1. sports
2. skating
3. figure skating

Isolation and Structure Elucidation of
Cytotoxic Natural Products with
Potential Anticancer Activity
Trong Duc Tran
(B.Eng)
Eskitis Institute for Cell and Molecular Therapies
Science, Environment, Engineering and Technology
Griffith University
Submitted in fulfilment of the requirements of the degree of
Master of Philosophy
July 2010
Declaration
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
……………………………….
……………………
Trong Duc Tran
Date
i
ii
Acknowledgements
First of all, I would like to express my sincere appreciation to my supervisors, Professor
Ronald J. Quinn and Dr. Ngoc B. Pham, for their patience, guidance and support. My
skills as a scientist have definitely matured under the supervision of Prof. Quinn and Dr.
Pham and for that I am most grateful.
I would like to thank Dr. Gregory Fechner for his kind supervision in training for the
bioassay screening. I also wish to thank Assoc. Prof. Anthony Carroll, Dr. Rohan Davis,
Dr. Yun Feng, Dr. Phuc Le, Dr. Hoan Vu, Dr. Harish Holla, Dr. Xinzhou Yang, Dr.
Sheng Yin and Ms. Lekha Suraweera for all academic and technical discussions; Assoc.
Prof. Andreas Hoffmann for the circular dichroism measurements; Ms. B. Aldred for
the provision of cancer cell lines; Dr. John Hooper of the Queensland Museum, Dr. Paul
Forster and Dr. G. Guymer of the Queensland Herbarium for the collection and
identification of biota samples. I thank all my friends Emma Barnes, Michelle Liberio
and Asiah Osman for many scientific and non-scientific conversations.
I acknowledge Education Australia Ltd for the provision of the “EAL Postgraduate
Research Student Mobility Scholarships” which allowed me to pursue my full-time
research.
iii
iv
Abstract
This project presented a strategy to select a subset of prefractionated fractions
for screening. Marine and plant biota samples were chosen based on their rare
taxonomies and mass spectroscopic data. A method to reduce 1155 fractions generated
from 105 selected samples to 330 UV active fractions was developed.
Cancer cell-based screening of 330 prefractionated fractions against four cancer
cell lines (A549, HeLa, LNCaP and PC3) and non-cancer cells (HEK) resulted in
nineteen active fractions belonging to fourteen biota samples (two plants and twelve
marine organisms). One plant and six marine animals were chosen for further
investigation. Subsequent mass-guided isolation led to the identification of forty-four
secondary metabolites, nine of which were not previously reported. Structures of these
compounds were elucidated by spectroscopic methods (1D and 2D-NMR, MS, CD and
specific optical rotation) and chemical methods.
In the plant sample Neolitsea dealbata, a total of nine alkaloids (55-63) were
isolated. A new aporphine, normecambroline (55), showed selective activity against
HeLa cells with an IC50 of 4.0 μM while its analogue, roemerine (56), displayed nonselective activity against all four cancer cell lines (A549, HeLa, LNCaP and PC3) and
two non cancer cell lines (HEK, NFF).
One of the marine samples, a specimen of the Potter Reef marine sponge
Diacarnus sp., showed the presence of terpene peroxides. Three known peroxide
compounds, sigmosceptrellin (27), deacarperoxide (28) and methyldiacarnoate (29),
were identified. Compound 27 inhibited cytotoxicity against all six cell lines with IC50
values ranging from 0.4 to 3.1 μM while the other two compounds were not active.
Chemical investigation of a marine specimen from Houghton Reef, Neopetrosia
exigua, resulted in the isolation of two cytotoxic compounds, mortuporamine C (42) and
a new 3-alkylpyridinium alkaloid, dehydrocyclostellettamine A (43). These two
compounds displayed activity against four cancer cell lines with IC50 values in
micromolar concentrations (3.0-13.7 μM).
v
Three new cyclodepsipeptides, neamphamide B (86), neamphamide C (87) and
neamphamide D (88), were found as constituents of a rare marine sponge Neamphius
huxleyi collected at Milln Reef, off Cape Grafton. Their structures were elucidated by
NMR spectroscopy and multiple stages of accurate mass measurements (ESI-FTICRMSn). Stereochemistry of residues of the peptides was determined by the Marfey amino
acid method and J-based configurational analysis. These compounds inhibited the cell
growth with IC50 values ranging from 91.3 to 366.1 nM.
Two previously unreported milnamide E (116) and hemiasterlin D (117)
together with nine known small peptides were isolated from a new sponge genus,
Pipestela candelabra. Compound 117 was identified as the first peptide skeleton
discovered in nature with a side chain containing 2-hydroxyacetic acid, tert-leucine and
N-methylvinylogous valine residues attached to the indole nitrogen. This compound
exhibited activity against HeLa cells with an IC50 of 1.8 nM. Cytotoxic results indicated
all hemiasterlin derivatives were approximately 100 fold more active against cancer cell
lines than the milnamide family.
A series of sixteen bromotyrosine alkaloids were identified from two Australian
sponges Suberea clavate and Pseudoceratina sp.. Two new bromotyrosine derivatives,
pseudoceralidinone A (148) and aplysamine 7 (149) were isolated from a specimen of
the Hook Reef lagoon sponge, Pseudoceratina sp.. Their absolute stereostructures were
determined by synthetic methods. Compound 149 inhibited moderate cytotoxicity (an
IC50 of 4.9 μM against PC3 cells) while compound 148 displayed no activity.
All isolated compounds were evaluated for their physico-chemical properties.
Results showed that thirty-six out of forty-four compounds (81.8%) passed Lipinski’s
rule and twenty-nine compounds (65.9%) displayed no violation against the
requirements of both Lipinski’s and Veber’s rule.
vi
Abbreviations
HPLC
High pressure liquid chromatography
RP
Reverse phase
18
C
octadecyl bonded silica
PAG
polyamide gel
PDA
Photo diode array
UV
ultra violet
IR
Infra red
CD
circular dichroism
[α]D
specific rotation
MS
mass spectrometry
MSn
multi stage mass spectrometry
ESI
electrospray ionization
LRESIMS
low resolution electrospray mass spectrum
HRESIMS
high resolution electrospray mass spectrum
FTICR
Fourier transform ion cyclotron resonance
MW
molecular weight
m/z
mass-ion ratio (z = 1)
amu
atomic mass unit
2D
two dimensional
NMR
nuclear magnetic resonance
COSY
correlation spectroscopy
HSQC
heteronuclear single quantum coherence
HMBC
heteronuclear multiple bond correlations
ROESY
rotating frame overhauser effect spectroscopy
TOCSY
total correlation spectroscopy
HSQMBC
heteronuclear single quantum multiple bond correlation
HSQC-TOCSY
heteronuclear multiple bond correlation total correlation
spectroscopy
2,3
JCH
2 or 3 bond hydrogen to carbon correlation
s
singlet
d
doublet
t
triplet
vii
m
multiplet
br
broad
ppm
parts per million
MHz
megahertz
DMSO
dimethylsulfoxide
DCM
dichloromethane
EtOH
ethanol
MeOH
methanol
CDCl3
deuterated chloroform
DMSO-d6
deuterated dimethylsylfoxide
CD3OD-d4
deuterated methanol
CD3OH-d3
deuterated methanol
TFA
trifluoroacetic acid
FA
formic acid
CE
Cotton effect
Ro5
the Rule of Five
IC50
concentration of a compound required to inhibit 50% of the receptor
population
SAR
structure-activity relationship
(Boc)2O
di-tert-butyl dicarbonate
EDCI
1-ethyl-3-(3-dimethylaminopropyl) carbodiimide
HOBt
Hydroxybenzotriazole
DMF
Dimethylformamide
viii
Table of Contents
Declaration
i
Acknowledgements
iii
Abstract
v
Abbreviations
vii
List of Figures/Plates
xiii
List of Tables/Lists
xvi
List of Schemes
xviii
List of Publications/Conference Poster
xix
Chapter 1
Introduction
1.1 The contribution of natural products to drug discovery
1
1.2 Natural products as potential anticancer agents
3
1.3 The application of “drug-like” properties into natural product discovery
7
1.3.1 The decline of interest on natural products from big pharmaceutical
companies
7
1.3.2 Physico-chemical properties
7
1.3.3 Do natural products have and “drug-like” properties?
9
1.3.4 A new paradigm in building a screening natural products library
10
1.4 Objective – Research plan
12
1.5 References
14
Chapter 2
Generation of prefractionated fractions for biological screening
2.1 Library analysis
19
2.1.1 Strategies for choosing biota samples
20
2.1.2 Strategies for selecting fractions
21
2.2 Re-generation of selected drug-like fraction library
22
2.3 Cytotoxicity screening
24
2.4 LC/MS analysis for bioactive fractions
26
2.5 Generation of hit list
26
2.5.1 Pipestela candelabra (QID6012360)
26
2.5.2 Neamphius huxleyi (QID6005333)
27
2.5.3 Pseudoceratina sp. (QID024253)
27
2.5.4 Suberea clavata (QID025573)
28
ix
2.5.5 Diacarnus sp. (QID6011588)
28
2.5.6 Neopetrosia exigua (QID6008038)
29
2.5.7 Neolitsea dealbata (QID018658)
29
2.5.8 Chondria sp. (QID6008434)
30
2.5.9 Aaptos aaptos (QID021901)
30
2.5.10 Diacarnus spinipoculum (QID012499)
31
2.5.11 Cystophora moniliformis (QID6015742)
31
2.5.12 Bohadschia marmorata (QID021666)
32
2.5.13 Erylus amissus (61846.9)
32
2.5.14 Solanum jucundum A.R.Bean (11044.8)
32
2.6 Screening results
33
2.7 References
35
Chapter 3
Previously isolated compounds from the Australian sponges
Diacarnus sp. and Neopetrosia exigua – A new 3-alkylpyridinium
alkaloid from N. exigua
3.1 Chemical investigation of the sponge Diacarnus sp.
37
3.1.1 Collection, Extraction and Isolation
37
3.1.2 Structure Elucidation
39
3.2 Chemical Investigation of the Sponge Neopetrosia exigua
40
3.2.1 Collection, Extraction and Isolation
42
3.2.2 Structure Elucidation and Discussion
43
3.3 Evaluation of “drug-like” properties of the isolated compounds
46
3.4 Biological activity
46
3.5 References
46
Chapter 4
Aporphinoid alkaloids from the Australian plant
Neolitsea dealbata
4.1 Introduction
51
4.2 Collection, Extraction and Isolation
55
4.3 Structure Elucidation and Discussion
57
4.3.1 Normecambroline (55)
57
4.3.2 Roemerine-Nα-oxide (57)
60
4.3.3 Roemerine-Nβ-oxide (58)
62
4.4 Evaluation of drug-like properties
65
x
4.5 Biological activity
65
4.6 References
67
Chapter 5
Cyclodepsipeptides from the Australian sponge
Neamphius huxleyi
5.1 Introduction
71
5.2 Collection, Extraction and Isolation
77
5.3 Structure Elucidation and Discussion
79
5.3.1 Neamphamide B (86)
79
5.3.2 Neamphamide C (87)
92
5.3.3 Neamphamide D (88)
98
5.4 Evaluation of drug-like properties
102
5.5 Biological activity
102
5.6 References
102
Chapter 6
Hemiasterlins and milnamides from the Australian sponge
Pipestela candelabra
6.1 Introduction
105
6.2 Collection, Extraction and Isolation
109
6.3 Structure Elucidation and Discussion
113
6.3.1 Milnamide E (116)
113
6.3.2 Hemiasterlin D (117)
119
6.4 Evaluation of “drug-like” properties
124
6.5 Biological activity
124
6.6 References
126
Chapter 7
Bromotyrosine alkaloids from the Australian sponges Suberea
clavate and Pseudoceratina sp.
7.1 Introduction
129
7.2 Collection, Extraction and Isolation
136
7.2.1 Collection, Extraction and Isolation for the sponge Suberea clavate
136
7.2.2 Collection, Extraction and Isolation for the sponge Pseudoceratina sp.
139
7.3 Structure Elucidation and Discussion
141
7.3.1 Pseudoceralidinone A (148)
141
7.3.2 Aplysamine 7 (149)
146
xi
7.3.3 Proposed biosynthesis of pseudoceralidinone A (148) and
aplysamin 7 (149)
152
7.4 Evaluation of drug-like properties
153
7.5 Biological activity
153
7.6 References
154
Chapter 8
Conclusion
161
General Experimental Details
163
APPENDIX
Appendix 1
1.1 Acquisition of Spectroscopy Data
163
1.2 Isolation of Natural Products
163
1.3 Physico-chemical Parameter Calculation
163
1.4 Biological screening
163
Appendix 2
Chapter 3 Experimental
165
Appendix 3
Chapter 4 Experimental
166
Appendix 4
Chapter 5 Experimental
168
Appendix 5
Chapter 6 Experimental
171
Appendix 6
Chapter 7 Experimental
174
Appendix 7
CD NMR data of isolated compounds in this research
xii
List of Figures/Plates
Figure 1.1
All available anticancer drugs, 1940s – 06/2006
3
Figure 1.2
Structure activity relationships for the sarcodictyins
6
Figure 1.3
Lipinski violations for 546 marketed oral drugs in
the FDA Orange Book 2007
8
Figure 1.4
Some processes generating prefractionated libraries
12
Figure 2.1
UV absorption analysis of some kinds of drugs (λ = 210-400 nm) 19
Figure 2.2
Examples of selecting fractions
22
Figure 2.3
HPLC profile of a biota sample in a prefractionation process
23
Figure 2.4
Analytical testing with simple chromophores
24
Figure 2.5
Mechanism of Alamar Blue indicator
25
Figure 2.6
Histogram of selected fractions
33
Figure 2.7
Total cancer cell-based screening results
34
Plate 3.1
Photograph of the sponge Diacarnus sp.
38
Plate 3.2
Photograph of the sponge N. Exigua
42
Figure 3.1
The 1H-NMR spectrum of 43 recorded at 600 MHz in DMSO-d6
44
Figure 3.2
The partial structures A, B and C of 43
44
Figure 3.3
Key HMBC correlations to establish the structure of 43
45
Figure 4.1
Some types of aporphinoid structures
51
Figure 4.2
The special feature of the aporphine system
52
Figure 4.3
Compounds previously isolated from the plant N. dealbata
53
Figure 4.4
The 1H-NMR spectrum of 55 recorded at 600 MHz in DMSO-d6
57
Figure 4.5
The partial structures A, B and C of compound 55
58
Figure 4.6
Key HMBC and ROESY correlations to establish
the structure of 55
58
Figure 4.7
CD spectra of compound 55 and 59
59
Figure 4.8
The 1H-NMR spectrum of 57 recorded at 600 MHz in DMSO-d6
60
Figure 4.9
Key HMBC and ROESY correlations to establish
the structure of 57
61
Figure 4.10
The 1H-NMR spectrum of 58 recorded at 600 MHz in DMSO-d6
62
Figure 4.11
Key HMBC and ROESY correlations to establish
Figure 4.12
the structure of 58
62
CD spectra of compound 56, 57 and 58
63
xiii
Figure 5.1
Two compounds isolated from the sponge Neamphius huxleyi
71
Figure 5.2
Four uncommon residues identified in 65 and 66
73
Plate 5.1
Photograph of the sponge Neamphius huxleyi
77
Figure 5.3
The 1H-NMR spectrum of 86 recorded at 600 MHz in MeOH-d3
79
Figure 5.4
Key HMBC and ROESY correlations to establish 2 partial
structures of 86
Figure 5.5
84
Key HMBC and ROESY correlations to establish
the structure of 86
84
2
Figure 5.6
FTMS spectrum of 86
86
Figure 5.7
Mechanism of Marfey’s reagent (FDAA)
86
Figure 5.8
Relative configurations of the β-OMeTyr, Agdha and
Htmha residues
89
1
Figure 5.9
The H-NMR spectrum of 87 recorded at 600 MHz in MeOH-d4
92
Figure 5.10
FTMS2 spectrum of 87
93
Figure 5.13
The 1H-NMR spectrum of 88 recorded at 600 MHz in MeOH-d4
98
Figure 5.14
Htmoa unit of 88 with key COSY/TOCSY and
HMBC correlations
99
Figure 6.1
Three residues characterised from milnamides and hemiasterlins
105
Figure 6.2
Structure activity relationship of 89
107
Figure 6.3
Two synthetic analogues of hemiasterlin
107
Plate 6.1
Photograph of the sponge Pipestela candelabra collected at
Wilson Reef, Coral Sea
Plate 6.2
109
Photograph of the sponge Pipestela candelabra collected at
Houghton Reef, Howick Group
111
Figure 6.4
The 1H-NMR spectrum of 116 recorded at 600 MHz in DMSO-d6 113
Figure 6.5
The partial structures A, B and combined A-B of 116
114
Figure 6.6
The partial structures C, D and combined C-D of 116
115
Figure 6.7
Key HMBC and ROESY correlations to establish
the structure of 116
116
Figure 6.8
CD spectrum of compounds 90 and 116
116
Figure 6.9
The 1H-NMR spectrum of 117 recorded at 600 MHz in DMSO-d6 119
Figure 6.10
The partial structures A, B, C and combined C-A-B of 117
120
Figure 6.11
The partial structures D and E of compound 117
121
Figure 6.12
Key HMBC and COSY correlations to establish
the structure of 117
121
xiv
Figure 7.1
Skeletons represented for simple bromotyrosine alkaloids
130
Figure 7.2
Skeletons represented for oxime bromotyrosine alkaloids
131
Figure 7.3
Skeletons represented for bastadin alkaloids
131
Figure 7.4
Skeletons represented for spirooxepinisoxazoline bromotyrosine
alkaloids
Figure 7.5
132
Skeletons represented for spirocyclohexadienylisoxazoline
alkaloids
133
Plate 7.1
Photograph of the sponge Suberea clavate
136
Plate 7.2
Photograph of the sponge Pseudoceratina sp.
139
Figure 7.6
The 1H-NMR spectrum of 148 recorded at 600 MHz in DMSO-d6 141
Figure 7.7
The partial structures A, B and C of 148
142
Figure 7.8
Key HMBC correlations to establish the structure of 148
142
Figure 7.9
Configurational correlation model for the (S)-MTPA and
(R)-MTPA esters
Figure 7.10
143
Δδ SR values for MTPA derivatives (155 and 156) and
their configurational correlation model
145
Figure 7.11
The 1H-NMR spectrum of 149 recorded at 600 MHz in DMSO-d6 146
Figure 7.12
The partial structures A, B and C of 149
147
Figure 7.13
Key HMBC correlations to establish the structure of 149
148
Figure 8.1
Histogram for all isolated compounds showing physico-chemical
parameters
Figure 8.2
161
Comparison in passing Lipinski rule and combined Lipinski with
Veber rules of all isolated compounds
xv
161
List of Tables/Lists
Table 2.1
Gradient timetable for LC/MS analysis of crude extracts
21
Table 2.2
Gradient timetable for prefractionation
23
Table 2.3
Gradient timetable for LC/MS analysis of active fractions
26
List 2.1
Taxonomic information of samples contains 19 active fractions
34
List 2.2
Chapters describing the investigated samples
34
Table 3.1
Taxonomy of sponges producing terpene peroxides
37
Table 3.2
Taxonomy of sponges producing 3-alkylpyridinium alkaloids
42
Table 3.3
NMR data for dehydrocyclostellettamine A (43) in
DMSO- d6
45
Table 3.4
Physico-chemical properties of isolated compounds
46
Table 3.5
Evaluation of cytotoxic potential of compounds 27-29 and 42-43
46
Table 4.1
NMR data for TFA salt of (6aR)-normecambroline (55) in
59
DMSO-d6
Table 4.2
NMR data for TFA salt of (6S,6aR)-roemerine-Nα-oxide (57) in
DMSO-d6
Table 4.3
61
NMR data for TFA salt of (6S,6aR)-roemerine-Nβ-oxide (58) in
63
DMSO-d6
Table 4.4
NMR data for TFA salt of compounds 56, 57 and 58 in DMSO-d6 64
Table 4.5
Physico-chemical properties of compounds 55-63
65
Table 4.6
Evaluation of cytotoxic potential of isolated compounds
66
Table 5.1
20 natural amino acids encoded by DNA
72
Table 5.2
Retention times of authentic FDAA-amino acids and
the hydrolysates of 86
87
Table 5.3
NMR comparison of the β-OMeTyr residue
88
Table 5.4
NMR data for neamphamide B (86)
89
Table 5.5
NMR data for neamphamide C (87) in CD3OD
95
Table 5.6
NMR data for neamphamide D (88) in CD3OD
100
Table 5.7
Physico-chemical properties of the isolated compounds
102
Table 5.8
Biological activity of the isolated compounds
102
Table 6.1
Taxonomy of sponges producing milnamides and hemiasterlins
106
Table 6.2
NMR data for FA salt of milnamide E (116) in DMSO-d6
117
Table 6.3
NMR data for FA salts of milnamide E (116), isolated
xvi
milnamide A (90) and referenced milnamide A in CD3CN-d3
118
Table 6.4
NMR data for TFA salt of hemiasterlin D (117) in DMSO- d6
123
Table 6.5
Physico-chemical properties of the isolatedcompounds
124
Table 6.6
Cytotoxicity evaluation of isolated compounds
124
Table 7.1
Taxonomy of sponges producing bromotyrosine alkaloids
129
Table 7.2
NMR data for the MTPA Esters of 154 in DMSO-d6
144
Table 7.3
NMR data for TFA salt of 148 in DMSO-d6
145
Table 7.4
NMR data for TFA salt of 149 in DMSO-d6
150
Table 7.5
NMR data for the MTPA Esters of 165a and 165b in DMSO-d6
151
Table 7.6
Physico-chemical properties of all isolated compounds
153
Table 7.7
Biological activity of some isolated compounds
154
xvii
List of Schemes
Scheme 1.1
Traditional way in natural product drug discovery
10
Scheme 1.2
Research program
13
Scheme 2.1
Diagram of selecting fractions for further investigations
20
Scheme 3.1
Extraction and Isolation Procedure for Diacarnus sp.
39
Scheme 3.2
Extraction and Isolation Procedure for N. exigua
43
Scheme 4.1
Extraction and Isolation Procedure for N. dealbata
56
Scheme 5.1
Extraction and Isolation Procedure for Neamphius huxleyi
78
Scheme 5.2
FTMS2 fragmentations and related neutral losses of 86
85
2,3
Scheme 5.3
FTMS
fragmentations and related neutral losses of 87
Scheme 5.4
FTMS2 fragmentations and related neutral losses of 87
(isolated m/z 788.4396)
Scheme 6.1
95
Extraction and Isolation Procedure for Pipestela candelabra
collected at Wilson Reef, Coral Sea
Scheme 6.2
110
Extraction and Isolation Procedure for Pipestela candelabra
collected at Houghton Reef, Howick Group
Scheme 7.1
94
112
Proposed biosynthesis of spirooxepinisoxazoline and
spirocyclohexadienylisoxazoline rings
133
Scheme 7.2
Extraction and Isolation Procedure for Suberea clavate
138
Scheme 7.3
Extraction and Isolation Procedure for Pseudoceratina sp.
140
Scheme 7.4
Modification of 148 to determine its absolute stereochemistry
144
Scheme 7.5
Total synthesis of oxime-protected aplysamin 7 (165)
148
Scheme 7.6
Isolation and stereochemical determination of enantiomers
Scheme 7.7
165a and 165b
149
A plausible biosynthesis of 148 and 149
152
xviii
List of Publications
1) Trong D. Tran, Ngoc B. Pham, Gregory Fechner and Ronald J. Quinn, Chemical
Investigation of Drug-like Compounds from the Australian Tree, Neolitsea dealbata,
Bioorganic & Medicinal Chemistry Letters (approved).
2) Trong D. Tran, Ngoc B. Pham, Gregory Fechner and Ronald J. Quinn, A Novel
Cytotoxic Peptide Hemiasterlin D and Milnamide E from the Australian Sponge
Pipestela candelabra (in preparation)
3) Trong D. Tran, Ngoc B. Pham, Gregory Fechner and Ronald J. Quinn, New
Cytotoxic Cyclic Depsipeptides from the Autralian Marine Sponge Neamphius huxleyi
(in preparation)
4) Trong D. Tran, Ngoc B. Pham, Gregory Fechner and Ronald J. Quinn, New
Bromotyrosine Alkaloids from the Australian Marine Sponge Pseudoceratina sp. (in
preparation)
Conference Poster
1) Drug-like alkaloids from the Australian tree, Neolitsea dealbata, RACI 2010 (Royal
Australian Chemical Institute's National Convention), Melbourne Convention Centre,
Melbourne, Australia (4th – 8th July 2010)
xix
Chapter 1
Introduction
1.1 The contribution of natural product to drug discovery
Natural product chemistry emerged in the 1800’s with the landmark isolation of
salicin (1) from the white willow bark, Salix alba in 1825.1 In the early stages, natural
product research mainly focused on medicinal plants which had been extensively
documented in written or verbal forms and passed down from generation to generation
over thousands of years.2,3 For instance, turmeric and ginger, two popular spices in
Asian countries, were known as effective ethnomedicines treating stomach-ache,
arthritis, asthmatic and tussive in the traditional Indian and Chinese medicines for over
3000 years.4,5 Not until the twentieth century were the mysteries of the biologically
active components in these spices disclosed scientifically.4 Curcumin (2) was first
isolated from turmeric in 18156 but its structure was not elucidated until 1910 as
diferuloylmethane.7 Curcumin and its analogues in turmeric were found to have
potential in the prevention and treatment of inflammation, cardiovascular disease,
Alzheimer's disease, memory deficits, arthritis and cancer.5 The isolation and structural
characterization of ginkgolide B (3) and its derivatives in ginger were first reported in
1967.8 Biological investigation of these compounds indicated they are the antiasthmatic
and antitussive components of ginger.5
H
O
HO
OH
O
O
O
O
HO
O
O
H3CO
HO
OH
H
OH
Salicin (1)
H3CO
OH
OH
N
H
HO
O
OCH3
O
HO
H
O
HO
OH
O
N
Quinine (4)
Ginkolide B (3)
Curcumin (2)
HO
OH
O
O
H
O
H
H
NH2
S
H3CO
N
H
O
N
N
HOOC
N
COOH
HO
O
H
H
S
N
H
O
N
O
COOH
Morphine (5)
Galanthamine (6)
Penicillin G (7)
O
Cephalosporin C (8)
Due to the development in the spectroscopic technology, natural product
research has gained a huge advance in its ability to undertake chemical identification.
So far, 15% of the 500,000 known plant species1 and about 5% of a total of 1.5 million
fungal species5 have been investigated chemically. Some of the identified compounds
1
have become indispensable drugs or the core structures of specific drugs, such as
quinine (4) (antimalarial), morphine (5) (anti-analgesic), galanthamine (6) (Alzheimer’s
disease), penicillins (7) (antibiotics) and cephalosporins (8) (antibiotics).
OH
OH
H
O
HO
H
O
O
O
HO
O
O
H
O
H
O
O
O
H
OH
O
O
OH
OH
OH
OH
OH
O
O
O
O
O
OH
OH
Marinisporolide A (11)
O
Marinomycin A (10)
O
O
OH
OH
HO
H
O
OH
OH
OH
O
OH
NH2
O
OH
Abyssomicin C (12)
Discodermolide (13)
It was not until the 1950s that natural product research commenced to focus on
marine sources.9 In the last 60 years, over 16,000 marine natural products were
discovered from marine microbes, algae and invertebrates and this number is increasing
rapidly.10 In comparison with terrestrial sources, research on marine organisms has
encountered more challenges. One of the most difficult tasks is deep-water sample
collection as it requires high-tech deep-water collection tools. The scarcity of biota, due
to the initial collection difficulty, often leads to more challenges in the following steps
of isolation, structure identification and pharmacological testing.10 For example, one
tonne of the sponge Lissodendoryx sp. was collected in order to obtain only 300 mg of
halichondrin B (9) for clinical testing.10 Despite these drawbacks, the novelty in
structures and interesting bioactivities of marine natural products have always appealed
to scientists in the development of new therapeutic agents.11 The discovery of polyenepolyol and polycyclic polyketides, such as marinomycin A (10), marinisporolide (11)
and abyssomicin C (12) from the new marine genera Marinispora and Salinispora has
provided new promising antibiotic candidates and might be complementing to penicillin
and its analogues discovered from fungi sources over 50 years ago.12 Some marine
products, such as halichondrin B (9), discodermolide (13), kahalalide F (17), dolastatin
10 (18) and okadaic acid (19) are now in preclinical and clinical trial for treating
cancer.13,14 According to Buss and Butler,15 over 20 drugs derived from actinomycetes,
bacteria, fungi, higher plants, marine invertebrates and vertebrates were introduced to
the market in the period 2003-2008. Moreover, there are 36 additional natural product-
2
O
OH
OH
O
O
O
H
O
O
OH
O
O
O
Halichondrin B (9)
O
OH
H
derived compounds currently in late stage drug development.15 Natural products thus
play a critical role in drug discovery programs.
1.2 Natural products as potential anticancer agents
Cancer is the second leading cause of death behind cardiovascular disease in the
US.16 It is predicted that cancer will become the leading cause of death in the human
population.17 According to World Health Organization (WHO),17 there were 7.9 million
deaths in 2007 due to cancer, of which about 72% occurred in low- and middle-income
countries. Lung cancer caused the highest mortality (1.4 million deaths/year), others
were stomach cancer (866,000 deaths/year), liver cancer (653,000 deaths/year), colon
cancer (677,000 deaths/year) and breast cancer (548,000 deaths/year).
Figure 1.1 All available anticancer drugs, 1940s – 06/200618
Where:
•
•
•
•
•
•
•
“B”: Biological; usually a large (>45 residues) peptide or protein either isolated from an
organism/cell line or produced by biotechnological means in a surrogate host.
“N”: Natural product.
“ND”: Derived from a natural product and is usually a semisynthetic modification.
“S”: Totally synthetic drug, often found by random screening/modification of an
existing agent.
“S*”: Total synthesis, but the pharmacophore is/was from a natural product.
“V”: Vaccine.
“NM”: Natural Product Mimic
While there are many kinds of cancer treatments such as surgery, radiation,
biological therapy and chemotherapy,19 chemotherapy remains the most effective
treatment for patients with solid tumours.20 Among many anticancer drugs used in
chemotherapy, drugs derived from natural products including modified natural products
or synthetic products with a natural pharmacophore hold an important position by
occupying 65% of anticancer drugs approved during the 1940s-2006 (Figure 1.1).18
According to the statistics of Newman et al.,18,21,22 there was an increase in the number
of anticancer drugs derived from natural products, from 54 in 1994 to 100 drugs in June
3
2006. Three main reasons for the success of natural products have been given.23 Firstly,
there is firm evidence that substances from natural sources formed during the
organism’s evolution have the capacity to correct the aberrant regulation of cellular core
machinery. Secondly, research on biologically active natural products can contribute to
the understanding of cancer mechanisms and therefore facilitate the development of the
compounds into therapies. Thirdly, being trialled through many centuries, traditional
therapies using natural products have selected those with no or less side effects.
Beside the well-known marketed anticancer drugs9 derived from terrestrial
natural sources (taxol (14), vinblastine (15) and vicristine (16)), marine organisms are a
promising source for potential anticancer agents. According to Alejandro’s survey,24-26
97 new marine secondary metabolites were reported to possess antitumor and cytotoxic
properties during 2001 to 2002.25 The number of new cytotoxic natural products derived
from marine sources went up to 150 and 136 in the periods 2003-200426 and 2005200624, respectively. A small natural peptide, kahalalide F (21), isolated from the
marine mollusc Elysia rufescens, showed potent cytotoxic activities in vitro against a
panel of twenty cell lines from different tumour types (liver, ovary, breast, colon and
prostate).27 The IC50 values of 21 on these cancer cell lines ranged from 0.3 to 5.3 μM
after 1-hour exposure and from 0.2 to 4 μM using continuous exposure, especially at
IC50 < 0.3μM against breast and prostate cancer cell lines. This compound has recently
been developed for phase I clinical trials in the US and Europe.28 Along with kahalalide
F (17), other marine natural products such as halichondrin B (9), dolastatin 10 (18) and
okadaic acid (19) are also currently in phase I or II clinical trials.24 Interestingly, a
tetrahydroisoquinoline alkaloid, ecteinascidin-743 (20) (Trabectedin, Yondelis®),
identified from Ecteinascidia turbinata has recently been granted Orphan Drug
designation from the European Commission and the FDA for soft tissue sarcomas and
ovarian cancer after preclinical studies over several years.24 One of the promising
anticancer drugs identified in the period from 2001 to 2006 was laurerditerpenol (21), a
novel diterpene isolated from the red alga Laurencia intricate by Kaleem and his coworkers in 2004.29 In vitro testing showed that this compound inhibited breast tumor
cells HIF-1 with an IC50 of 0.4 μM blocking the induction of nuclear HIF-1α protein. Its
mechanisms of action on human cancer cells have been further investigated.
4
N
OH
O
O
O
HN
OH
O
NH
OH
N
OCH3
O
H
O
H3CO
H
O
O
O
N
OH
OH
O
R
O
O
O
O
OCH3
O
H
OH
O
O
O
HN
O
NH
N
NH2
HN
Vinblastine (15): R = CH3
Vincristine (16): R = CHO
Taxol (14)
NH
H
N
O
O
O
NH
NH
H
N
HN
O
O
HN
OH
O
O
H
O
O
O
HO
OH
O
O
O
O
OH
H
N
HN
O
O
H
H
O
OH
Okadaic acid (19)
Kahalalide F (17)
HO
OMe
NH
MeO
O
AcO
O
H
N
Me
H
N
N
Me
HO
O
S
H
H
OH
NMe
N
N
O
OCH3 O
O
OCH3 O
N
S
H
O
Dolastatin 10 (18)
O
OH
Ecteinascidin 743 (20)
Laurenditerpenol (21)
Besides providing drugs for diseases, natural product research also provides
pharmacophores for developing better bioactive derivatives as well as designing new
analogues with greater synthetic accessibility.30 So far many analogues of taxol (14)
have been synthesized, of which 6 analogues are in phase II clinical trial, 4 analogues
are currently in phase I clinical trial and further 23 analogues are in preclinical
development.31 Based on the core structure of the marine secondary metabolite,
sarcodictyins A (22), which was found to have taxol-like activities in tubulin
polymerization and microtubule stabilization, Nicolaou et al.32 constructed a large
combinatorial library of sarcodictyin analogues and explored their structure activity
relationship (SAR) against tumour cells (Figure 1.2). The results revealed that the
activity of sarcodictyins A analogues with an ester group at C3 was higher than an
amide and the reduction of the ester to give an alcohol led to the loss of its activity.
Furthermore, if a hydroxy group at C4 were replaced by a ketone group, the biological
activity would appear more tolerable. Another position found to play a crucial
cytotoxicity role was an ester side chain at C8. It was found that either replacing a
natural urocanic acid side chain with an acetate group or a phenyl carbamate group or
changing a natural imidazole substituent for pyridine, thiazole or oxazole would result
in the complete loss of activity.
5
Side chain is important
for activity
O
O
N
8
H
N
Both nitrogens are
necessary for activity
O
1
4
H
3
OH
Ketal substitutions
are well tolerated
O
O
Esters have higher activity than amides,
reduction of the ester to the alcohol
leads to loss of activity
Sarcodictyin A (22)
Figure 1.2 Structure activity relationships for the sarcodictyins32
Bryostatin 1 (23), isolated from the marine bryozoan Bugula neritina with the
yield of 1.4x10-4 %, was found to inhibit a variety of cancer cells.33 A SAR study
indicated that the rings A and B of 23 might be the active pharmacophore and
substituents on these rings could be removed without losing activity. This research also
determined that the key pharmacophoric elements of bryostatin in binding protein
kinase C were at C1, C19 and C26.34 A synthetic compound, bryostatin analogue A
(24), which was synthesized after 19 steps in 2% of yield, possesses a simpler structure
than the parent bryostatin.35 This analogue showed a higher affinity binding to the
protein kinase C with a Ki of 0.25 nM compared with 1.35 nM for bryostatin and greater
potency in vitro against a subset of the NCI’s panel of cell lines.36 MeO
B
O
H
HO
O
O
OAc
H
1
O
O
19
H
O
O
A
H
OH
H
H
H
H
O
O
OH H
O
OH O
O
C
O
26
OH
OH O
O
OH
OMe
OMe
O
O
Bryostatin 1 (23)
Bryostatin analogue A (24)
6
1.3 The application of “drug-like” properties into natural product discovery
1.3.1 The decline of interest on natural products from big pharmaceutical companies
Despite having made great contributions to drug discovery, in the past few years
natural product research has lost financial investments from most big pharmaceutical
companies. These companies have either scaled down or terminated their natural
product operations. Reasons have been given to explain the decline of interest in natural
products. Natural products could not compete with other drug discovery methods in
delivering large-scale compound supply within a short time frame and in delivering
compounds that passed the criteria of oral bioavailability, a significant requirement in a
hit-to-lead program. To continue taking part in the fight against diseases, natural
product research needs to address these two main problems. Recent improvements in
instrumentation, robotics, screening technology have shortened the hit identification
period for natural products. Large scale supply of natural products still requires the
development of synthetic routes in many cases. Oral bioavailability has been one of the
big hurdles to natural products. Natural product researchers need to prove to
pharmaceutical companies that their novel natural product compounds can be absorbed
into the human bloodstream and are worthwhile to be further developed into a drug. The
prediction of the amount of drug actually absorbed from a given dose into the
bloodstream can be addressed using animal models, or in silico models; logD and pKa
by chromatography; permeability profiles by artificial membrane; intestinal drug
transport by Caco-2 cell membrane.37 However, these techniques require time and good
lab facility and are more suitable when the drug is in the last stage of discovery.
1.3.2 Physico-chemical properties
Upon studying the solubility and permeability of 2245 drug candidates from the
databases of the World Drug Index, the United States Adopted Name and International
Non-proprietary Name reaching the phase II clinical process, Lipinski’s group proposed
the “rule of five” (Ro5) or Lipinski’s rule as key predictors for oral bioavailability of a
compound. According to this rule, an oral-acting drug-like molecule should satisfy four
parameters38
− The calculated logarithm of the n-octanol/water partition coefficient (ClogP) of
less than 5
− Less than 5 hydrogen bond donors (sum of OH and NH).
− Less than 10 hydrogen bond acceptors (expressed as the sum of O and N).
− Molecular weight of less than 500 Da.
7
Lipinski’s rule also indicated that oral bioavailability does not apply to natural
products or any molecules recognized by an active transport systems. If a compound
fails two or more parameters, there is a high probability that oral activity problems will
occur.
The first parameter, logP, has been considered as the lord of the “rule of five”. It
is defined as the ratio of un-ionized drug distributed between octanol and water phases
at equilibrium.39 This factor plays an important role in solubility, permeability, plasma
protein binding, metabolic turnover and toxicity of drugs. A high logP prevents the
active compounds from reaching the site of action due to poor distribution. In contrast, a
low logP can cause the compounds to be absorbed by active transports before they reach
target cells.40,41
Hydrogen bonds increase solubility in water and must be broken in order for the
compound to permeate through the lipid bilayer membrane. Thus, an increasing number
of hydrogen bonds reduce partitioning from the aqueous phase into the lipid bilayer
membrane. Hydrogen bond donors and acceptors also participate in providing hydrogen
bonding specifically between ligand and receptor.42
Molecular weight also has a significant influence on the bioavailability of a
compound. When molecular size increases, the solubility of the compound will reduce
since a larger cavity must be formed in water to solubilize this compound. Moreover,
increasing the molecular weight leads to a decrease in the compound concentration at
the surface of the intestinal epithelium, thus affecting the absorption. Increasing the
molecular size also hampers passive diffusion through the tightly packed aliphatic side
chains of the bilayer membrane.43
C: 7%
D: 1%
B: 11%
A: 81%
Figure 1.3 Lipinski violations for 546 marketed oral drugs in the FDA Orange Book
2007 44
Where
A: No violation of Lipinski rules
C: Violation of two Lipinski rules
B: Violation of one Lipinski rule
D: Violation of three Lipinski rules
Recently, Fotouhi and his colleagues in the Roche Research Center44 have
analysed the physico-chemical properties of the currently marketed oral drugs in the
8
FDA Orange Book 2007 (Figure 1.3). The analysis showed over 90% of oral drugs
passed the Ro5.
Veber complements Lipinski’s rule with two other parameters, polar surface
area (PSA) and number of rotatable bonds (NROT) which can be considered as two
more requirements of drug-like molecules from the physico-chemical property
viewpoint. By observing the solubility and permeability of over 1100 drug candidates in
rats, Veber and co-workers45 observed that a molecule with PSA less than 140Å2 and
NROT less than 10 have better oral activity compared to the others. Polar surface area
(PSA) is a parameters affecting absorption.46 Owing to the polar surface, the molecule
can form hydrogen bonds or Van der Waal bonds with other compounds. A rotatable
bond, indicating a chemical’s flexibility, is defined as any single non-ring bond, bound
to non-terminal heavy atoms. Permeability of a molecule will be significantly reduced if
the NROT is over 10. Lipinski’s and Veber’s rules have guided drug discovery
researchers in their hit-to-lead decision making process.
