 [3,8-Dihydroxy-7-methoxyserrulat-14-en-methyl-19-benzoate]