1.3.3 Do natural products have drug-like properties?
Although natural products were cited as an exception to the Ro5, database
surveys of the drug-like nature of natural products as well as their relationships with
marketed drugs and synthetic molecules have been investigated using statistical
analyses, computational molecular modelling. The results of analysing 126,140 unique
entries in the Dictionary of Natural Products (version 2005) showed that 60% of natural
products had no Ro5 violations.47 In detail, natural products had median calculated
logPs between 2 to 3 (reaching a peak at 2.5); median HBAs were in the range of 3-4;
HBDs were maximum at 0-1 and MWs peaked at 300 Da. This report also agreed with
previous studies by Miklos Feher40 and Thomas Henkel48 on three natural product
databases Bioactive Natural Product Database – Szenzor Management (version 1996),
Available Chemicals Directory (ADC) and Dictionary of Natural Products (version
1996). In 2008, Ertl et al.42 used another database set containing 10,968 drug molecules,
670,536 combinatorial compounds and 3,287 natural products to investigate
physicochemical properties. Ertl et al concluded that both drugs and natural products
had similar lipophilicity with a maximum logP at 3.0, whereas the logP of combinatorial
molecules had a maximum at 4.0. A Gaussian distribution of logP indicated that
compounds from natural sources were less hydrophobic than the synthetic molecules.
Similar result was obtained when a research group from the Novartis Institutes analysed
115,590 deglycosylation natural products and 290,000 structures from their in-house
9
collection of commercially available synthetic compounds.49 All of the above research
shows that at least 60% of isolated natural products can be orally bioavailable, and
natural products are closer to drugs than synthetic molecules. However, acquiring
natural products is costly and a method to build a natural product library enriched with
drug-like property is vital.
1.3.4 A new paradigm in building a natural products library
Source materials
Bioassay screening
Extracts
Active fractions
Isolated compounds
Bioassay screening
Bioassay screening
Active extracts
Chromatographic
fractions
Active compound(s)
Scheme 1.1 Traditional way in natural product drug discovery5
Traditional natural product drug discovery using crude extracts and bioassay
guided isolation (Scheme 1.1) is one of the reasons hampering the interest in natural
products from pharmaceutical companies. Although this conventional routine does not
require a big effort in preparing samples and supplying a high degree of chemical
diversity, crude extracts have some drawbacks. Firstly, non-specific interference from
fatty acids, polyphenols and other salts in crude extracts can cause false-positive or
false-negative results in bioassay screenings. Secondly, minor components can be
missed out in a complex mixture because their concentrations might be below the
detection threshold of a biological screening or be masked by other compounds in a
complex mixture. Thirdly, chemically unattractive compounds are often isolated.
Fourthly, it takes a great deal of labour and time to identify and isolate bioactive
components.50-52
Instead of crude extracts, pure compound libraries of natural products have been
constructed.47,53 With structures in hand, compounds violating the physico-chemical
properties can be eliminated before screening. However, current pure compound
libraries can not represent all desired chemical diversity of natural products and minor
metabolites may be overlooked. Time and resource for the generation of these libraries
are still an issue.50,54
10
Prefractionated natural product libraries have been considered as a new and
effective approach. Several prefractionation methods have been described using single
or multi-step solid-phase extraction (SPE),55-57 flash column chromatography or
preparative high performance liquid chromatography (prep-HPLC),58,59counter-current
chromatography (CCC),60-62 centrifugal partition chromatography (CPC)63,64 and ultra
performance liquid chromatography (UPLC).52 By using this method, extremely
hydrophilic and hydrophobic components (sugars, salts, nucleotides, fatty acid…) are
eliminated. With 4-5 compounds per fraction, higher dose of compounds can be tested.
Minor components are also detected and the desired chemical diversity is still secured.
On analysing the HTS results obtained from eleven screening campaigns, Merlion
Pharmaceuticals reported the activity of prefractionated fractions was increased by
twelve times over crude extracts. They also found 80% of the primary screening hit
fractions were active while their associated crude extract showed no activity even at
four-time screening dose.51 The similar outcome of screening prefractionated library
(79.9%) was also achieved by Wyeth Pharmaceuticals.50 These results prove that
prefractionated libraries have more advantages than crude extract and pure compound
libraries.
Several pharmaceutical companies and research groups have reported their
methods for building prefractionated libraries in natural product drug discovery.5052,56,58,59,65-67
Generally, these methods used reverse-phase HPLC to separate fractions
cooperating with other detectors such as photodiode array (PDA), mass spectroscopy
(MS) or evaporative light scattering (ELS) to identify, dereplicate and isolate
biologically active constituents. Wyeth Pharmaceuticals generated a 200,000 fraction
library from microorganism sources (10 fractions per extract) (Figure 1.4a).50 Also
working in microbial metabolites, Merlion’s process was simpler with 4 fractions per
extract. In this way, they established a library with 120,000 fractions for HTS (Figure
1.4b).51 In research on marine organisms, Ireland’s group has used SPE in a primary
separation for eliminating salts and other highly hydrophilic components, followed by
reverse-phase HPLC to produce 80 fractions per sample (Figure 1.4c).67 Compared with
microorganism or marine extracts, plant extracts contain tannins and other polyphenols
affecting the bioassay screening results, prefractionation for these terrestrial sources is
thus more sophisticated. Sequoia Science presented a multi-step process for generating
their 36,000 fraction library from plant extracts in which 200 fractions were created
from one sample (Figure 1.4d).58 Also from plants, Yan’s group described a less
complicated method to generate their library. A plant extract was treated through SPE
11
prior to loading on a HPLC column to separate 24 fractions (Figure 1.4e). This group
demonstrated they could provide 62,000 fractions from 2600 unique natural product
samples per year.52
Microorganism
Microorganism
Marine samples
At least 2 fermentation
conditions combined
MeOH extracts
MeOH extracts
MeOH extracts
Prep-HPLC
(4 fractions/extract)
SPE
(5 fractions/sample)
Prep-HPLC
(10 fractions/extract)
HTS
(120,000 fractions)
LC/MS prefractionation
(20x4=80 fractions/sample)
HTS
(200,000 fractions)
a) Wyeth Pharmaceuticals
HTS
(15,360 fractions)
b) Merlion Pharmaceuticals
Plant samples
Plant samples
Organic extracts
(EtOAc, EtOH, MeOH)
c) Ireland’s group
EtOH extracts
Aqueous extracts
SPE
Flash chromatography
(Si)
Flash chromatography
(C18)
4 fractions/extract
1 fraction/extract
LC/MS prefractionation
(24 fractions/sample)
HTS
(62,000 fractions)
HPLC
(40x5=200 fractions/sample)
HTS
(36,000 fractions)
d) Sequoia Sciences
e) Yan’s group
Figure 1.4 Current processes for generating prefractionated libraries.
The overview shows that methods vary from collection of a small number of
fractions (4-10 fractions) per sample to a larger number of fractions (80-200 fractions)
per sample. Although these cleaner fractions are screened, large screening points per
sample will limit the number of samples being investigated. A better model for the
fraction generation is required.
12
1.4 Objective – Research plan
The objective of this study is to identify drug-like natural products from both
plant and marine sources which exhibit cytotoxicity on cancer cell lines. The aims of the
project are (Scheme 1.2):
Aim 1: Identification of prefractionated fractions showing activity in a cancer
cell-based screening. A small lead-like enhanced fraction library will be prepared. 330
fractions are chosen to fit in one 384-well plate. These fractions will then be screened
against a panel of four cancer cells (lung cancer – A549, cervical cancer – HeLa,
prostate cancer – LNCaP and PC3) and non-cancer cell line (human embryonic kidney
293 – HEK). Details are discussed in chapter 2.
Aim 2: Isolation and structure elucidation of active components and their
analogues. Large scale isolation and structure elucidation are performed for biota
samples whose fractions are in the hit list. Active components and their analogues will
be
identified
for
structure-activity
relationship.
Purification
and
structural
characterization are reported in chapters 3-7.
Aim 3: Evaluations of drug-like properties and biological activity. Physicochemical properties (MW, clogP, HBA, HBD, PSA and NROT) of all isolated
compounds are calculated by using JChem 2.5.2 software. The potential and selective
anticancer activity of isolated compounds is also evaluated by comparing inhibitions of
pure compounds against cancer cell lines (A549, HeLa, LNCaP and PC3) and noncancer cell lines (HEK, NFF). These evaluations are reported in chapters 3-7.
Samples
Generate prefractionated
fractions
Cancer cell-based
screening
Large scale isolation
Active fractions
Structure elucidation
Biological evaluation
Drug-like evaluation
Hit compounds
Scheme 1.2 Research program
13
Aim 1
Aim 2
Aim 3
1.5 References
(1)
Fabricant, D. S.; Farnsworth, N. R. Environ. Health Perspect. 2001, 109, 69-75.
(2)
Escohotado, A. A brief history of drugs: from the stone age to the stoned age;
Park street Press: Rochester, 1996.
(3)
Cragg, G. M.; Newman, D. J. Ann. N. Y. Acad. Sci. 2001, 953a, 3-25.
(4)
Patwardhan, B.; Warude, D.; Pushpangadan, P.; Bhatt, N. eCAM 2005, 2, 465473.
(5)
Colegate, S. M.; Molyneux, R. J. Bioactive natural products - Detection,
isolation and structural determination; 2 ed.; CRC Press, 2007.
(6)
Vogel, E.; Pelletier, S. J. Pharm 1815, 2, 50.
(7)
Lampe, V. Ber Dtsch Chem Ges 1910, 43, 2163-2170.
(8)
Maruyama, M.; Terahara, A.; Itagaki, Y.; Nakanishi, K. Tetrahedron Lett. 1967,
8, 299-302.
(9)
Neidle, S.; Thurston, D. E. Nat. Rev. Cancer. 2005, 5, 285-296.
(10)
Bhakuni, D. S.; Rawat, D. S. Bioactive marine natural products; Anamaya: New
Delhi, 2005.
(11)
Baker, D. D.; Chu, M.; Oza, U.; Rajgarhia, V. Nat. Prod. Rep. 2007, 24, 12251244.
(12)
Fenical, W.; Jensen, P. R. Nat. Chem. Biol. 2006, 12, 666-673.
(13)
Simmons, T. L. Mol. Cancer Ther. 2005, 4, 333-342.
(14)
Cragg, G. M.; Grothaus, P. G.; Newman, D. J. Chem. Rev. 2009, 109, 30123043.
(15)
Buss, A. D.; Butler, M. S. Natural Product Chemistry for Drug Discovery;
Royal Society of Chemistry: Cambridge, 2010.
(16)
Gibbs, J. B. Science 2000, 287, 1969-1973.
(17)
http://www.who.int.
(18)
Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2007, 70, 461-477.
(19)
http://www.cancer.gov/cancertopics/treatment/types-of-treatment.
(20)
Savage, P.; Stebbing, J.; Bower, M.; Crook, T. Nature clinical practice:
Oncology 2009, 6, 43-52.
(21)
Cragg, G. M.; Newman, D. J.; Snader, K. M. J. Nat. Prod. 1997, 60, 52-60.
(22)
Newman, D. J.; Cragg, G. M.; Snader, K. M. J. Nat. Prod. 2003, 66, 1022-1037.
(23)
Boik, J. Natural Compounds in Cancer Therapy; 1st ed.; Oregon Medical Press,
LLC: Minnesota, 2001.
(24)
Mayer, A. M. S.; Gustafson, K. R. Eur. J. Cancer 2008, 44, 2357-2387.
14
(25)
Mayer, A. M. S.; Gustafson, K. R. Eur. J. Cancer 2004, 40, 2676-2704.
(26)
Mayer, A. M. S.; Gustafson, K. R. Eur. J. Cancer 2006, 2241-2270.
(27)
Sewell, J. M. Eur. J. Cancer 2005, 41, 1637-1644.
(28)
Nuijen, B.; Nuijen, B.; Bouma, M.; Talsma, H.; Manada, C.; Jimeno, J. M.;
Lazaro, L. L.; Bult, A.; Beijnen, J. H. Drug Dev. Ind. Pharm. 2001, 27, 767-780.
(29)
Mohammed, K. A. J. Nat. Prod. 2004, 66, 2002-2007.
(30)
Paterson, I.; A.Anderson, E. Science 2005, 310, 451-453.
(31)
Cragg, G. M.; Kingston, D. G. I.; Newman, D. J. Anticancer agents from natural
products; CRC press, 2005.
(32)
Nicolaou, K. C.; Winssinger, N.; Vourloumis, D.; Ohshima, T.; Kim, S.;
Pfefferkorn, J.; Xu, J. Y.; Li, T. J. Am. Chem. Soc. 1998, 120, 10814-10826.
(33)
Schaufelberger, D. E.; Koleck, M. P.; Beutler, J. A.; Vatakis, A. M.; Alvarado,
A. B.; Andrews, P.; Marzo, L. V.; Muschik, G. M.; Roach, J.; Ross, J. T.;
Lebherz, W. B.; Reeves, M. P.; Eberwein, R. M.; Rodgers, L. L.; Testerman, R.
P.; Snader, K. M.; Forenza, S. J. Nat. Prod. 1991, 54, 1265–1270.
(34)
Wender, P. A.; Debrabander, J.; Harran, P. G.; Jimenez, J.-M.; Koehler, M. F.
T.; Lippa, B.; Park, C. M.; Siedenbiedel, C.; Pettit, G. R. Proc. Natl. Acad. Sci.
USA 1998, 95, 6624–6629.
(35)
Wender, P. A.; Baryza, J. L.; Bennett, C. E.; Bi, F. C.; Brenner, S. E.; Clarke, M.
O.; Horan, J. C.; Kan, C.; Lacote, E.; Lippa, B.; Nell, P. G.; Turner, T. M. J. Am.
Chem. Soc. 2002, 124, 13648-13649.
(36)
Wender, P. A.; Baryza, J. L.; Brenner, S. E.; Clarke, M. O.; L.Craske, M.;
Horan, J. C.; Meyer, T. Curr. Drug Discov. Tech. 2004, 1, 1-11.
(37)
Waterbeemd, H. V. D.; Lennernäs, H.; Artursson, P. Methods and Principles in
Medicinal Chemistry 2004.
(38)
Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery
Rev. 1997, 23.
(39)
Leeson, P. D.; Springthorpe, B. Nat. Rev. Drug Discovery 2007, 6, 881-890.
(40)
Feher, M.; Schmidt, J. M. J. Chem. Inf. Comput. Sci. 2003, 43, 218-227.
(41)
Ganesan, A. Curr. Opin. Chem. Biol. 2008, 12, 306-317.
(42)
Ertl, P.; Roggo, S.; Schuffenhauer, A. J. Chem. Inf. Model. 2008, 48, 68–74.
(43)
Kerns, E. H.; Di, L. Drug-like Properties: Concepts, Structure Design and
Methods; Elsevier Inc, 2008.
(44)
Fotouhi, N.; Gillespie, P.; Goodnow, R. Expert Opin. Drug Discov. 2008, 3,
733-744.
15
(45)
Veber, D. F.; Johnson, S. R.; Cheng, H.-Y.; Smith, B. R.; Ward, K. W.; Kopple,
K. D. J. Med. Chem. 2002, 45, 2615-2623.
(46)
Vogel, H. G.; Hock, F. J.; Maas, J.; Mayer, D.; Hock, F. G. Drug Discovery and
Evaluation: Safety and Pharmacokinetic Assays; Springer, 2006.
(47)
Quinn, R. J.; Carroll, A. R.; Pham, N. B.; Baron, P.; Palframan, M. E.;
Suraweera, L.; Pierens, G. K.; Muresan, S. J. Nat. Prod. 2008, 71, 464-468.
(48)
Henkel, T.; Brunne, R. M.; Muller, H.; Reichel, F. Angew. Chem. Int. Ed. 1999,
38, 643-647.
(49)
Ertl, P.; Roggo, S.; Schuffenhauer, A. J. Chem. Infor. Model 2008, 48, 68-74.
(50)
Wagenaar, M. M. Molecules 2008, 13, 1406-1426.
(51)
Appleton, D. R.; Buss, A. D.; Butler, M. S. Chimia 2007, 61, 327-331.
(52)
Tu, Y.; Jeffries, C.; Ruan, H.; Nelson, C.; Smithson, D.; Shelat, A. A.; Brown,
K. M.; Li, X. C.; Hester, J. P.; Smillie, T.; Khan, I. A.; Walker, L.; Guy, K.;
Yan, B. J. Nat. Prod. 2010, 73, 751-754.
(53)
Bindseil, K. U.; Jakupovic, J.; Wolf, D.; Lavayre, J.; Leboul, J.; Pyl, D. V. D.
Drug Discov. Today 2001, 6, 840-847.
(54)
Lumley, J. A. QSAR Comb. Sci. 2005, 24, 1066-1075.
(55)
Cardellina, J. H.; Munro, M. H. G.; Fuller, R. W.; Manfredi, K. P.; McKee, T.
C.; Tischler, M.; Bokesch, H. R.; Gustafson, K. R.; Beutler, J. A.; Boyd, M. R.
J. Nat. Prod. 1993, 56, 1123.
(56)
Schmid, I.; Sattler, I.; Grabley, S.; Thiericke, R. J. Biolmol. Screening 1999, 4,
15-25.
(57)
Thiericke, R. Autom. Methods Manag. Chem. 2000, 22, 149-157.
(58)
Eldridge, G. R.; Vervoort, H. C.; Lee, C. M.; Cremin, P. A.; Williams, C. T.;
Hart, S. M.; Jonhson, M. O.; Zeng, L. Anal. Chem. 2002, 74, 3963-3971.
(59)
Abel, U.; Koch, C.; Speitling, M.; Hansske, F. G. Curr. Opin. Chem. Biol. 2002,
6, 453-458.
(60)
Wu, S.; Yang, L.; Gao, Y.; Liu, X.; Liu, F. J. Chromatogr. A 2008, 1180, 99107.
(61)
Alvi, K. A.; Peterson, J.; Hofmann, B. J. Ind. Microbiol. 1995, 15, 80.
(62)
Lu, Y.; Berthod, A.; Hu, R.; Maa, W.; Pan, Y. Anal. Chem. 2009, 81, 40484059.
(63)
Armbruster, J. A.; Borris, R. P.; Jiminez, Q.; Zamora, N.; Castillo, G. T.; Harris,
G. H. J. Liq. Chromatogr. Relat. Technol. 2001, 24, 1827.
16
(64)
Ingkaninan, K.; Hazekamp, A.; Hoek, A. C.; Balconi, S.; Verpoorte, R. J. Liq.
Chromatogr. Relat. Technol. 2000, 23, 2195.
(65)
Jia,
Q.
Generating and Screening a Natural Product Library for
Cyclooxygenase and Lipoxygenase Dual Inhibitors. Studies in Natural Products
Chemistry, 2003; Vol. 29.
(66)
Bitzer, J.; Kopcke, B.; Stadler, M.; Hellwig, V.; Ju, Y. M.; Seip, S.; Henkel, T.
Chimia 2007, 61, 332-338.
(67)
Bugni, T. S.; Harper, M. K.; McCulloch, M. W. B.; Reppart, J.; Ireland, C. M.
Molecules 2008, 13, 1372-1383.
17
18
Chapter 2
Generation of prefractionated fractions for
biological screening
2.1 Library analysis
The Eskitis Institute designed successfully a prefractionated method to generate
eleven lead-like enhanced fractions for each sample. A library of 202,983 fractions was
produced from over 18,000 marine and terrestrial samples. Examination of the HPLC
chromatograms has shown that there are some fractions in the lead-like area displaying
no UV absorption.
100%
80%
60%
40%
20%
0%
A ntic anc er A ntibac terial
drugs
drugs
A ntiv iral
drugs
A ntiparas itic
drugs
A ntif ungal
drugs
Natural products and modified natural products
Synthetic with natural product pharmacophore
Figure 2.1 UV absorption analysis of some kinds of drugs (λ = 210-400 nm)
An analysis on UV-absorption of drugs for cancer and infectious diseases
(excluding biological drugs and vaccines) was carried out (Figure 2.1). Using drug lists
reported by Newman and Cragg (anticancer – 164 entities from 1940s to 12/2007;1
antibacteria – 98 entities, antivirus – 28 entities, antiparasite – 13 entities and antifungi
– 41 entities from 01/1981 to 06/20062), the UV absorptions of these drugs were then
collected from databases including Dictionary of Natural Products, SciFinder Scholar,
CAPlus and National Center for Biotechnology Information (NCBI)). Results showed
that over 90% of drugs related to natural products have UV activity. In particular, for
anticancer drugs, 75 out of 79 natural products and modified natural products (NND,
94.9%) and all 43 synthetic derived from natural products (SNM, 100%) have UV
absorptions ranging from 210-400 nm, which can be detected by a photodiode array
detector in HPLC. These results encouraged us to choose UV absorption as a criterion
19
for selecting fractions. Fractions showing UV activity in prefractionated HPLC
chromatograms were selected for screening (Scheme 2.1).
Samples based on MS data
1) Strong MS signals from crude extracts
2) Rare taxonomies
3) Available in-house HPLC chromatograms
4) UV active profiles
Samples based on taxonomy
1) Known taxonomies
2) Available in-house HPLC chromatograms
3) UV active profiles
LC/MS of crude extracts
105 chosen samples
1) Strong UV absorption in a lead-like area
2) Strong MS signals in a lead-like area
C18-HPLC
1155 lead-like enhanced fractions
330 selected fractions
1) Strong UV absorption
Cytotoxic screening
91 samples (311 inactive fractions)
14 samples (19 active fractions)
LC/MS of active fractions
7 investigated samples (11 active fractions)
1) Strong activity
2) Strong MS signals
3) Rare taxonomies
4) Over 20g of available materials
Scheme 2.1 Diagram of the procedure to select fractions for screening and chemical
investigation
2.1.1 Strategies for choosing biota samples
In order to generate a subset of 330 fractions fitting in one 384-well microtitre
plate (other remaining wells were used for controls), two routes were utilised to choose
biota samples.
Route 1 - Taxonomy: The aim of using taxonomy is to obtain new derivatives of
known cytotoxic natural products. A literature search of cytotoxic natural products in
Dictionary of Natural Products (version 2009) resulted in 1770 compounds with
molecular weights less than 500 Da. Taxonomic information on these compounds was
20
extracted and compared with available taxonomies in our in-house database. Only
samples with available prefractionated HPLC profiles were selected for further
investigation.
Route 2 - Mass spectrometry data: This second method aims to choose rare
taxonomic biota containing novel compounds which can be detected by MS. An inhouse MS database of 6,000 methanol crude extracts was analysed to select samples
having strong MS signals. Only samples which were originated from rare taxonomies
and had available prefractionated HPLC profiles were selected for further investigation.
The methanol extracts of all chosen samples from the two routes were then
subjected to LC/MS analysis to get more information for the selection process. Twenty
microliters of the crude extract were loaded to the HPLC column (Phenomenex Luna,
C18, 3μm, 4.6 mm x 50 mm) controlled by MassLynx 4.1 software. MS data were
collected in both positive and negative modes.
Table 2.1 Gradient timetable for LC/MS analysis of crude extracts
1
2
3
4
5
Where
Time
0.0
0.5
10.5
11.0
12.0
%A
100
100
0
100
100
%B
0
0
100
0
0
A: 0.1% formic acid in water
Flow
1.0
1.0
1.0
1.0
1.0
Curve
1
6
6
6
6
B: 0.1% formic acid in methanol
The aim of this LC/MS analysis was to cut-off the number of samples. Only
samples showed strong MS signals in an area where predicted logP values in a C18HPLC column are less than 5 (eluting from 1.5 min (10% methanol) to 7.5 min (70%
methanol))3 were chosen as candidates for the next selection step.
2.1.2 Strategies for selecting fractions
HPLC chromatograms of the chosen samples were extracted from the in-house
database. Fractions were then selected on the criteria of UV activity. The chosen
number of fractions per sample varied depending on the distribution of UV active peaks
in the lead-like area whose gradient program was previously set up in the Eskitis
Institute . Figure 2.2 shows two examples of selecting fractions.
21
Figure 2.2 Examples of selecting fractions
2.2 Re-generation of selected drug-like fraction library
Plant and marine samples (200mg) were extracted sequentially with hexane (2
ml x 2), dichloromethane (2 ml x 3) and methanol (2 ml x 3). Solid phase extraction
(Oasis® HLB cartridge 6cc/200mg, 30µm) was performed with the combined DCM and
MeOH extracts to remove highly hydrophobic compounds before the solvents were
evaporated using a centrifugal evaporator. Filtering through 1g of polyamide gel (PAG)
prior to evaporating solvents was only applied for plant samples in order to remove
polyphenols which may cause false positive results in bioassay screening. The dry crude
extracts were then dissolved in 800 μL DMSO to make a stock concentration of 250
μge/μL (stock concentration = staring material (μg)/volume of DMSO (μL)). One
hundred microliters of this aliquot were subjected to prefractionation while the
remaining material was stored. Extractables from 25 mg of the dry weight starting
material (250 μge/μL x 100 μL) were loaded onto the HPLC column. The
prefractionation was performed on a HPLC column (Phenomenex Onyx Monolithic
C18, 4.6 mm x 100 mm) with a gradient as shown in Table 2.2 above.
Eleven fractions from 2 to 7 minutes (Figure 2.3) were collected into deep
plastic 96-well plates (2mL x 96). The most polar components eluting between 0 and 2
minutes and also extremely non-polar components eluting after 7 minutes were
discarded. According to previous study in the Eskitis Institute, lead-like enhanced
fractions were only collected from 2 to 7 minutes. These fractions were then dried using
a centrifugal evaporator. Selected fractions were dissolved in 250 μL DMSO to make a
concentration of 100 μge/μL. One hundred microliters of the selected peaks were
transferred into 384-well microtitre plates for screening, while the remaining was dried
down and stored for chemical analysis by LC/MS.
22
1 2 3 4
5
6
8
7
9
10 11
11 fractions
Figure 2.3 HPLC profile of a biota sample in a prefractionation process
Table 2.2 Gradient timetable for prefractionation
No.
1
2
3
4
5
6
7
8
9
Where
Time
0.01
3.00
3.01
6.50
7.00
7.01
8.00
9.00
11.00
Flow
4.00
4.00
3.00
3.00
3.00
4.00
4.00
4.00
4.00
%B
10.0
50.0
50.0
100.0
100.0
100.0
100.0
10.0
10.0
%C
90.0
50.0
50.0
0.0
0.0
0.0
0.0
90.0
90.0
Curve
6
6
6
5
6
6
6
6
6
B: 0.1% Trifluoroacetic acid in methanol
C: 0.1% Trifluoroacetic acid in water
Since most active natural products have been isolated as minor compounds
which are often less than 0.1% by dry weight of the sample,4,5 a concern arose that these
minor compounds might be missed out by our selection using UV absorption. An
investigation for the detection threshold was performed. An analytical testing was
performed with compounds having simple chromophores (1 aromatic ring or 2 double
bonds) at different loadings onto the HPLC column (100 μg, 10 μg, 1 μg and 0.1 μg)
(HPLC program shown in Table 2.2). The result revealed that 2-cyclohexenone (2
double bonds in the structure) had a detection limit of 1 μg/injection. The detection limit
of compounds with aromatic ring systems, 2-bromophenol, methyl-4-hydroxy-benzoate,
2-(5-bromo-2-methyl-1H-indol-3-yl) acetic acid and benzophenone, was ten-fold lower
(0.1 μg/injection) (Figure 2.4). With a total loading of 25 mg dry weight in our
prefractionation process, our approach could detect compounds in the range of 1 μg to
0.1 μg. This meant minor components up to 0.001% of dry weight were included for
this study.
23
Load: 100 μg
O
Load: 10 μg
Load: 1 μg
O
Load: 100 μg
OH
OMe
Br
Load: 100 μg
HO
Load: 10 μg
Load: 10 μg
Load: 1 μg
Load: 1 μg
Load: 0.1 μg
Load: 0.1 μg
O
COOH
Br
Load: 100 μg
Load: 100 μg
Load: 10 μg
Load: 10 μg
Load: 1 μg
Load: 1 μg
Load: 0.1 μg
Load: 0.1 μg
N
H
Figure 2.4 Analytical testing with simple chromophores
2.3 Cytotoxicity screening
There are two types of bioassays for cytotoxicity screening, enzyme assays and
cell-based assays. Compared with enzyme assays, cell-based assays are more amenable
to natural product screening.4 Besides the simple method using spectrophotometric or
turbidimetric to detect the activity, the occurrence of molecular interactions which can
be assessed within the context of a living cellular environment is considered as another
important advantage of this type of assays.5 Among some dyes using in cell-based
assays, a cell health indicator, Alamar Blue, is widely used for the HTS because of its
stability, non-toxicity and non-costliness.6 Chemically, resazurin (25) is the active
component of Alamar Blue reagent, which is blue in colour and virtually nonfluorescent. When the cells are alive, they maintain a reducing environment in the
cytosol of the cell. Upon entering the cells, resazurin is reduced to resorufin (26), a
reddish compound with highly fluorescent. Herein Alamar Blue was employed for the
in vitro cytotoxicity screening.
24
O
O
O
O
O
O
N
N
O
resazurin (25)
resorufin (26)
Figure 2.5 Mechanism of Alamar Blue indicator
Human embryonic kidney (HEK 293) and human lung adenocarcinoma cells
(A549) were grown in media – DMEM supplemented with 10% Foetal Bovine Serum
(FBS), 100 Unit/ml Penicillin and 100 μg/mL Streptomycin. Cervical adenocarcinoma
cells (Hela), prostate adenocarcinoma cells (LNCaP and PC3) were grown in media –
RPMI supplemented with 10% FBS, 100 Unit/ml Penicillin and 100 μg/mL
Streptomycin. Cells were grown under 5% CO2 in a humidified environment at 37oC.
The cytotoxicity of selected fractions was measured after 72-hour incubation
using an Alamar Blue proliferation assay. Forty-five microlitres of media containing
1000 cells for 72-hour timepoint were added to a 384-well microtitre plate (Falcon
black clear 384 TC microtitre plates). Plates were incubated overnight at 37°C, 5% CO2
and 80% humidity to allow cells to adhere. Stock concentrations of samples were
prepared at 100 μge/μL in DMSO. Test agents were then diluted 1 in 10 in media. Five
microliters of diluted agent were added to the cells to give a total volume (50 μL). Final
concentration tested was 1 μge/μL (final DMSO concentration of 1%). Each fraction
was tested in triplicate. After 72-hour incubation at 37°C 5% CO2 and 80% humidity,
cell proliferation was measured with the addition of 10 μL of a 60% Alamar Blue
solution in media to each well of the microtitre plate to give a final concentration of
10% Alamar blue. The plates were incubated at 37°C 5% CO2 and 80% humidity
overnight. The fluorescence of each well was measured at excitation 535 nm and
emission 590 nm on the Victor II. Vincristine sulfate was used during screening as a
control compound.
The following formula was used to calculate percent inhibition of test agents in
the cytotoxic assay:
⎛
⎞
(FI 590 of test agent − FI 590 of control )
⎟ × 100
Inhibition (%) = ⎜1 −
⎜ (FI 590 of 100% reduced Alamar blue − FI 590 of control ) ⎟
⎝
⎠
Where: FI 590 = Fluorescent Intensity at 590nm emission (535nm excitation)
For the fractions which were active in the first screening, they were again
evaluated activities with a ten-fold decreasing dose (0.1 μge/μL).
25
2.4 LC/MS analysis for bioactive fractions
The remaining drug-like fractions showing activity in in vitro screening were
subjected to a second LC/MS analysis for active component profiles, (C18 column
(Phenomenex Luna, 2 mm x 30 mm) at 0.2 mL/min). Depending on the position of
these fractions in the initial HPLC prefractionated profiles, an individual HPLC
program was developed for better separations as well as good resolutions obtained in
MS spectra. These LC/MS data were used to guide large-scale isolation (Table 2.3).
Table 2.3 Gradient timetable for LC/MS analysis of active fractions
1
2
3
4
5
6
7
Fractions 1-3
Time %A %B
0.0
98
2
1.0
98
2
8.0
50
50
10.0
20
80
12.0
20
80
12.1
98
2
14.0
98
2
Fractions 4-8
Time %A %B
0.0
90
10
1.0
70
30
5.0
70
30
9.0
20
80
12.0
0 100
12.1
90
10
14.0
90
10
8
Where
A: 0.1% FA in water
Fractions 9-11
Time
%A
%B
0.0
90
10
0.5
55
45
1.5
50
50
3.5
50
50
5.5
40
60
9.5
0
100
12.0
0
100
12.1
90
10
14.0
90
10
Flow
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
Curve
1
6
6
6
6
6
6
6
6
B: 0.1% FA in methanol
2.5 Generation of hit list
There were a total of 1155 fractions generated from 105 plant and marine biota
samples and 330 fractions were selected for testing. Of the 330 fractions, there were 19
active fractions from 14 samples and 311 inactive fractions from 91 samples. The
HPLC traces of the 14 active samples are shown below. For these 14 samples, screened
active fractions were highlighted in green and screened inactive fractions in grey.
Screened active peak(s)
Activity >80%
Screened peak(s) (not active)
Activity >40% and <80%
2.5.1 Pipestela candelabra (QID6012360)
Fractions
9
10
A549
23 -3
75
2
Active fractions
9
10
HeLa
85
5
99
91
LNCaP
81
23
103 47
(+) m/z
499, 513, 537, 523
527, 525, 539, 580, 628
26
PC3
71
1
88 42
HEK
43 12
85 31
(-) m/z
511
525, 578, 626
According to Soest et al.,7 Pipestela is a new genus of Axinellidae which has
been discovered recently and no literature has been reported on its chemical
investigation. Moreover, this fraction showed strong cytotoxic activity. Therefore, this
sponge was chosen for further study.
2.5.2 Neamphius huxleyi (QID6005333)
Fractions
9
10
A549
0
103
105
HeLa
83
1
105
64
102
Active fraction
10
LNCaP
104
(+) m/z
788, 795
85
PC3
47
9
107
104
HEK
81
8
109
105
(-) m/z
786, 793
Neamphius is a rare genus by Dictionary of Natural Products as well as
SciFinder Scholar search. Only two compounds have been isolated from this genus,
including neamphine (MW 167) and a peptide neamphamide A (MW 1688). LC/MS
analysis suggested the active fraction 10 contained compounds with different molecular
weights. This sponge demonstrated high activities against cancer cells at a concentration
0.01 μge/μl. Possessing high activity and rare taxonomy, this marine sponge was chosen
to be investigated.
2.5.3 Pseudoceratina sp. (QID024253)
Fractions
3
5
6
7
8
9
10
A549
1
-5
-4
-4
-4
-2 -2
17
1
HeLa
2
-1
-2
-2
0
-1
3
82
5
Active fractions
9
10
LNCaP
17
13
30
9
21
42
13
95
29
(+) m/z
668, 476, 506, 494
776, 803
27
PC3
13
18
16
10
13
9
2
55
4
HEK
15
11
14
8
13
8
4
31
7
(-) m/z
666, 474, 504, 295, 760
774, 1112, 1194
55
The genus Pseudoceratina is a well-known source for bromotyrosine alkaloids
with 55 compounds isolated. This sponge also showed strong activity and was chosen
for chemical investigation.
2.5.4 Suberea clavata (QID025573)
Fractions
1
6
7
8
9
10
11
A549
-1
-1
-4
-2
-2
10 -1
-2 -1
Active fractions
10
11
HeLa
-1
0
-2
-1
-1
16
1
1
2
LNCaP
23
26
29
31
10
83
30
59
14
(+) m/z
494, 480, 819, 834
480, 494, 743, 770
PC3
9
7
10
10
10
12
9
3
1
HEK
4
2
3
2
-2
24
5
10
-4
(-) m/z
760, 817, 832, 862, 897, 1094, 1112
817, 1112, 1095, 1032, 1080
The genus Suberea has been known to possess bromine derivatives. There have
been 22 compounds isolated according to Dictionary of Natural Products. However, no
compound matched with the mass data in active fractions. It is interesting that the active
peak has the ion peak at (-) m/z 1112, which is similar to that of the sponge
Pseudoceratina sp. This project satisfied all requirements of good and selective
activities as well as strong signals in the mass spectrum. Hence, it was selected.
2.5.5 Diacarnus sp. (QID6011588)
Fractions
7
9
10
11
A549
-2
-1
-3
1
31
3
HeLa
-1
-2
58
-1
104
2
28
LNCaP
26
25
29
2
58 18
PC3
11
7
59
3
84 12
HEK
5
1
9
1
75 13
Active fractions
10
11
(+) m/z
353
353, 393
(-) m/z
351
All 24 compounds from this genus reported in Dictionary of Natural Products
have different molecular weights compared with compounds in this active fraction. As it
had good and selective activities, this sponge was chosen for investigation.
2.5.6 Neopetrosia exigua (QID6008038)
Fractions
9
10
A549
102 -3
-2
HeLa
80
2
0
Active fraction
9
LNCaP
70
8
34
PC3
60
3
12
(+) m/z
244, 324
HEK
49
1
8
(-) m/z
322
Dictionary of Natural Products showed 8 compounds reported from this genus.
The activity of this fraction was quite selective with lung (A549) and cervical (HeLa)
cell lines. Moreover, the fraction contained several small molecules and their signals
displayed very strongly in the (+) ESI-MS. Therefore, this biota sample was chosen for
further chemical investigation.
2.5.7 Neolitsea dealbata (QID018658)
Fractions
4
5
6
7
8
9
A549
35
-2
39
101 39
-5
7
Active fraction
7
HeLa
29
17
45
83 27
8
11
(+) m/z
280, 296
LNCaP
21
25
48
59
19
18
20
PC3
14
12
52
44
12
21
34
HEK
15
8
34
45
31
-2
19
(-) m/z
278, 294
Only three triterpenoids (cycloneolitsin, taraxerone, taraxerol) and three
sesquiterpenoids (linderadine, pseudoneolinderane and linderalactone) have been
29
reported for this species. Mass data revealed alkaloids with molecular weight of less
than 500 Da, satisfying one parameter of the Ro5. This plant was chosen for further
study.
2.5.8 Chondria sp. (QID6008434)
Fractions
5
8
9
11
A549
1
-2
-2
-1 -5
HeLa
-2
-2
1
0
3
LNCaP
29
28
27
36
8
Active fraction
11
PC3
14
22
15
51 12
(+) m/z
295, 211, 185
HEK
6
6
0
17
4
(-) m/z
267
There were 44 compounds isolated from this genus Chrondria. Mass data
showed that compounds in the active peak did not match with any compounds isolated
previously. This alga showed strong signals in mass spectrum and selective activity
against prostate cancer cells PC3. However, it was not chosen due to its moderate
activity.
2.5.9 Aaptos aaptos (QID021901)
Fractions
2
3
4
9
A549
-1
-1
-1 -2
-2
HeLa
-1
-2
1
2
-1
Active fraction
LNCaP
17
30
70
5
14
(+) m/z
215, 229
4
PC3
28
27
34
1
5
HEK
12
5
42
2
4
(-) m/z
213, 227
According to Dictionary of Natural Products, aaptamine (MW 228) and 9-De-Omethylaaptamine (MW 214) have been identified as active components. MS result of
30
the active fraction suggested it contained compounds with MW of 228 and 214. Thus,
this sponge was not investigated.
2.5.10 Diacarnus spinipoculum (QID012499)
Fractions
8
9
10
11
A549
-4
-2
-3
-2
-4
Active fraction
11
HeLa
-1
-2
-1
-1
3
LNCaP
31
20
32
31 13
(+) m/z
409, 431, 407, 632
PC3
25
23
24
59
5
HEK
9
13
6
11 -4
(-) m/z
391, 407, 423, 425, 427
The genus Diacarnus is well-known for peroxide compounds. Most of the
compounds in fraction 11 had the same MW as diacarnoxide A, C (MW 408) or
tasnemoxide A, B (MW 408), diacarnoxide D (MW 406), epimuqubilin A (MW 392)
and aikupikoxide A (MW 424). No chemistry was pursued for this sponge.
2.5.11 Cystophora moniliformis (QID6015742)
LLP
9
10
11
A549
-4
-3
-4
-2
Active fraction
11
HeLa
2
-2
-5
2
LNCaP
25
31
57 17
(+) m/z
243, 261, 279, 301
PC3
6
7
-3
0
HEK
-3
3
-4
-3
(-) m/z
277
This alga was reported with 91 compounds, including polyphenol and
terpenoids. MS of the fraction showed an ion with the same mass as 8,13-Epoxy-14,15dinor-12-labden-3-ol (MW 278). Due to a moderate activity and possibly known
compounds, this alga was not investigated further.
31
2.5.12 Bohadschia marmorata (QID021666)
Fractions
9
10
11
A549
-1
103
2
103 -2
HeLa
0
104
3
105
1
Active fractions
10
11
LNCaP
20
103 13
103 20
(+) m/z
486, 470, 618
486, 470
PC3
10
107
9
107
4
HEK
9
108
-3
104
7
(-) m/z
1084, 1102, 1118, 1130, 1456
1102, 1148, 1426, 1456
Eleven compounds have been isolated from this genus. They were steroids and
sterol glycosides. Two active fractions were found containing ion peak at (+) m/z 486,
which matched with the previously isolated ternaygenin (MW 485). Despite strong
activities, this marine sponge was not selected.
2.5.13 Erylus amissus (61846.9)
Fractions
11
Active fraction
11
A549
103
-2
HeLa
104
26
(+) m/z
939, 953, 969, 993
LNCaP
104
11
PC3
106
2
HEK
109 38
(-) m/z
937, 951, 967, 981, 1011, 1023, 1077, 1096
Dictionary of Natural Products revealed 38 compounds isolated from the genus
Erylus. Most of them were sterol glycosides and glycolipids. Mass data showed strong
signals in negative mode. The active fraction showed compounds with ion peaks at (-)
m/z 937, 951, 967, 981 and 1096 suggesting to be eryloside H (MW 938), eryloside J
(MW 952), eryloside G (MW 968), eryloside I (MW 982) and a glycolipid erylusamine
TA (MW 1097). This biota was not chosen due to its unselective toxicity.
2.5.14 Solanum jucundum A.R.Bean (11044.8)
Fractions
2
3
A549
-12
4
HeLa
14
11
LNCaP
12
3
32
PC3
7
6
HEK
9
23
4
5
6
7
8
9
8
-2
8
-5
34
103
-3
20
2
12
1
12
105
30
22
-4
19
-9
103
3
Active fraction
13
(+) m/z
868
9
19
8
21
4
22
106
-1
6
-7
25
-6
31
110
-6
(-) m/z
866, 912, 1028
Genus Solanum contains steroidal glycosides, which have been known to be
cytotoxic. Over 485 compounds have been isolated from this genus. Mass spectroscopic
data of the active fraction showed strong signals in negative mode rather than positive
one. These peaks displayed the same fragmentation pattern suggesting they had the
same core structure. Although the activities were good, they were not selective.
Therefore, this plant was not pursued.
2.6 Screening results
In summary, 330 selected fractions were selected by UV profiles from 105
samples (90 fractions from 24 plants and 240 fractions from 81 marine samples). The
histogram (Figure 2.6) showed that UV active compounds from marine samples were
mostly in fractions 9-11 while the UV active fractions from plant sources were
distributed mainly in fractions 6-7.
60
Number of fractions
50
40
Marine fractions
30
Plant fractions
20
10
0
1
2
3
4
5
6
7
8
9
10
11
Fraction number
Figure 2.6 Histogram of selected fractions
33
Bioassay screening showed 19 out of 330 fractions (5.8%) having activity
against at least one type of cancer cell lines (Figure 2.7). In detail, there were 8 active
fractions against lung cancer cells A549 (2.4%), 12 active fractions against cervical
cancer cells HeLa (3.6%), 14 active fractions against prostate cancer cells PC3 (4.2%)
and 16 active fractions against prostate cancer cells LNCaP (4.8%).
Non-active fractions
94.2%
Active fractions
5.8%
Figure 2.7 Total cancer cell-based screening results
Taxonomic information of bioactive samples is shown in List 2.1. Samples
chosen for further chemical investigation were selected based on their bioactivity,
amount of material (>20g) and taxonomy. Thus, 7 samples have been investigated (List
2.2) and discussed in Chapters 3-7.
List 2.1 Taxonomic information of samples contains 19 active fractions
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
QID
6012360
6005333
024253
025573
6011588
6008038
018658
6008434
021901
012499
6015742
021666
61846.9
11044.8
Family
Axinellidae
Alectonidae
Pseudoceratinidae
Pseudoceratinidae
Podospongiidae
Petrosiidae
Lauraceae
Rhodomelaceae
Suberitidae
Podospongiidae
Cystoseiraceae
Holothuriidae
Geodiidae
Solanaceae
Genus
Pipestela
Neamphius
Pseudoceratina
Suberea
Diacarnus
Neopetrosia
Neolitsea
Chondria
Aaptos
Diacarnus
Cystophora
Bohadschia
Erylus
Solanum
Species
candelabra
huxleyi
sp.
clavata
sp.
exigua
dealbata
sp.
aaptos
spinipoculum
moniliformis
marmorata
amissus
jucundum
Samples chosen to be investigated
List 2.2 Chapters describing the investigated samples
QID
6011588
6008038
018658
6005333
6012360
025573
024253
Class
Demospongiae
Demospongiae
N.A
Demospongiae
Demospongiae
Demospongiae
Demospongiae
Order
Poecilosclerida
Haplosclerida
N.A
Hadromerida
Halichondrida
Verongida
Verongida
Family
Podospongiidae
Petrosiidae
Lauraceae
Alectonidae
Axinellidae
Pseudoceratinidae
Pseudoceratinidae
34
Genus
Diacarnus
Neopetrosia
Neolitsea
Neamphius
Pipestela
Suberea
Pseudoceratina
Species
sp.
exigua
dealbata
huxleyi
candelabra
clavata
sp.
Chapter
3
3
4
5
6
7
7
2.7 References
1.
G. M. Cragg, P. G. Grothaus and D. J. Newman, Chem. Rev., 2009, 109, 30123043.
2.
D. J. Newman and G. M. Cragg, J. Nat. Prod., 2007, 70, 461-477.
3.
F. Lombardo, M. Y. Shalaeva, K. A. Tupper, F. Gao and M. H. Abraham, J.
Med. Chem., 2000, 43, 2922-2928.
4.
F. E. Koehn, High Impact Technologies for Natural Products Screening,
Natural Compounds as Drugs, Birkhauser, Basel, Switzerland, 2008.
5.
D. D. Baker, M. Chu, U. Oza and V. Rajgarhia, Nat. Prod. Rep., 2007, 24, 12251244.
6.
G. J. R. Zaman, Drug Discov. Today, 2004, 9, 828-830.
7.
B. Alvarez, J. N. A. Hooper and R. W. M. V. Soest, Memoir Queensl Mus, 2008,
52, 105-118.
35
36
Chapter 3
Previously isolated compounds from the Australian
sponges Diacarnus sp. and Neopetrosia exigua – A new
3-alkylpyridinium alkaloid from N. exigua
This chapter reports the chemical investigation of the CH2Cl2/MeOH extracts of
the Australian sponges Diacarnus sp. and Neopetrosia exigua whose fractions showed
activities in a cytotoxic screening. Mass-guided identification afforded the isolation of
four known compounds, sigmosceptrellin A (27), diacarperoxide A (28) and methyl
diacarnoate A (29) from Diacarnus sp. and motuporamine C (42) from Neopetrosia
exigua. An unreported alkaloid, dehydrocyclostellettamine A (43) was also identified
from N. exigua.
3.1 Chemical investigation of the sponge Diacarnus sp.
The genus Diacarnus, one of six genera Diacarnus,1-6 Latrunculia,7 Mycale,8,9
Plakortis,10,11 Prianos12,13 and Sigmosceptrella14,15 (Table 3.1), has been known as a
rich source of terpene peroxides in marine organisms.5 According to Dictionary of
Natural Products (version 2009),16 twenty one out of twenty four natural products
isolated from this genus were norterpene and norsesterterpene peroxides. These terpene
peroxides have been known to possess a variety of biological activities including
cytotoxic,4,6,17 ichthyotoxic,14,15 sea urchin egg cell division inhibiting,12 antimicrobial,18
antiviral,17 antitoxoplasmodic19 and antimalarial activities.3
Table 3.1 Taxonomy of sponges producing terpene peroxides
Genus
Diacarnus
Latrunculia
Mycale
Plakortis
Prianos
Sigmosceptrella
Unidentified18
Species
cf. spinopoculum,4 levii,1,2,6 megaspinorhabdosa5
sp.,7 brevis20
sp.,9 ancorina,8 cf. spongiosa21
sp.,10 zyggompha11
sp.12,13
laevis14,15
3.1.1 Collection, Extraction and Isolation
A specimen of Diacarnus sp. was collected at the depth of 22m, Potter Reef,
Queensland, Australia. It was identified as Diacarnus sp. (phylum Porifera, class
Demospongiae, order Poecilosclerida, family Podospongiidae). A voucher specimen
37
QMG321995 has been deposited at the Queensland Museum, South Brisbane,
Queensland, Australia.
Plate 3.1 Photograph of the sponge Diacarnus sp.
A freeze dried sample of Diacarnus sp. (5.0g) was extracted exhaustively with
hexane (200ml), dichloromethane (2 x 250ml) and methanol (250ml), respectively. The
dichloromethane extract was evaporated to yield a yellow residue (0.7g). This crude
extract was pre-absorbed onto diol (1.0g) and packed dry into a small cartridge, which
was connected to a diol preparative HPLC column (5μm, 21.2 x 150mm). A linear
gradient from 100% hexane to 50% hexane – 50% isopropanol was performed over 60
minutes at a flow rate of 9.0 ml/min and 60 fractions (1.0 minute each) were collected.
Mass-guided identification showed fractions 10 to 20 contained the ion peaks of interest
in (+)-LRESIMS at (+)m/z 353 and 393. These fractions was then combined and
subjected on a diol preparative HPLC column (5μm, 21.2 x 150mm) in 30 minutes.
Three compounds, sigmosceptrellin A (27, 4.0mg, 0.08% dry wt), diacarperoxide A (28,
0.8mg, 0.016% dry wt) and methyl diacarnoate A (29, 0.4mg, 0.008% dry wt) were
isolated (Scheme 3.1).
38
Diacarnus sp. (5.0g)
a)
DCM (0.7g)
b)
Fractions 10-20
(+)-m/z 353, 393
Compound 27
(4.0mg, 0.08%)
Fractions 10-14
Compound 28
(0.8mg, 0.016%)
Fraction 18
c)
Compound 29
(0.4mg, 0.008%)
Fraction 20
a) Extraction with DCM
b) A linear gradient from 100% hexane to 50% hexane – 50% isopropanol in 60 minutes, a flow
rate of 9.0 ml/min, 1.0 min/fraction
c) A linear gradient from 100% hexane to 80% hexane – 20% isopropanol in 30 minutes, a flow
rate of 9.0 ml/min, 1.0 min/fraction
Scheme 3.1 Extraction and Isolation Procedure for Diacarnus sp.
3.1.2 Structure Elucidation
Structures of sigmosceptrellin A (27), diacarperoxide A (28) and methyl
diacarnoate A (29) were elucidated on the basis of their spectroscopic data (MS and
NMR). The NMR data and optical rotation values of 27 ([α]D25 +51.4o (c 0.8, CHCl3);
28, ([α]D25 +40.8o (c 0.1, CHCl3) and 29, ([α]D25 -25.1o (c 0.1, CHCl3) were similar to
those reported in the literatures.2,5,9,14,15
O
O
O
OH
O
O
H
O
O
O
O O
O
O O
Sigmosceptrellin A (27)
Diacarperoxide A (28)
39
Methyl diacarnoate A (29)
3.2 Chemical Investigation of the Sponge Neopetrosia exigua
Chemical investigation of the genus Neopetrosia was first reported by Oku et
22
al. in 2003. A new tetrahydroisoquinoline alkaloid, renieramycin J (30) isolated from
this sponge demonstrated potent cytotoxicity against 3Y1, HeLa and P388 cells with the
IC50 values in a range of nanomole. Upon further investigation on this genus, two
peptides, neopetrosiamide A (31) and B (32),23 along with other three alkaloids,
njaoamine G (33), njaoamine H (34) and 1,2,3,4-tetrahydroquinolin-4-one (35)24 were
discovered. These compounds also showed strong cytotoxicity on a cell-based assay.
Chemistry of the marine sponge N. exigua, in particular, was first studied in 2004.25 A
series of six sterols, three galactosyl diacylglycerols and four alkaloids (36-39) were
identified. Bioassay testing revealed these alkaloids possessed a moderate cytotoxicity
against HL-60 cells but they were inactive towards the five test microorganisms. An
alkaloid exiguamine A (40) was identified from the Papua New Guinea sponge N.
exigua in 2006.26 This compound was known as one of the most potent indoleamine2,3-dioxygenase (IDO) inhibitors with a Ki of 210 nM. The most recent component of N.
exigua found in the Australian sponge N. exigua was a novel pentacyclic hydroquinone,
exiguaquinol (41). With the inhibition of Helicobacter pylori glutamate racemase
(MurI) at an IC50 of 4.4 μM, compound 41 was the first H. pylori MurI inhibitor derived
from natural product sources.27
Some natural products isolated from the genus Neopetrosia
O
HO
O
NH2
H
H
OH
N
N
N
OH
O
O
O
OH
N
O
N
R
N
H
O
Renieramycin J (30)
Njaoamine G (33): R = H
Njaoamine H (34): R = OH
40
1,2,3,4-tetrahydro
quinolin-4-one (35)
Ph
Ph
O
O
O
O
H2N
O
H
N
H
N
N
N
H
N
H
N
H
O
HN
O
O
S
Ph
S
HN
O
NH
O
HN
OH
O
S
NH
O
OH
S
HO
R
O
HN
HN
OH
HN
O
O
O
HN
O
H
N
H
N
O
O
O
N
S
OH
NH2
O
O
Ph
HN
O
S
O
H
N
O
O
O
O
H
N
NH
H
N
N
H
N
H
O
N
H
N
O
O
O
Neopetrosiamide A (31): R =
-
O
S+
Neopetrosiamide B (32): R =
-
O
S+
HO
HN
HN
HN
NH2
NH2
HN
N
N
N
O
O
O
OH
OH
O
O
O
N
N
N
Araguspongine M (36) Araguspongine B (37)
NH2
Araguspongine D (38)
Dopamine (39)
O
H2N
O
N
N
OSO3
O
O
O
OH
N
H
H
HO
N
O
N
O
O
OH
O3S
Exiguaquinol (41)
Exiguamine A (40)
3-Alkylpyridinium alkaloids (3-APs) have been considered as chemical markers
for the systematic determination of haplosclerid sponges (Table 3.2).28 So far, over
seventy 3-APs have been identified from marine sponges.29 Depending on the length,
branching and termination of the alkyl chains, 3-APs have been divided into five
groups, monomeric 3-APs, linear dimeric 3-APs, cyclic dimeric 3-APs, tricyclic 3-APs
and 3-AP polymers. These metabolites showed cytotoxicity at concentrations of a few
micrograms per millilitre,30 antibacterial activity31 and enzyme inhibition.32 Turk and
his co-workers29 found that potent bioactivities of linear 3-AP compounds were more
active than the cyclic forms. The tendency in biological activities increased with the
41
degree of polymerization of 3-APs and 3-APs in quaternary ammonium ions were more
active than neutral or tertiary amine forms.
Table 3.2 Taxonomy of sponges producing 3-alkylpyridinium alkaloids
Genus
Amphimedon
Callyspongia
Calyx
Cribrochalina
Haliclona
Niphates
Pachychalina
Reniera
Stelletta
Theonella
Xestospongia
Species
sp. , compressa,34 viridis35
sp.,36 fibrosa,37 ridleyi38
podatypa39
sp.40
viscosa41
sp.42
sp.43
sarai44
maxima45
swinhoei46
wiedenmayeri47
33
3.2.1 Collection, Extraction and Isolation
A specimen of Neopetrosia exigua was collected at the depth of 20m, Houghton
Reef, Howick Group, Queensland, Australia. It was identified as Neopetrosia exigua
(phylum Porifera, class Demospongiae, order Haplosclerida, family Petrosiidae). A
voucher specimen QMG320774 has been deposited at the Queensland Museum, South
Brisbane, Queensland, Australia.
Plate 3.2 Photograph of the sponge N. Exigua
A freeze dried sample of Neopetrosia exigua (5.0g) was extracted exhaustively
with hexane (250ml), dichloromethane (2 x 250ml) and methanol (2 x 250ml),
respectively. The dichloromethane and methanol extracts were combined and then the
solvents were evaporated to yield a brown residue (0.5g). This crude extract was preabsorbed onto C18 (1.0g), packed dry into a small cartridge and connected to a C18
preparative HPLC column (5μm, 21.2 x 150mm). A linear gradient from 100% water
(0.1% TFA) to 100% methanol (0.1% TFA) was performed over 60 minutes at a flow
42
rate of 9.0 ml/min and 60 fractions (1.0 minute each) were collected. Mass-guided
identification demonstrated fractions 36 to 44 contained the ion peaks of interest in (+)LRESIMS at (+)m/z 244 and 324. Fractions 36-44 were then combined and subjected
on the C18 preparative HPLC column (5μm, 21.2 x 150mm) yielding motuporamine C
(42, 4.5mg, 0.09% dry wt) and dehydrocyclostellettamine A (43, 3.0mg, 0.06% dry wt)
(Scheme 3.2).
Neopetrosia exigua (5.0g)
a)
DCM/MeOH (0.5g)
b)
Fractions 36-44
(+)-m/z 244, 324
Compound 4
(4.5mg, 0.09%)
Fractions 20-22
c)
Compound 5
(3.0mg, 0.06%)
Fractions 34-35
a) Extraction with DCM and MeOH
b) A linear gradient from 100% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes, a
flow rate of 9.0 ml/min, 1.0 min/fraction
c) A linear gradient from 20% methanol (0.1% TFA) – 80% water (0.1% TFA) to 65% methanol
(0.1% TFA) – 35% water (0.1% TFA) in 60 minutes, a flow rate of 9.0 ml/min, 1.0 min/fraction
Scheme 3.2 Extraction and Isolation Procedure for N. exigua
3.2.2 Structure Elucidation and Discussion
3.2.2.1 Motuporamine C (42)
Structure of compound 42 was characterised by its spectroscopic data (MS, 1H,
13
C and 2D-NMR). Compared with NMR data in the literature, it was found that 42 was
motuporamine C previously reported.48-50
N
N
H
NH2
Motuporamine C (42)
3.2.2.2 Dehydrocyclostellettamine A (43)
Compound 43 was isolated as light yellowish gum. The (+)-HRESIMS
displayed a divalent molecular ion at m/z 244.2066 (calcd (+)m/z (z = 2) 244.2060, Δ2.5
ppm) which was consistent for the molecular formula C34H52N22+.
43
Figure 3.1 The 1H-NMR spectrum of compound 43 recorded at 600 MHz in DMSO-d6
Combined data from the
13
C-NMR and HSQC spectra indicated that 43 was
superimposable with the presence of seventeen carbons, including one quaternary
carbon (δC 142.3 ppm), six sp2 tertiary carbons (δC 145.2, 144.1, 142.4, 142.3, 131.4,
127.5 and 127.2 ppm) and ten secondary carbons (δC 60.7, 31.6, 30.7, 29.0, 28.8, 28.6,
28.4, 27.3, 26.6 and 25.5 ppm). The chemical shifts and coupling constant patterns of
protons at δΗ 9.02 (brs), 8.93 (d, J=6.0 Hz), 8.48 (d, J=7.8 Hz) and 8.07 (t, J=6.0, 7.8
Hz) were indicative of a 1,3-disubstituted alkylpyridinium system (fragment A).41,51,52
This assignment was further confirmed by HMBC correlations from H-2 (δΗ 9.02ppm)
to C-4 (δC 145.2 ppm) and C-6 (δC 142.4 ppm); from H-5 (δΗ 8.07 ppm) to C-6 (Figure
3.2).
COSY correlations afforded the establishment of two spin systems –C7–C8–C9–
(fragment B) and –C13–C14–C15=C16–C17–C18– (fragment C).
3
JHC long range
correlations from H-7 (δΗ 4.54 ppm) to C-9 (δC 25.5 ppm); from H-13 (δΗ 1.19 ppm)
and H-17 (δΗ 2.39 ppm) to C-15 (δC 131.4 ppm); from H-14 (δΗ 1.88 ppm) and H-18
(δΗ 2.84 ppm) to C-16 (δC 127.3 ppm) supported the assignment of these two long chain
moieties (Figure 3.2). The Z geometry was assigned for Δ15,16 due to the observed J
coupling constant (J=12.0 Hz) between H-15 and H-16.52,53
4
5
15
3
6
N
2
Fragment A
7
9
13
16
18
HMBC
COSY
Fragment C
Fragment B
Figure 3.2 The partial structures A, B and C of compound 43
44
Detailed analysis of HMBC data demonstrated HMBC correlations from H-7 to
C-2 (δC 144.1 ppm) and C-6 (δC 142.4 ppm), H-18 to both C-2 and C-4 (δC 145.2 ppm)
and H-17 to C-3 (δC 142.3 ppm). This information was evident to the connectivity of
fragments B and C to C-1 and C-3 in fragment A (Figure 3.3). A chain of three
remaining secondary carbons –C10–C11–C12– was assigned to attach to C-9 on the basis
of their 1H and
13
C resonances. Hence, the complete structure of 43 was finally
established as a symmetrical structure. With the length of the alkyl chains and the
number of monomers, compound 43 was identified as a new 3-alkylpyridinium,
dehydrocyclostellettamine A.
18'
4
13'
9'
11'
3
6
N
7'
N
2'
2
HMBC
7
3'
17
9
15
13
11
16
4'
18
Figure 3.3 Key HMBC correlations to establish the structure of 43
Table 3.3 NMR data for dehydrocyclostellettamine A (43) in DMSO-d6 at 30oC
Position
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
δCa
δH (mult., J, int.)b
144.1
142.4
145.2
127.5
142.3
60.7
30.7
25.5
29.0
28.8
28.4
28.6
26.6
131.4
127.3
27.3
31.6
9.02 (s, 1H)
8.48 (d, 7.8, 1H)
8.07 (t, 6.0, 1H)
8.93 (d, 6.0, 1H)
4.54 (br.s, 1H)
1.89 (br.s, 1H)
1.22 (m, 2H)
1.23 (m, 2H)
1.23 (m, 2H)
1.23 (m, 2H)
1.19 (br.s, 2H)
1.88 (bs, 2H)
5.36 (dd, 7.2, 12.0, 2H)
5.34 (dd, 6.6, 12.0, 2H)
2.39 (dd, 6.6, 7.2, 2H)
2.84 (t, 7.2, 2H)
gCOSY
(H no.)
5
4, 6
5
8
7, 9
8
13, 15
14, 16
15, 17
16, 18
14
gHMBC
(C no.)
4, 6, 7, 18
2, 6, 18
3, 6
2, 4, 5, 7
2, 6, 8, 9
7, 10
15
13, 15, 16
14
17
3, 15, 16
2, 4, 16
4
6
N
N
2
7
9
11
13
45
15
16
18
3.3 Evaluation of “drug-like” properties of the isolated compounds
The physico-chemical properties of all isolated compounds were evaluated by JChem software. Four parameters of the Ro5 (MW, logP, HBA and HBD) and two
others (PSA and NROT) were calculated (Table 3.4). The results showed all five
compounds complied with both Lipinski’s and Veber’s rules. Although compound 27 is
quite hydrophobic with ClogP value of 6.42 and more than 5, it is considered to be
orally active since its other remaining parameters pass the Lipinski’s and Veber’s rules.
Table 3.4 Physico-chemical properties of isolated compounds
Compounds
Formula
MW
logP
HBA
HBD
PSA
NROT
27
28
29
42
43
C25H42O4
C20H32O5
C20H32O5
C20H41N3
C34H52N22+
406.60
352.47
352.47
323.56
488.79
6.42
4.20
4.20
3.64
1.87
4
4
4
3
0
1
0
0
2
0
55.76
61.83
61.83
41.29
7.76
5
6
6
7
0
Predicted
Bioavailability
9
9
9
9
9
3.4 Biological activity
Three out of five compounds were active in in vitro cytotoxicity assays.
Compound 27 showed the strongest inhibition against all cancer and non-cancer cells
with IC50 values from 0.4 to 3.0 μM while its peroxide derivatives (28 and 29) were not
active. The long chain in the structures of 42 and 43 might be a reason affecting their
cytotoxicity towards all cell lines. In particularly, these compounds did not demonstrate
selective activity.
Table 3.5 Evaluation of cytotoxic potential of compounds 27-29 and 42-43
Compound
27
28
29
42
43
Vincristine
sulfate
A549
0.5±0.3
>100
90.2±8.5%
3.5±1.1
13.6±5.0
HeLa
0.4±0.2
>100
68.7±12.1%
6.8±0.4
10.6±1.2
Cell lines/ IC50 ± SD (μM)
LNCaP
PC3
3.1±0.8
0.9±0.2
>100
90±10%
>100
95±5%
3.0±0.2
5.4±0.6
9.9±2.0
13.7±3.5
0.020±0.002
0.036±0.003
0.006±0.001
0.015±0.001
HEK
1.3±0.1
>100
90.5±4.6%
4.0±0.5
6.8±0.6
NFF
1.2±0.8
>100
92.3±10.9%
3.3±1.2
9.7±10.0
0.030±0.004
0.017±0.003
3.5 References
(1)
D'Ambrosio, M.; Guerriero, A.; Deharo, E.; Debitus, C.; Munoz, V.; Pietra, F.
Hel. Chim. Acta 1998, 81, 1285-1292.
(2)
D'Ambrosio, M.; Guerriero, A.; Debitus, C.; Waikedre, J.; Pietra, F.
Tetrahedron Lett. 1997, 38, 6285-6288.
(3)
Youssef, D. T. A.; Yoshida, W. Y.; Kelly, M.; Scheuer, P. J. J. Nat. Prod. 2001,
64, 1332-1335.
46
(4)
Sperry, S.; Valeriote, F. A.; Corbett, T. H.; Crews, P. J. Nat. Prod. 1998, 61,
241-247.
(5)
Ibrahim, S. R. M.; Ebel, R.; Wray, V.; Muller, W. E. G.; Ebel, R. E.; Proksch, P.
J. Nat. Prod. 2008, 71, 1358-1364.
(6)
Dai, J.; Liu, Y.; Zhou, Y. D.; Nagle, D. G. J. Nat. Prod. 2007, 70, 130-133.
(7)
Ovenden, S. P. B.; Capon, R. J. J. Nat. Prod. 1999, 62, 214-218.
(8)
Capon, R. J.; Macleod, J. K. J. Nat. Prod. 1987, 50, 225-229.
(9)
Capon, R. J.; Rochfort, S. J.; Ovenden, S. P. B. J. Nat. Prod. 1997, 60, 12611264.
(10)
Feng, Y.; Davis, R. A.; Sykes, M.; Avery, V. M.; Camp, D.; Quinn, R. J. J. Nat.
Prod. 2010, 73, 716-719.
(11)
Berrué, F.; Thomas, O. P.; Bon, C. F.-L.; Reyes, F.; Amade, P. Tetrahedron
2005, 61, 11843-11849.
(12)
Manes, L. V.; Bakus, G. J.; Crews, P. Tetrahedron Lett. 1984, 25, 931-934.
(13)
Kashman, Y.; Rotem, M. Tetrahedron Lett. 1979, 19, 1707-1708.
(14)
Albericci, M.; Lempereur, M. C.; Braekman, J. C.; Daloze, D.; Tursch, B.;
Declercq, J. P.; Germain, G.; Meerssche, M. V. Tetrahedron Lett. 1979, 29,
2587-2690.
(15)
Albericci, M.; Braekman, J. C.; Daloze, D.; Tursch, B. Tetrahedron 1982, 38,
1881-1890.
(16)
In Dictionary of Natural Products on CD-Rom Chapman and Hall ed.; CRC
Press: London, 2009.
(17)
Tanaka, J.; Higa, T.; Suwanborirux, K.; Kokpol, U.; Bernardinelli, G.; Jefford,
C. W. J. Org. Chem. 1993, 58, 2999-3002.
(18)
Capon, R. J.; Macleod, J. K. Tetrahedron 1985, 41, 3391-3404.
(19)
El-Sayed, K. A.; Hamann, M. T.; Hashish, N. E.; Shier, W. T.; Kelly, M.; Khan,
A. A. J. Nat. Prod. 2001, 64, 522-524.
(20)
Capon, R. J.; MacLeod, J. K.; Willis, A. C. J. Org. Chem. 1987, 52, 339-342.
(21)
Capon, R. J. J. Nat. Prod. 1991, 54, 190-195.
(22)
Oku, N.; Matsunaga, S.; Soest, R. W. M. V.; Fusetani, N. J. Nat. Prod. 2003, 66,
1136-1139.
(23)
Williams, D. E.; Austin, P.; Marrero, A. R. D.; Soest, R. V.; Matainaho, T.;
Roskelley, C. D.; Roberge, M.; Andersen, R. J. Org. Lett. 2005, 7, 4173-4176.
(24)
Sorek, H.; Rudi, A.; Benayahu, Y.; Kashman, Y. Tetrahedron Lett. 2007, 48,
7691-7694.
47
(25)
Liu, H.; Mishima, Y.; Fujiwara, T.; Nagai, H.; Kitazawa, A.; Mine, Y.;
Kobayashi, H.; Yao, X.; Yamada, J.; Oda, T.; Namikoshi, M. Mar. Drugs 2004,
2, 154-163.
(26)
Brastianos, H. C.; Vottero, E.; Patrick, B. O.; Soest, R. V.; Matainaho, T.; Mauk,
A. G.; Andersen, R. J. J. Am. Chem. Soc. 2006, 128, 16046-16047.
(27)
Leone, P. D. A.; Carroll, A. R.; Towerzey, L.; King, G.; McArdle, B. M.; Kern,
G.; Fisher, S.; Hooper, J. H. A.; Quinn, R. J. Org. Lett. 2008, 10, 2585-2588.
(28)
Andersen, R. J.; Soest, R. W. M. V.; Kong, F. Alkaloids: Chemical and
Biological Perspectives; Pergamon Press: New York, 1996; Vol. 10.
(29)
Turk, T.; Sepcic, K.; Mancini, I.; Guella, G. Studies in Natural Products
Chemistry: Bioactive Natural Products (Part O); Elsevier, 2008; Vol. 35.
(30)
Turk, T.; Frangez, R.; Sepcic, K. Mar. Drugs 2007, 5, 157-167.
(31)
Oliveira, J. H. H. L. D.; Seleghim, M. H. R.; Timm, C.; Grube, A.; Köck, M.;
Nascimento, G. G. F.; Martins, A. C. T.; Silva, E. G. O.; Souza, A. O. D.;
Minarini, P. R. R.; Galetti, F. C. S.; Silva, C. L.; Hajdu, E.; Berlinck, R. G. S.
Mar. Drugs 2006, 4, 1-8.
(32)
Sepcic, K.; Turk, T. Progress in Molecular and Subcellular Biology, Subseries
Marine Molecular Biotechnology; Springer: Berlin, Heidelberg, 2006.
(33)
Hirano, K.; Kubota, T.; Tsuda, M.; Mikami, Y.; Kobayashi, J. Chem. Pharm.
Bull. 2000, 48, 974-977.
(34)
Albrizio, S.; Ciminiello, P.; Fattorusso, E.; Magno, S.; Pawlik, J. R. J. Nat.
Prod. 1995, 58, 647-652.
(35)
Kelman, D.; Kasman, Y.; Rosenberg, E.; Ilan, M.; Ifrach, I.; Loya, Y. Aquat.
Microb. Ecol. 2001, 24, 9-16.
(36)
Wang, G. Y. S.; Kuramoto, M.; Uemura, D.; Yamada, A.; Yamaguchi, K.;
Yazawa, K. Tetrahedron Lett. 1996, 37, 1813-1816.
(37)
Coleman, M. T. D.; Faulkner, D. J.; Dubowchik, G. M.; Roth, G. P.; Polson, C.;
Fairchild, C. R. J. Org. Chem. 1993, 58, 5925-5930.
(38)
Scott, R. H.; Whyment, A. D.; Foster, A.; Gordon, K. H.; Milne, B. F.; Jaspars,
M. J. Membrane. Biol. 2000, 176, 119-131.
(39)
Stierle, D. B.; Faulkner, D. J. J. Nat. Prod. 1991, 54, 1134-1136.
(40)
Matsunaga, S.; Shinoda, K.; Fusetani, N. Tetrahedron Lett. 1993, 34, 59535954.
(41)
Volk, C. A.; Kock, M. Org. Biomol. Chem. 2004, 2, 1827-1830.
(42)
Quinoa, E.; Crews, P. Tetrahedron Lett. 1987, 28, 2467-2468.
48
(43)
Oliveira, J. H. H. L. D.; Grube, A.; Kock, M.; Berlinck, R. G. S.; Macedo, M.
L.; Ferreira, A. G.; Hajdu, E. J. Nat. Prod. 2004, 67, 1685-1689.
(44)
Sepcic, K.; Guella, G.; Mancini, I.; Pietra, A.; Serra, M. D.; Menestrina, G.;
Tubbs, K.; Macek, P.; Turk, T. J. Nat. Prod. 1997, 60, 991-996.
(45)
Fusetani, N.; Asai, N.; Matsunaga, S.; Hirota, H. Tetrahedron Lett. 1994, 35,
3967-3970.
(46)
Kobayashi, J.; Zeng, C.; Ishibashi, M.; Shigemori, H.; Sasaki, T.; Mikami, Y. J.
Chem. Soc. Perkin Trans. 1 1992, 1291-1294.
(47)
Sakemi, S.; Totton, L. E.; Sun, H. H. J. Nat. Prod. 1990, 53, 995-999.
(48)
Williams, R. E.; Lassota, P.; Andersen, R. J. J. Org. Chem. 1998, 63, 48384841.
(49)
Fürstner, A.; Rumbo, A. J. Org. Chem. 2000, 65, 2608–2611.
(50)
Williams, D. E.; Craig, K. S.; Patrick, B.; McHardy, L. M.; Soest, R. V.;
Roberge, M.; Andersen, R. J. J. Org. Chem. 2002, 67, 245-258.
(51)
Xu, N. J.; Sun, X.; Yan, X. J. Chinese Chem. Lett. 2007, 18, 947-950.
(52)
Casapullo, A.; Pinto, O. C.; Marzocco, S.; Autore, G.; Riccio, R. J. Nat. Prod.
2009, 72, 301-303.
(53)
Oku, N.; Nagai, K.; Shindoh, N.; Terada, Y.; Soest, R. W. M. V.; Matsunaga, S.;
Fusetani, N. Bioorg. Med. Chem. Lett. 2004, 14, 2617-2620.
49
50
Chapter 4
Aporphinoid alkaloids from the Australian plant
Neolitsea dealbata
4.1 Introduction
Aporphinoid alkaloids are distributed in plant families, Annonaceae,
Hernandiaceae, Lauraceae, Menispermaceae, Monimiaceae and Ranunculaceae.1-4 Up to
now, more than 500 aporphine alkaloids have been isolated.5 Structurally, aporphinoid
skeletons are constructed by a system of 4 rings A, B, C and D,6 and have been
categorised into five classes including aporphines, proaporphines, oxoaporphines,
dehydroaporphines and dimeric aporphines (Figure 4.1).5 Among these classes,
aporphines have been known as the largest class which can be divided into three groups:
noraporphines when the nitrogen is secondary, aporphines when the nitrogen is tertiary
and aporphine salts if the nitrogen is quaternary.5 According to the structural study on
the aporphine alkaloids, Shamma and Guinaudeau7 found that positions 1 and 2 are
often oxygenated by hydroxy, methoxy or methylenedioxy groups. In some cases, these
groups appear at C-3, C-8, C-9, C-10 or C-11 and rarely encounter oxygenation at C-7
or C-4.
2
O
5
6a
O
1
H3CO
H3CO
B
A
C
NH
N
N
HO
H3CO
H
H
11
O
D
8
O
OCH3 H3CO
O
O
Glaziovine (46)
(Proaporphine)
Lysicamine (45)
(Oxoaprophine)
Anonaine (44)
(Aporphine)
N
N
N
OH
HO
O
HO
H
H
Dehydronstephalagine 47)
(Dehydroaporphine)
Khyberine (48)
(Demiric aporphine)
Figure 4.1 Some types of aporphinoid structures5,7
Possessing a chiral centre at position 6a, aporphines are optically active. A
stereochemical survey of Craig et al.8 for nine aporphines indicated it is possible to
assign absolute configuration of the aporphine skeleton by an optical rotatory dispersion
(ORD) experiment. The S-series showed a high amplitude positive Cotton Effect (CE)
at 235-245 nm while the R-series displayed the opposite sign. They also took a notice
51
that a (S)-configuration was assigned if aporphine structures had substituents at C-1, C2, C-10 and C-11, and a (R)-configuration would be verified if these compounds had no
substituents at carbons 10 and 11. Following this stereochemical study, this group also
found that an circular dichroism (CD) experiment is a better method to determine
aporphine stereochemistry.9 Examining CD spectra of nineteen aporphine alkaloids
showed that (6aS)-aporphines had a maximum positive CE at 240nm and (6aS)aporphines possessed the strong negative CE at the same area wavelength. Craig also
indicated the substituents in ring D can affect the configuration of a twisted biphenyl
system in aporphines. If this ring contains two or more oxygenated substituents,
aporphines will belong to the (S)-configuration while with one or no substituent on this
ring, they may belong to either two stereochemical series. The relationship between the
chiral centre at the position 6a and the twisted biphenyl system is shown as below.9
3
3a
A
B
N
C
1b
1
R
1a
H
5
6a
11a
N
R
H
7a
D
10
S-configuration
R-configuration
Figure 4.2 The special feature of the aporphine system9
Aporphine alkaloids have been known to exhibit antivirus, antibacterial, antiplatelet, antioxidant, cytotoxic and anticancer activities.5 For cytotoxicity and anticancer
activities, structure activity relationships of aporphine compounds have been
progressively investigated.5 Cordell et al.10 found that a methylenedioxy group attached
to carbons 1 and 2 in ring A might play an important role in the cytotoxicity of
aporphine alkaloids. However, the results of screening 53 isoquinoline alkaloids against
five cancer cell lines (A549, HCT8, KB, P388 and L1210)11 revealed only two out of
four aporphine alkaloids containing a methylenedioxy group were active against at least
one of five cancer cell lines with IC50 values less than 4 μg/ml and two other compounds
had no activities. This study indicated the methylenedioxy group was important but not
sufficient in cytotoxic properties. Compared with ring A, there was no evidence to
prove that the N-methylation or N-acetylation at position 6 in ring B had an influence on
cytotoxicity. Furthermore, a research on cytotoxicity of aporphine N-oxides also
displayed that the N-oxide moiety in ring B did not enhance activity.11 Interestingly, an
oxo functional group in ring C was found as a key element of the pharmacophore for
oxoaporphines.11 Structure and cytotoxic relationships of ring D have not been
investigated so far.
52
Neolitsea, a genus of the Lauraceae family, contains 80 species distributed in
tropical areas from Asia, Malesia to Australia.12 Until now, some species in this genus
have been thoroughly investigated N. acuminatissima,13 N. buisanensis,14 N. cuipala,15
N. Parvigemma,16,17 N. pubescens,18 N. pulchella,19 N. sericea,20 N. villosa21 and N.
zeylanica.22 According to Dictionary of Natural products version 2010,23 there were 20
aporphines among 92 compounds isolated from the genus Neolitsea. However, studies
of the chemistry of N. dealbata are limited to the isolation of three sesquiterpenoids
(linderadine (49), pseudoneolinderane (50) and linderalactone (51))24 and three
triterpenoids (taraxerone (52), taraxerol (53) and cycloneolitsin (54)).24,25
O
O
O
O
O
O
O
O
O
O
O
O
Linderadine (49) Pseudoneolinderane (50) Linderalactone (51)
H
H
H
HO
O
O
H
H
Taraxerone (52)
H
Taraxerol (53)
Cycloneolitsin (54)
Figure 4.3 Compounds previously isolated from the plant N. dealbata
This chapter details the chemical investigation of the N. dealbata (R.Br.) Merr.
bark, one of two species in the genus Neolitsea distributed commonly in Australia. One
new aporphine, (6aR)-normecambroline (55) along with eight known alkaloids (6aR)roemerine
(56),26
actinodaphnine
roemerine-Nα-oxide
(59),27
(57),
(6aS)-laurolitsine
roemerine-Nβ-oxide
(60),28,29
(6aS)-boldine
(58),
(6aS)-
(61),30
(1S)-
norjuziphine (62)31,32 and (1S)-juziphine (63)32 were isolated. The structure of 55 was
characterized by NMR and MS data. Its absolute stereochemistry was elucidated by a
CD experiment. Although roemerine N-oxide has been previously reported as a
constituent of Papaver glaucum33 and Papaver gracile34 in the Papaveraceae family, the
full absolute configurations of two isomeric roemerine-N-oxides were not solved.
Herein, absolute stereostructures of roemerine-Nα-oxide (57) and roemerine-Nβ-oxide
(58) were first determined by CD and NMR analysis. Cytotoxicity of these alkaloids
was also reported.
53
List of aporphine alkaloids isolated from N. dealbata
3a
O
1b
1
O
R3
H
N
6a
1a
R2O
5
H
R2
R1O
H
11
7a
R1
R3
H3CO
H
CF3COO
OH
H3CO
OH
R1 = OH;
R2 = R3 = H
(55)
(56)
R1 = R2 = H; R3 = CH3
R1 = H;
R2 = CH3; R3 = OH (57)
R1 = H;
R2 = OH; R3 = CH3 (58)
R
OH
CF3COO
CF3COO
9
H
N
N
R1 + R2 = CH2; R3 = H
(59)
R2 = R3 = H (60)
R1 = CH3;
R1 = R3 = CH3; R2 = H
(61)
54
R=H
(62)
R = CH3 (63)
4.2 Collection, Extraction and Isolation
The bark of N. dealbata (R.Br.) Merr. was collected at Thomas Road, 1.5km
West of Yungaburra, Australia on 28 May 1998. A voucher sample (AQ605446) has
been lodged at the Queensland Herbarium, Brisbane, Australia.
The dried ground bark of Neolitsea dealbata (10g) was extracted exhaustively
with hexane, dichloromethane and methanol, respectively. The dichloromethane and
methanol extracts were combined and filtered through polyamide gel (PAG) to remove
tannins and then the solvents were evaporated to yield a dark brown residue (1.087g).
This crude extract was purified through a C18 preparative HPLC column. Isocratic
condition of 10% methanol (0.1% TFA) – 90% (0.1% TFA) water was maintained for
the first 10 minutes at a flow rate of 9 ml/min, and then a linear gradient to 100%
methanol (0.1% TFA) was performed over 40 minutes at a flow rate of 9 ml/min.
Isocratic condition of 100% methanol (0.1% TFA) was maintained for a further 10 min
at a flow rate of 9 ml/min, 60 fractions (1 minute each) were collected. Compound 59
was obtained from fractions 30-33. Fractions 34 to 38 contained an ion peak of interest
at (+)m/z 280. The C18 preparative HPLC column with linear gradient from 30%
methanol (0.1% TFA) – 70% water (0.1% TFA) to 60% methanol (0.1% TFA) – 40%
water (0.1% TFA) was used to isolate this compound. Among 80 fractions collected,
fractions 27 to 35 contained compound 56 (4.5mg, 0.045% dry wt). The peaks observed
at (+)m/z 296 in fractions 39-41 and 47-48 were also purified yielding compound 57
(0.6 mg, 0.006% dry wt) and compound 58 (0.8 mg, 0.008% wt). Since fractions 24 to
29 had many strong peaks in the (+)-LRESIMS at (+)m/z 282, 286, 300, 314 and 328,
these fractions were combined before further purified with the linear gradient from 5%
methanol (0.1% TFA) – 95% water (0.1% TFA) to 60% methanol (0.1% TFA) – 40%
water (0.1% TFA) in 80 minutes on preparative C18 HPLC column. Four compounds
representing for four ion peaks in mass spectra were isolated compound 60 (4.7 mg,
0.047% dry wt), compound 61 (1.2 mg, 0.012% dry wt), compound 55 (1.0 mg, 0.01%
dry wt), compound 63 (0.6 mg, 0.006% dry wt) and compound 62 (1.0 mg, 0.01% dry
wt) (Scheme 4.1).
55
Neolitsea dealbata (10g)
a)
DCM/MeOH extract (1.4g)
b)
Extract after PAG (1.1g)
c)
Fractions 30-33
Fractions 24-29
Compound 59
13.6mg, 0.136%
Fractions 21-28
d)
Compound 60
(4.7mg, 0.047%)
Fractions 25-30
Compound 55
(1.0mg, 0.01%)
Fractions 54-58
Compound 61
(1.2mg, 0.012%)
Fractions 32-33
Compound 62
(1.0mg, 0.01%)
Fractions 61-63
Compound 63
(0.6mg, 0.006%)
Fractions 45-48
Fractions 34-38
(+)-m/z 280, 296
e)
Compound 56
(4.5mg, 0.045%)
Fractions 27-35
Compound 58
(0.6mg, 0.006%)
Fractions 39-41
Compound 57
(0.8mg, 0.008%)
Fractions 47-48
a) Extraction with DCM and MeOH
b) Extract after being filtered through polyamide gel (PAG)
c) Isocratic condition of 10% methanol (0.1% TFA) – 90% water (0.1% TFA) in 10 minutes, a linear
gradient to 100% methanol (0.1% TFA in 40 minutes, isocratic condition of 100% methanol (0.1% TFA)
in the last 10 minutes, a flow rate of 9.0 ml/min, 1.0 min/fraction
d) A linear gradient from 5% methanol (0.1% TFA) – 95% water (0.1% TFA) to 60% methanol (0.1%
TFA) – 40% water (0.1% TFA) in 80 minutes, a flow rate of 4.0 ml/min, 1.0 min/fraction
e) A linear gradient from 30% methanol (0.1% TFA) – 70% water (0.1% TFA) to 60% methanol (0.1%
TFA) – 40% water (0.1% TFA) in 60 minute, a flow rate of 4.0 ml/min, 1.0 min/fraction.
Scheme 4.1 Extraction and Isolation Procedure for N. dealbata
56
4.3 Structure Elucidation and Discussion
4.3.1 Normecambroline (55)
Compound 55, [α]25D -284.4o (c 0.05, C2H5OH) was purified as a quite brown
amorphous solid. The (+)-HRESIMS revealed a signal for [M+H]+ at (+)m/z 282.1135
(calcd (+)m/z 282.1125, Δ 3.5ppm), corresponding to the molecular formula C17H15NO3.
Figure 4.4 The 1H-NMR spectrum of 55 recorded at 600 MHz in DMSO-d6
Combined data from HSQC and HMBC spectra of 55 indicated the presence of
seventeen carbons including eight quaternary carbons (δC 156.4, 147.4, 142.5, 130.2,
124.3, 122.5, 121.0 and 115.4 ppm), four sp2 tertiary carbons (δC 129.0, 115.0, 113.4
and 107.4 ppm), one sp3 tertiary carbon (δC 51.9 ppm) and four secondary carbons (δC
101.0, 40.3, 31.4 and 24.7 ppm). A singlet proton H-3 at δH 6.83 ppm showed strong
HMBC correlations with other quaternary aromatic carbons C-1 (δC 142.5 ppm), C-2
(δC 147.4 ppm), C-1a (δC 115.4 ppm) and C-1b (δC 121.0 ppm) suggesting this proton
was in a penta-substituted aromatic ring. A methylene carbon (δC 101.0 ppm)
corresponding to two geminal protons at δΗ 6.21 ppm (1H, d, J=1.2 Hz) and δΗ 6.06
ppm (1H, d, J=1.2Hz) in the HSQC spectrum was assigned to a methylenedioxy group
due to their typical resonances.29 HMBC data indicated that these protons correlated to
both C-1 and C-2. This evidence facilitated the attachment of this methylenedioxy
group to the aromatic carbons C-1 and C-2 (fragment A).
A spin system of aliphatic carbons CH2-CH2-NH-CH-CH2- was deduced due to
their COSY correlations. The long range correlations observed in the HMBC spectrum
from H-4 (δH 2.90/3.07 ppm) to C-5 (δC 40.3 ppm), H-5 (δH 3.28/3.62 ppm) and H-7 (δH
2.80/3.00 ppm) to C-6a (δC 51.9 ppm) supported the locations of these carbons. Further
57
analysis of the COSY spectral data demonstrated a proton at δH 9.02 ppm also
correlated with another exchangeable proton at δH 9.47 ppm. Consequently, it was
supposed that the -NH- group in fragment B was protonated as a TFA salt. The
downfield resonances of C-5, C-6a, H-5 and H-6a also confirmed this assigment
(Figure 4.5).
With typical J coupling constants and observed COSY correlations, three
remaining aromatic protons H-8 (δH 7.18 ppm, d, J=7.8 Hz), H-9 (δH J=6.72 ppm, dd,
J=8.4 and 2.4 Hz) and H-11 (δH 7.50 ppm, d, J=2.4 Hz) were assigned in a threesubstituted benzene system. Although no HMBC correlation was observed from a
remaining proton at δH 9.53 ppm, this signal was ascribed as a hydroxyl group attaching
directly to C-10 due to the typically downfield resonance of the ipso-carbon C-10 (δC
156.4 ppm). The last moiety of 55 (fragment C) was then established (Figure 4.5).
3
11a
3a
O
H2
N
4
1b
O
1
7
6a
1a
HO
Fragment A
Fragment B
10
7a
8
HMBC
COSY
Fragment C
Figure 4.5 The partial structures A, B and C of compound 55
The assemblage of fragments A and B was deduced due to HMBC correlations
from H-4 (δH 2.90 ppm) to C-3 (δC 107.4 ppm) and C-3a (δC 124.3 ppm). A ROESY
signal between H-3 and H-4 (δH 2.90 ppm) further supported this connectivity. The
observed HMBC correlation from H-7 (δH 3.00 ppm) to C-8 (δC 129.0 ppm) as well as
the ROESY correlation between H-7 (δH 3.00 ppm) and H-8 (δH 7.18 ppm) facilitated
the connectivity between fragments B and C. The presence of HMBC correlations from
H-4 and H-7 to C-1b led to the formation of a covalent bond between C-6a and C-1b.
Two incompleted bonds at C-1a and C-11a were linked together to ensure the
consistence of its molecular weight with MS data. The complete structure of 55 was
finally elucidated (Figure 4.6).
4
3
O
1b
O
7
11a
HO
NH2
6a
1a
H
HMBC
ROESY
8
Figure 4.6 Key HMBC and ROESY correlations to establish the structure of 55
58
The absolute stereochemistry of 55 was determined by the CD experiment. The
observed maxima of negative cotton effect at 238.2 (-14.6) nm and two positive cotton
effects at 274.0 (+5.4) and 202.2 (+9.6) nm in the CD spectrum (Figure 4.7) as well as
the measured optical rotation [α ]D of -284.4o (c 0.05, C2H5OH) evidenced that this
25
compound had an R-configuration at C-6a.9 Compound 55 was identified as a new
aporphine alkaloid, (6aR)-normecambroline.
20
Molecular ellipcity
(deg.cm2 /dmol)
15
59
10
5
0
200
-5
250
300
-10
350
400
55
-15
-20
wavelength, λ (nm)
Figure 4.7 CD spectra of compound 55 and 59
Table 4.1 NMR data for TFA salt of (6aR)-normecambroline (55) in DMSO-d6
δC
δH (mult., J, int.)
1
2
3
4
142.5
147.4
107.4
24.7
5
40.3
7
31.4
8
9
10
11
3a
6a
7a
11a
1a
1b
-OCH2O-
129.0
115.0
156.4
113.4
124.3
51.9
122.5
130.2
115.4
121.0
101.0
NH
-
OH
-
6.83 (s, 1H)
3.07 (td, 12.0, 6.0, 1H)
2.90 (dd,16.8, 4.2, 1H)
3.62 (overlap H2O)
3.28 (overlap H2O)
3.00 (dd, 14.4, 4.8, 1H)
2.80 (t, 14.4, 1H)
7.18 (d, 7.8, 1H)
6.72 (dd, 8.4, 2.4)
7.50 (d, 2.4, 1H)
4.38 (br.t, 1H)
6.21 (d, 1.2,1H)
6.06 (d, 1.2,1H)
9.47 (d, 7.8, 1H)
9.02 (d, 10.2, 1H)
9.53 (br.s, 1H)
Position
gCOSY
(H no.)
5
5
4, NH
4, NH
6a
6a
9
8, 11
9
7, NH
5, 6a
-
59
ROESY
(H no.)
4
5
3, 5
4
4, 6a
6a, 8
9, 7
5, 7
-
gHMBC
(C no.)
1, 1b, 2, 4, 1a
3a, 5, 1b
1b, 3a, 3
3a
3a, 6a
8, 11a, 1b, 6a
1b, 6a, 11a
10, 11a, 7
7a, 10, 11
7a, 9, 10
1, 2
1, 2
-
4.3.2 Roemerine-Nα-oxide (57)
Compound 57 was purified as a light yellow amorphous solid. The (+)HRESIMS showed a signal for [M+H]+ at m/z 296.1296 (calcd 296.1281, Δ 5.0ppm),
corresponding to the molecular formula C18H17NO3.
Figure 4.8 The 1H-NMR spectrum of 57 recorded at 600 MHz in DMSO-d6
The 1H-NMR spectrum revealed this compound has the same aporphine skeleton
as compound 56. Detailed analysis of a series of 1D and 2D-NMR resulted in the
presence of a methylenedioxy group attached to C-1 and C-2 in ring A, N-methyl in ring
B and un-substituted ring D of the aporphine system. Basically, the structure of 57 was
similar to that of 56 (Table 4.4). Compared with 56, compound 57 possessed downfield
shifts at C-5 (ΔδH5 = 0.33 and 0.62 ppm, ΔδC5 = 11.1 ppm), C-6a (ΔδH6 = 0.64 ppm,
ΔδC6 = 9.3 ppm) and an N-methyl group (ΔδH = 0.65 ppm and ΔδC 14.1 ppm). These
differences could be due to the presence of an oxygen atom attached to the N-methyl
group, which are often found in N-oxide aporphine alkaloids. This suggestion was
supported by the 16-amu-mass difference between 56 and 57. The observation of a
singlet downfield signal at δH 12.48 ppm which showed a HMBC correlation to C-6a
indicated a hydroxy group should be connected directly to the nitrogen atom. The
presence of this hydroxy group was observed since trifluoroacetic acid (TFA) was used
as a buffer in the HPLC resulting in the protonation of the N-oxide group. This evidence
indicated compound 57 was isolated as a TFA salt of roemerine-N-oxide.
The observation of two positive cotton effects at 270 nm, and 207 nm as well as
a negative one at 231 nm in the CD spectrum (Figure 4.12) suggested 57 had an Rconfiguration.9 A significant ROESY correlation was observed between H-6a and N-
60
CH3 indicating that an anti-arrangement must exist between H-6a and N-oxide (Figure
4.9). From this evidence, the absolute stereochemistry of this compound was assigned
as (6S,6aR)-roemerine-Nα-oxide.
4
3
5
O
OH
O
1b
N
6a
7
11a
HMBC
ROESY
H
8
Figure 4.9 Key HMBC and ROESY correlations to establish the structure of 57
Table 4.2 NMR data for TFA salt of (6S,6aR)-roemerine-Nα-oxide (57) in DMSO-d6
Position
1
2
3
4
5
7
8
9
10
11
3a
6a
7a
11a
1a
1b
–OCH2O–
N-CH3
N-OH
6.88 (s, 1H)
3.25 (overlap H2O)
3.01 (dd,4.8, 18.0, 1H)
gCOSY
(H no.)
5
5
4
3, 5
3, 5
gHMBC
(C no.)
1, 2, 4, 1b
3a
3, 3a, 1b
62.5
4.06 (td, 6.0, 12.6, 1H)
4.02 (td, 4.8, 12.6, 1H)
4
4
4
4, 6a
3a, 6a, 4, N-CH3
4
27.6
3.58 (dd, 4.2, 14.4, 1H)
3.13 (t, 14.4, 1H)
6a
6a
8, 6a, N-CH3
6a, 8
6a, 1b, 11a
1b, 6a, 7a
128.6
128.1
127.8
126.3
123.5
69.8
131.9
131.4
115.6
119.3
101.1
7.44 (d, 7.8, 1H)
7.36 (td, 7.2, 1.2, 1H)
7.40 (t, 7.8, 1H)
8.02 (d, 7.8, 1H)
5.08 (dd, 3.6, 13.8, 1H)
6.24 (d, 0.6, 1H)
6.05 (d, 0.6, 1H)
9
8, 10
9, 11
10
7
-
9, 7
8, 10
9, 11
10
5, N-CH3
-
10, 7
7a, 11
8
9, 11a, 7a, 1a
1, 2
1, 2
54.7
3.72 (s, 3H)
12.45 (s, 1H)
6a
7, 5
6a, 5, 4, 1b
6a
δC a
δH (mult., J, int.)b
143.1
147.6
107.2
23.2
-
61
ROESY
(H no.)
-
4.3.3 Roemerine-Nβ-oxide (58)
Compound 58 was obtained as a light yellow amorphous solid. The (+)HRESIMS showed a signal for [M+H]+ at m/z 296.1295 (calcd 296.1281, Δ 4.7ppm),
corresponding to the molecular formula C18H17NO3.
Figure 4.10 The 1H-NMR spectrum of 58 recorded at 600 MHz in DMSO-d6
The 1H-NMR data of 58 was quite similar to those of compound 57. Compound
58 also had the same molecular weight as compound 57. However, there were small
differences in chemical shifts of the neighbouring protons and carbons of the nitrogen
atom between 58 and 57, such as the downfield resonances of C-5 (ΔδC5 = 1.6 ppm) and
H-5 (ΔδH5 = 0.05 ppm) as well as C-6 (ΔδC6 = 1.9 ppm) and H-6 (ΔδH6 = 0.01 ppm) and
the upfield resonances of N-CH3 (ΔδC = 7.8 ppm and ΔδH = 0.33 ppm). From the
evidence above, compound 58 was isomeric to compound 57. The R-configuration at C6a was assigned since its CD spectrum was quite similar to that of compounds 57
(Figure 4.12). Due to the lack of a correlation between N-CH3 and H-6a in ROESY
spectrum, roemerine-Nβ-oxide was suggested for compound 58. Therefore, the final
structure of compound 58 was (6R,6aR)-roemerine-Nβ-oxide.
3
4
O
5
O
1b
11a
6a
7
N
OH
H
HMBC
ROESY
8
Figure 4.11 Key HMBC and ROESY correlations to establish the structure of 58
62
20
Molecular ellipcity
(deg.cm2 /dmol)
15
10
(6aR)-roemerine (56)
5
0
200
-5
(6S, 6aR)-roemerine-Nα-oxide (57)
250
300
350
400
(6R, 6aR)-roemerine-Nβ-oxide (58)
-10
-15
-20
wavelength, λ (nm)
Figure 4.12 CD spectra of compound 56, 57 and 58
Table 4.3 NMR data for TFA salt of (6S,6aR)-roemerine-Nβ-oxide (58) in DMSO-d6
25.7
6.92 (s, 1H)
3.30 (overlap H2O)
3.14 (dd, 4.8, 18.6, 1H)
gCOSY
(H no.)
5
5
4
3, 5
3, 5
gHMBC
(C no.)
1, 2, 4, 1b
3a
3, 3a, 1b
5
64.1
4.11 (dd, 6.0, 11.4, 1H)
4.07 (dd, 5.4, 13.2, 1H)
4
4
4
4, 6a
3a, 6a, 4, N-CH3
4
7
27.5
3.44 (overlap H2O)
3.22 (t, 13.8, 1H)
6a
6a
8, 6a, N-OH
6a, 8
6a, 1b, 11a
1b, 6a, 7a
8
9
10
11
3a
6a
7a
11a
1a
1b
128.4
128.2
127.7
126.3
122.2
71.7
131.7
129.9
115.3
119.6
9
8, 10
9, 11
10
7
-
9, 7
8, 10
9, 11
10
–OCH2O–
101.4
7.46 (d, 7.8, 1H)
7.35 (td, 7.8, 1.8, 1H)
7.41 (t, 7.8, 1H)
8.04 (d, 7.8, 1H)
5.08 (dd, 4.2, 14.4, 1H)
6.24 (s, 1H)
6.05 (s, 1H)
10, 7
7a, 11
8
9, 11a, 7a, 1a
1, 2
1, 2
N-CH3
N-OH
46.9
δC
δH (mult., J, int.)
1
2
3
143.1
148.1
107.3
4
Position
3.39 (overlap H2O)
12.87 (s, 1H)
-
63
ROESY
(H no.)
-
5, N-OH
6a, 5
6a, 5, 4, 1b
6a
Table 4.4 NMR data for TFA salt of compounds 56, 57 and 58 in DMSO-d6
(6aR)-roemerine (56)
δC
Position
(6S, 6aR)-roemerine-Nα-oxide (57)
δH b (mult., J, int.)
δC
(6R, 6aR)-roemerine-Nβ-oxide (58)
δH b (mult., J, int.)
δC b
δH (mult., J, int.)
1
142.9
-
143.1
-
143.1
-
2
147.7
-
147.6
-
148.1
-
3
107.3
6.85 (s, 1H)
107.2
6.88 (s, 1H)
107.3
6.92 (s, 1H)
25.3
3.22 (m, 1H)
23.2
3.25 (overlap H2O)
25.7
3.30 (overlap H2O)
4
5
7
3.01 (dd, 4.8, 18.0, 1H)
2.96 (dd, 17.5, 3.5, 1H)
51.4
3.74 (dd, 12.0, 5.4, 1H)
62.5
3.40 (overlap H2O)
30.0
3.55 (dd, 13.8, 4.2, 1H)
4.06 (td, 6.0, 12.6, 1H)
3.14 (dd, 4.8, 18.6, 1H)
64.1
4.02 (td, 4.8, 12.6, 1H)
27.6
2.86 (t, 14.0, 1H)
3.58 (dd, 4.2, 14.4, 1H)
4.11 (dd, 6.0, 11.4, 1H)
4.07 (dd, 5.4, 13.2, 1H)
27.5
3.13 (t, 14.4, 1H)
3.44 (overlap H2O)
3.22 (t, 13.8, 1H)
8
128.0
7.40 (d, 7.8, 1H)
128.6
7.44 (d, 7.8, 1H)
128.4
7.46 (d, 7.8, 1H)
9
127.9
7.35 (t, 7.8, 1H)
128.1
7.36 (td, 7.2, 1.2, 1H)
128.2
7.35 (td, 7.8, 1.8, 1H)
10
128.0
7.40 (t, 7.8, 1H)
127.8
7.40 (t, 7.8, 1H)
127.7
7.41 (t, 7.8, 1H)
11
126.3
8.01 (d, 7.8, 1H)
126.3
8.02 (d, 7.8, 1H)
126.3
8.04 (d,7.8, 1H)
3a
124.4
-
123.5
-
122.2
-
6a
60.5
4.43 (br.s, 1H)
69.8
5.07 (dd, 3.6, 13.8, 1H)
71.7
5.08 (dd, 4.2, 14.4, 1H)
7a
132.5
-
131.9
-
131.7
-
11a
132.5
-
131.4
-
130.9
-
1a
115.2
-
115.6
-
115.3
-
1b
121.1
-
119.3
-
119.6
-
-OCH2O-
101.2
6.23 (br.s, 1H)
101.1
6.24 (d, 0.6, 1H)
101.4
6.24 (br.s, 1H)
N-CH3
40.6
3.07 (d, 4.2, 3H)
54.7
3.72 (s, 3H)
46.9
3.39 (overlap H2O)
NH
-
10.65( br.s, 1H)
-
-
-
-
N-OH
-
-
-
12.45 (s, 1H)
-
12.87 (s, 1H)
6.07 (br.s, 1H)
6.05 (d, 0.6, 1H)
64
6.05 (br.s, 1H)
4.4 Evaluation of drug-like properties
All of the isolated compounds (55-63) were evaluated as “drug-like” molecules
by J-Chem software. Four parameters of the Ro5 (MW, logP, HBA and HBD) and two
other parameters (PSA and NROT) were calculated (Table 4.5).
Table 4.5 Physico-chemical properties of compounds 55-63 (in a neutral form)
Compounds
Formula
MW
logP
HBA
HBD
PSA
NROT
55
56
57
58
59
60
61
62
63
C17H15NO3
C18H17NO2
C18H17NO3
C18H17NO3
C18H17NO4
C18H19NO4
C19H21NO4
C17H19NO3
C18H21NO3
281.31
279.33
295.33
295.33
311.33
313.35
327.37
285.34
299.36
2.19
3.32
2.20
2.20
1.89
1.63
2.57
2.53
3.09
4
3
4
4
5
5
5
4
4
2
0
0
0
2
3
2
3
2
50.72
21.70
45.34
45.34
59.95
70.95
62.16
61.72
52.93
0
0
0
0
1
2
2
3
3
Predicted
Bioavailability
9
9
9
9
9
9
9
9
9
Molecular weights of these compounds range from 281 Da to 327 Da matching
with the maximum molecular weight distribution of natural products reported by
Henkel35 and Feher36 groups. The results of calculated logP and HBA are also
consistent with Feher’s analysis36 in which logP of compounds from natural sources
reached the peak at 2-3 units while HBA distributed around 3-5 units. Notably, most
isolated compounds here have two more HBDs than HBDs of natural products reported
by Feher et al..36 However, all compounds comply with four properties of Lipinski’s
rule.37 As the results, all isolated compounds also satisfy two criteria of Veber’s rules
with the PSA approximately distributing from 50 to 60 Å2 and NROT ranging from 0 to
3 units. Interestingly, all these compounds not only fulfil physico-chemical properties
for drug-like molecules but also satisfy the requirements for lead-like compounds,
including MW ≤ 460, -4 ≤ logP ≤ 4.2, HBA ≤ 9, HBD ≤ 5 and NROT ≤ 10.38
4.5 Biological activity
Cytotoxic screening was carried out on a panel of human cancer cell lines,
including lung adenocarcinoma cancer cells (A549), cervical adenocarcinoma cells
(HeLa) and prostate cancer cells (LNCaP and PC3) as well as two human non-cancer
cell lines, human embryonic kidney 293 (HEK 293) and neonatal foreskin fibroblast
cells (NFF). Although cytotoxic activities of compounds 56 and 59 have been
reported,5,26,27 their activities have not been evaluated against these cell lines previously.
Table 4.6 Evaluation of cytotoxic potential of isolated compounds
65
Compound
55
56
57
58
59
60
61
62
63
Paclitaxel
Vincristine
sulphate
a
A549
28.2±5.5
3.4±0.3
>100
>100
16.0±4.9
>100
>100
>100
>100
0.0038
±0.0008
HeLa
4.0±0.5
16.6±3.5
>100
>100
28.1±3.5
>100
>100
>100
>100
0.0089
±0.0006
IC50 ± SD (μM) or % Inhibition ± SD
LNCaP
PC3
HEK
13.8±3.1
23.0±4.0
59.7±2.2
12.7±2.9
7.7±1.4
15.9±4.7
a
>100
>100
61.3±10.5%
>100
>100
49±6%
27.4±5.1
36.4±1.0
55.1±10.9%
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
0.0039±0.000
5
0.0029±0.0
0.011±
006
0.002
NFF
42.6±6.8
15.2±2.9
83.1±14.7%
66.6±9.2%
>100
>100
>100
>100
>100
0.0047±0.0010
The inhibition was calculated at the concentration of 100 μM.
The screening result showed that only three compounds 55, 56 and 59 out of
nine isolated alkaloids were active. The new compound, normecambroline (55),
demonstrated the highest potency against HeLa cells (IC50 of 4.0 μM) and was less
active against other cancer cell lines with IC50 values ranging from 13.8 to 28.2 μM.
Interestingly, it had selective cytotoxicity against HeLa cells with approximately tenfold higher potency than the two non-cancer cells HEK 293 and NFF. In comparison
with 55, compound 56 was less active against the cervical cancer cells (HeLa) with an
IC50 of 16.6 μM; but exhibited the most significant activity against lung cancer cells
(A549) with an IC50 of 3.4 μM which was 8-fold more potent compared to 55. Among
the three active compounds, compound 59 showed moderate activities on the growth of
the five cancer cell lines. This compound has been reported to have other biological
activities, including antibacterial and antifungal,39 antiparasite,30 antiplatelet.40 Notably,
when the 1,2-methylenedioxy substituent was absent from the structures of compound
60 and 61, there was no inhibition. This result again confirmed the 1,2-methylenedioxy
functional group was required for the expression of cytotoxicity of the aporphine
alkaloids as suggested by Cordell et al..10 However, compounds 57 and 58 also had no
cytotoxicity even though their structures contain the 1,2-methylenedioxy group whereas
their parent compound, roemerine (56), demonstrated general cytotoxic activities
against all cell lines tested. This phenomenon was previously found in some active
compounds, such as codeine, ethylmorphine, morphine and thebaine of which the Noxides derivatives were less toxic than the parent compounds.41 From the physicochemical properties (Table 4.6) the N-oxide analogues (compounds 57 and 58) would be
expected to have similar capacity to enter the cell as compounds 55 and 56. The Noxides may be involved in detoxification mechanisms or preventing DNA binding.
66
So far the knowledge about the mechanisms of aporphine alkaloids against
cancer cells is limited with studies pointing to cell cycle arrest, DNA damage and
inhibition of topoisomerase II as the cause of cytotoxicity.42-44 Typically it is considered
that the mechanism of action results from the intercalation of the molecule into DNA
due to the compound planar structure.5 However, here we have seen that the twisted
biphenyl skeleton of aporphines produces cytotoxicity9,45 and the lack of the 1,2methylenedioxy group or the inclusion of N-oxide into the molecule cause a loss of
activity.
4.6 References
(1)
Guinaudeau, H.; Leboeuf, M.; Cavé, A. J. Nat. Prod. 1979, 42, 325-360.
(2)
Guinaudeau, H.; Leboeuf, M.; Cavé, A. J. Nat. Prod. 1983, 46, 761-835.
(3)
Guinaudeau, H.; Leboeuf, M.; Cavé, A. J. Nat. Prod. 1988, 51, 389-474.
(4)
Guinaudeau, H.; Leboeuf, M.; Cavé, A. J. Nat. Prod. 1994, 57, 1033-1135.
(5)
Stévigny, C.; Bailly, C.; Leclercq, J. Q. Curr. Med. Chem. - Anticancer Agents
2005, 5, 173-182.
(6)
Shamma, M.; Slusarchyk, W. A. Chem. Revs. 1964, 64, 59-79.
(7)
Shamma, M.; Guinaudeau, H. Tetrahedron 1984, 40, 4795-4822.
(8)
Craig, J. C.; Roy, S. K. Tetrahedron 1965, 21, 395-399.
(9)
Ringdahl, B.; Chan, R. P. K.; Craig, J. C. J. Nat. Prod. 1981, 44, 80-85.
(10)
Likhitwitayawuid, K.; Angerhofer, C. K.; Chai, H.; Pezzuto, J. M.; Cordell, G.
A. J. Nat. Prod 1993, 56, 1468-1478.
(11)
Wu, Y. C.; Liou, Y. F.; Lu, S. T.; Chen, C. H.; Chang, J. J.; Lee, K. H. Planta
Med. 1989, 55, 163-165.
(12)
http://en.wikipedia.org/wiki/Neolitsea.
(13)
Chang, F. R.; Hsieh, T. J.; Huang, T. L.; Chen, C. Y.; Kuo, R. Y.; Chang, Y. C.;
Chiu, H. F.; Wu, Y. C. J. Nat. Prod. 2002, 65, 255-258.
(14)
Wu, S. L.; Li, W. S. Phytochemistry 1991, 30, 4160-4162.
(15)
Boruah, P.; Bhuyan, P. D.; Dutra, S. C.; Mohan, S.; Mathur, R. K. Folia
Microbiol. 1999, 44, 385-387.
(16)
Chen, K. S.; Hsieh, P. W.; Hwang, T. L.; Chang, F. R.; Wu, Y. C. Nat. Prod.
Res. 2005, 19, 283-286.
(17)
Li, W. S.; McChesney, J. D. J. Nat. Prod. 1990, 53, 1581-1584.
(18)
Johns, S. R.; Lamberton, J. A.; Sioumis, A. A. Aust. J. Chem. 1969, 22, 13111312.
67
(19)
Hui, W. H.; Luk, K.; Loo, S. N.; Arthur, H. R. J. Chem. Soc. 1971, 2826-2829.
(20)
Sharma, M. C.; Ohira, T.; Yatagai, M. Phytochemistry 1993, 33, 721-722.
(21)
Li, W. S.; Duh, C. Y. Phytochemistry 1993, 32, 1503-1507.
(22)
Joshi, B. S.; Kamat, V. N.; Govindachari, T. R. Tetrahedron 1967, 21, 261-265.
(23)
In Dictionary of Natural Products on CD-Rom; ; Chapman and Hall ed.; CRC
Press: London, 2008.
(24)
Xiujun, W., The University Of Alabama In Huntsville, 2005.
(25)
Labriola, R.; Ourisson, G. Tetrahedron 1971, 27, 1901-1908.
(26)
You, M.; Wickramaratne, D. B. M.; Silva, G. L.; Chai, H.; Chagwedera, T. E.;
Farnsworth, N. R.; Cordell, G. A.; Kinghorn, A. D.; Pezzuto, J. M. J. Nat. Prod.
1995, 58, 598-604.
(27)
Stévigny, C.; Block, S.; Pauw-Gillet, M. C. D.; Hoffmann, E. d.; G.Llabrès;
Adjakidjé, V.; Quetin-Leclercq, J. Planta Med. 2002, 68, 1042-1044.
(28)
Nakasato, T.; Nomura, S. Chem. Pharm. Bull. 1959, 7, 780-784.
(29)
Yang, J. H.; Li, L.; Wang, Y. S.; Zhao, J. F.; Zhang, H. B.; Luo, S. D. Hel.
Chim. Acta 2005, 88, 2523-2526.
(30)
Hoet, S.; Stévigny, C.; Block, S.; Opperdoes, F.; Colson, P.; Baldeyrou, B.;
Lansiaux, A.; Bailly, C.; Quetin-Leclercq, J. Planta Med. 2004, 70, 407-413.
(31)
Chen, J. J.; Chang, Y. L.; Teng, C. M.; Lin, W. Y.; Chen, Y. C.; Chen, I. S.
Planta Med. 2001, 67, 423-427.
(32)
Bohlke, M.; Guinaudeau, H.; Angerhofer, C. K.; Wongpanich, V.; Soejarto, D.
D.; Farnsworth, N. R. J. Nat. Prod. 1996, 59, 576-580.
(33)
Phillipson, J. D.; Gray, A. I. J. Nat. Prod. 1981, 44, 296-307.
(34)
Sarryar, G.; Mat, A.; Unsal, C.; Ozhatay, N. Acta Pharm. Turc. 2002, 44, 159168.
(35)
Henkel, T.; Brunne, R. M.; Muller, H.; Reichel, F. Angew. Chem. Int. Ed. 1999,
38, 643-647.
(36)
Feher, M.; Schmidt, J. M. J. Chem. Inf. Comput. Sci. 2003, 43, 218-227.
(37)
Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery
Rev. 1997, 23.
(38)
Hann, M. M.; Oprea, T. I. Curr. Opin. Chem. Biol. 2004, 8, 255-263.
(39)
Tsai, I. L.; Liou, Y. F.; Lu, S. T. Gaoxiong Yi Xue Ke Xue Za Zhi 1989, 5, 132145.
(40)
Wu, Y. C.; Chang, F. R.; Chao, Y. C.; Teng, C. M. Phytother. Res. 1998, 12,
S39-S41.
68
(41)
Phillipson, J. D. Xenobiotica 1971, 1, 419-447.
(42)
Tang, H.; Wang, X. D.; Wei, Y. B.; Huang, S. L.; Huang, Z. S.; Tan, J. H.; An,
L. K.; Wu, J. Y.; Chan, A. S. C.; Gu, L. Q. Eur. J. Med. Chem. 2008, 43, 973980.
(43)
Ma, J.; Jones, S. H.; Marshall, R.; Johnson, R. K.; Hecht, S. M. J. Nat. Prod.
2004, 67, 1162-1164.
(44)
Hsieh, T. J.; Liu, T. Z.; Chern, C. L.; Tsao, D. A.; Lu, F. J.; Syu, Y. H.; Hsieh, P.
Y.; Hu, H. S.; Chang, T. T.; Chen, C. H. Food Chem. Toxicol. 2005, 43, 11171126.
(45)
Brown, G. M.; Hall, L. H. Acta Cryst. 1977, B33, 2051-2057.
69
70
Chapter 5
Cyclodepsipeptides from the Australian sponge
Neamphius huxleyi
5.1 Introduction
The genus Neamphius, one of the six genera of the sponge family Alectonidae
(Alectona, Delectona, Dotona, Neamphius, Scolopes and Spiroxya) was first described
by Sollas in 1888.1 Up to now, only two compounds have been identified from the
marine sponge Neamphius huxleyi. Neamphine (64) was first discovered as a novel
sulphur skeleton whose structure was solved by single crystal X-ray diffraction analysis
in 1991.2 Compound 64 inhibited the growth of L1210 cells with an IC50 less than
10μg/mL. The second metabolite, neamphamide A (65), has been recently isolated and
reported to possess an anti-HIV activity with an IC50 of 28 nM.3
H 2N
O
O
O
O
H2N
NH
N
OH
NH
OH
O
S
N
O
N
O
H2N
N
O
NH
OH
O
O
H
N
OH
HN
O
HN
Neamphine (64)
NH
O
O
O
H 2N
O
OH
N
H
NH
O
H
N
O
O
N
HN
NH2
O
HN
H 2N
NH
Neamphamide A (65)
Figure 5.1 Two compounds previously isolated from the sponge Neamphius huxleyi
Structural analysis of 65 revealed the presence of eight amino acid residues,
including two units of L-asparagine (Asn), one D-arginine (Arg), one L-Nmethylglutamine (NMeGln), one L-homoproline (L-Hpr), one L-leucine (Leu) and two
D-allo-threonine (aThr) along with four unnormal residues β-methoxytyrosine
(βOMeTyr),
3,4-dimethylglutamine
(3,4-diMeGln),
4-amino-7-guanidino-2,3-
dihydroxyheptanoic acid (Agdha) and 3-hydroxy-2,4,6-trimethylheptanoic acid (Htmha)
(Figure 5.2). The structure of 65 shared these atypical residues with thirty one peptides
and depsipeptides identified from marine sponges, callipeltin A (66)4 and its
derivatives5-8 from the New Caledonian sponge Callipelta sp.4,5 and the Vanuatu sponge
Latrunculia sp.;6-8 papuamides A-D (67-70) and theopapuamide (71) from the Papua
New Guinea sponge Theonella mirabilis and T. swinhoei;9,10 mirabamides A-D (72-75)
from the Federated Stated of Micronesia sponge Siliquariaspongia mirabilis;11 and
71
homophymines A-E and A1-E1 (76-85) from the New Caledonian sponge Homophymia
sp.12,13
Table 5.1 20 natural amino acids encoded by DNA14
No.
Amino acids
1
Alanine
Structure
No.
11
O
(Ala)
Amino acids
Structure
Leucine
O
(Leu)
OH
OH
NH2
2
Arginine
(Arg)
3
NH2
NH
H 2N
N
H
4
Methionine
(Met)
O
S
14
O
HO
Phenylalanine
O
(Phe)
OH
OH
NH2
NH2
15
O
HS
Glutamic
acid
Proline
O
(Pro)
OH
OH
16
O
HO
Serine
(Ser)
OH
O
HO
O
NH2
17
O
H2 N
Threonine
OH
Glycine
(Gly)
OH
NH2
18
O
H2N
O
(Thr)
OH
NH2
8
OH
NH2
Glutamine
(Gln)
NH
O
(Glu)
7
OH
NH2
NH2
6
OH
NH2
Cystein
(Cys)
13
OH
O
5
H 2N
NH2
O
Aspartic acid
(Asp)
O
NH2
H 2N
O
Lysine
(Lys)
OH
Asparagine
(Asn)
12
O
Tryptophan
O
(Trp)
OH
OH
NH2
N
H
9
Histidine
(His)
N
HN
10
Isoleucine
19
O
Tyrosine
(Tyr)
OH
NH2
OH
NH2
HO
20
O
(Ile)
O
Valine
O
(Val)
OH
NH 2
OH
NH2
The βOMeTyr residue (Figure 5.2) was first characterised from callipeltin A
(66) in 1996.4 Compared with other amino acids detected by amino acid analysis
methods, βOMeTyr was unstable during the acid degradation.15 Its absolute
72
stereochemistry was still unidentified until all four stereoisomers of βOMeTyr were
successfully synthesized from L- and D-tyrosine in 2005.15 These synthetic
diastereomers were then transformed into the β-methoxyaspartate derivatives via
oxidative ozonolysis. The oxidized 66 was also hydrolysed. Its hydrolysates were then
derivatized with Marfey’s reagent. Absolute configurations of β-methoxytyrosine
residue in 66 was assigned as (2R,3R) which were deduced from observing Marfey
derivatives of (2R,3R)-β-methoxyaspartate. Another different synthetic approach to
solve the absolute stereochemistry of the βOMeTyr residue in papuamide A (67) was
also reported in 2005.16 By comparison in the 1H-NMR data of four synthetic βOMeTyr
diastereomers with those of the authentic natural product residue (2R,3R)-βOMeTyr in
67 was determined. Also in the same year, stereochemistry was studied for the
βOMeTyr unit in neamphamide A (2).17 This peptide was oxidized with RuCl3-NaIO4
followed by hydrolysis in HCl. The TFA derivatives of the hydrolysates were produced
and then subjected to GC/MS analysis. Four stereoisomers of dimethyl-Ntrifluoroacetyl-3-methoxyaspartate were synthesized as the standards. Interestingly, the
retention time revealed that compound 65 contained a (2S,3R)-βOMeTyr residue whose
configurations were different from those of the (2R,3R)-βOMeTyr unit in 66 and 67.
O
O
NH
H2N
HN
O
HO
β-methoxytyrosine (βOMeTyr)
NH
O
3,4-dimethylglutamine (3,4-diMeGln)
OH
H
N
H2N
NH
OH
OH
O
4-amino-7-guanidino-2,3-dihyroxy
heptanoic acid (Agdha)
Figure 5.2
O
3-hydroxy-2,4,6-trimethyl
heptantoic acid (Htmha)
Four uncommon residues identified in 65 and 66
The absolute configurations of the 3,4-diMeGln, Agdha and Htmha residues
were first investigated in 1996 when they were characterised as residues of 66.4 A pure
3,4-diMeGln amino acid was isolated from the hydrolysates of 66. A positive Cotton
effect at 217 nm indicated this glutamine-derived amino acid had an S-configuration at
C-2. NOE effects demonstrated a proton H-2 and both two methyl groups were located
on the same side of the molecule. This information facilitated the assignment of
(2S,3S,4R)-3,4-diMeGln.
73
A pure Agdha was also isolated from the hydrolysates of 66.4 Oxidative workup
and hydrolysis of Agdha with hydrochloric acid resulted in an L-arginine residue.
Modelling the structure for minimum energy as well as calculated J-coupling constants
led to the proposal of (2R,3R,4S)-Agdha. During the time from 2000 to present, there
have been four research groups, including Joullié’s,18 Lipton’s,19-21 Rao’s22,23 and
Kim’s24 groups focusing on the synthesis of this aberrant Agdha amino acid. Their
results were consistent with the previous report in 1996.
The absolute stereochemistry of the pure Htmha residue was first assigned as
(2R,3R,4S) by addressing Mosher’s method and J-coupling analysis. However, on the
progress toward the total synthesis of 66, D’Auria and co-workers25 revised its absolute
stereochemistry as (2R,3R,4R) by comparing NMR data of their synthetic stereomers
with their natural hydrolysate. By different synthetic approach, the absolute
configurations of Htmha was again confirmed as (2R,3R,4R) in 2003.26
Synthetic procedures were also developed to investigate the stereochemistry of
Agdha and Htmha moities in neamphamide A (65).17 The results demonstrated these
units had the same absolute configurations with those previously studied for 66.
List of cyclodepsipeptides bearing similar unusual residues with compound 2
OH
O
OH
H
N
O
NH
NH
HN
OH
N
O
O
H
N
OH
O
R2
OH
HN
O
NH2
O
N
H
NH
O
H
N
O
H
N
H2N
O
O
OH
O
O
N
H
HO
HO
NH
NH
HN
O
O
O
O
O
H2N
HN
O
O
N
N
HN
NH
NH2
O
O
O
O
O
O
Callipeltin A (66)
O
O
O
NH
NH
O
OR1
H
N
HN
N
H2N
NH
O
R3
OH
Papuamide A (67): R1 = R2 = H; R3 = CH3
Papuamide B (68): R1 = R2 = R3 = H
OH
O
H2N
OH
O
H
N
HO
OH
N
H
OH
O
O
O
HN
H
N
O
HO
N
H
O
HN
O
O
HN
NH
O
O
O
N
HO
H2N
O
O
NH
O
OH
H2N
O
N
R
Theopapuamide (71)
OH
74
O
O
O
H
N
O
O
H
N
O
O
NH
N
O
O
N
H
NH
O
N
O
O
NH
O
O
O
O
O
NH2
Papuamide C (69): R = CH3
Papuamide D (70): R = H
O
OH
H2N
HN
NH2
N
H
NH2
OH
O
OH
OH
O
HO
OH
H
N
H
N
N
H
N
H
HO
O
HN
O
HN
O
H2N
O
O
O
O
O
NH2
HN
NH2
HN
R2
R2
HN
HN
O
O
N
NH
O
NH
O
O
O
N
O
NH
O
O
O
NH
O
OR1
H
N
NH
Mirabamide A (72): R1 =
Me
O
HO
HO
NH
O
OR1
H
N
N
N
O
O
O
O
O
O
O
O
O
O
OH
Me
O
HO
HO
Mirabamide B (73): R1 =
; R2 = Cl
OH
OH
; R2 = Cl
OH
Mirabamide C (74): R1 = H; R2 = Cl
Mirabamide D (75): R1 =
Me
O
HO
HO
; R2 = H
OH
R2
Homophymine A (76): R1 = OH
Homophymine A1 (77): R1 = NH2
NH2
O
OH
O
OH
O
Homophymine B (78): R1 = OH
Homophymine B1 (79): R1 = NH2
OH
O
O
H
N
R1
OH
O
O
N
H
O
N
H
O
HN
O
Homophymine C (80): R1 = OH
Homophymine C1 (81): R1 = NH2
HN
OH
R2
O
Homophymine D (82): R1 = OH
Homophymine D1 (83): R1 = NH2
NH2
O
N
NH
O
N
O
H
N
H2N
OH
NH
O
NH2
O
Homophymine E (84): R1 = OH
Homophymine E1 (85): R1 = NH2
O
O
O
O
O
OH
NH2
O
Biological investigation of these peptides showed they exhibited cytotoxic,10,13
antifungal11 and antiviral3,4,9,12 activities. Nevertheless, anti-HIV property has been
considered as their typical activity with the EC50 values in the range of nanomole (50%
effective concentration), such as neamphamide A (65) (EC50 28 nM),3 callipeltin A (66)
(EC50 0.01 μg/mL),4 papuamide A (67) (EC50 3.6 ng/mL),9 homophymine A (76) (EC50
75 nM).12 Since these peptides share some structural homology, the βOMeTyr and 3,4diMeGln residues have been assumed to play a role in their anti-HIV activity.10,11 The
investigation of their structure-activity relationship has been in progress.12
This chapter describes the isolation and structure elucidation of three new
cyclodepsipeptides, named neamphamide B (86), C (87) and D (88) from the Australian
sponge Neamphius huxleyi. The absolute stereochemistry of some amino acids was
determined by a Marfey amino acid method while the relative configurations of other
75
residues were suggested by a J-based configurational analysis. Their cytotoxic activity
against a panel of human cell lines was also reported.
List of cyclodepsipeptides isolated from the Australian sponge Neamphius huxleyi
Hpr
O
∗
∗
∗
H 2N
NH
O
OH
aThr2
N
O
NH
H
N
NH
∗
O
O
H2N
βOMeTyr
R2
Asn
∗
Agdha
OH
∗
OH
O
O
H
N
∗
O
O
OH
N
H
HN
O
O
HN
H2N
NH
O
O
aThr1
O
N
HN
R1
O
O
3,4-diMeGln
HN
Arg
H2N
NH
Neamphamide B (86): R1 = NH2 ; R2 = CH3
Neamphamide C (87): R1 = OH ; R2 = CH3
Neamphamide D (88): R1 = NH2 ; R2 = C2H5
76
OH
∗
Leu
5.2 Collection, Extraction and Isolation
A specimen of Neamphius huxleyi was collected at the depth of 22m, Milln
Reef, off Cape Grafton, Queensland, Australia in 2002. It was identified as Neamphius
huxleyi
(phylum
Porifera,
class
Demospongiae,
order
Hadromerida,
family
Alectonidae). A voucher specimen QMG319323 has been deposited at the Queensland
Museum, South Brisbane, Queensland, Australia.
Plate 5.1 Photograph of the sponge Neamphius huxleyi
A freeze dried sample of Neamphius huxleyi (5g) was extracted exhaustively
with hexane (250 ml), dichloromethane (250 ml) and methanol (2 x 250 ml),
respectively. The dichloromethane and methanol extracts were combined and
evaporated solvents to yield a yellow residue (0.5 g). This crude extract was preabsorbed onto C18 (1.0 g) and packed dry into a small cartridge, which was connected to
a C18 preparative HPLC column (5 μm, 21.2 x 150 mm). A linear gradient from 100%
water (0.1% TFA) to 100% methanol (0.1% TFA) was performed over 60 minutes at a
flow rate of 9 ml/min and 60 fractions (1.0 minute each) were collected. Mass-guided
identification demonstrated fractions 26 to 29 contained the ion peaks of interest at
(+)m/z 1589, 1576, 1575, 795 and 788 in (+)-LRESIMS. 1H-NMR spectra of these
fractions revealed they were still impure. Fractions 26 to 28 were then combined and
chromatographed on a Betasil C18 column (5 μm, 150 x 21.2 mm) from 45% methanol
(0.1% FA) – 55% water (0.1% FA) to 100% methanol (0.1% FA) in 60 minutes.
Compounds 86 (12 mg, 0.24% dry wt) and 87 (4 mg, 0.08% dry wt) were obtained.
Fraction 29 was also purified on a Betasil C18 column (5 μm, 4.6 x 150 mm) from 50%
methanol (0.1% FA) – 50% water (0.1% FA) to 100% methanol (0.1% FA) in 30
minutes yielding compound 88 (1.2 mg, 0.024% dry wt) (Scheme 5.1).
77
Neamphius huxleyi (5g)
a)
DCM/MeOH extract (0.5g)
b)
Fractions 26-28
(+)-m/z 1576, 1575, 788
Fractions 29
(+)-m/z 1589, 795
d)
c)
Compound 86
(12.0mg, 0.24%)
Fractions 45-47
Compound 87
(4.0mg, 0.08%)
Fractions 50-51
Compound 88
(1.2mg, 0.024%)
Fractions 21-22
a) Extraction with DCM and MeOH
b) A linear gradient from 100% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes, a
flow rate of 9.0 ml/min, 1.0 min/fraction
c) A linear gradient from 45% methanol (0.1% FA) – 55% water (0.1% FA) to 100% methanol
(0.1% TFA) in 60 minutes, a flow rate of 9.0 ml/min, 1.0 min/fraction
d) A linear gradient from 50% methanol (0.1% FA) – 50% water (0.1% FA) to 100% methanol
(0.1% TFA) in 30 minute, a flow rate of 1.0 ml/min, 1.0 min/fraction.
Scheme 5.1 Extraction and Isolation Procedure for Neamphius huxleyi
78
5.3 Structure Elucidation and Discussion
5.3.1 Neamphamide B (86)
Neamphamide B (86), [α]25D -5.5 (c 0.08, MeOH), was purified as a colourless
amorphous powder. The (+)-HRESIMS displayed a divalent molecular ion [M+2H]2+ at
m/z 787.9463 (calcd (+) m/z 787.9487, Δ -3.0 ppm) and a molecular ion [M+H]+ at m/z
1574.8935 which was consistent for the molecular formula C71H119N19O21 (calcd (+)
m/z 1574.8901, Δ 2.2 ppm).
Figure 5.3 The 1H-NMR spectrum of compound 86 recorded at 600 MHz in MeOH-d3
Compound 86 showed poor solubility in acetone, pyridine and acetonitrile. The
1
H and
13
C NMR spectra in DMSO-d6 or combined solvent systems with DMSO-d6
(CD3CN-d3 – DMSO-d6; acetone-d6 – DMSO-d6; pyridine-d5 – DMSO-d6) exhibited
poorly broadened, doubled or tripled peaks with two or three sets of signals. By
employing MeOH-d4, the 1H and
13
C-NMR spectra were recorded as a single set of
well-resolved resonances. Therefore, MeOH-d4 was used for a series of 1D and 2DNMR experiments (1H-NMR,
13
C-NMR, COSY, ROESY, TOCSY, HSQC, HMBC,
HSQC-TOCSY and HSQMBC) for structure elucidation for compound 86. However,
exchangeable protons could not be observed in MeOH-d4. Thus, MeOH-d3 was then
utilised as an alternative NMR solvent for 1H-NMR, 13C-NMR, COSY, ROESY, HSQC
and HMBC spectra to confirm the structure. Good resolutions with sharp exchangeable
79
signals were only obtained after converting this compound from its formic acid salt to
the neutral form with ammonium hydroxide.
Combined data from the
13
C-NMR and HSQC spectra indicated that this
compound contained fourteen carbonyls between δ171.2 and δ181.0 ppm; four
quaternary carbons at δC 158.8, 158.8, 158.7 and 129.7 ppm; four aromatic carbons (δC
131.1, 131.1, 116.3 and 116.3 ppm); six methine carbons adjacent to oxygen (δC 84.3,
79.4, 75.6, 73.0, 72.5 and 67.1 ppm); ten α-methine carbons corresponding to ten αmethine protons from δH 3.92 to 5.40 ppm; one methoxy group at δC 57.1 ppm (δH 3.14
ppm); one N-methyl signal at δC 30.8 ppm (δH 3.00 ppm); ten methyl carbons (δC 24.9,
23.7, 22.9, 21.5, 20.3, 17.4, 15.7, 15.6, 14.9 and 14.8 ppm) along with another 21
aliphatic carbons.
OH
A para-substituted benzene ring was deduced due to
7
6
5
COSY correlations of two aromatic protons at δH 7.22 (d,
4
O
1
3
O
2
HN
HMBC
2H, J = 8.4 Hz) and 6.79 (d, 2H, J = 8.4 Hz). Carbons at
TOCSY
δC 129.7 and 158.7 ppm were located at C-4 and C-7,
respectively based on 3JCH HMBC correlations of protons
Fragment A
H-5 and H-6. A downfield resonance of the carbon C-7
led to the assignment of a hydroxy group attached directly to it. COSY spectral data
also revealed a correlation system –NH-CH(α)-CH(β)– from H-2 (δH 5.02 ppm) to NH
(δH 8.23 ppm) and H-3 (δH 4.47 ppm). The HMBC correlations from H-2 to C-4 and H3 to both C-5 and C-9 allowed the connectivity between this spin system and the phenyl
group. Since a methoxy group at δH 3.14 ppm (δC 57.1 ppm) showed HMBC correlation
to C-3, a β-methoxytyrosine residue (βOMeTyr) was assigned for fragment A.
H2 N
O
5
4
3
2
N
1
O
Possessing a typical chemical resonance and showing HMBC signals with
the N-methyl group (δC 30.8 ppm and δH 2.99 ppm) as well as a carbonyl
at δC 171.2 ppm, a carbon C-2 (δC 56.7 ppm, δH 4.81 ppm) was assigned
as an α-carbon in an N-methyl amino acid. The observed COSY and
Fragment B TOCSY correlations from H-3 (δ 1.53/1.62 ppm) to H-2 and H-4 (δ
H
H
1.66/1.82 ppm) along with HMBC correlations from protons H-3 and H-4 to the same
carbonyl carbon at δC 177.5 ppm supported the establishment of an N-methyl glutamine
(NMeGln) residue for fragment B.
80
5
Leucine amino acid (Leu) was also elucidated as a partial structure of
4
compound 86 (fragment C) due to the observation of the HMBC
3
2
N
H
1
correlation from H-2 (δH 4.61 ppm) to C-1 (δC 175.2 ppm) and COSY
O
correlations from H-2 to an exchangeable protons NH (δH 7.25 ppm) and
Fragment C
H-3 (δH 1.30, 1.78 ppm); from H-4 (δH 1.76 ppm) to H-3, H-5 (δH 0.91
ppm) and 4-CH3 (δH 0.96 ppm). TOCSY correlations and the signals of 3JCH long range
correlation in the HMBC spectrum were also evidenced for this assignment.
A methine carbon at δC 53.4 ppm was deduced as an αNH
H2N
H
N
6
3
5
4
carbon since its proton H-2 (δH 4.41 ppm) correlated with C-
2
NH
1
O
1 (δC 174.3 ppm) and an exchangeable proton NH (δH 7.84
ppm). COSY and TOCSY data indicated the correlations of
Fragment D
protons H-2 (δH 4.42 ppm), H-3 (δH 1.64/1.94 ppm), H-4 (δH
1.60 ppm), H-5 (δH 3.18 ppm) and 5-NH (δH 7.30 ppm) whose positions were
confirmed by HMBC correlations. The observation of HMBC correlation from H-5 to a
quaternary carbon at δC 158.8 ppm which is a typical chemical shift of guanidine
carbon,4 facilitated the establishment of an arginine residue (Arg) for fragment D.
4
A proton H-2 (δH 3.92 ppm) corresponding to an α-carbon (δC 64.0
O
1
3
O
ppm) showed 3JHH coupling with an exchangeable proton NH (δH 8.28
2
ppm) and H-3 (δH 4.33 ppm) as well as a 5JHH coupling with H-4 (δH
NH
1.29 ppm) in the TOCSY spectrum. Furthermore, HMBC analysis
Fragment E
revealed H-2 and H-3 had correlations with the same carbonyl (δC
172.2 ppm). The downfield resonances of H-3 and C-3 (δC 67.1 ppm) suggested they
were adjacent to an oxygen atom. All information afforded the establishment of a
threonine residue (Thr1) for fragment E.
Another threonine residue (Thr2) was also elucidated for fragment F
4
O
1
3
O
2
NH
Fragment F
after detailed analysis of 2D-NMR spectral data. A spin system NHCH-CH-CH3 was determined by the COSY and TOCSY spectra. The
location of C-1 was assigned at δC 173.3 ppm due to the 2JCH long
range correlation observed from H-2 (δH 5.40 ppm).
81
Combined COSY and TOCSY spectral data allowed the connection of
4
3
5
6
N
2
an alkyl chain from H-2 (δH 5.26 ppm) to H-6 (δH 2.87/3.85 ppm). The
downfield resonances of C-2 (δC 53.4 ppm) and C-6 (δC 44.8 ppm) led
1
O
to the suggestion of the linkage -N- between them. This assignment was
Fragment G
supported by the HMBC correlation from H-2 to C-6. A homoproline
residue (Hpr), uncommon amino acid, was also constructed as a partial structure of 86
(fragment G).
Detailed analysis of 2D-NMR spectral data demonstrated HMBC correlations
from NH-fragment A to C-1-fragment B; from both NCH3 and H-2 in fragment B to C1-fragment C; from NH-fragment C to C-1-fragment D; from both NH and H-2 in
fragment D to C-1-fragment E; from both NH and H-2 in fragment E to C-1-fragment
F and from H-3-fragment F (δH 5.26 ppm) to C-1-fragment G (δC 171.3 ppm). The
observed ROESY correlations from H-2-fragment A to H-6-fragment G (δH 3.85 ppm);
NCH3-fragment B to H-2-fragment C; NH-fragment D to H-2-fragment E and H-4fragment F (δH 1.21 ppm) to H-2-fragment G further supported the establishment of a
macrocyclic ring system containing fragments F-G-A-B-C-D-E (Figure 5.4).
The COSY and TOCSY spectra again revealed a spin system of -
NH
H 2N
4
5
3
NH-CH-CH(CH3)-CH(CH3)-, from proton H-2 (δH 4.24 ppm) to
2
1
O
an exchangeable proton NH (δH 9.15 ppm) and H-3 (δH 2.40
Fragment H
ppm), from H-3 to H-4 (δH 2.79 ppm) and 3-CH3 (δH 1.20 ppm),
O
from H-4 to 4-CH3 (δH 1.30 ppm). These positions were also secured by HMBC
correlations. The observation of HMBC correlations from H-2 and H-1 to a carbonyl at
δC 174.1 ppm suggested the location of C-1 while another downfield resonance at δC
181.0 ppm correlating with 5-NH2 (δH 6.98/7.66 ppm), H-3 and 4-CH3 was assigned for
an amide carbon C-5. All information resulted in the establishment of fragment H.
A long chain CH-CH-CH-CH2-CH2-CH2 was also
NH
H2N
8
H
N
6
NH
established by the COSY and TOCSY correlations. The
3
5
7
O
4
2
O
Fragment I
1
O
HMBC correlations from H-2 (δH 3.98 ppm) and H-3
(δH 3.63 ppm) to a carbonyl at δC 176.7 ppm allowed
the location of C-1. A guanidine carbon was assigned
for a quaternary carbon at δC 158.8 ppm due to its typical chemical shift and its 3JCH
long range correlation with H-7 (δH 3.19 ppm). Possessing the typical downfield
82
resonances, carbons C-2 (δC 73.0 ppm) and C-3 (δC 75.6 ppm) were proposed to be
adjacent to an hydroxy group while a carbon C-4 (δC 51.3 ppm) was suggested to bind
directly to a nitrogen atom. A COSY correlation between H-4 (δH 4.18 ppm) and NH
(δH 7.68 ppm) further supported this suggestion. Consequently, fragment I was
elucidated
as
an
abnormal
amino
acid
residue
4-amino-7-guanidino-2,3-
dihydroxyheptanoic acid (Agdha).
O
H2N
4
An asparagine residue (Asn) was established from the COSY
1
2
correlations from H-2 (δH 4.72 ppm) to H-3 (δH 2.79/2.87 ppm) and
3
O
NH
an exchangeable proton NH (δH 8.25 ppm) as well as HMBC signals
Fragment J
from H-2 to C-1 (δC 174.0 ppm), C-3 (δC 37.9 ppm) and C-4 (δC
174.9 ppm). Further analysis of COSY and HMBC spectra demonstrated that a proton at
δH 7.06 ppm correlated with another exchangeable proton at δH 7.66 ppm and carbon C3 comfirming the presence of the asparagine (fragment J).
OH
7
5
6
4
O
3
1
2
Four remaining methyl groups at δH 1.09, 0.98, 0.94 and 0.87
ppm in the 1H-NMR spectrum were assigned respectively for 2CH3, 4-CH3, 6-CH3 and H-7 (fragment K) based on their COSY
Fragment K
and TOCSY correlations. The observation of HMBC signals
from H-2 (δH 2.62 ppm) to C-3 (δC 79.4 ppm) and 2-CH3 (δC 15.7 ppm); from H-3 (δH
3.51 ppm) to C-4 (δC 34.5 ppm), C-5 (δC 40.6 ppm) and 4-CH3 (δC 17.4 ppm); from H-5
(δH 1.18 ppm) to C-6 (δC 26.4 ppm), C-7 (δC 22.9 ppm) and 6-CH3 (δC 24.9 ppm) also
confirmed these positions. The location of C-1 (δC 179.1 ppm) was deduced from its
2,3
JCH correlations with H-2 and H-3 in the HMBC spectrum. Carbon C-3 was suggested
to be bound to a hydroxy group due to its typical downfield resonance. A final subunit
of compound 86 was established as a 3-hydroxy-2,4,6-trimethylheptanoic acid residue
(Htmha).
Fragments H-I-J-K were assembled due to the observation of 3JCH long range
correlations from both NH and H-2 in fragment H to C-1-fragment I; from NH and H-4
in fragment I to C-1-fragment J and from NH and H-2 in fragment J (δH 4.72 ppm) to
C-1-fragment K (δC 179.1 ppm) (Figure 5.4).
83
G
A
K
O
N
OH
O
J
O
F
O
O
O
O
N
H
H2N
NH
O
HN
B
O
O
E
HN
NH
OH
H
O
N
O
H
N
H 2N
NH2
H
N
I
NH
D
OH
O
O
HN
NH
OH
O
C
O
HN
H2N
H2N
O
HMBC
NH
ROESY
Figure 5.4 Key HMBC and ROESY correlations to establish 2 partial structures of 86
Detailed analysis of HMBC and ROESY spectral data showed the HMBC
correlations from NH-fragment F to C-1-fragment H and the ROESY correlation
between NH-fragment F and H-2-fragment H. These supported the connectivity
between a macrocyclic ring F-G-A-B-C-D-E and a side chain H-I-J-K (Figure 5.5).
Comparing with MS data, a hydroxy groups was assigned for the position C-3-fragment
E. The final structure of 23 was established as shown in figure 5.5.
O
H2N
NH
O
OH
O
O
O
NH
OH
O
H
N
H 2N
O
H
N
OH
O
HN
O
H2N
NH
O
O
O
HN
HMBC
O
OH
N
H
NH
OH
N
O
N
HN
NH2
O
O
ROESY
HN
H2N
NH
Figure 5.5 Key HMBC and ROESY correlations to establish the structure of 86
The complete structure of neamphamide B (86) was confirmed by utilizing
electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry
(ESI-FTICR-MS). Direct FTMS2 analysis of the neamphamide B singly charged parent
ion (m/z 1574.8935) produced two singlet fragments at m/z 1057.5692 and 674.4205
(Scheme 5.2). The mass difference between the MS1 pseudomolecular ion and the MS2
fragment ion a1 (m/z 1057.5692) is 517.3243 Da which corresponds to the loss of a
neutral molecule b1 (C22H43N7O7). Likewise, the loss of an unique elemental formula
C42H64N10O12 (b2) was also deduced from the observation of the highly abundant
84
fragment ion at 674.4205 (a2). The fragments obtained by ESI-FTICR-MSn experiment
further confirmed the structural establishment of compound 86.
a2
O
H2 N
NH
O
H
N
H2 N
Elemental n+
Exp. Mass
(Exact Mass)
Δ = Error ppm
O
OH
N
OH
O
O
NH
OH
OH
NH
H
N
O
O
O
O O
OH
N
H
HN
O
O
O
HN
C71H120N19O21+
1574.8935
(1574.8901)
Δ = 2.2
NH
O
H2 N
N
HN
NH2
O
O
HN
a1
H 2N
NH2
O
N
O
OH
O
O
H2 N
O
O
NH
OH
NH
OH
MS 2
H
N
H2 N
HN
MS 2
OH
O
H2N
O
O
OH
O O
OH
HN
H2 N
NH
O
O
O
O
O
H2 N
OH
NH
OH
O
H
N
NH2
NH2
H
N
OH
a2: C29H56N9O9+
674.4205
(674.4195)
Δ = 1.5
O
HN
NH2
O
NH
N
HN
O
H2 N
NH2
b2: C42H64N10O12
900.4730
(900.4705)
Δ = 2.8
NH
O
H 2N
HN
O
O
HN
O
N
HN
N
N
H
O
HN
b1: C22H43N7O7
517.3243
(517.3224)
Δ = 3.7
O
NH
O
O
NH2
NH
O O
OH
a1: C49H77N12O14+
1057.5692
(1057.5677)
Δ = 1.4
NH2
O
H2 N
O
Scheme 5.2 FTMS2 fragmentations and related neutral losses of 86
85
Figure 5.6 FTMS2 spectrum of 86
Stereochemical determination of compound 86
The Marfey method which was first reported in 198427 has been proven to be a
reliable, accurate and highly sensitive technique for amino acid and peptide analysis.28
This method uses 1-fluoro-2,4-dinitrophenyl-5-L-alanine amide (FDAA or Marfey’s
reagent), to react with the amine group on an amino acid, peptide or target molecule by
nucleophilic substitution of the aromatic fluorine to produce a diastereomer (Figure
5.7). Due to the formation of an intramolecular hydrogen bond between the carboxy and
carboxyamide group in the L-L diastereomer which is considered to be stronger than that
in the L-D diastereomer, these isomeric compounds can be separated using reversedphase HPLC.27
NO2
O
NO2
H
N
O
NH2
O 2N
O
NO2
H
N
NH2
H2N
NH2
OH
R
O2 N
F
O
O 2N
HN
L,D amino acid
OH
R
L,L-FDAA derivative
O
HN
OH
FDAA
O
H
N
R
L,D-FDAA derivative
Figure 5.7 Mechanism of Marfey’s reagent (FDAA)
The Marfey analysis was applied to determine absolute stereochemistry of
amino acid residues in 86. The acid hydrolysate of 86 (6N HCl, 120oC, 16h) and
authentic amino acids were derivatized with Marfey’s reagent – L-FDAA before being
carried out by an LC/MS analysis.3,27 Chromatographic comparison of the acid
hydrolysate with appropriate amino acid standards revealed the presence of D-Arg, LAsn, L-NMeGln, L-Hpr, L-Leu and two D-allo-Thr (D-aThr) in 86.
86
Table 5.2 Retention times of authentic FDAA-amino acids and the hydrolysates of 86
Standards
86
L-
D-
L-
D-
L-
D-
L-
D-
L-
D-
L-
D-
L-
D-
Arg
Arg
Asn
Asn
Hpr
Hpr
Leu
Leu
NMeGln
NMeGln
Thr
Thr
aThr
aThr
16.32
20.71
21.08
27.03
26.35
28.95
30.09
25.16
25.89
21.68
23.27
21.24
16.35
20.74
16.18
27.08
28.93
25.19
The ion peak of the β-OMeTyr amino acid was not observed either in the acid
hydrolysate solution or its Marfey derivatives. This problem was previously met upon
the hydrolysis of those peptides containing β-OMeTyr unit under a strong acidic
condition.3,4,9,11 In order to avoid the decomposition of this amino acid, Oku17 and
Zampella15 suggested a method to transform β-OMeTyr to β-OMeAsp prior to
employing an amino acid analysis. However, with the limited materials in hand, this
approach was unable to be explored. Instead, J-based NMR configurational analysis
was applied in the assignment of its stereochemistry (Figure 5.8).29-31 3JH-H coupling
constants were obtained from the 1H-NMR spectrum while accurate measurements of
2,3
JH-C were extracted from 2D-NMR spectra (HSQC-TOCSY and HSQMBC). A
ROESY experiment was also carried out in order to assist the establishment of final
configurations. The large coupling value of 9.6Hz between β-OMeTyr-H2 and βOMeTyr-H3 pointed to their anti orientation relationship. A ROESY experiment was
then applied to determine the relative threo or erythro stereochemistry in this case. The
observation of a ROESY correlation from β-OMeTyr-NH at δ 8.23ppm to β-OMeTyrH5 at δ 7.22ppm and no ROESY signal observed between β-OMeTyr-NH and βOMeTyr-OMe supported the interpretation of the erythro relative configuration for this
residue. Here, the configurational assignment for the β-OMeTyr unit in 86 was different
from that in neamphamide A (65) although they were from the same taxonomy. It is
noted that the absolute stereocenters of the β-OMeTyr residue in neamphamide A and
papuamide B (68) were solved previously based on the GC/MS retention time of a
modified natural product residue with those of four synthetic stereoisomers as
standards.17 The β-OMeTyr-H3 in 65 which was determined as (R)- β-OMe-L-Tyr
showed a broad singlet signal at δΗ3 5.03ppm while the equivalent proton in 68 assigned
for (R)- β-OMe-L-Tyr showed a doublet at δΗ3 4.24ppm with a vicinal coupling constant
of 9.3 Hz. So far, only two out of four stereoisomers of the β-OMeTyr amino acid, (R)β-OMe-L-Tyr and (R)-β-OMe-D-Tyr, have been discovered in natural products.32 In
comparison with published NMR data of the known absolute configurational β-OMeTyr
residue in other peptides (Table 5.3), it was seen that the coupling patterns and
87
22.54
22.58
resonances of the β-OMeTyr in 86 (δΗ3 4.47ppm, J = 9.6Hz) were similar to those
reported for (R)-β-OMe-D-Tyr. Hence, it is highly possible that the absolute
configuration of the β-OMeTyr residue in 86 is (R)-β-OMe-D-Tyr or (2S,3R)-3methoxytyrosine.
Table 5.3 NMR comparison of the β-OMeTyr residue
3
OMe
O
Natural products
(S)
(R)
NH
OH
(R)-β-OMe-L-Tyr
OMe
O
1
(R)
(R)
4
Neamphamide A (65)3
Callipeltin A (66)4
Papuammide B (68)9
Mirabamide A (72)11
Neamphamide B (86)
JH2-H3
(Hz)
-a
9.5b
9.3b
9.7c
9.6b
δH2
(ppm)
4.67
5.01
5.18
5.31
5.02
δC2
(ppm)
60.3
53.5
53.4
52.1
52.9
δH3
(ppm)
5.03
4.55
4.24
4.29
4.49
δC3
(ppm)
81.9
84.0
84.7
84.6
84.0
Absolute
configuration
(2S,3R)17
(2R,3R)15
(2R,3R)17
(2R,3R)11
(2R,3R)
a
NH
OH
(R)-β-OMe-D-Tyr
Broad singlet
Recored in MeOH-d4 or MeOH-d3
c
Recored in CD3CN-H2O (5:1)
b
The J-based NMR configurational analysis was also used to determine the
relative stereochemistry of Agdha and Htmha (Figure 5.8). A large 3JH2-H3 of 9.6 Hz in
the Agdha residue indicated an anti orientation between these protons. The observed
ROE effect between Agdha-2OH and Agdha-H4 supported the C2-C3 erythro
configuration. A gauche relationship between Agdha-H3 and Agdha-H4 was deduced
due to their small coupling constant 3JH3-H4 of 1.2Hz. Moreover, these two protons
showed small 2JH-C values (2JH3-C4 = -1.8 Hz; 2JH4-C3 = -2.4 Hz) indicating the anti
orientations of two pairs, Agdha-H3 versus Agdha-NH and Agdha-H4 versus Agdha3OH. The small vicinal coupling constant 3JH4-C2 of 1.8Hz further confirmed the
assignment of the threo configuration at C3-C4 in the Agdha moiety. Herein, the
determinations of the anti erythro arrangement for C2-C3 and the gauche- threo
arrangement for C3-C4 using J-based NMR analysis were consistent with those
absolute stereocenters in Agdha residue of neamphamide A (65)17 and callipeltin A
(66)4 previously studied. They also closely matched values calculated by a quantum
computational method.33 Hence, the configurations of the Agdha residue in 86 were
established as (2R*,3R*,4S*)-Agdha.
The medium magnitude of 3JH2-H3 = 7.2Hz in the Htmha residue accounted for a
pair of gauche/anti equilibrating rotamers at two stereocenters C2 and C3 (Figure 5.8).
A large value of 2JH2-C3 (-6.0Hz) and a small value of 3JH3-C1 (3.0Hz) suggested two
concurrent gauche arrangements of Htmha-H2 versus Htmha-3OH and Htmha-H3
versus Htmha-C1. Moreover, the intermediate magnitude of the heteronuclear coupling
constants (2JH3-C2 = 3.0Hz and 3JH3-C8 = 3.6Hz) also indicated the anti/gauche- pair of
erythro series which was the most consistent with the coupling constant patterns among
88
those of four equilibrating rotamers.31 The observation of a ROE correlation between
Htmha-H8 and Htmha-H4 added further strength to this interpretation. A gauche
rotamer between Htmha-H3 and Htmha-H4 was deduced due to an obtained small
vicinal coupling of 3JH2-H3 = 3.6Hz. Other values of heteronuclear coupling constants
(Figure 5.8) were in agreement with a gauche+ rotamer in the erythro relative
stereochemistry. The configurations of 86 were, therefore, assigned as (2R*,3R*,4R*)Htmha.
O
1
OMe
(R*)
O
OH
(R*)
4
(R*)
1
NH
(S*)
OH
OH
C1
H2
JH2-H3 = 9.6Hz (large)
2
JH2-C3 = -5.4Hz (large)
3
JH2-C4 = 3.0Hz (small)
3
JH3-C1 = 3.0Hz (small)
NH H
2
OH
erythro (C2-C3)
OH
HO
3
C1
N
H
H2
H3
C1
C4
C5
OH
H3 OH
HO
H3
3
JH2-H3 = 9.6Hz (large)
JH2-C3 = -6.0Hz (large)
2
JH3-C2 = -5.4Hz (large)
3
JH2-C4 = 3.0Hz (small)
3
JH3-C1 = 3.0Hz (small)
2
C1
C4
H2
erythro (C2-C3)
gauche-
H2 HN
anti
OH
HO
C1
N
H2
H OCH3
H3
H C2
H4
N
OCH3
H3
C1
anti
anti
OH
NH2
Agdha
OCH3
H3
H3
N
H
HN
β-OMeTyr
H3CO
NH
5
(R*)
H4 NH
3
JH3-H4 = 1.2Hz (small)
JH3-C4 = -1.8Hz (small)
2
JH4-C3 = -2.4Hz (small)
3
JH3-C5 = 3.0Hz (small)
3
JH4-C2 = 1.8Hz (small)
2
C2
C5
H3 OH
threo (C3-C4)
threo
O
OH
5
1
(R*)
(R*)
(R*)
8
9
Htmha
H3 OH
C1
C1
HO
H2
C4
C4
3
H2
H3
erythro (C2-C3)
C8
H3
H2
C1
OH
JH2-H3 = 7.2Hz (medium)
JH2-C3 = -6.0Hz (large)
2
JH3-C2 = -3.0Hz (medium)
3
JH3-C1 = 3.0Hz (small)
3
JH3-C8 = 3.6Hz (medium)
2
C4
C8
gauche-
anti
H4
C9
C2
HO
H4
C2
H3
gauche+
3
JH3-H4 = 3.6Hz (small)
JH3-C4 = -0.6Hz (small)
2
JH4-C3 = -6.0Hz (large)
3
JH3-C5 = 3.0Hz (small)
2
C5
C5
H3 OH
erythro (C3-C4)
ROESY correlation
Figure 5.8 Relative configurations of the β-OMeTyr, Agdha and Htmha residues
Table 5.4 NMR data for neamphamide B (86)
Position
1
2
3
4
5, 9
6, 8
7
3-OCH3
NH
OH
δCa
171.4 c
53.4
84.3
129.7
131.1
116.3
158.7
57.1
gTOCSYb
gCOSYa
(H no.)
(H no.)
βOMeTyr
δH (mult., J, int.) a
ROESYa
(H no.)
gHMBCa
(C no.)
5.02 (t, 9.6, 1H)
4.47 (d, 9.6, 1H)
3, NH
2
3
2
5, 9, Hpr-6
5, 9, 3-OCH3
1, 3, 4
1, 2, 5, 9, 3-OCH3
7.22 (d, 8.4, 2H)
6.79 (d, 8.4, 2H)
6, 8
5, 9
6, 8
5, 9
2, 3, NH, 3-OCH3
3, 7
4, 7
3.13 (s, 3H)
8.23 (d, 9.6, 1H)
9.27 (br.s)
2
3, 5, 9
5, 9
6, 8
3
NMeGln-1
89
NMeGln
1
2
3
171.2c
56.7
25.6
4
32.5
5
NCH3
NH2
177.5
30.8
1
2
3
175.2
51.2
40.0
4
5
4-CH3
NH
26.3
21.5
23.7
4.81 (overlap H2O)
1.53 (m, 1H)
1.62 (m, 1H)
1.66 (m, 1H)
1.82 (m, 1H)
2.99 (s, 3H)
Ha 6.60 (br.s)
Hb 7.06 (br.s)
3
2
2
3, 4
2, 4
1, 3, Leu-1, NCH3
1, 2, 5
2, 3
2, 3, 5
2, 3, 5
2, Leu-2, Leu-3, Leu-5
2, Leu-1
4
5, NMeGln-NCH3
NMeGln-NCH3
1, 3, 4
Hb
Ha
Leu
4.61 (m, 1H)
1.30 (m, 1H)
1.78 (m, 1H)
1.76 (m, 1H)
0.91 (d, 6.6, 3H)
0.96 (d, 6.6, 3H)
7.25 (d, 7.2, 1H)
3, NH
2
2
5, 4-CH3
4
4
2
3, 4, 5, 4-CH3
2, 3, 4
2, 3, 4
NH
2, NMeGln-NCH3
4
3, 4-CH3
3, 4
Arg-1
Arg
1
2
3
174.3 d
53.4
28.5
4
5
6
NH
5-NH
26.3
42.2
158.8
1
2
3
4
NH
3-OH
172.2
64.0
67.1
20.3
1
2
3
4
NH
173.3
56.9
72.5
14.9
1
2
3
171.3 c
53.4
26.7
4
21.5
5
6
26.6
44.8
4.41 (td, 5.4, 8.4, 1H)
1.64 (m, 1H)
1.94 (m, 1H)
1.62 (m, 2H)
3.17 (m, 2H)
3, NH
2
2
5
4, 5-NH
7.84 (d, 8.4, 1H)
7.30 (t, 4.8, 1H)
2
5
3, 4, 5
1, 3, 4, aThr1-1
1
1, 2, 5
2, 5
4, 6
aThr1-2
aThr1-1
4, Arg-NH
1, 3, 4, aThr2-1
1
2, 3
2, 3, aThr2-1
4
aThr1
3.92 (t, 3.6, 1H)
4.33 (m, 1H)
1.29 (d, 6.6, 3H)
8.28 (d, 3.6, 1H)
5.81 (d, 4.7, 1H)
3, NH
2, 4
3
2
3
3, 4
2, 4
2, 3
2
2, aThr2-NH
aThr2
5.40 (dd, 1.8, 9.6, 1H)
5.61 (td, 2.4, 6.0, 1H)
1.21 (d, 6.6, 3H)
8.87 (d, 9.6, 1H)
3, NH
2, 4
3
2
3, 4
2, 4
2, 3
Hpr-2
3,4-diMeGln-2,
aThr2-NH
1, 3, 4
4, Hpr-1
2, 3
2, 3,4-diMeGln-1
Hpr
1
2
174.2
59.1
3
4
5
3-CH3
4-CH3
NH
5-NH2
37.7
43.5
181.0
15.6
14.8
Agdha
1
2
3
4
5
6
7
8
2-OH
3-OH
NH
176.7
73.0
75.6
51.3
30.4
26.6
42.4
158.8
5.26 (d, 4.2, 1H)
1.60 (m, 1H)
2.15 (d, 13.8, 1H)
1.48 (m, 1H)
1.69 (m, 1H)
1.66 (m, 2H)
2.87 (m, 1H)
3.85 (d, 12.6, 1H)
3
3, 4
2, 4
3
2, 4, 5, 6
6
5
5
4.24 (dd, 4.8, 10.2, 1H)
3, NH
3, 3-CH3
4-CH3, aThr2-NH
2.40 (m, 1H)
2.79 (m, 1H)
2, 3-CH3
3, 4-CH3
2, 4, 3-CH3
3, 4-CH3
5-NH2
1.20 (d, 6.6, 3H)
1.30 (d, 6.6, 3H)
9.15 (d, 4.8, 1H)
Ha 6.98 (br.s)
Hb 7.66 (br.s)
3
4
2
Hb
Ha
2, 3
4
3.98 (dd, 5.4, 9.6, 1H)
3.63 (dd, 6.6, 9.6, 1H)
4.18 (dd, 4.2, 7.8, 1H)
1.59 (m, 1H)
1.72 (m, 1H)
1.65 (m, 2H)
3.19 (m, 2H)
3, 2-OH
2, 4, 3-OH
3, 5, NH
4
4
7
6, 7-NH
3
2
5, 6, 7
4
4
4
7
5.90 (d, 4.8, 1H)
5.65 (d, 6.0, 1H)
7.68 (d, 4.2, 1H)
2
3
2
3, 4, 5
3,4-diMeGln
aThr2-4
1, 3, 4, 6
βOMeTyr-2
d
Agdha-2
4
2, 3, 4
3, 4, 5
2, 3, Agdha-1
5
5
3,4-diMeGln-NH
5
7, 2-OH
3
1, 3, 4
2
4
6, 8
4
1, 3
2, 4
4, Asn-1
2, Asn-NH
90
1, 3, 4, 3-CH3,
Agdha-1
1, 2, 5, 3-CH3
2, 3, 5
7-NH
7.36 (t, 4.8, 1H)
7
Asn
1
2
3
174.0 d
52.2
37.9
4
NH
4-NH2
174.9 d
1
2
3
4
5
6
7
2-CH3
4-CH3
6-CH3
3-OH
179.1
45.0
79.4
34.5
40.6
26.4
22.9
15.7
17.4
24.9
4.72 (m, 1H)
2.79 (m, 1H)
2.87 (m, 1H)
3, NH
2
2
8.25 (d, 7.8, 1H)
Ha 7.06 (br.s)
Hb 7.64 (br.s)
2
Hb
Ha
3
2
2
1, 3, 4, Htmha-1
2, 4
2, 4
Htmha-2, Agdha-NH
2, 3, Htmha-1
3
4
4-CH3, Asn-NH
2-CH3, 4-CH3
2-CH3
2, 4-CH3, 6-CH3
1, 3, 2-CH3
1, 2, 5, 4-CH3
4-CH3, 6-CH3
3, 4
2, 3, 5, 7
4, 5, 7
5, 6, 6-CH3
1, 2, 3
3, 4, 5, 2-CH3
5, 6, 7
2
Htmha
2.62 (m, 1H)
3.51 (m, 1H)
1.76 (m, 1H)
1.18 (m, 2H)
1.65 (m, 1H)
0.87 (d, 6.6, 3H)
1.09 (d, 7.2, 3H)
0.98 (d, 7.2, 3H)
0.94 (d, 6.6, 3H)
4.65 (d, 6.0, 1H)
3, 2-CH3
2, 4, 3-OH
4-CH3
4, 6
3, 2-CH3
2, 4, 2-CH3
4-CH3
6, 6-CH3
6
2
4
6
3
2, 3
3, 4, 6, 4-CH3
a
Recorded in CD3OD
Recorded in CD3OH
c, d
These assignments are interchangeable
b
Htmha
Hpr
O
∗
Asn
H2N
NH
∗
O
OH
aThr2
N
O
H
N
NH
∗
O
O
H2N
βOMeTyr
∗
NH
∗
Agdha
OH
∗
OH
O
O
H
N
∗
O
N
H
HN
O
O
aThr1
H2N
NH
O
NMeGln
O
HN
O
N
HN
NH2
O
O
3,4-diMeGln
HN
Neamphamide B (86)
O
OH
H2N
NH
91
Arg
OH
∗
Leu
5.3.2 Neamphamide C (87)
Neamphamide C (87), [α]24D -8.1 (c 0.08, MeOH), was isolated as a colourless
amorphous solid. The (+)-HRESIMS showed a divalent molecular ion [M+2H]2+ at m/z
788.4396 (calcd (+) m/z 788.4407, Δ -1.4 ppm) and a molecular ion peak at m/z
1575.8726 ([M+H]+) corresponding to the molecular formula C71H118N18O22 (calcd
1575.8741, Δ -0.9 ppm).
Figure 5.9 The 1H-NMR spectrum of compound 87 recorded at 600 MHz in MeOH-d4
The (+)-HRESIMS data indicated that neamphamide C differs from
neamphamide B for the presence of an OH group in the place of a NH2 group.
Consequently, one of the three amino acid residues containing a carboxyamide group in
neamphamide B (86) (NMeGln, 3,4-DiMeGln and Asn) should be replaced by a
carboxyl group in neamphamide C (87). The 1D and 2D-NMR data measured in
CD3OD of both two compounds were similar. Attempts to interpret the NMR spectral
data of 87 in CD3OH were unsuccessful since the data did not provide well-resolved
exchangeable signals. The Marfey analysis of 87 demonstrated the presence of D-Arg,
L-Asn, L-Hpr, L-Leu, two D-aThr and one L-N-methylglutamic acid (L-NMeGlu)
residue while the trace of an L-NMeGln was invisible. This indicated compound 87
differs from 86 at the L-NMeGlu residue. Detailed NMR analysis of 87 revealed some
key HMBC correlations from NMeGlu-NCH3 to Leu-C1, Arg-H2 to aThr1-C1, aThr1-
92
H2 to aThr2-C1, 3,4-diMeGln-H2 to Agdha-C1, aThr2-H3 to Hpr-C1 and Asn-H2 to
Htmha-C1; and a key ROESY correlation between Hpr-H6 and βOMeTyr-H2
suggesting amino acids in 87 were aligned in the same sequence with those in 86. In
order to confirm this result, structural elucidation of 87 by multiple stages of the ESIFTICR-MSn were employed.
Figure 5.10 FTMS2 spectrum of 87 (isolated m/z 1575.8726)
The sequential fragmentations observed for neamphamide C (87) are described
in Schemes 5.3 and 5.4. The singly charged parent ion of 87 (m/z 1575.8726) was
fragmented into two significant ions at m/z 1058.5536 and 674.4149 which was similar
to the MS2 fragment pattern of 86. Compared with the fragment a1 in compound 86, a
fragment a3 here was excess one amu while elemental formulae assigned for a4 and b4
were similar to those in 86. This indicated the OH group replaced the NH2 group in the
residue NMeGln (Scheme 5.3). A FTMS3 study for the m/z 1058.5536 ion was carried
out. A significant singly charged ion observed at m/z 480.2566 demonstrated the loss of
b5 moiety which contains the NMeGlu residue (Scheme 5.3).
Similar evidence for the presence of the NMeGln residue was obtained when a
doubly charged parent ion (m/z 788.4396) was isolated for the MS2 stage of
fragmentation (Scheme 5.4). Two other important fragments, one singly charged ion a6
(m/z 501.3037) and one doubly charged ion a7 (m/z 499.2925), were observed in the
FTMS2 spectrum. Detailed analysis of MS data indicated that these fragments were
formed as a result of the elimination of two neutral moieties b6 and b7 from the parent
93
molecule. All information from ESI-FTICR-MS definitely confirmed the presence of
the NMeGlu residue in compound 87.
The β-OMeTyr, agdha and htmha stereocenters were assumed to be identical to
the corresponding chiral centers in 86 on the basis of similar chemical shifts and ROE
patterns. Hence, the complete structure of 87 was finally established.
Scheme 5.3 FTMS2,3 fragmentations and related neutral losses of 87
94
O
H2N
NH
N
OH
O
O
O
H
N
H2N
O
a7
OH
NH
OH
OH
NH2
C71H120N18O222+
788.4396 ==>1576.8792
(788.4407 = =>1576.8814)
Δ = -1.4
O
H
N
O
N
H
O
O O
OH
HN
O
O
O
HN
a6
NH
O
H2N
N
HN
OH
O
O
HN
H2N
NH2
O
N
OH
HO
O O
MS 2
O
N
O
OH
O
O
O O
OH
N
H
HN
HN
H2N
b7: C28H42N4O9
578.2942
(578.2952)
Δ = -1.7
O
N
HN
OH
O
O
O
HN
H2N
H2N
NH
OH
O
O
O
H2N
MS 2
NH
O
N
O
H2N
NH
O
O
b6: C49H78N12O15
1074.5682
(1074.5710)
Δ = -2.6
H2N
H
N
O
NH
OH
NH
OH
OH
NH2
H
N
O
O
OH
N
H
O
HN
O
HN
a7: C43H78N14O132+
499.2925 = => 998.5850
(499.2931 = => 998.5862)
Δ = -1.2
H2N
O
O
HN
H2N
NH
O
H2N
O
H
N
H2N
O
NH
OH
NH
OH
OH
NH
O
a6: C22H41N6O7+
501.3037
(501.3031)
Δ = 1.2
Scheme 5.4 FTMS2 fragmentations and related neutral losses of 87 (isolated m/z
788.4396)
Table 5.5 NMR data for neamphamide C (87) in CD3OD
Position
1
2
3
4
5, 9
6, 8
7
3-OCH3
δC
171.5 a
53.4
84.4
129.7
131.0
116.3
158.7
57.2
δH (mult., J, int.)
gCOSY
(H no.)
βOMeTyr
ROESY
(H no.)
gHMBC
(C no.)
5.02 (t, 9.6, 1H)
4.48 (d, 9.0, 1H)
3
2
5, 9, Hpr-6
5, 9, 3-OCH3
1, 3, 4
1, 2, 5, 9, 3-OCH3
7.23 (d, 8.4, 2H)
6.79 (d, 8.4, 2H)
6, 8
5, 9
2, 3, 3-OCH3
3, 7
4, 7
3, 5, 9
3
3.14 (s, 3H)
NMeGlu
1
2
3
171.2 a
56.9
25.6
4.81 (overlap H2O)
1.54 (m, 1H)
3
2
1, 3, NCH3, Leu-1
1, 2, 5
95
4
32.5
5
NCH3
177.4
30.8
1.62 (m, 1H)
1.65 (m, 1H)
1.82 (m, 1H)
2
2, 3, 5
2, 3, 5
3.00 (s, 3H)
2, Leu-2, Leu-3, Leu-5
2, Leu-1
NMeGlu-NCH3
NMeGlu-NCH3
1, 3, 4
2, NMeGlu-NCH3
3, 4-CH3
3, 4
Leu
1
2
3
175.2
51.2
40.0
4
5
4-CH3
26.3
21.5
23.7
1
2
3
174.4 b
53.5
28.5
4
5
6
26.3
42.2
158.8
1
2
3
4
172.2
63.9
67.2
20.4
4.61 (dd, 3.0, 10.8, 1H)
1.30 (m, 1H)
1.78 (m, 1H)
1.76 (m, 1H)
0.92 (d, 6.6, 3H)
0.96 (d, 6.6, 3H)
3
2
2
5, 4-CH3
4
4
Arg
4.42 (dd, 4.8, 8.4, 1H)
1.64 (m, 1H)
1.94 (m, 1H)
1.60 (m, 2H)
3.18 (m, 2H)
3
2
2
5
4
1, 3, aThr1-1
1
1, 2, 4, 5
4, 6
aThr1
3.92 (d, 3.0, 1H)
4.33 (dd, 3.6, 7.2, 1H)
1.30 (d, 6.6, 3H)
3
2, 4
3
4
2
1, 3, 4, aThr2-1
1
2, 3
aThr2
1
2
3
4
173.3
56.9
72.5
14.9
5.40 (d, 1.8, 1H)
5.61 (td, 2.4, 6.0, 1H)
1.21 (d, 6.6, 3H)
3
2, 4
3
4
2
1, 3, 4
4, Hpr-1
2, 3
aThr2-4
1, 3, 4, 6
Hpr
1
2
3
171.4 a
53.6
26.6
4
21.5
5
6
26.6
44.9
1
2
3
4
5
3-CH3
4-CH3
Agdha
1
2
3
4
5
174.3 b
59.2
38.4
43.2
180.7
15.1
14.9
176.5
72.9
75.7
51.4
30.1
6
7
8
26.5
42.4
158.8
1
2
3
173.9 b
52.1
37.6
4
174.9 b
1
2
3
4
5
6
7
2-CH3
4-CH3
179.1
45.4
79.9
34.3
40.5
26.4
21.8
14.9
17.6
5.27 (d, 4.2, 1H)
1.60 (m, 1H)
2.15 (d, 13.8, 1H)
3
1.50 (m, 1H)
1.69 (m, 1H)
1.66 (m, 2H)
2.86 (m, 1H)
3.85 (d, 12.6, 1H)
3
2, 4
6
5
5
βOMeTyr-2
3,4-diMeGln
4, 3-CH3
5
4.25 (d, 9.6, 1H)
2.40 (m, 1H)
2.81 (m, 1H)
3
2, 3-CH3
3, 4-CH3
1.22 (d, 6.6, 3H)
1.29 (d, 6.6, 3H)
3
4
2
2, 3, 4
3, 4, 5
3.99 (d, 9.6, 1H)
3.63 (dd, 1.2, 9.6, 1H)
4.19 (dd, 4.2, 7.8, 1H)
1.60 (m, 1H)
1.72 (m, 1H)
1.65 (m, 2H)
3.20 (m, 2H)
3
2, 4
3, 5
4
1, 3, 4
1, 2
5, 6
2
2
1, 3, 4, 3-CH3, Agdha-1
1, 2, 4, 5, 3-CH3
2, 3, 5
4
4
7
6
5, 6, 8
Asn
4.73 (m, 1H)
2.79 (m, 1H)
2.87 (m, 1H)
3
2
2
1, 3, 4, Htmha-1
2, 4
2, 4
Htmha
a, b
2.66 (p, 7.2, 1H)
3.51 (dd, 3.6, 8.4, 1H)
1.78 (m, 1H)
1.20 (m, 2H)
1.65 (m, 1H)
0.88 (d, 6.6, 3H)
1.10 (d, 7.2, 3H)
0.99 (d, 7.2, 3H)
3, 2-CH3
2, 4
4-CH3
4, 6
4-CH3
2-CH3, 4-CH3
2-CH3, 6-CH3
1, 3, 2-CH3
1, 2, 4, 5, 4-CH3
3, 4-CH3
3, 4, 6, 4-CH3
6
2
4
2-CH3, 4-CH3, 6-CH3
3, 4, 7
2, 3, 7
5, 6, 6-CH3
1, 2, 3
3, 4
These assignments are interchangeable
96
Htmha
Hpr
O
∗
Asn
H2N
NH
∗
O
OH
aThr2
N
O
H
N
NH
∗
O
O
H2N
βOMeTyr
∗
NH
∗
Agdha
OH
∗
OH
O
O
H
N
∗
O
N
H
HN
O
NH
O
NMeGlu
O
O
aThr1
HN
H2N
O
N
HN
OH
O
O
3,4-diMeGln
HN
Neamphamide C (87)
O
OH
H2N
NH
97
Arg
OH
∗
Leu
5.3.3 Neamphamide D (88)
Neamphamide D (88), [α]25D -18.9 (c 0.08, MeOH), was obtained as a
colourless amorphous powder. The (+)-HRESIMS displayed a signal at m/z 1588.9052
corresponding to the molecular formula C72H121N19O21 (calcd (+) m/z 1588.9057, Δ -0.3
ppm), 14 mass units more than that for 86, corresponding to a methylene group excess.
Figure 5.13 The 1H-NMR spectrum of compound 88 recorded at 600 MHz in MeOH-d4
NMR analysis indicated the components in 88 was similar to those in 86 with
the presence of Arg, Asn, NMeGln, Hpr, Leu, Thr, β-OMeTyr, 3,4-DiMeGln and Agdha
in the structure. The only difference between 88 and 86 is the replacement of the 3hydroxy-2,4,6-trimethylheptanoic acid unit (Htmha) in 86 by a 3-hydroxy-2,4,6trimethyloctanoic acid residue (Htmoa) in 88. The spin system of the Htmoa moiety was
deduced from COSY and TOCSY analysis (Figure 5.14). The observed key HMBC
correlation from Htmoa-H8 (δH 0.90ppm) to Htmoa-C6 (δC 32.6ppm) as well as a triplet
pattern of Htmoa-H8 (J = 7.2Hz) further confirmed this assignment. Compound 88 was
then subjected to the Marfey’s analysis as the same procedure described for the
stereochemical characterization of amino acid residues in 86. This experiment resulted
in the same absolute configurations of amino acids with those in 86, D-Arg, L-Asn, LNMeGln, L-Hpr, L-Leu and two D-allo-Thr (D-aThr). Due to insufficient sample
quantity, the HSQMBC and HSQC-TOCSY spectra of 88 were not obtained wellresolved enough for J-based configurational method. Although the stereochemical
98
information on the Agdha and Htmoa residues in 88 was not solved, on the basis of
biogenetic considerations these moieties were proposed to be similar to those in 86.
O
OH
6
1
8
COSY/TOCSY
HMBC
Htmoa
Figure 5.14 Htmoa unit of 88 with key COSY/TOCSY and HMBC correlations
99
Table 5.6 NMR data for neamphamide D (88) in CD3OD at 600 MHz
Position
δCa
1
2
3
4
5, 9
6, 8
7
3-OCH3
171.3 b
53.0
84.2
129.7
130.8
116.1
158.7
56.9
δH (mult., J, int.)
gCOSY
(H no.)
gTOCSY
(H no.)
ROESY
(H no.)
gHMBC
(C no.)
βOMeTyr
5.03 (d, 9.6, 1H)
4.47 (d, 9.6, 1H)
3
2
3
2
5, 9, Hpr-6
5, 9, 3-OCH3
1, 3, 4
1, 2, 5, 9, 3-OCH3
7.23 (d, 8.4, 2H)
6.79 (d, 8.4, 2H)
6, 8
5, 9
6, 8
5, 9
2, 3, 3-OCH3
3, 7
4, 7
3, 5, 9
3
3.14 (s, 3H)
NMeGln
1
2
3
170.8 b
56.5
25.6
4
32.0
5
NCH3
177.5
30.5
4.82 (overlap H2O)
1.55 (m, 1H)
1.63 (m, 1H)
1.65 (m, 1H)
1.82 (m, 1H)
3
2
3, 4
2
1, 3, Leu-1, NCH3
1, 2, 5
2, 3, 5
2, 3, 5
3.00 (s, 3H)
2, Leu-2, Leu-3, Leu-5
2, Leu-1
NMeGln-NCH3
NMeGln-NCH3
1, 3, 4
Leu
1
2
3
175.2
50.9
39.6
4
5
4-CH3
26.0
21.0
23.4
1
2
3
174.4 c
53.0
27.9
4
5
6
26.0
41.9
158.5
1
2
3
4
171.9
63.6
66.9
20.1
4.60 (dd, 2.4, 11.4, 1H)
1.31 (m, 1H)
1.81 (m, 1H)
1.76 (m, 1H)
0.92 (d, 6.6, 3H)
0.96 (d, 6.6, 3H)
3
2
2
3, 4, 5, 4-CH3
2
2
4
4
4
4
2
NMeGln-NCH3
3, 4-CH3
3, 4, 5
Arg
4.43 (dd, 4.8, 9.0, 1H)
1.63 (m, 1H)
1.95 (m, 1H)
1.62 (m, 2H)
3.18 (m, 2H)
3
2
2
5
4
3, 4, 5
2
2
1, 3, aThr1-1
4
6
aThr1
3.93 (d, 3.0, 1H)
4.33 (dd, 3.6, 6.6, 1H)
1.30 (d, 6.6, 3H)
3
2, 4
3
3, 4
2, 4
2, 3
4
2
1, 3, 4
1
2, 3
3, 4
2, 4
3
4
1, 3, 4
2
2, 3
aThr2-4
1, 6
aThr2
1
2
3
4
173.3
56.4
72.0
14.6
5.39 (d, 1.2, 1H)
5.61 (td, 2.4, 6.6, 1H)
1.21 (d, 6.0, 3H)
3
2, 4
3
Hpr
b
1
2
3
170.9
53.3
26.3
4
20.9
5
6
26.3
44.6
1
2
3
4
5
3-CH3
4-CH3
174.3 c
58.9
38.1
42.6
180.6
14.6
14.7
1
2
3
4
5
176.4
72.6
75.4
51.0
30.1
6
7
8
26.6
42.1
158.5
1
2
174.0 c
51.8
5.26 (d, 4.2, 1H)
1.61 (m, 1H)
2.16 (d, 13.2, 1H)
1.47 (m, 1H)
1.69 (m, 1H)
1.66 (m, 2H)
2.88 (m, 1H)
3.87 (d, 12.6, 1H)
3
2
3
3, 4, 5
2
2, 4
3
6
5
5
4, 5
4, 5
5
3
3,4-diMeGln
4.25 (d, 9.6, 1H)
2.40 (m, 1H)
2.80 (m, 1H)
3
2, 4, 3-CH3
3, 4-CH3
3, 3-CH3
2, 3-CH3
3, 4-CH3
4, 3-CH3
1, 3, 4, 3-CH3
2
3-CH3
1.21 (d, 6.6, 3H)
1.29 (d, 6.6, 3H)
3
4
2, 3
4
2
2, 3, 4
3, 4, 5
3, 4
2, 4
3, 5, 6, 7
4
4
7
6
4
1, 3, 4
1, 2
Agdha
3.99 (d, 9.0, 1H)
3.63 (dd, 0.6, 9.0, 1H)
4.19 (dd, 0.6, 9.0, 1H)
1.60 (m, 1H)
1.73 (m, 1H)
1.65 (m, 2H)
3.19 (m, 2H)
3
2, 4
3, 5
4
4
7
6
2
6, 8
Asn
4.72 (t, 6.0, 1H)
3
3
100
1, 3, 4, Htmoa-1
3
37.3
4
174.9 c
1
2
3
4
5
178.9
45.0
79.7
33.5
37.5
6
7
8
2-CH3
4-CH3
6-CH3
32.6
25.3
11.2
14.5
17.5
21.0
2.78 (dd, 5.4, 16.2, 1H)
2.89 (dd, 6.0, 16.2, 1H)
2
2
2
2
2, 4
2, 4
Htmoa
2.65 (p, 7.2, 1H)
3.51 (dd, 3.0, 9.0, 1H)
1.79 (m, 1H)
1.36 (m, 1H)
1.09 (m, 1H)
1.47 (m, 1H)
1.52 (m, 2H)
0.90 (t, 7.2, 3H)
1.10 (d, 6.6, 3H)
1.00 (d, 7.2, 3H)
0.93 (d, 6.0, 3H)
3, 2-CH3
2, 4
5, 4-CH3
6
4, 6
5, 7, 6-CH3
6, 8
7
2
4
6
3, 2-CH3
2, 4, 2-CH3
5, 4-CH3
6
1, 3, 2-CH3
1, 2, 5, 4-CH3
2-CH3, 4-CH3
5, 6-CH3
6, 8
5, 6, 7
2, 3
4, 5
4, 5, 6
6
1, 2, 3
3, 4, 5
5, 6, 7
3, 4
3
C chemical shifts obtained from correlations observed in gHSQC and gHMBC spectra, referenced to residual CD3OH at δC 49.15
ppm and δH 3.31 ppm
a 13
b, c
These assignments are interchangeable
Htmoa
Hpr
O
∗
Asn
H2N
NH
∗
O
OH
aThr2
N
O
H
N
NH
∗
O
O
H2N
βOMeTyr
∗
NH
∗
Agdha
OH
∗
OH
O
O
H
N
∗
O
N
H
HN
O
NH
O
NMeGln
O
O
aThr1
HN
H2N
O
N
HN
NH2
O
O
3,4-diMeGln
HN
Neamphamide D (88)
O
OH
H2N
NH
101
Arg
OH
∗
Leu
5.4 Evaluation of drug-like properties
The negative values of calculated logP as well as the high values of PSA (> 657
2
Å ) indicate that these compounds are very hydrophilic. This reason can be explained by
the possession of many hydrogen bond donors (26-27 HBDs) and acceptors (22 HBAs)
in the peptide structures which increase peptides’ solubility in water. From the results of
physico-chemical properties, three isolated compounds do not comply with the
requirements of Lipinski’s and Veber’s rules for drug-like molecules.
Table 5.7 Physico-chemical properties of the three isolated cyclodepsipeptides
Compounds
Formula
MW
LogP
HBD
HBA
PSA
86
87
88
C71H119N19O21
C71H118N18O22
C72H121N19O21
1574.82
1575.80
1588.85
-9.20
-13.33
-8.76
26
27
26
22
22
22
663.17
657.38
663.17
NROT
36
36
37
Predicted
Bioavailability
No
No
No
5.5 Biological activity
The anticancer activity of neamphamides B-D (86-88) was evaluated on a panel
of human cancer and non-cancer cell lines (Table 5.8). The results demonstrated that
these cyclic depsipeptides possessed potent cytotoxic activities against both cancer and
non-cancer cells with IC50 values ranging from 87.6 to 366.1 nM. With comparable IC50
values in all six cell lines, the difference observed between Htmha and Htmoa or
between NMeGln and NMeGlu residues did not affect the cytotoxic activity.
Table 5.8 Biological activity of the three isolated compounds
Compound
86
87
88
Vincristine sulfate
A549
125.2±11.2
130.5±3.5
91.3±17.6
20.2±1.9
HeLa
113.5±8.1
172.1±18.8
208.4±6.3
35.8±3.5
Cell lines/ IC50 ± SD (nM)
LNCaP
PC3
228.3±9.6 170.1±12.5
188.2±10.1
113.4±4.1
108.8±7.5 134.1±20.3
6.1±0.7
15.3±1.1
HEK
204.7±25.7
157.3±30.8
87.6±12.6
30.5±3.7
NFF
366.1±18.2
262.2±15.9
152.0±9.2
16.9±3.2
5.6 References
(1)
Hooper, J. N. A.; Soest, R. W. M. V. Systema Polifera: A Guide to the
Classification of Sponges; Kluwer Academic & Plenum: New York, 2002.
(2)
Silva, E. D. D.; Racok, J. S.; Andersen, R. J. Tetrahedron Lett. 1991, 32, 27072710.
(3)
Oku, N.; Gustafson, K. R.; Cartner, L. K.; Wilson, J. A.; Shigematsu, N.; Hess,
S.; Pannell, L. K.; Boyd, M. R.; McMahon, J. B. J. Nat. Prod. 2004, 67, 14071411.
(4)
Zampella, A.; D'Auria, M. V.; Paloma, L. G.; Casapullo, A.; Minale, L.;
Debitus, C.; Henin, Y. J. Am. Chem. Soc. 1996, 118, 6202-6209.
102
(5)
D'Auria, M. V.; Zampella, A.; Paloma, L. G.; Minale, L.; Debitus, C.;
Roussakis, C.; Bert, V. L. Tetrahedron 1996, 52, 9589-9596.
(6)
Zampella, A.; Randazzo, A.; Borbone, N.; Luciani, S.; Trevisi, L.; Debitus, C.;
D'Auria, M. V. Tetrahedron Lett. 2002, 43, 6163-6166.
(7)
Sepe, V.; D'Orsi, R.; Borbone, N.; D'Auria, M. V.; Bifulco, G.; Monti, M. C.;
Catania, A.; Zampella, A. Tetrahedron 2006, 62, 833-840.
(8)
D'Auria, M. V.; Sepe, V.; D'Orsi, R.; Bellotta, F.; Debitus, C.; Zampella, A.
Tetrahedron 2007, 63, 131-140.
(9)
Ford, P. W.; Gustafson, K. R.; McKee, T. C.; Shigamatsu, N.; Maurizi, L. K.;
Pannell, L. K.; Williams, D. E.; Silva, E. D. D.; Lassota, P.; Allen, T. M.; Soest,
R. V.; Andersen, R. J.; Boyd, M. R. J. Am. Chem. Soc. 1999, 121, 5899-5909.
(10)
Ratnayake, A. S.; Bugni, T. S.; Feng, X.; Harper, M. K.; Skalicky, J. J.;
Mohammed, K. A.; Andjelic, C. D.; Barrows, L. R.; Ireland, C. M. J. Nat. Prod.
2006, 69, 1562-1586.
(11)
Plaza, A.; Gustchina, E.; Baker, H. L.; Kelly, M.; Bewley, C. A. J. Nat. Prod
2007, 70, 1753-1760.
(12)
Zampella, A.; Sepe, V.; Luciano, P.; Bellotta, F.; Monti, M. C.; D'Auria, M. V.;
Jepsen, T.; Petek, S.; Adeline, M. T.; Laprévôte, O.; Aubertin, A. M.; Debitus,
C.; Poupat, C.; Ahond, A. J. Org. Chem. 2008, 73, 5319-5327.
(13)
Zampella, A.; Sepe, V.; Bellotta, F.; Luciano, P.; D'Auria, M. V.; Cresteil, T.;
Debitus, C.; Petek, S.; Poupat, C.; Ahond, A. Org. Biomol. Chem. 2009, 7,
4037-4044.
(14)
Dewick, P. M. Medicinal Natural Products: A Biosynthetic Approach; 3rd ed.;
John Wiley & Sons, 2009.
(15)
Zampella, A.; D'Orsi, R.; Sepe, V.; Casapullo, A.; Monti, M. C.; D'Auria, M. V.
Org. Lett. 2005, 7, 3585-3588.
(16)
Makino, K.; Nagata, E.; Hamada, Y. Tetrahedron Lett. 2005, 46, 6827-6830.
(17)
Oku, N.; Krishnomoorthy, R.; Benson, A. G.; Ferguson, R. L.; Lipton, M. A.;
Phillips, L. R.; Gustafson, K. R.; McMahon, J. B. J. Org. Chem. 2005, 70, 68426847.
(18)
Liang, B.; Carroll, P. J.; Joullié, M. M. Org. Lett. 2000, 2, 4157-4160.
(19)
Acevedo, C. M.; Kogut, E. F.; Lipton, M. A. Tetrahedron 2001, 57, 6353-6359.
(20)
Thoen, J. C.; Ramos, A. I. M.; Lipton, M. A. Org. Lett. 2002, 4, 4455-4458.
(21)
Cranfill, D. C.; Ramos, A. I. M.; Lipton, M. A. Org. Lett. 2005, 7, 5881-5883.
103
(22)
Chandrasekhar, S.; Ramachandar, T.; Rao, B. V. Tetrahedron: Asymmetry 2001,
12, 2315-2321.
(23)
Kumar, A. R.; Rao, B. V. Tetrahedron Lett. 2003, 44, 5645-5647.
(24)
Jeon, J.; Hong, S. K.; Oh, J. S.; Kim, Y. G. J. Org. Chem. 2005, 71, 3310-3313.
(25)
Zampella, A.; D'Auria, M. V. Tetrahedron: Asymmetry 2002, 13, 1237-1239.
(26)
Turk, J. A.; Visbal, G. S.; Lipton, M. A. J. Org. Chem. 2003, 68, 7851-7844.
(27)
Marfey, P. Carlsberg Res. Commun. 1984, 49, 591-596.
(28)
Bhymer, C.; Bayon, M. M.; Caruso, J. A. J. Sep. Sci. 2003, 26, 7-19.
(29)
Matsumori, N.; Kaneno, D.; Murata, M.; Nakamura, H.; Tachibana, K. J. Org.
Chem. 1999, 64, 866-876.
(30)
Riccio, R.; Bifulco, G.; Cimino, P.; Bassarello, C.; Paloma, L. G. Pure Appl.
Chem. 2003, 75, 295-308.
(31)
Bifulco, G.; Dambruoso, P.; Paloma, L. G.; Riccio, R. Chem. Rev. 2007, 107,
3744-3779.
(32)
In Dictionary of Natural Products on CD-Rom Chapman and Hall ed.; CRC
Press: London, 2009.
(33)
Bassarello, C.; Zampella, A.; Monti, M. C.; Paloma, L. G.; D'Auria, M. V.;
Riccio, R.; Bifulco, G. Eur. J. Org. Chem. 2006, 604-609.
104
Chapter 6
Hemiasterlins and milnamides from the Australian
sponge Pipestela candelabra
6.1 Introduction
Hemiasterlins and milnamides are well-known as two related cytotoxic
tripeptide families from marine sponges. Both hemiasterlin (or milnamide B) (89)1 and
milnamide A (90)2 were first isolated and reported in 1994. Characteristic structural
features of these peptides are the presence of a dipeptide side chain of two unnatural
amino acids, tert-leucine and N-methylvinylogous valine (Figure 6.1). This dipeptide
fragment is incorporated with another untypical amino acid residue, tri- or
tetramethylated tryptophan, to produce the hemiasterlin molecules or connected to a
tetrahydro-β-carboline residue to form the milnamide skeleton.
O
O
N
NH
N
H
Methylated tryptophan residue
OH
N
H
O
tert-leucine residue
N-methylvinylogous valine residue
Figure 6.1 Three residues characterised from milnamides and hemiasterlins
Structural analysis using chemical degradation methods showed that the tertleucine and N-methylvinylogous valine moieties in hemiasterlins and also in
milnamides possessed the same L-configurations.3 Attempts in solving absolute
configurations of the trimethylated tryptophan by amino acid analysis and other
spectroscopic methods were unsuccessful until a semi-synthetic product hemiasterlin
ester (96) was successfully crystallized in 1996.4 Detailed X-ray diffraction analysis of
96 supported the stereochemical determination of the trimethylated tryptophan as an Sconfiguration. Thus, complete stereochemistry of hemiasterlin and its analogues was
assigned as (11S,14S,17S).
The absolute stereochemistry of the tetrahydro-β-carboline in the milnamide
family was solved by a X-ray analysis in 2004 when single crystals of milnamide C (94)
was successfully formed.5 In the same year, the enantioselective total synthesis of
milnamide A (1) was reported6 which again confirmed the total absolute stereochemical
configurations for milnamide derivatives. Interestingly, the S-configurations were found
for all stereocenters in both milnamides and hemiasterlins.
So far, four hemiasterlin deriveatives (hemiasterlin (89),1 hemiasterlin A (91),3,7
hemiasterlin B (92),3 hemiasterlin C (93)7) and three milnamide analogues (milnamide
105
A (90),2 milnamide C (94)5 and milnamide D (95)5,8) were identified from 4 sponge
genera (Auletta,2,7 Cymbastela,3,8 Hemiasterella1 and Siphonochalina7) (Table 6.1).
Table 6.1 Taxonomy of sponges producing milnamides and hemiasterlins
Genus
Auletta2,7
Cymbastela3,8
Hemiasterella1
Siphonochalina7
Species
constricta
sp.
minor
sp.
Compounds
89, 90, 93
89, 90, 92, 93, 94
90
90, 93, 95
List of milnamides and hemiasterlins isolated from marine sponges
R2
O
O
O
O
N
N
H
NH
N
OH
N
H
O
N
N
N
OH
O
R
R1
Hemiasterlin/Milnamide B (89): R1 = R2 = CH3
Hemiasterlin A (91): R1 = H, R2 = CH3
Hemiasterlin B (92): R1 = R2 = H
Hemiasterlin C (93): R1 = CH3, R2 = H
Milnamide A (90): R = -CH2Milnamide C (94): R = C=O
Milnamide D (95): R= -CH=
Exploring cytotoxic activities of the hemiasterlin family demonstrated that these
peptides exhibited potent cytotoxic activity against a panel of cancer cell lines (A495,
MCF7, OVCAR-3, P388, HT29, COLO205, SF539 and U373) with the IC50 values in
the range of ng/ml.2,3,7 They also showed the activity in a tubulin polymerization assay
at concentrations less than 1 μg/mL7,8 which were more potent than either of the
anticancer drugs Taxol or Vincristine.9,10 Compared with other known cytotoxins and
antimitotic agents from natural sources, the hemiasterlin family possesses a simple
skeleton which makes them attractive targets for total synthesis.
Total synthesis of 89 was first reported in 1997.11 Synthesis of hemiasterlin
analogues and investigation of their structure activity relationship (SAR) on antimitosis
and cytotoxicity were later studied by the groups of Andersen12 and Yamashita.13 It was
found that the gem-dimethylation at the β-carbon and N-methylation at the α-amino
group on the tryptophan residue, the double bond and C-15 isopropyl substituent on the
vinylogous valine residue were all extremely critical for potent biological activity.
Moreover, the L-configurations of all amino acid residues in the peptides were also
required for maximal cytotoxicity. Although the indole moiety also played an important
role on the bioactivity, their cytotoxic activity could improve if this moiety was
replaced by the benzene ring (Figure 6.2).12 From the SAR information, a synthetic
analogue of hemiasterlin (97) was discovered and showed higher activity than 89 in the
tubulin polymerization assay as well as in the cell proliferation assay.13 Both 89 and 97
have been licensed by Wyeth and are in clinical trials for anticancer drugs.14
106
Figure 6.2 Structure activity relationship of 8912
O
O
O
O
N
N
H
NH
N
N
H
O
O
NH
OH
O
N
Synthetic hemiasterlin methyl ester (96)
Synthetic analogue of hemiasterlin (97)
Figure 6.3 Two synthetic analogues of hemiasterlin
Among the four genera identified hemiasterlins and milnamides, chemical
investigation of the sponges from the genera Auletta and Cymbastela also demonstrated
that these sponges not only shared these hemiasterlin and milnamide compounds but
also provided a source of small depsipeptides, geodiamolides and jaspamide.3,5,15,16
These classes of depsipeptides also have a potent cytotoxic activity. The list of
geodiamolides isolated from the marine genera Auletta and Cymbastela is presented
below.
List of geodiamolides isolated from the marine genera Auletta and Cymbastela
R1
OH
R2
H
N
N
O
O
O
R3
O
O
NH
Geodiamolide A5,15 (98): R1 = I, R2 = CH3, R3 = CH3
Geodiamolide B15 (99): R1 = Br, R2 = CH3, R3 = CH3
Geodiamolide C15 (100): R1 = Cl, R2 = CH3, R3 = CH3
Geodiamolide L15 (107): R1 = I, R2 = CH2OH, R3 = CH3
Geodiamolide M15 (108): R1 = Br, R2 = CH2OH, R3 = CH3
Geodiamolide N15 (109): R1 = Cl, R2 = CH2OH, R3 = CH3
Geodiamolide O15 (110): R1 = I, R2 = CH3, R3 = CH2OH
Geodiamolide P15 (111): R = Br, R2 = CH3, R3 = CH2OH
Geodiamolide Q15 (112): R = Cl, R2 = CH3, R3 = CH2OH
R1
OH
R2
H
N
N
O
O
O
O
O
NH
Geodiamolide D5,16 (101): R1 = I, R2 = CH3
Geodiamolide E5,16 (102): R1 = Br, R2 = CH3
Geodiamolide F16 (103): R1 = Cl, R2 = CH3
Geodiamolide R15 (113): R1 = I, R2 = CH2OH
Neosiphoniamolide15 (114): R1 = I, R2 = CH(CH3)2
107
R
HO
Br
OH
H
N
N
O
O
O
O
O
NH
NH
Geodiamolide G15 (104): R = I
Geodiamolide J15 (105): R = Br
Geodiamolide K15 (106): R = Cl
H
N
N
O
O
O
O
O
NH
Jaspamide3,16 (115)
O
The sponge genus Pipestela has recently been characterised as a new genus
which shares some shape features and molecular DNA sequences with the other ten
genera in the Axinellidae family
17
: Auletta, Axinella, Cymbastela, Dragmacidon,
Dragmaxia, Pararhaphoxya, Phakellia, Phycopsis, Ptilocaulis and Reniochalina.18 This
genus is distributed only across the north-eastern region of Australia and is very
common from the Great Barrier Reef, Coral Sea, Papua New Guinea, Solomon Islands
and Vanuatu. Pipestela candelabra is more popular than the four other assigned species,
P. rara, P. occidentalis, P. hooperi and P. terpenensis.17
According to Dictionary of Natural Products (version 2010), no secondary
metabolite has been reported from the genus Pipestela. This chapter gives details about
the chemical investigation of the marine sponge Pipestela candelabra along with the
evaluation of the cytotoxicity of isolated compounds. Eleven compounds were isolated
and structurally characterised. A new milnamide E (116) and a novel hemiasterlin D
(117) were identified along with nine known compounds which were previously
reported: hemiasterlin (89),1 hemiasterlin A (91),3 milnamide A (90),2 milnamide C
(94),5 milnamide D (95),8 geodiamolide D (101),16 geodimolide E (102),16 geodiamolide
F (103)16 and jaspamide (115).19
108
6.2 Collection, Extraction and Isolation
A specimen of Pipestela candelabra was collected at the depth of 36m, Wilson
Reef, Coral Sea, Queensland, Australia. It was identified as Pipestela candelabra
(phylum Porifera, class Demospongiae, order Halichondrida, family Axinellidae). A
voucher specimen QMG320597 has been deposited at the Queensland Museum, South
Brisbane, Queensland, Australia.
Plate 6.1 Photograph of the sponge Pipestela candelabra collected at
Wilson Reef, Coral Sea
A freeze dried sample of Pipestela candelabra (10.0g) was extracted
exhaustively with hexane (250ml), dichloromethane (2 x 250ml) and methanol (2 x
250ml), respectively. The dichloromethane and methanol extracts were combined and
then evaporated solvents to yield a yellow residue (1.7g). This crude extract was preabsorbed onto C18 (1.0g) and packed dry into a small cartridge, which was connected to
a C18 preparative HPLC column (5μm, 21.2 x 150mm). A linear gradient from 100%
water (0.1% TFA) to 100% methanol (0.1% TFA) was performed over 60 minutes at a
flow rate of 9.0 ml/min and 60 fractions (1.0 minute each) were collected. Mass-guided
identification demonstrated fractions 37 to 43 contained the ion peaks of interest in (+)LRESIMS at (+)m/z 513, 525, 527, 537, 580 and 628. 1H-NMR test for these fractions
led to the identification of a pure compound 91 (4mg, 0.04% dry wt) in fraction 38.
Fractions 37, 39 and 40 were then combined and subjected on the Hypersil BDS C18
column (5 μm, 250 x 10 mm) with a sixty-minute linear gradient from 40% methanol
(0.1% TFA) – 60% water (0.1% TFA) to 100% methanol (0.1% TFA) at a flow rate of
4ml/min. Compounds 95 (0.5mg, 0.005% dry wt), 116 (0.3mg, 0.003% dry wt) and
compound 91 (2mg, 0.02% dry wt) were purified. Other fractions from 41 to 43 were
also combined and loaded on the Hypersil BDS C18 column (5 μm, 250 x 10 mm) with a
109
linear gradient from 50% methanol (0.1% TFA) – 50% water (0.1% TFA) to 100%
methanol (0.1% TFA) at a flow rate of 4ml/min in sixty minutes, yielding 102 (2mg,
0.02% dry wt), 101 (4mg, 0.04% dry wt) and 89 (7.5mg, 0.075% dry wt), respectively
(Scheme 6.1).
Pipestela candelabra (10.0g)
a)
DCM/MeOH extract (0.5g)
b)
Fractions 37, 39-40
(+)-m/z 513, 525, 537
Compound 95
(0.5mg, 0.05%)
Fraction 34
Fractions 41-43
(+)-m/z 527, 580, 628
c)
d)
Compound 102
(2.0mg, 0.02%)
Fractions 30-31
Compound 116
(0.3mg, 0.003%)
Fraction 47
Compound 91
(6.0mg, 0.06%)
Fraction 38 (b)
Fractions 38-41 (c)
Compound 89
(7.5mg, 0.075%)
Fractions 37-40
Compound 101
(4.0mg, 0.04%)
Fractions 33-35
a) Extraction with DCM and MeOH
b) A linear gradient from 100% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes, a
flow rate of 9.0 ml/min, 1.0 min/fraction
c) A linear gradient from 40% methanol (0.1% TFA) – 60% water (0.1% TFA) to 100% methanol
(0.1% TFA) in 60 minutes, a flow rate of 4.0 ml/min, 1.0 min/fraction
d) A linear gradient from 50% methanol (0.1% TFA) – 50% water (0.1% TFA) to 100% methanol
(0.1% TFA) in 60 minute, a flow rate of 4.0 ml/min, 1.0 min/fraction.
Scheme 6.1 Extraction and Isolation Procedure for Pipestela candelabra
collected at Wilson Reef, Coral Sea
Compound 116 was elucidated as a new compound by NMR experiments (see
6.3.1). Since its yield of isolation was not enough for further chemical and biological
investigation, re-collection of biota was performed. Taxonomy searching and chemical
analysis (HPLC profiles and LC/MS data) in the in-house library resulted in another
sample of Pipestela candelabra which was collected at the depth of 20m, Houghton
Reef, Howick Group, Queensland, Australia. A voucher specimen QMG320790 has
been deposited at the Queensland Museum, South Brisbane, Queensland, Australia.
This sample was then investigated in thirty-gram scale.
110
Plate 6.2 Photograph of the sponge Pipestela candelabra collected at
Houghton Reef, Howick Group
Another freeze dried sample of Pipestela candelabra (30.0g) was extracted
exhaustively with hexane (750ml), dichloromethane (4 x 250ml) and methanol (4 x
250ml), respectively. The dichloromethane and methanol extracts were combined and
then evaporated solvents to yield a yellow residue (7.50g). This crude extract was preabsorbed onto C18 (10g) and packed dry into a cartridge (25 x 50mm), which was
connected to a C18 preparative HPLC column (5μm, 50 x 150mm). A linear gradient
from 100% water (0.1% TFA) to 100% methanol (0.1% TFA) was performed over 90
minutes at a flow rate of 20 ml/min and 180 fractions (0.5 minute each) were collected.
Mass-guided identification demonstrated fractions 111 to 132 contained the ion peaks of
interest in (+)-LRESIMS at (+)m/z 513, 525, 527, 536, 537, 539, 580, 628, 709 and 853.
Fractions 111 to 114 were then combined and further purified by a Betasil C18 column
(5μm, 150 x 21.2mm) using a linear gradient from 45% methanol (0.1% TFA) – 55%
water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes. Compounds 95
(1.0mg, 0.0033% dry wt), 91 (12.0mg, 0.0400% dry wt) and 117 (0.8mg, 0.0026% dry
wt) were obtained. A fifteen milligram mixture of compounds 90 and 116 in fractions
115 to 124 was subjected on a Hypersil BDS C18 column (5 μm, 250 x 10 mm) with a
sixty-minute linear gradient from 50% methanol (0.1% FA) – 50% water (0.1% FA) to
100% methanol (0.1% FA) yielding 116 (2.5mg, 0.0083% dry wt) and 90 (2.0mg,
0.0067% dry wt). The combined fractions 125 to 132 were also chromatographed on a
Betasil C18 column (5 μm, 150 x 21.2 mm) from 50% methanol (0.1% TFA) – 50%
water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes. Six compounds were
purified, including compound 89 (18.0mg, 0.06% dry wt), 103 (1.0mg, 0.0033% dry
wt), 102 (7.0mg, 0.0233% dry wt), 101 (15.0mg, 0.05% dry wt), 115 (1.4mg, 0.0047%
dry wt) and 94 (0.5mg, 0.0017% dry wt), respectively (Scheme 6.2).
111
Pipestela candelabra (30.0g)
a)
DCM/MeOH extract (7.5g)
b)
Fractions 111-114
(+)-m/z 513, 537, 853
c)
Compound 95
(1.0mg, 0.0033%)
Fractions 23-24
Compound 117
(0.8mg, 0.0026%)
Fraction 32
Compound 91
(12.0mg, 0.0400%)
Fractions 25-27
Fractions 125-132
(+)-m/z 527, 536, 553, 580, 608, 709
Fractions 115-124
(+)-m/z 525, 539
e)
d)
Compound 89
(18.0mg, 0.0600 %)
Fractions 20-25
Compound 116
(2.5mg, 0.0083%)
Fractions 39-41
Compound 90
(2.0mg, 0.0067%)
Fractions 43-47
Compound 102
(7.0mg, 0.0233%)
Fraction 29
Compound 103
(1.0mg, 0.0033%)
Fractions 20-25
Compound 115
(1.4mg, 0.0047%)
Fraction 43
Compound 101
(15.0mg, 0.05%)
Fractions 31-33
Compound 94
(0.5mg, 0.0017%)
Fraction 45
a) Extraction with DCM and MeOH
b) A linear gradient from 100% water (0.1% TFA) to 100% methanol (0.1% TFA) in 90 minutes, a flow rate of 20 ml/min, 0.5 min/fraction
c) A linear gradient from 45% methanol (0.1% TFA) – 55% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes, a flow rate of 9ml/min, 1.0
min/fraction
d) A linear gradient from 50% methanol (0.1% FA) – 50% water (0.1% FA) to 100% methanol (0.1% FA) in 60 minute, a flow rate of 4ml/min, 1.0 min/fraction
e) A linear gradient from 50% methanol (0.1% TFA) – 50% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minute, a flow rate of 9ml/min, 1.0
min/fraction
Scheme 6.2 Extraction and Isolation Procedure for Pipestela candelabra collected at Houghton Reef, Howick Group
112
6.3 Structure Elucidation and Discussion
6.3.1 Milnamide E (116)
Compound 116, [α]25D +10.8 (c 0.02, MeOH), was obtained as a white solid.
A molecular ion in the (+)-HRESIMS at m/z 525.3418 was consistent for the
molecular formula C30H44N4O4 (calcd (+) m/z 525.3435, Δ -3.2ppm), corresponding
to 11 unsaturations.
Figure 6.4 The 1H-NMR spectrum of 116 recorded at 600 MHz in DMSO-d6
Combined data from the 13C-NMR and HSQC spectra revealed this structure
contained three carbonyls (δC 171.0, 169.7 and 168.6 ppm); five olefinic carbons (δC
138.2, 119.5, 118.6, 117.9 and 110.9 ppm); five quaternary aromatic carbons (δC
135.9, 131.8, 131.7, 125.4 and 113.0 ppm) and two other quaternary carbons (δC 34.7
and 34.3 ppm); three tertiary carbons adjacent to a heteroatom (δC 72.1, 55.8 and
53.4 ppm) and one tertiary carbon in alkyl chain (δC 28.8 ppm); one secondary
carbon close to a heteroatom (δC 48.6 ppm) and nine primary carbons (δC 43.0, 30.8,
30.0, 26.4, 26.4, 26.4, 24.1, 19.3, 18.8 and 13.5 ppm) out of which the two signals at
δC 43.0 and 30.8 ppm were typical for N-methyl groups.20
113
The 2D-NMR spectra were used to establish the partial structure of 116. A
benzene moiety in fragment A was firstly deduced based on an ABCD spin system in
the COSY spectrum with signals at δ 7.47 (d, J = 8.4 Hz), 6.87 (t, J = 7.2, 7.8 Hz),
6.95 (t, J = 7.2, 7.8 Hz) and 7.24 (d, J = 8.4 Hz) corresponding to H-5, H-6, H-7 and
H-8.20 These protons showed 3JCH long range correlations with two quaternary
carbons C-4b (δC 125.4 ppm) and C-8a (δC 135.9 ppm). An exchangeable proton at
δΗ 10.62 ppm also displayed the HMBC correlations with carbons C-4b, C-8a as well
as C-4a (δC 113.0 ppm) and C-9a (δC 131.8 ppm) suggesting the presence of an
indole ring.20 A ROESY correlation between NH (δ10.62 ppm) and H-8 confirmed
the establishment of the indole system in fragment A.
Key HMBC correlations from δH 3.62/3.88 ppm (H-1) to C-3 and C-19; from
δH 3.40 ppm (H-3) to C-1, C-4, C-19, C-20 and C-20’; and also from δH 2.43 ppm
(H-19) to C-1 and C-3 indicated the relationships between these carbons in fragment
B. The downfield resonances of these carbons C-1 (CH2, δC 48.6 ppm), C-3 (CH, δC
72.1 ppm) and C-19 (N-CH3, δC 43.0 ppm) suggested the presence of a nitrogen atom
to link these carbons together. In addition, the observation of the ROESY
correlations from H-19 to H-1 and H-3 provided further evidence for this assignment.
A location of the quaternary carbon C-4 (δC 34.7 ppm) was assigned due to strong
HMBC correlations from H-3, H-20 (δH 1.37 ppm) and H-20’(δH 1.30 ppm) to C-4.
All information facilitated the construction of fragment B.
Further analysis of the 2D-NMR data showed the proton H-1 correlated with
C-4a and C-9a. The HMBC signals were also observed from H-3, H-20 and H-20’ to
C-4a. Moreover, the NH in fragment A had the ROESY correlation with H-1.
Therefore the connectivity of fragments A and B were established (Figure 6.5).
Figure 6.5 The partial structures A, B and combined A-B of compound 116
The downfield resonance H-12 (δH 4.75 ppm), corresponding to the carbon C12 (δC 53.4 ppm), showed a strong HMBC signal to the carbonyl C-13 (δC 171.0
ppm) and correlated with an exchangeable proton H-11 (δH 7.90 ppm) suggesting this
114
methine group was an α-carbon of an amino acid residue.21 The characteristic of a
tert-butyl group was assigned by the signal at δH 0.90 ppm (H-21) equivalent to 9
integrated protons in the 1H-NMR spectrum, and its HMBC correlation to the
quaternary carbon C-12a (δC 34.3 ppm). Due to the long range HMBC correlation
from H-21 to C-12, a tert-leucine residue was suggested for fragment C (Figure 6.6).
The system containing C-15, C-16, C-15a, C-23 and C-23’ was located due to
the observation of the COSY correlations from H-16 (δH 6.59 ppm) to H-15 (δH 4.89
ppm); from H-15a (δH 1.91 ppm) to both H-15 and two methyl groups H-23 (δH 0.77
ppm) and H-23’ (δH 0.67 ppm). The positions of C-17 (δC 131.7 ppm), C-18 (δC
168.6 ppm) and C-24 (δC 13.5 ppm) were then assigned by the 3JCH long range
correlations from H-15 to C-17 and from H-16 to C-18 and C-24. This assignment
was supported by the downfield shift of C-16 as the β-position in an α,β-unsaturated
carbonyl.20 The E-configuration was suggested for Δ16-17 by the absence of the
ROESY correlation was observed between H-16 and H-24 as in other milnammide
structural class. The downfield resonance of the methine carbon C-15 (δC 55.8 ppm)
and the HMBC correlation from H-15 to the N-methyl group C-22 (δC 30.8 ppm)
determined the location of this N-methyl group. From this evidence, the final part of
116 (fragment D) was elucidated as an N-methylvinylogous valine residue (Figure
6.6). Fragments C and D were assembled due to the HMBC correlation from H-22 to
C-13 and the observed ROESY correlation between H-12 and H-22 (Figure 6.6).
Figure 6.6 The partial structures C, D and combined C-D of compound 116
Since fragments B and C shared the same carbonyl C-10 (δC 169.7 ppm)
observed in the HMBC spectrum, two incomplete bonds in the combined fragments
A-B and C-D were suggested to connect through this carbonyl linkage. This
assignment was again confirmed by the ROESY correlation from H-3 to H-10. The
complete structure of 116 was finally established (Figure 6.7).
115
Figure 6.7 Key HMBC and ROESY correlations to establish the structure of 116
Determination of absolute stereochemistry of compound 116
Marfey’s analysis of 116 revealed the presence of an L-tert-Leu residue.
Structurally, compound 116 differed from 90 only by the lack of the N-methyl
substituent in the β-carboline ring. It is of interest to note that the absolute
stereochemistry of 90 has recently been confirmed by the total synthesis.6 The 1HNMR data of 90 isolated from the sponge Pipestela candelabra was consistent with
that previously reported for milnamide A.2,6 Furthermore, the negative Cotton Effects
(CE) at 223, 240 nm and a positive CE at 270 nm in the circular dichroism (CD)
spectrum also confirmed the same configuration of 90 compared with those of both
natural and synthetic milnamide A.6 Hence, the absolute stereochemistry of 90 here
was assigned as (3S, 12S, 15S). It has been well established that the absolute
configuration of 116 can be determined from the signs of optical rotation and the
negative Cotton Effects in the CD spectrum as published for milnamide A.6
Milnamide E (116) has an [α]25D +10.8 (c 0.02, MeOH), as similar sign to those
priviously reported for milnamide A ([α]25D +28.8 (c 0.50, CH2Cl2)),2 milnamide C
([α]28D +68.0 (c 0.06, MeOH))5 and milnamide D ([α]25D +156.0 (c 0.39, MeOH)).8
More significantly, compound 116 has two prominent negative CEs at 220 and 235
nm along with a positive CE at 270 nm, which are of the same sign to those of 90.6
All information facilitated the absolute stereostructure of 116 as (3S,12S,15S).
6
Molecular ellipcity
(deg.cm2/dmol)
4
90
2
0
200
-2
220
240
260
280
300
-4
116
-6
-8
Wavelength, λ (nm)
Figure 6.8 CD spectrum of compounds 90 and 116
116
Table 6.2
Position
NMR data for FA salt of milnamide E (116) in DMSO-d6 at 600 MHz
δC
1
48.6
3
4
4a
4b
5
6
7
8
8a
9a
10
11
12
12a
13
15
15a
16
17
18
19
20
20’
21, 21’, 21”
22
23
23’
24
NH
72.1
34.7
113.0
125.4
118.6
117.9
119.5
110.9
135.9
131.8
169.7
53.4
34.3
171.0
55.8
28.8
138.2
131.7
168.6
43.0
30.0
24.1
26.4
30.8
19.3
18.8
13.5
-
δH
(mult., J, int.)
3.62 (d, 14.4, 1H)
3.88 (d, 14.4, 1H)
3.40 (s, 1H)
7.47(d, 8.4, 1H)
6.87 (t, 7.2, 7.8, 1H)
6.95 (t, 7.2, 7.8, 1H)
7.24 (d, 8.4, 1H)
7.90 (d, 9.6, 1H)
4.75 (d, 9.0, 1H)
4.89 (t, 9.6, 10.2, 1H)
1.91 (m, 1H)
6.59 (d, 9.6, 1H)
2.43 (s, 3H)
1.37 (s, 3H)
1.30 (s, 3H)
0.90 (s, 9H)
2.87 (s, 3H)
0.77 (d, 6.6, 3H)
0.67 (d, 6.6, 3H)
1.76 (s, 3H)
10.62 (s, 1H)
gCOSY
(H no.)
ROESY
(H no.)
NH
19
11, 19, 20
NH
3
21, 22
16
15a, 22
1, 3
12, 16
1, 8
6
5, 7
6, 8
7
12
11
15a, 16
15, 23, 23’
15
15a
15a
-
gHMBC
(C no.)
3, 4a, 9a, 19
3, 4a, 9a, 19
1, 4, 4a, 10, 19, 20, 20’
4a, 4b, 7, 8a
4b, 8
5, 8a
4b, 6
10
12a, 13, 21, 21’, 21”
13, 15a, 16, 17, 22, 23
15
15a, 18, 24
1, 3, 9a
3, 4, 4a, 20’
3, 4, 4a, 20
12, 12a
13, 15
15, 15a, 23’
15, 15a, 23
16, 17, 18
4a, 4b, 8a, 9a
21'
21
20'
4b
8a
20
N
H
13
N
H
1
N
16
(S)
(E)
18
(S)
N
9a
O
12a
(S)
10
3
4a
21"
22
O
O
19
23
15a
Milnamide E (116)
117
23'
24
OH
Table 6.3
NMR data for FA salts of milnamide E (116), isolated milnamide A
(90) and referenced milnamide A6 in CD3CN-d3 at 600 MHz
Position
1
3
5
6
7
8
11
12
15
16
19
20
20'
21, 21’, 21”
22
23
23’
24
N-CH3
a1
Milnamide E (29)
δH (mult., J, int.)
3.91 (d, 15.0, 1H)
3.71 (d, 15.0, 1H)
3.12 (s, 1H)
7.58 (d, 8.4, 1H)
6.97 (t, 7.2, 7.8, 1H)
7.04 (t, 7.2, 7.8, 1H)
7.31 (d, 7.8, 1H)
6.62 (d, 9.6, 1H)
4.75 (9.0, 1H)
4.93 (t, 9.6, 10.2, 1H)
6.62 (d, 9.6, 1H)
2.48 (s, 3H)
1.43 (s, 3H)
1.41 (s, 3H)
0.86 (s, 9H)
2.93 (s, 3H)
0.82 (d, 6.6, 3H)
0.75 (d, 6.6, 3H)
1.82 (s, 3H)
Milnamide A (1)
δH (mult., J, int.)
3.92 (d, 14.4, 1H)
3.79 (d, 15.0, 1H)
3.14 (s, 1H)
7.61 (d, 7.8, 1H)
6.99 (t, 7.8, 7.2, 1H)
7.10 (t, 7.2, 1H)
7.32 (d, 7.8, 1H)
6.39 (d, 9.0, 1H)
4.74 (d, 9.6, 1H)
4.92 (t, 9.6, 10.8, 1H)
6.65 (d, 9.0, 1H)
2.53 (s, 3H)
1.43 (s, 3H)
1.42 (s, 3H)
0.84 (s, 3H)
2.90 (s, 3H)
0.82 (d, 6.6, 3H)
0.73 (d, 7.2, 3H)
1.76 (s, 3H)
3.58 (s, 3H)
H-NMR data for 90 referenced from literature 6
118
Milnamide A (1) a
δH (mult., J, int.)
3.93 (d, 1H)
3.79 (d, 1H)
3.15 (s, 1H)
7.61 (d, 1H)
7.00 (t, 1H)
7.12 (t, 1H)
7.33 (d, 1H)
6.51 (d, 1H)
4.76 (d, 1H)
4.98 (t, 1H)
6.66 (d, 1H)
2.53 (s, 3H)
1.43 (s, 3H)
1.41 (s, 3H)
0.85 (s, 3H)
2.92 (s, 3H)
0.83 (d, 3H)
0.75 (d, 3H)
1.78 (s, 3H)
3.59 (s, 3H)
6.3.2 Hemiasterlin D (117)
Compound 117, [α]24D -52.3 (c 0.02, MeOH), was obtained as a white
amorphous solid. The (+)-HRESIMS showed signals for [M+H]+ at m/z 853.5470
and [M+Na]+ at m/z 875.5296 which were consistent for the molecular formula
C46H72N6O9 (calcd (+) m/z 853.5434, Δ 4.2 ppm).
Figure 6.9 The 1H-NMR spectrum of 117 recorded at 600 MHz in DMSO-d6
HSQC spectral data of 117 established the presence of 32 carbons bound to
protons including 7 olefinic carbons, 8 methine and 17 methyl carbons. The 1H-NMR
data indicated the presence of a 1,2-disubstituted benzene ring with four out of seven
olefinic protons at δΗ 8.11 (H-5, d, J = 7.2 Hz), 7.11 (H-6, t, J = 7.2, 7.8 Hz), 7.15
(H-7, t, J = 7.2, 7.8 Hz) and 7.54 (H-8, d, J = 8.4 Hz).20 Quaternary carbons at C-4
and C-9 were located at δC 125.5 and 136.7 ppm, respectively due to the 3JCH HMBC
correlations of the protons in the benzene ring. The other singlet aromatic proton H-2
(δΗ 7.20 ppm, δC 125.1 ppm) also correlated with C-3, C-4 and C-9 suggesting the
establishment of a 3-substituted indole system (Figure 6.10).20 In addition, the
HMBC correlation from H-5 to C-3 (δC 117.7 ppm) gave further evidence
confirming the assignment of the fragment A.
A spin system of CH-NH-CH3 was deduced due to the COSY correlations
from an exchangeable proton NH (δH 8.88 ppm) to H-11 (δH 4.47 ppm) and H-21 (δH
2.24 ppm) as well as the downfield resonances of a methine carbon C-11 (δC 66.7
119
ppm) and a methyl C-21 (δC 33.2 ppm). Further analysis of the COSY spectral data
demonstrated that the proton at δH 8.88 ppm also correlated with another
exchangeable proton at δH 7.51 ppm indicating the -NH- linkage was protonated as a
TFA salt. The observation of HMBC correlations from protons in two geminal
methyl groups H-22 (δH 1.37 ppm) and H-22’ (δH 1.38 ppm) to carbons at δC 66.7
and 37.7 ppm led to the location of the quaternary carbon C-10 (δC 37.7 ppm).
Therefore, fragment B was elucidated (Figure 6.10). In addition, the proton H-22
also had the 3JCH correlation with carbon C-3 in fragment A. The ROESY spectral
data indicated that H-22’ correlated with H-2. All information was evident for the
connection of fragment B to fragment A (Figure 6.10).
Fragment C was confirmed based on the HMBC correlation from a singlet
proton at δH 6.34 ppm, corresponding to a methine carbon C-27 (δC 76.2 ppm), to the
carbonyl C-28 (δC 167.7 ppm) (Figure 6.10). HMBC signals from H-27 to C-2 and
C-9 as well as ROESY correlations from H-27 to H-2 and H-8 further confirmed the
assemblage of fragment C to fragment A (Figure 6.10).
Figure 6.10 The partial structures A, B, C and combined C-A-B of compound 117
A typical tert-butyl group containing three methyl carbons at δC 26.1 ppm (C23, 23’ and C-23”) and the quaternary carbon at δC 34.7 ppm (C-14a) were
elucidated due to HMBC correlation from H-23 (δH 1.01 ppm) to C-14a. The
downfield resonance H-14 (δH 4.87 ppm) correlated with an exchangeable proton H13 (δH 8.90 ppm) with an 3JHH coupling constant of 8.4 Hz and showed a strong
HMBC signal to the carbonyl C-15 (δC 170.0 ppm) suggesting this methine group
was an α-carbon of an amino acid residue. The observation of HMBC correlations
from H-14 to C-14a and C-23 along with ROESY correlations from H-13 and H-14
to H-23 led to the assignment of a tert-leucine subunit. The presence of the Nmethylvinylogous valine residue was again observed in 117 by the COSY
correlations from δH 6.68 ppm (H-18) to δH 4.93 ppm (H-17); from δH 2.03 ppm (H17a) to both H-17 and two methyl groups at δH 0.81 and 0.74 ppm (H-25 and H-25’)
as well as the 3JCH long range correlations from H-17 to C-19 (δC 131.8 ppm); from
120
H-18 to C-20 (δC 168.6 ppm) and C-26 (δC 13.2 ppm). The position of the N-methyl
group (δH 3.04 ppm and δC 30.9 ppm) was deduced by the observation of the HMBC
signal from H-24 to C-17 and the ROESY correlation between H-24 and H-18. The
HMBC correlation from H-24 to C-15 as well as the ROESY correlation between H24 and H-14 evidenced for the assemblage of the tert-leucine and Nmethylvinylogous valine residues in the fragment D (Figure 6.11).
Another set of moiety containing the tert-leucine and N-methylvinylogous
valine residues (Figure 6.11) which was similar to fragment D was also established
based mainly on the HMBC, COSY and ROESY correlations (fragment E).
Figure 6.11 The partial structures D and E of compound 117
No HMBC correlation was observed to a carbonyl carbon C-12 (δC 165.5
ppm) except for the exchangeable proton H-13. However, this carbon C-12 was
supposed as a linkage of fragment B and D since the observation of the ROESY
correlation between H-11 to H-13 and also the typical downfield shift of the αcarbon in an amino acid residue (Figure 6.12).
Figure 6.12 Key HMBC and COSY correlations to establish the structure of 117
Further analysis of the 2D-NMR data for the proton H-29 in fragment E
revealed that this exchangeable proton had the HMBC correlations to C-28 and
correlated with H-27 in the ROESY spectrum. Consequently, fragment E was
assembled to fragment C at the carbonyl C-28. A hydroxyl group was suggested to
bind directly to C-27 due to the downfield resonance of the proton H-27 (δH 6.34
ppm) and ensuring the molecular weight for this compound. The complete structure
of 117 was finally established (Figure 6.12).
121
Determination of absolute stereochemistry of compound 117
L-tert-Leu
D-tert-Leu
33.49
33.54
35.12
Standard
Hemiasterlin D (117)
Hydrolysis of 117 with 6N HCl was performed in 8 hours followed by
reaction with the Marfey’s reagent and LC/MS analysis.22,23 The presence of two Ltert-Leu residues in 117 was established since only a single peak corresponding to Ltert-Leu was observed by ion-selective monitoring for its FDAA derivatives.
Attempts to determine the configurations of other units in 117 by CD analysis were
unsuccessful. In order to assign the absolute stereochemistry of the Nmethylvinylogous valine moiety, Coleman et al.3 used ozonolysis to transform the Nmethylvinylogous valine moiety to N-methylvaline prior to employing Marfey
analysis. However this approach was unable to explore with the limited material in
hand (0.8 mg). It is noticed that 117 has the same sign of optical rotation with
hemiasterlin (89) ([α]25D -72.6 (c 0.06, MeOH), lit. [α]25D -95 (c 0.06, MeOH)1),
hemiasterlin A (91) ([α]25D -60.3 (c 0.25, MeOH), lit. [α]D -45 (c 0.25, MeOH)3) and
hemiasterlin C (93) ([α]D -18.8 (c 0.11, MeOH)7 (literature did not report the optical
value of hemiasterlin B (92)3). From the specific rotation, the absolute configurations
assigned for two L-tert-Leu residues as well as similar NMR data, it can be assumed
that chiral centres in two dipeptide fragments containing tert-leucine and Nmethylvinylogous valine and also in the trimethylated tryptophan residue are similar
to those in other hemiasterlin analogues. Presently, stereochemical information at C27 has been unidentified. However, possessing a side chain containing 2hydroxyacetic acid, tert-leucine and N-methylvinylogous valine residues attached to
the indole nitrogen, hemiasterlin D is known as a first peptide skeleton discovered in
nature.
23'
23
22
4
N
40
HO
33a
39
O
36
O
12
N
H
NH
O
18
N
14
20
O
F3 C
O
O
37"
37
37'
Hemiasterlin D (117)
122
26
17a
25
21
N
38
24
NH2
2
OH
30
34
10
27
O
23"
O
15
3
9
39'
22'
25'
OH
Table 6.4
NMR data for TFA salt of hemiasterlin D (117) in DMSO- d6
Position
δC
2
3
4
5
6
7
8
9
10
11
12
13
14
14a
15
17
17a
18
19
20
21
22
22’
23, 23’, 23”
24
25
25’
26
27
28
29
30
30a
31
33
33a
34
35
36
37, 37’, 37”
38
39
39’
40
NH
125.2
117.9
125.8
120.2
118.9
121.1
110.8
136.8
37.7
66.7
165.8
55.3
34.7
170.3
56.0
28.5
137.9
131.8
168.6
33.2
26.7
21.8
26.1
30.9
18.9
18.9
13.2
76.2
168.0
53.8
34.8
170.6
55.4
28.4
137.8
132.0
168.5
25.7
30.3
18.5
18.5
13.2
-
δH
(mult., J, int.)
7.20 (s, 1H)
8.11 (d, 7.2, 1H)
7.11 (t, 7.2, 1H)
7.15 (t, 7.2, 7.8, 1H)
7.54 (d, 8.4, 1H)
4.47 (d, 10.2, 1H)
8.90 (d, 7.8, 1H)
4.87 (d, 8.4, 1H)
4.93 (t, 9.6, 10.2, 1H)
2.03 (m, 1H)
6.68 (d, 9.0, 1H)
2.24 (t, 4.8, 3H)
1.37 (s, 3H)
1.38 (s, 3H)
1.01 (s, 9H)
3.04 (s, 3H)
0.81 (d, 6.0, 3H)
0.80 (d, 6.0, 3H)
1.80 (s, 3H)
6.34 (s, 1H)
8.10 (br.s, 1H)
4.73 (d, 9.0, 1H)
5.01 (t, 10.2, 1H)
1.97 (m, 1H)
6.64 (d, 9.6, 1H)
0.88 (s, 9H)
2.94 (s, 3H)
0.79 (d, 6.6, 3H)
0.74 (d, 6.0, 3H)
1.80 (s, 3H)
8.88 (bs, 1H)
7.51 (bs, 1H)
gCOSY
(H no.)
6
5, 7
6, 8
7
NH
14
13
17a, 18
17, 25, 25’
17
NH
30
29
33a, 34
33
33
11
-
123
ROESY
(H no.)
22, 27
10, 22
27
4, 13, 21, 22
11, 23
24, 23
25, 26
18, 24
24, 17a, 25
11
5, 11
2
14, 24
14, 17a, 18, 23
17
2, 8, 29
27, 37
37, 38
39, 40
34, 38
33a, 38
29, 30, 38, 40
30, 33a, 34, 37
33
-
gHMBC
(C no.)
3, 4, 9, 10 , 27
7, 9
4, 8
5, 9
4, 6
12
12, 14a, 15, 23
18, 19
20, 26
11
3, 10, 11, 22’
3, 10, 11, 22
14, 14a
15, 17
17, 17a
17, 17a
18, 19, 20
2, 9, 28
28
30a, 31, 37
31, 33a, 34, 35
36, 40
30, 30a
31, 33
33, 33a
33, 33a
34, 35, 36
-
6.4 Evaluation of “drug-like” properties
In comparison with neamphamide B-D (86-88) which were predicted to be
very hydrophilic, almost all the peptides isolated from the sponge Pipestela
candelabra have logPs ranging from 0.51 to 4.21 satisfying drug-like criteria. Three
compounds (89, 90 and 117) are assigned to be more flexible than the others since
their number of rotatable bonds were over 10. Although peptides were cited as an
exception to the Ro5, the results of “drug-like” evaluation here indicate that eight
compounds pass the physico-chemical requirements (Table 6.5).
Table 6.5 Physico-chemical properties of the isolatedcompounds
Compound
Formula
MW
LogP
HBD
HBA
PSA
89
90
91
94
95
101
102
103
115
116
117
C30H46N4O4
C31H46N4O4
C29H44N4O4
C31H44N4O5
C31H45N4O4
C27H38IN3O6
C27H38BrN3O6
C27H38ClN3O6
C36H45BrN4O6
C30H44N4O4
C46H72N6O9
526.71
538.72
512.68
552.70
537.71
627.51
580.51
536.06
709.67
524.69
853.10
0.74
2.77
0.51
4.21
1.35
3.38
3.32
3.06
5.04
2.74
2.31
3
2
4
2
2
3
3
3
4
3
6
5
5
5
5
4
6
6
6
6
5
10
103.67
94.88
114.53
111.95
94.65
125.04
125.04
125.04
140.83
125.74
210.61
NROT
Predicted
Bioavailability
11
8
11
8
8
2
2
2
3
8
20
9
9
9
9
9
9
9
6.5 Biological activity
Cytotoxicity of geodiamolides E-F (101-103) and jaspamide (115) towards a
panel of sixty cancer cell lines were previously studied in National Cancer Institute
(NCI).24 These depsipeptides demonstrated an extreme activity against all cancer
cells tested. Thus, these compounds were not evaluated anticancer activity in this
research. Here, cytotoxic activity of milnamides and hemiasterlin families was
evaluated against four cancer cell lines and two non cancer cell lines.
Table 6.6 Cytotoxicity evaluation of isolated compounds
IC50 ± SE (nM)a
Compound
A549
HeLa
LNCaP
PC3
HEK
NFF
<1.7
<1.7
<1.7
<1.7
<1.7
<1.7
89
90
736.1±30.1 1326.7±430.0
572.3±10.3
2660.7±128.9
1408.0±517.0
2604.8±469.5
<1.7
<1.7
<1.7
<1.7
<1.7
91
12.3±3.2
94
3256.0±136.6
2102.2±615.1
1629.2±243.6
12490.2±197.3 3674.5±469.5 7345.1±143.7
95
418.5±13.5
747.8±59.8
927.6±20.3
564.0±33.6
417.3±59.6
2468.1±577.0
116
248.2±50.0
412.6±25.1
287.0±1.5
441.4±26.1
498.3±24.7
1041.9±180.0
117
5.0±1.8
2.2±0.9
3.5±1.2
8.2±2.9
11.1±0.9
1.8±0.6
Vincristine
20.2±3.1
7.2±1.2
6.1±0.8
15.3±2.9
18.2±2.5
16.9±3.1
sulphate
a
Each IC50 or % Inhibition was determined as the mean ± SE of two independent experiments with triplicate determinations for
each concentration
A new compound, hemiasterlin D (117) showed the highest potency against
HeLa cells with an IC50 of 1.8 nM and was less active against other cells with IC50
values ranging from 2.2 to 11.1 nM. In comparison with 117, milnamide E (116)
displayed approximately 100 fold less cytotoxicity and also showed the most
124
inhibition against HeLa cells (IC50 of 248.2 nM). When an N-methyl was introduced
in the β-carboline ring converting 116 to 90, the activity was reduced 2-6 times
suggesting the N-9 of the β-carboline may be one of the important positions affecting
the cytotoxicity of milnamide compounds.
Among three milnamides derivatives (90, 94 and 95) possessing the same Nmethyl-β-carboline but different functional groups at C-1, milnamide C (94) was
observed to exhibit the weakest activity against all six cell lines while milnamide D
(95) displayed the most activity with 2-4 fold more potency than 90 and 5-10 fold
stronger than 94. This indicated that a positive charge at the N-2 may have an
influence on the inhibition of the cell growth by interacting with a protein target.
Replacing the methylene in 90 or an imine in 95 with a carbonyl group to give 94 led
to a decrease in activity suggesting that the substitution at this carbon interferes with
the binding of the compound to an active site on a protein.
With nanomolar IC50 values observed for three hemiasterlin compounds, it
was worth noting that the opening of the β-carboline system in milnamides to give
the hemiasterlin skeleton was beneficial for cytotoxic activity. Here our results again
confirmed the potent cytotoxicity of hemiasterlin (89) and hemiasterlin A (91) as
previous reports.3,7 On the examination of hemiasterlin’s bioactivity, Anderson et al.9
found these cytotoxic tripeptides induced mitotic arrest and produced abnormal
mitotic spindles. Structure activity relationships (SAR) of hemiasterlin analogues
were indicated that the tryptophan β-dimethylation and N-methylation on the αamino group, the double bond Δ18,19 and C-17 isopropyl substituent on the
vinylogous valine residue were extremely critical for potent biological activity.
Moreover, the L-configurations of all amino acid residues in the peptides were also
required for maximal cytotoxicity.12,13 It is interesting to note that the previous SAR
investigation had only examined hydrogen and a methyl group at the indole nitrogen
and 89 was found 7-fold stronger activity than 91.12 Although a new hemiasterlin D
(117) here was less potent than its two analogues, the results showed that
cytotoxicity was still tolerated when replacing the N-methylindole of the
tetramethyltryptophan residue with a long side chain. This result may offer a new
position (N1) for hemiasterlin (89), which is in clinical trial,14 to be developed for
more selective activity and bioavailability.
125
6.6 References
(1)
Talpir, R.; Benayahu, Y.; Kasman, Y. Tetrahedron Lett. 1994, 35, 44534456.
(2)
Crews, P.; Farias, J. J.; Emrich, R.; Keifer, P. A. J. Org. Chem. 1994, 59,
2932-2934.
(3)
Coleman, J. E.; Silva, E. D. d.; Kong, F.; Andersen, R. J. Tetrahedron 1995,
51, 10653-10662.
(4)
Coleman, J. E.; Patrick, B. O.; Andersen, R. J.; Rettig, S. J. Acta Cryst. 1996,
C52, 1525-1527.
(5)
Sonnenschein, R. N.; Farias, J. J.; Tenney, K.; Mooberry, S. L.; Lobkovsky,
E.; Clardy, J.; Crews, P. Org. Lett. 2004, 6, 779-782.
(6)
Liu, C.; Masuno, M. N.; MacMillan, J. B.; Molinski, T. F. Angew. Chem. Int.
Ed. 2004, 43, 5951-5954.
(7)
Gamble, W. R.; Durso, N. A.; Fuller, R. W.; Westergaard, C. K.; Johnson, T.
R.; Sachkett, D. L.; Hamel, E.; II, J. H. C.; Boyd, M. R. Bioorg. Med. Chem.
1999, 7, 1611-1615.
(8)
Chevallier, C.; Richardson, A. D.; Edler, M. C.; Hamel, E.; Harper, M. K.;
Ireland, C. M. Org. Lett. 2003, 5, 3737-3739.
(9)
Anderson, H. J.; Coleman, J. E.; Andersen, R. J.; Roberge, M. Cancer
Chemother. Pharmacol. 1997, 39, 223-226.
(10)
Bai, R.; Durso, n. A.; Sackett, D. L.; Hamel, E. Biochemistry 1999, 38,
14302-14310.
(11)
Andersen, R. J.; Coleman, J. E.; Piers, E.; Wallace, D. J. Tetrahedron Lett.
1997, 38, 317-320.
(12)
Nieman, J. A.; Coleman, J. E.; Wallace, D. J.; Piers, E.; Lim, L. Y.; Roberge,
M.; Andersen, R. J. J. Nat. Prod. 2003, 66, 183-199.
(13)
Yamashita, A.; Norton, E. B.; Kaplan, J. A.; Niu, C.; Loganzo, F.;
Hernandez, R.; Bayer, C. F.; Annable, T.; Musto, S.; Discafani, C.; Zark, A.;
Kaloustian, S. A. Bioorg. Med. Chem. Lett. 2004, 14, 5317-5322.
(14)
Kingston, D. G. I. J. Nat. Prod. 2009, 72, 507-515.
(15)
Coleman, J. E.; Soest, R. V.; Andersen, R. J. J. Nat. Prod. 1999, 62, 11371141.
(16)
Silva, E. D. d.; Andersen, R. J.; Allen, T. M. Tetrahedron Lett. 1990, 31, 489492.
126
(17)
Alvarez, B.; Hooper, J. N. A.; Soest, R. W. M. V. Mem. Queensl. Mus. 2008,
52, 105-118.
(18)
Hooper, J.; Soest, R. V. Systema Porifera: A Guide to the Classification of
Sponges; Kluwer Academic/Plenum: New York, 2002.
(19)
Zabriskie, T. M.; Klocke, J. A.; Ireland, C. M.; Marcus, A. H.; Molinski, T.
F.; Faulkner, D. J.; Xu, C.; Clardy, J. C. J. Am. Chem. Soc. 1986, 108, 31233124.
(20)
Silverstein, R. M.; Webster, F. X.; Kiemle, D. Spectrometric Identification of
Organic Compounds 7th ed.; John Wiley & Son, 2005.
(21)
Breitmaier, E. Structure Elucidation by NMR in Organic Chemistry. A
practical guide; 3rd ed.; John Wiley & Son, 2002.
(22)
Oku, N.; Gustafson, K. R.; Cartner, L. K.; Wilson, J. A.; Shigematsu, N.;
Hess, S.; Pannell, L. K.; Boyd, M. R.; McMahon, J. B. J. Nat. Prod. 2004,
67, 1407-1411.
(23)
Marfey, P. Carlsberg Res. Commun. 1984, 49, 591-596.
(24)
http://pubchem.ncbi.nlm.nih.gov/assay.
127
128
Chapter 7
Bromotyrosine alkaloids from the Australian sponges
Suberea clavate and Pseudoceratina sp.
7.1 Introduction
Bromotyrosine derivatives are common secondary metabolites from sponges in
the order Verongida and some marine ascidians.1 There have been so far over 280
bromotyrosine marine alkaloids reported, out of which over 90% compounds were
derived from four sponge families Aplysinidae, Aplysinellidae, Ianthellidae and
Pseudoceratinidae.2 Depending upon the rearrangement or partial reduction of aromatic
rings or side chains, structures of bromotyrosine alkaloids have been divided into six
main classes including simple bromotyrosine derivatives, oximes, bastadins,
spirooxepinisoxazolines, spirocyclohexadienylisoxazolines and other structural classes
in the families geodiamolides, jaspamides, polyandrocarpamides, polycitones,
polycitrins and chelonins.3
Table 7.1
Genus
Agelas
Aiolochroia
Anomoianthella
Aplysina
Aplysinella
Druinella
Ianthella
Iotrochota
Jaspis
Oceanapia
Poecillastra
Psammaplysilla
Pseudoceratina
Suberea
Thorectopsamma
Trachyopsis
Tylodina
Verongia
Verongula
Taxonomy of sponges producing bromotyrosine alkaloids
Species
4
oroides
crassa5
popeae6
aerophoba,7 archeri,8 caissara,9 cavernicola,10 fistularis forma
fulva,11 geradogreeni,12 insularis,13 lacunosa,14 laevis,15 thiona.6
rhax,16
sp.17
ardis,18 basta,19 flabelliformis,20 quadrangulata,21
birotulata,22
wondoensis,23
sp.24
wondoensis,23
arabica,25 purpurea,26 purea,27
crassa,28 durissima,29 verrucosa,30
praetensa,31
xana,32
aplysinoides,33
fungina34
aerophoba,35 archeri,36 aurea,37 cauliformis,38 cavernicola,39
fistularis,40 lacunosa,41 rigida,42
gigantea,43
Structures of simple bromotyrosine alkaloids contain only one full or partial
reduced aromatic ring (Figure 7.1). The first member of this class was 2,6-dibromo-4-
129
acetamide-4-hydroxycyclohexadienone (118) isolated from the methanol extract of the
sponge Verongia fistularis in 1967.40 Aeroplysinin-1 (119) from V. aerophoba,18 the
first naturally 1,2-dihydroarene-1,2-diol skeleton, was also classified in this structural
class. Its absolute stereochemistry was solved by an X-ray diffraction analysis.18,44
Aeroplysinin-2 (120), the first bromotyrosine alkaloid with a lactone functionality, was
characterised on the basis of NMR experiments and chemical synthesis.18,45,46 In
addition, a series of aplysinadiene (121) and its analogues containing an α,β-unsaturated
side chain, 2-(3,5-dibromo-4-hydroxyphenyl)acetic acid (122) and other aromatic
bromotyrosine derivatives as well as N,N-dimethyl-3’,5’-dibromotyramine (123) and its
tyramine derivatives were also classified as members of simple bromotyrosine
alkaloids.
OCH3
O
Br
Br
OCH3
Br
Br
HO
NH2
O
HO
HO
HO
O
CN
O
2,6-dibromo-4-acetamide-4
(+)-aeroplysinin-1 (119)
-hydroxycyclohexadienone (118)
Br
Br
H
Br
Br
O
aeroplysinin-2 (120)
Br
O
HO
HO
HO
COOH
Br
aplysinadiene (121)
Br
N
Br
2-(3,5-dibromo-4-hydroxy
phenyl)acetic acid (122)
N,N-dimethyl-3',5'dibromotyramine (123)
Figure 7.1 Skeletons represented for simple bromotyrosine alkaloids
Biosynthetically, oxime bromotyrosine alkaloids have been known as
derivatized products of the amine functionality.3 Three structural groups were
categorised for this class. The first group contains a bromotyrosine oxime joining to a
histamine unit (verongamine (124)47). The second group have a bromotyrosine oxime
cooperating with a bromotyramine moiety or a side chain of two or three carbons
(aplysamine 4 (125)48). The last group consists of one or two bromotyrosine oximes
connecting with a disulfide chain in a cysteine residue ((E,E)-psammaplin A (126) and
(E,Z)-psammaplin A (127)49). The geometries of the oxime functionality were
determined by observing the chemical resonance of C-7 in the
13
C-NMR data. If its
chemical shift is about 35 ppm, a Z-configuration will be elucidated. In cases the carbon
resonance is from 27 to 28 ppm, an E-configuration should be assigned.49 So far, the Egeometry has been determined for most bromotyrosines in this class (Figure 7.2).3
130
Br
H3CO
H3CO
HO
HO
N
N
(E)
7
Br
δC 28.7
(E)
H
N
7
H
N
Br
Br
NH
δC 28.8 O
N
O
O
NH2
Br
verogamine (124)
aplysamine 4 (125)
OH
O
δC 26.7
Br
S
7
(E)
N
H
(E)
S
N
HO
OH
N
H
N
O
7'
Br
δC 26.7
HO
psammaplin A (126)
HO
O
δC 27.5
Br
S
7
(E)
N
H
(Z)
7'
S
N
HO
OH
N
H
N
O
Br
δC 35.7
HO
psammaplin A (127)
Figure 7.2 Skeletons represented for oxime bromotyrosine alkaloids
The bastadin class (Figure 7.3) has been named for secondary metabolites
containing two (hemibastadins) or four bromotyrosine units (bastadins).3 This class was
mostly found from sponges in the genera Ianthella20,21,50-52 (hemibastadin 1 (128)19 and
bastadin 1 (129)53) and Psammaplysilla54 and also from the ascidian Botryllus.55
Br
OH
OH
N
H
N
Br
O
O
OH
HO
OH
N
Br
Br
HO
O
H
N
Br
Br
O
N
H
HO
N
OH
hemibastadin 1 (128)
bastadin-1 (129)
Figure 7.3 Skeletons represented for bastadin alkaloids
Searching for the bromotyrosine structural core using Dictionary of Natural
Products (version 2010) resulted in eight compounds in the spirooxepinisoxazoline
bromotyrosine class consisting of psammaplysins A-F and ceratinamides A-B.56-60
Among these compounds, psammaplysin A (130) and B (131) from the Red Sea sponge
Psammaplysilla purpurea were first identified.56 Not until 1985, the final structure of
the
spirooxepinisoxazoline
skeleton
was
completely
solved
as
a
spiro[4.6]dioxazundecane after single crystals of psammaplysin A acetamide acetate
(132) were developed and characterised by X-ray diffraction method (Figure 7.4).61
131
Br
O
O
Br
N
H3CO
H
N
Br
O
HO
O
Br
NHR2
R1
psammaplysin A (130):
R1 = H ; R2 = H
psammaplysin B (131):
R1 = OH ; R2 = H
psammaplysin A acetamide acetate (132): R1 = H ; R2 = Ac
Figure 7.4 Skeletons represented for spirooxepinisoxazoline bromotyrosine alkaloids
Spirocyclohexadienylisoxazoline alkaloids have been considered as the typical
class in bromotyrosine derivatives derived from marine sponges (Figure 7.5).3 This type
of
skeleton
includes
mono-spirocyclohexandienylisoxazolines
spirocyclohexandienylisoxazolines.
and
bis-
Mono-spirocyclohexandienylisoxazolines
have
been divided into four groups, simple mono-spirocyclohexandienylisoxazolines
(purealidin R (133)27), linear side chain mono-spirocyclohexadienylisoxazolines
(purealidin
L
(134)27),
(araplysillin-1(135)25)
bromotyrosine
and
mono-spirocyclohexandienyl-isoxazolines
histamine
mono-spirocyclohexandienylisoxazolines
(aerophobin-1 (136)62). In comparison between the last two groups, the bromotyrosine
mono-spirocyclohexandienylisoxazoline
spirocyclohexandienylisoxazoline
bromotyramine
residue
while
skeleton
moiety
attached
structures
to
of
contains
a
bromotyrosine
the
histamine
a
or
a
mono-
spirocyclohexandienylisoxazolines have been defined as containing a spirocyclohexandienylisoxazolines
and
a
histamine
or
2-amino-homohistamine
unit.
Bis-
spirocyclohexandienylisoxazoline alkaloids are derived from bromotyrosine with two
spirocyclohexandienylisoxazoline rings connecting through a nitrogenous side chain
(caissarine B (137)9) or a bromotyramine unit (19-deoxyfistularin-3 (138)39). Relative
stereochemistry of the spirocyclohexandienylisoxazolines can be determined by
observing the chemical shifts of H-1 and H-7. Trans-configuration is deduced from the
resonances at δH 4.2 ppm for H-1 and at δH 3.8 and 3.1 ppm for H-7. In case, if H-1 and
H-7 resonance at δH 4.2 and 3.42 ppm (2H), cis-isomer will be assigned. Since absolute
configurations of aerothionin (145) as (1R,6S) were determined by the X-ray diffraction
analysis,63 its CD spectrum with two prominent positive Cotton Effects at 248 and 285
nm became a conventional reference to elucidate absolute stereochemistry of other
spirocyclohexandienylisoxazoline bromotyrosine derivatives.63-65
132
OCH3
Br
(1R)
OCH3
OCH3
Br
Br
Br
Br
Br
(6S)
HO
HO
HO
O
O
7
N
N
NH2
Br
O
NH
H
N
O
N
N
H
NH2
O
purealidin R (133)
NH2
H
N
O
O
purealidin L (134)
Br
araplysillin-1 (135)
OCH3
OCH3
OCH3
Br
Br
Br
HO
Br
Br
Br
HO
O
N
O
H
N
H
N
O
O
OH
H
N
N
N
H
N
O
OH
N
O
aerophobin-1 (136)
caissarine B (137)
OCH3
Br
Br
HO
O
OH
H
N
N
Br
O
O
O
Br
N
N
H
O
OH
19-deoxyfistularin-3 (138)
Br
Br
OCH3
Figure 7.5 Skeletons represented for spirocyclohexadienylisoxazoline alkaloids
Biogenesis of the spirooxepinisoxazoline and spirocyclohexadienylisoxazoline
subunits was proposed by Clardy et al. in 1985.61 The bromotyrosine residue is
transformed into the spirooxepinisoxazoline or spirocyclohexadienylisoxazoline ring
under an oximino epoxidation and followed by epoxide ring opening as shown in
arrows for the split line from structure 3 to the following structures (Scheme 7.1).
OH
Br
HO
Br
HO
Br
Br
Br
OH
Br
Br
Br
Br
R
R
N
R
N
H
OH
OH
N
HO
N
O
O
O
O
O
Br
O
NH2
HO
O
OH
OH
O
O
O
OH
O
OH
HO
OH
O
N
Br
Br
Br
Br
R
HO
O
O
HO
R
N
R
N
O
O
Scheme 7.1 Proposed biosynthesis of spirooxepinisoxazoline and
spirocyclohexadienylisoxazoline rings61
133
Br
Br
Bromotyrosine alkaloids have been known to possess a wide range of
bioactivities including the histamine H2 and H3 and the adrenergic α1D and α2A receptor
antagonists,66,67 anti-human immunodeficiency virus 1 (HIV-1),68 anti-angiogenic,69
antimicrobial,70,71 antifouling,60 isoprenylcysteine carboxyl methyltransferase11,12 and
mycothiol-S-conjugate amidase24,72 inhibitory, cytotoxic73 and anticancer16-18 activities.
Among these biological activities, antimicrobial and cytotoxicity were considered as
two typical activities of bromotyrosine derivatives.3
This chapter reports the chemistry of two marine sponges in the family
Pseudoceratinidae, Suberea clavate and Pseudoceratina sp. along with the cytotoxic
evaluation for all isolated compounds. Ten known compounds were isolated from the
Australian sponge Suberea clavate including purealidin L (134),27 N,N,N-trimethyl-3,5dibromotyrosine (139),74 purealidin O (140),27 aerophobins 2 (141),62 aplysinamisine 2
(142),75 11,19-dideoxyfistularin 3 (143),29 11-hydroxyaerothionin (144),29 aerothionin
(145),76 homoaerothionin (146),76 fistularin 3 (147).77,78 Chemical investigation of the
Australian
sponge
Pseudoceratina
sp.
afforded
two
new
compounds,
pseudoceralidinone A (148) and aplysamine 7 (149) together with six known
compounds, aerophobins 2 (141),62 fistularin 3 (147),77,78 3-methylmaleimide-5-oxime
(150),79
5-[3,5-dibromo-4-[(2-oxo-5-oxazolidinyl)methoxy]phenyl]-2-oxazolidinone
(151),41 fistularin 2 (152)77 and (3,5-dibromo-2-hydroxy-4-methoxyphenyl)-acetic acid
(153).12 Their cytotoxicity was also evaluated against cancer and non-cancer cell lines.
134
List of bromotyrosine alkaloids isolated from the Australian sponges
Suberea clavate and Pseudoceratina sp.
OCH3
OCH3
OH
Br
Br
Br
Br
HO
O
Br
Br
HO
O
H
N
NH
H
N
N
N
N
H
NH2
N
N,N,N-trimethyl-3,5
-dibromotyrosine (139)
purealidin L (134)
OCH3
purealidin O (140)
OCH3
OCH3
Br
Br
NH2
HO
O
N
Br
HO
O
H
N
Br
Br
HO
N
NH
HO
O
Br
NH2
N
H
NH
H
N
N
O
O
aerophobins 2 (141)
O
H
N
NH2
N
O
NH
aplysinamisine 2 (142)
Br
Br
O
H
N
O
N
H
OH
N O
HO
Br
11-hydroxyaerothionin (144)
OCH3
OCH3
Br
Br
Br
Br
H3CO
H
N
Br
OH
O
O
Br
HO
N
O
O
Br
O
OH
N
H
N
H
N
N
O
O
O
N
H
O
Br
Br
N O
HO
aerothionin (145)
11,19-dideoxyfistularin 3 (143)
Br
OCH3
OCH3
Br
OCH3
Br
Br
Br
Br
Br
H3CO
OH
HO
O
Br
OH
O
O
O
OH
N
O
N
O
N
H
N
O
OH
homoaerothionin (146)
Br
fistularin 3 (147)
Br
N
H
N
Br
O
O
H
N
H
N
N
OH
Br
O
N
O
O
N
Br
Br
Br
NH
N
H
O
OH
N
H
OH
N
OH
O
pseudoceralidinone A (148)
O
O
aplysamine 7 (149)
3-methylmaleimide
-5-oxime (150)
OCH3
Br
HN
Br
OCH3
Br
O
Br
Br
HO
O
O
O
OH
H
N
N
Br
Br
NH
OH
O
O
O
O
5-[3,5-dibromo-4-[(2-oxo-5-oxazolidinyl)
methoxy]phenyl]-2-oxazolidinone (151)
Br
Fistularin 2 (152)
135
NH
O
O
O
(3,5-dibromo-2-hydroxy-4
-methoxyphenyl)-acetic acid (153)
7.2 Collection, Extraction and Isolation
7.2.1 Collection, Extraction and Isolation for the sponge Suberea clavate
A specimen of Suberea clavate was collected at the depth of 18m, Fairfax
Island, northwest side of reef, Queensland, Australia in 1998. It was identified as
Suberea clavate (phylum Porifera, class Demospongiae, order Verongida, family
Pseudoceratinidae). A voucher specimen QMG314013 has been deposited at the
Queensland Museum, South Brisbane, Queensland, Australia.
Plate 7.1 Photograph of the sponge Suberea clavate
A freeze dried sample of Suberea clavate (5g) was extracted exhaustively with
hexane (250 ml), dichloromethane (250 ml) and methanol (2 x 250 ml), respectively.
The dichloromethane and methanol extracts were combined and then evaporated
solvents to yield a yellow residue (1.0 g). This crude extract was pre-absorbed onto C18
(1.0 g) and packed dry into a small cartridge, which was connected to a C18 preparative
HPLC column (5 μm, 21.2 x 150 mm). A linear gradient from 100% water (0.1% TFA)
to 100% methanol (0.1% TFA) was performed over 60 minutes at a flow rate of 9
ml/min and 60 fractions (1.0 minute each) were collected. Compounds 139 (12.0 mg,
0.24% dry wt), 140 (1.5 mg, 0.03% dry wt) and 143 (11.0 mg, 0.22% dry wt) were
obtained from the first separation in fractions 16 and 17; 33; 41 and 42, respectively.
Fractions 21 to 29 were combined and chromatogramed on a Betasil C18 column (5 μm,
250 x 10 mm) from 20% methanol (0.1% TFA) – 80% water (0.1% TFA) to 60%
methanol (0.1% TFA) – 40% water (0.1% FA) in 60 minutes. Compounds 134 (2.0 mg,
0.04% dry wt), 141 (0.5 mg, 0.01% dry wt) and 142 (0.8 mg, 0.016% dry wt) were
purified. Mass-guided identification demonstrated fractions 35 to 40 contained the ion
peaks of interest in (-)-LRESIMS at (-)m/z 817, 831, 833 and 1112. These fractions
were then combined and loaded onto a Betasil C18 column (5 μm, 150 x 21.2 mm) using
a linear gradient from 55% methanol (0.1% TFA) – 45% water (0.1% FA) to 100%
136
methanol (0.1% TFA) in 60 minutes. Compounds 144 (7.0 mg, 0.14% dry wt), 145
(12.0 mg, 0.24% dry wt), 146 (2.0 mg, 0.04% dry wt) and 147 (1.0 mg, 0.02% dry wt)
were then isolated (Scheme 7.2).
137
Suberea clavate (5.0g)
a)
DCM/MeOH extract (1.0g)
b)
Compound 139
(12.0mg, 0.24%)
Fractions 16-17
Compound 140
(1.5mg, 0.03%)
Fraction 32
Fractions 125-132
Fractions 125-132
(-)-m/z 817, 831, 833, 1112
Compound 143
(11.0mg, 0.22%)
Fractions 41-42
c)
d)
Compound 134
(2.0mg, 0.04%)
Fractions 20-21
Compound 142
(0.8mg, 0.016%)
Fractions 33-34
Compound 141
(0.5mg, 0.01%)
Fraction 30
Compound 145
(12.0mg, 0.24%)
Fractions 26-28
Compound 144
(7.0mg, 0.14%)
Fractions 23-24
Compound 147
(1.0mg, 0.02%)
Fraction 35
Compound 146
(2.0mg, 0.04%)
Fraction 45
a) Extraction with DCM and MeOH
b) A linear gradient from 100% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes, a flow rate of 9.0 ml/min, 1.0 min/fraction
c) A linear gradient from 20% methanol (0.1% TFA) – 80% water (0.1% TFA) to 60% methanol (0.1% TFA) – 40% water (0.1% TFA) in 60 minutes, a flow
rate of 4.0 ml/min, 1.0 min/fraction
d) A linear gradient from 55% methanol (0.1% TFA) – 45% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minute, a flow rate of 9.0 ml/min, 1.0
min/fraction
Scheme 7.2 Extraction and Isolation Procedure for Suberea clavate
`
138
7.2.2 Collection, Extraction and Isolation for the sponge Pseudoceratina sp.
A specimen of Pseudoceratina sp. was collected at the depth of 9.4m, Hook
Reef lagoon, Queensland, Australia in 1999. It was identified as Pseudoceratina sp.
(phylum Porifera, class Demospongiae, order Verongida, family Pseudoceratinidae). A
voucher specimen QMG315237 has been deposited at the Queensland Museum, South
Brisbane, Queensland, Australia.
Plate 7.2 Photograph of the sponge Pseudoceratina sp.
A freeze dried sample of Pseudoceratina sp. (5g) was extracted exhaustively
with hexane (250 ml), dichloromethane (250 ml) and methanol (2 x 250 ml),
respectively. The dichloromethane and methanol extracts were combined and then
evaporated solvents to yield a yellow residue (1.5 g). This crude extract was preabsorbed onto C18 (1.0 g) and packed dry into a small cartridge, which was connected to
a C18 preparative HPLC column (5 μm, 21.2 x 150 mm). A linear gradient from 100%
water (0.1% TFA) to 100% methanol (0.1% TFA) was performed over 60 minutes at a
flow rate of 9 ml/min and 60 fractions (1.0 minute each) were collected. Compound 150
(5.0 mg, 0.1% dry wt), was obtained from fraction 15. Fractions 39 to 48 contained an
ion peak of interest in (-)-LRESIMS at (-)m/z 1112. These fractions were therefore
combined and chromatogramed on a Betasil C18 column (5 μm, 150 x 21.2 mm) using a
linear gradient from 55% methanol (0.1% TFA) – 45% water (0.1% FA) to 100%
methanol (0.1% TFA) in 60 minutes. Compound 147 (40.0 mg, 0.8% dry wt) was
purified. Fractions 23-37 in the first large-scale isolation were also combined and
loaded onto a C18 preparative HPLC column (5 μm, 21.2 x 150 mm) with the 60minute-linear gradient from 20% methanol (0.1% TFA) – 80% water (0.1% TFA) to
90% methanol (0.1% TFA) – 10% water (0.1% TFA). A new bromotyrosine,
pseudoceralidinone A (148, 12 mg, 0.24% dry wt) and three known compounds, 151
139
(21 mg, 0.42% dry wt), 141 (18 mg, 0.36% dry wt) and 152 (0.5 mg, 0.01%) were
isolated. Further purification for fractions 27 to 39 on the same reverse phased HPLC
column with the linear gradient from 20% methanol (0.1% TFA) – 80% water (0.1%
TFA) to 80% methanol (0.1% TFA) – 20% water (0.1% TFA) in 60 minutes yielded
two compounds, a new aplysamine 7 (149, 4.5 mg, 0.09% dry wt) and 153 (1.0 mg,
0.02% dry wt) (Scheme 7.3).
Pseudoceratina sp. (5.0g)
a)
DCM/MeOH extract (1.5g)
b)
Compound 150
(5.0mg, 0.1%)
Fraction 15
Fractions 39-48
(-)-m/z 1112
Fractions 23-37
e)
Compound 147
(40.0mg, 0.8%)
Fractions 33-42
c)
Compound 148
(12.0mg, 0.24%)
Fractions 16-17
Compound 141
(18.0mg, 0.36%)
Fractions 25-26
Compound 151
(21.0mg, 0.42%)
Fractions 20-21
Fractions 27-39
Compound 152
(0.5mg, 0.01%)
Fraction 40
d)
Compound 153
(1.0mg, 0.02%)
Fraction 27
Compound 149
(4.5mg, 0.09%)
Fractions 24-25
a) Extraction with DCM and MeOH
b) A linear gradient from 100% water (0.1% TFA) to 100% methanol (0.1% TFA) in 60 minutes, a
flow rate of 9.0 ml/min, 1.0 min/fraction
c) A linear gradient from 20% methanol (0.1% TFA) – 80% water (0.1% TFA) to 90% methanol
(0.1% TFA) – 10% water (0.1% TFA) in 60 minutes, a flow rate of 9.0 ml/min, 1.0 min/fraction
d) A linear gradient from 20% methanol (0.1% TFA) – 80% water (0.1% TFA) to80% methanol
(0.1% TFA) – 20% water (0.1% TFA) in 60 minute, a flow rate of 4.0 ml/min, 9.0 min/fraction.
e) A linear gradient from 55% methanol (0.1% TFA) – 45% water (0.1% TFA) to100% methanol
(0.1% TFA) in 60 minute, a flow rate of 9.0 ml/min, 9.0 min/fraction.
Scheme 7.3 Extraction and Isolation Procedure for Pseudoceratina sp.
140
7.3 Structure Elucidation and Discussion
7.3.1 Pseudoceralidinone A (148)
Compound 148, [α]25D +3.5 (c 0.1, MeOH), was purified as a colourless
amorphous solid. The (+)-LRESIMS of 148 displayed an isotopic cluster of ions
[M+H]+ at m/z 421, 423 and 425, which indicated the molecule contained two bromines.
A pseudomolecular ion peak in the (+)-HRESIMS at m/z 420.9746 (calcd (+) m/z
420.9757, Δ -2.6 ppm) allowed the molecular formula C14H18Br2N2O3 to be assigned to
compound 148.
Figure 7.6 The 1H-NMR spectrum of 148 recorded at 600 MHz in DMSO-d6
Analysis of the 13C-NMR and HSQC spectra indicated this molecule contained
one carbonyl carbon (δC 158.2 ppm), four quaternary carbons (δC 152.0, 139.0, 117.8
and 117.8 ppm), two sp2 methines (δC 130.4 and 130.4 ppm), one sp3 methine (δC 74.4
ppm), four methylenes (δC 70.3, 54.3, 47.0 and 24.8 ppm) and two methyls (δC 42.3 and
42.3 ppm).
Collected data from the 1H-NMR and COSY spectra demonstrated the presence
of a system -CH-CH2-NH-. A 5-substituted oxazolidinone ring (Fragment A) was
suggested for the spin system -CH-CH2-NH- due to the observation of HMBC
correlations from H-3 (δH 7.75ppm), H-4 (δH 3.36 and 3.85ppm) and H-5 (δH 5.59ppm)
to an upfield carbonyl carbon C-2 (δC 158.2ppm) (Figure 7.7).
141
A two-proton singlet signal at δH 7.71 ppm which had HMBC correlations with
quaternary carbons C-6 (δC 139.0 ppm), C-8 (δC 117.8 ppm), C-9 (δC 152.0 ppm) and
C-10 (δC 117.8 ppm) was assigned for the symmetrical aromatic protons often found in
a 3,4-dibromotyrosine structural class.80 Hence, fragment B was elucidated (Figure 7.7).
Further analysis of COSY spectral data indicated another spin system (CH3)2NH-CH2-CH2-CH2- with the correlations between N-CH3 (δH 2.84 ppm) and an
exchangeable proton H-16 (δH 9.63 ppm), from H-15 (δH 3.36 ppm) to H-16 and H-14
(δH 2.18 ppm) and from H-14 to H-13 (δH 4.04 ppm). Their locations were also
supported by HMBC correlations from N-CH3 to C-15 (δC 54.3 ppm) as well as from H14 to C-15 and C-13 (δC 74.4 ppm). The downfield resonance of a methylene C-13 (δC
74.4 ppm) suggested this carbon was bound to an oxygen atom. All evidence facilitated
the establishment of a fragment C for an N,N-dimethylaminopropanol residue (Figure
7.7).
Br
10
3
5
NH
14
16
9
NH
O
8
O
2
O
Fragment A
6
Br
13
15
HMBC
COSY
Fragment C
Fragment B
Figure 7.7 The partial structures A, B and C of 148
Detailed analysis of the HMBC spectrum revealed the HMBC correlations from
H-5 to C-7 and C-11 as well as from H-4 to C-6. This supported the connectivity
between fragments A and B (Figure 7.8). Fragments B and C were also assembled due
to the 3JCH long range correlation from H-13 to C-9. The final structure of 148,
pseudoceralidinone A, was established.
Br
NH
13
O
9
4
Br
5
HMBC
NH
O
O
Figure 7.8 Key HMBC correlations to establish the structure of 148
Determination of absolute stereochemistry for compound 148
There are several methods for the assignment of the absolute stereochemistry of
organic compounds including specific optical rotation, circular dichroism (CD), optical
rotatory dispersion (ORD) and more modern methods such as X-ray crystallography and
NMR spectroscopy. Among them, NMR spectroscopic analysis of MTPA esters
142
(Mosher method) have been proven a reliable and useful technique in order to determine
the absolute configuration of secondary alcohols or secondary amines.81
L1 shielded
L1
OMe
L2
O
(S)
L1
H
L2
OMe
Ph
CF3
O
O
CF3
OMTPA
Hx
(S)-MTPA ester
SR
=
S
-
R
Hy
Hz
L2 shielded
<0
Hc
>0
H
Hb
Ha
MeO
L1
L2
L2
Ph
O
(R)
L1
H
MeO
CF3
O
O
(R)-MTPA ester
MTPA Model
CF3
Figure 7.9 Configurational correlation model for the (S)-MTPA and (R)-MTPA esters
The Mosher method first reported by Mosher et al. in 197382,83 uses α-methoxyα-(trifluoromethyl)phenylacetic
acid
(MTPA)
or
α-methoxy-α-(trifluoromethyl)
phenylacetic chloride acid (MTPA-Cl) as an auxiliary reagent. This method begins with
the esterification (amidation) of the alcohol (amine) with two enantiomers of MTPA
(MTPA-Cl). The 1H-NMR or
19
F-NMR spectra of two diastereomeric esters are
acquired and compared prior to making a final assignment. The convenient and
important chemical process relies on a difference in steric bulkiness of the substituents
of the two β carbons (next to the carbonyl carbon). In an NMR solution, the carbinyl
proton, ester carbonyl and trifluoromethyl groups of the MTPA unit are oriented on the
same plane (Figure 7.9). Due to the anisotropic effect of an phenyl group in the MTPA
unit and the β-substituents, the different chemical shifts between two diastereomeric
esters can be observed. For the (S)-MTPA ester, the protons of L1 are shielded by the
phenyl ring while those on L2 are unaffected. Conversely, the protons of L2 are shielded
whereas those on L1 remain unaffected in the (R)-MTPA ester.82,83 Based on ΔδSR
obtained by subtracting the chemical shifts (δR) of protons in the (R)-MTPA ester from
those (δS) in the (S)-MTPA ester, Kakisawa et al.84 later suggested a molecular model
for the assignment of absolute configuration in which positive and negative ΔδSR values
were found on the right and left sides of the MTPA plane, respectively (Figure 7.9).
This model was used to examine the absolute stereochemical structures of a series of
organic compounds and proved the reliability and validation in the establishment of the
absolute stereochemistry.84-88
Herein, the absolute configuration of 148 was examined by the Mosher analysis.
Compound 148 was first hydrolysed with hydrochloride acid 6N using microwaveassisted conditions at 140oC in 10 minutes and followed by Boc-protection in sodium
143
bicarbonate for 30 minutes to produce 154 (the yield of 46% in two steps) (Scheme 7.4).
In case the microwave condition can change the chirality of 148, an optical rotation for
154 was carefully measured. In comparison with the specific optical rotation of 148
([α]25D +3.5 (c 0.1, MeOH)), 154 had the same sign of the optical rotation value ([α]25D
+3.1 (c 0.1, MeOH)) indicating that the configuration of 154 was stable under
microwave irradiation. Mosher analysis was then performed on 154 to determine its
absolute stereochemistry at the position C-5.
Br
N
Br
NH
148
Br
Br
1) HCl 6N, microwave
140oC, 10min
O
O
N
(R) or (S)-MTPA-Cl
pyridine, r.t, 24h
O
2) (Boc)2O, NaHCO3
r.t, 30min
Br
O
O
Br
NHBoc
154
N
OH
R = (S)-MTPA (155)
R = (R)-MTPA (156)
NHBoc
OR
Scheme 7.4 Modification of 148 to determine its absolute stereochemistry
Table 7.2 NMR data for the MTPA Esters of 154 in DMSO-d6
Position
3
4
5
7, 11
13
14
15
16
N-CH3
2’
Ar, mtpa
Mtpa-OCH3
a
(S)-MTPA ester (155)
δH (mult., J in Hz) a gCOSY (H no.)
7.13 (t, 5.5)
4
3
Ha 3.32 (m)
3
Hb 3.45 (m)
5.96 (t, 6.0)
4
7.62 (s)
4.01 (t, 6.0)
14
2.24 (m)
13, 15
3.31 (m)
14
a
2.79 (s)
1.32 (s)
7.46 (m)
Ar, mtpa
7.49 (m)
Ar, mtpa
3.49 (s)
-
(R)-MTPA ester (156)
gCOSY (H no.)
δH (mult., J in Hz) a
7.12 (t, 5.5)
4
Ha 3.30 (m)
3
3
Hb 3.44 (m)
5.95 (t, 6.0)
4
7.63 (s)
4.01 (t, 6.0)
14
2.24 (m)
13, 15
3.33 (m)
14
a
2.83 (s)
1.32 (s)
7.46 (m)
Ar, mtpa
7.50 (m)
Ar, mtpa
3.49 (s)
-
ΔδSR=δS-δR
(ppm)
+0.01
+0.02
+0.01
+0.01
-0.01
0.00
0.00
-0.02
-0.04
0.00
Not observed
Br
16
O
NH
11
O
13
5
Br
N
H
7
OR
R = (S)-MTPA (155)
R = (R)-MTPA (156)
After checking the Δδ
SR
2
O
1'
2'
3
values which were qualitatively proportional to the
distance from the MTPA moiety, the substituents about the chiral centre were prioritised
according to the Cahn-Ingold-Prelog rules and the absolute stereochemistry at C-5 was
assigned as an S-configuration (Figure 7.10).
144
Δδ>0
-0.04
N
-0.04
NHBoc
Br
-0.01
0.00
-0.02
O
H
Ha
0.00
+0.02
Br
H
H
+0.01
0.00
O
Hb
OMTPA
0.00
+0.01
N
O
H
OMTPHA+0.01
Br
(S)
H
0.00
N
O
-0.01
Br
Δδ<0
Figure 7.10 Δδ SR values for MTPA derivatives (155 and 156) and their configurational
correlation model
Table 7.3 NMR data for TFA salt of pseudoceralidinone A (148) in DMSO-d6
Position
δC
2
3 (N)
4
158.2
47.0
5
6
7, 11
8, 10
9
13
14
15
16 (N)
N-CH3
74.4
139.0
130.4
117.8
152.0
70.3
24.8
54.3
42.3
δH
(mult., J, int.)
gCOSY
(H no.)
7.75 (s, 1H)
Ha 3.36 (m, 1H)
Hb 3.85 (t, 9.0, 1H)
5.59 (t, 7.8, 1H)
7.71 (s, 2H)
4.04 (t, 6.0, 2H)
2.18 (m, 1H)
3.36 (m, 2H)
9.63 (br.s, 1H)
2.84 (d, 4.2, 6H)
ROESY
(H no.)
4
4b, 5
4a, 5
4a, 4b
14
13, 15
14, 16
15, N-CH3
16
4a, 4b
3
3, 5
7, 11
`
5
14
Br
NH
O
9
10
13
15
6
O
F3C
Br
8
3
NH
5
O
O
2
O
Peudoceralidinone A (148)
145
gHMBC
(C no.)
2, 4, 5
2, 5, 6
2, 5, 6
2, 4, 6, 7, 11
5, 8, 9, 10
9, 14, 15
13, 15
13, 14, N-CH3
15
7.3.2 Aplysamine 7 (149)
Compound 149, [α]25D +8.1 (c 0.08, MeOH), was obtained as a colourless
amorphous solid. The (+)-LRESIMS of 149 exhibited an isotopic cluster of ions
[M+H]+ at m/z 664, 666, 668 and 670 indicating the presence of three bromine atoms.
The molecular formula of 149 was determined to be C23H28Br3N3O5 by observing a
pseudomolecular ion peak in the (+)-HRESIMS at m/z 663.9662 (calcd (+)-m/z
663.9652, Δ 1.5 ppm).
Figure 7.11 The 1H-NMR spectrum of 149 recorded at 600 MHz in DMSO-d6
The
13
C-NMR spectrum revealed that structure of 149 contained 23 carbons.
Combined with the HSQC data, these carbons included one carbonyl carbon (δC 163.0
ppm), eight quaternary carbons (δC 153.7, 151.3, 150.8, 143.0, 130.3, 117.0, 117.0 and
110.2 ppm), five sp2 methines (δC 132.9, 130.4, 130.4, 129.1 and 112.5 ppm), one sp3
methine (δC 69.4 ppm), five methylenes (δC 70.1, 54.3, 46.1, 27.6 and 24.8 ppm) and
three methyls (δC 56.1, 42.3 and 42.4 ppm).
Analysis of COSY and HMBC spectral data supported the establishment of a
3,5-dibromo-4-(3-dimethylamino)propyloxyphenyl residue (Fragment A) which was
also found in pseudoceralidinone A (148). A broad signal at δH 9.50 ppm representing
146
for an exchangeable proton in NH demonstrated COSY correlations with two N-methyl
groups and H-22, suggesting this tertiary amine was protonated by TFA (Figure 7.12).
The observation of COSY correlations from H-11 (δH 3.27 and 3.39 ppm) to an
exchangeable proton H-10 (δH 7.79 ppm) and H-12 (δH 4.67 ppm) led to the assignment
of a spin system -NH-CH2-CH-. In addition, the downfield resonance of the methine C12 (δC 69.4 ppm) indicated this carbon was adjacent to an oxygen atom. Thus, fragment
B was established (Figure 7.12).
Further analysis of the 1H-NMR and HMBC data displayed the presence of a
1,3,4-trisubstituted aromatic ring containing H-2 (δH 7.37, d, 1.8), H-5 (δH 6.99, d, 8.4)
and H-6 (δH 7.12, dd, 8.4, 1.8). 3JCH long range correlations in the aromatic system
allowed the locations of quaternary carbons C-1 (δC 130.3 ppm), C-3 (δC 110.2 ppm)
and C-4 (δC 153.7 ppm). The position of a methoxy group (δH 3.80 ppm) was deduced
due to its HMBC correlation to C-4. The typical upfield resonance of the quaternary
carbon C-3 in a benzene ring was indicative of its attachment to a bromine atom.89,90 An
isolated deshielded methylene proton H-7 (δH 3.71 ppm) correlated with C-1, C-2 (δC
132.9 ppm), C-6 (δC 112.5 ppm), C-8 (δC 151.3 ppm) and C-9 (δC 163.0 ppm) in the
HMBC spectrum suggesting the side chain substituent connected directly to C-1. A
sharp and downfield signal at δH 11.94 ppm was assigned for an N-OH group which was
supported by its HMBC correlation with the typical quaternary carbon C-8 (δC 151.3
ppm) representing for the -C=N- signal.91 With the upfield chemical shift of carbon C-7
observed at δC 27.6 ppm, an E configuration for the oxime functionality was suggested
as the corresponding value for a (Z)-geometry would be >35 ppm.89,91,92 From this
evidence, fragment C was constructed (Figure 7.12).
Br
NH
22
O
11
17
O
16
20
13
Br
Fragment A
N
H
12
15
7
9
N
O
3
Br
4
O
1
8
OH 6
HMBC
COSY
Fragment C
Fragment B
Figure 7.12 The partial structures A, B and C of 149
Key HMBC correlations from H-11 to C-13 (δC 143.0 ppm); from H-12 to C-14
(δC 130.4 ppm) and C-18 (δC 130.4 ppm) led to the connectivity from fragment A to
fragment B. Fragments B and C were also assembled due to the observation of 3JCH
long range correlation from H-10 to C-9 (Figure 7.13). In order to ensure the
consistency with molecular weight, a hydroxyl group attaching to the methine carbon C147
12 was assigned. Consequently, the complete structure of aplysamine 7 (149) was
established.
Br
NH
O
18
13
Br
14
O
HMBC
Br
12
N
H
OH
10
9
N
OH
O
Figure 7.13 Key HMBC correlations to establish the structure of 149
Determination of absolute stereochemistry of compound 149
OH
NH2
O
O
O
O
Br2, CH3COOH
r.t, 3h
1) (CF3CO)2O, 90oC, 18h Br
OH 2) TFA 70%, r.t, 16h
Br
NH2
O
80%
157
OH
OH
O
46%
159
158
O
Br
BnONH2.HCl
EtOH, reflux, 4h
O
64%
OH
N
BnO
160
N
Br
HO
1) Br2, HCl 6M, r.t, 1h
HO
2) Boc2O, NaOH 10%, r.t, 30min
NH2
NHBoc
N
O
Br
50%
163
162
Br
TFA:DCM (1:1)
r.t, 30min
NHBoc
OH
OH
161
90%
Br
K2CO3, KI, Me2CO:ACN (1:1)
reflux, 16h
Br
80%
OH
Cl
N
Br
O
N
O
O
10, HOBt, EDCI, DMF, r.t, 2h
Br
NH2
60%
OH
Br
Br
N
H
OH
164
N
165
OBn
O
Scheme 7.5 Total synthesis of oxime-protected aplysamin 7 (165)
Total synthesis was addressed in order to determine the absolute stereochemistry
of 149 (Scheme 7.5). The alpha protected 3-bromo-4-methoxyphenylpyruvic acid oxime
(160) was prepared from O-methyltyrosine (157). Compound 157 was converted to 3bromo-4-methoxyphenylpyruvic acid (159) via three steps93 and then treated with Obenxylhydroxylamine in ethanol at reflux temperature for 4 hours to give 160 with a
yield of 64%.94 A racemate of commercially available octopamine (161) was chosen as
a starting material for synthesis of the final product (165). Dibromination of 161 with
bromine in hydrochloride acid at room temperature gave the dibromo derivative after 1
hour. This product was then protected by (Boc)2O in sodium hydroxide with an overall
yield in two steps of 80%. O-alkylation of 162 with N-(3-chloropropyl)-N,Ndimethylamine hydrochloride using potassium carbonate, potassium iodide in acetoneacetonitrile (1:1) at reflux temperature to obtain 163 (50%).95 Removal of the Boc group
148
in 163 was achieved in 50% TFA in DCM to produce 164 (90%). This compound was
coupled with 160 using EDCI/HOBt in dried DMF solvent to give 165 with the yield of
75%.94,96
Br
N
O
O
Br
Br
N
H
OH
N
OBn
O
Chiral-HPLC (Lux 5μm, Amylose-2)
Br
Br
N
N
O
O
O
O
Br
Br
OH
Br
Br
N
H
N
H
OH
N
OBn
(R) or (S)-MTPA-Cl
pyridine, r.t, 24h
Br
O
N
O
O
O
Br
Br
N
H
Br
Br
OR
N
H
N
OBn
OR
O
R = (S)-MTPA (166a)
R = (R)-MTPA (167a)
N
OBn
O
R = (S)-MTPA (166b)
R = (R)-MTPA (167b)
Br
Br
N
O
O
O
O
Br
Br
OH
Br
Br
N
H
165a
O
(R) or (S)-MTPA-Cl
pyridine, r.t, 24h
Br
N
OBn
165b, [α]24D = -6.7o (c 0.08, MeOH)
165a, [α]24D = +5.2o (c 0.08, MeOH)
N
N
O
N
H
OH
N
OBn
O
N
OBn
O
165b
Scheme 7.6 Isolation and stereochemical determination of enantiomers 165a and 165b
A racemic mixture of 165 was subjected to a chiral HPLC column and resolved
into two enantiomers (165a and 165b). Absolute stereochemistry of these two
stereoisomers was then assigned by Mosher ester analysis as the same procedure with
compound 154 (Table 7.5). The results indicated that compound 165a with an optical
rotation value of +5.2 (c 0.08, MeOH) was determined to have a S-configuration while
its isomer (165b) owing an optical rotation value of -6.7 (c 0.08, MeOH) was assigned
as an R-configuration. Hence, the absolute configuration of 2 can be determined based
on the sign of optical rotation from these isomeric oxime-protected of aplysamin 7
(165a and 165b). Possessing a similar sign of the measured optical rotation value with
that of (S)-oxime-protected aplysamin 7 (165a), compound 149 ([α]24D +8.1 (c 0.08,
MeOH)) was characterised as a S-configuration.
149
Table 7.4 NMR data for TFA salt of aplysamin 7 (149) in DMSO-d6
δC
δH
(mult., J, int.) b
gCOSY
(H no.)
1
2
3
4
5
6
7
8
9
10 (N)
11
130.3
132.9
110.2
153.7
112.5
129.1
27.6
151.3
163.0
46.1
12
13
14, 18
15, 17
16
20
21
22
23 (NH)
O-CH3
N-OH
N-CH3
69.4
143.0
130.4
117.0
150.8
70.1
24.8
54.3
56.1
42.4
7.37 (d, 1.8, 1H)
6.99 (d, 8.4, 1H)
7.12 (d, 8.4, 1.8, 1H)
3.71 (s, 2H)
7.79 (t, 6.0, 1H)
3.27 (m, 1H)
3.39 (m, 1H)
4.67 (t, 6.0, 1H)
7.56 (s, 2H)
3.99 (t, 6.0, 2H)
2.16 (m, 2H)
3.35 (underneath water peak)
9.43 (br.s)
3.80 (s, 3H)
11.94 (s, 1H)
2.84 (d, 4.2, 6H)
6
6
2, 5
11
10, 12
10, 12
11
21
20, 22
21, 23
22, N-CH3
23
Position
ROESY
(H no.)
7
O-CH3
7
2, 6, N-OH
12
10, 14, 18
11, 12
22
5
2, 6, 7
22
Br
NH
O
22
16
17
O
O
Br
F3C
7
12
O
15
N
H
13
OH
1
Br
4
O
N
OH
150
3
9
gHMBC
(C no.)
3, 4, 6, 7
1, 3, 4
2, 4, 5, 7
1, 2, 6, 8, 9
9, 11
9, 12, 13
9, 12, 13
11, 13, 14, 18
12, 15, 16, 17
16, 21, 22
20, 22
20, 21, N-CH3
4
8
22
Table 7.5 NMR data for the MTPA Esters of 165a and 165b in DMSO-d6
(S)-MTPA
ester (166a)
δH (mult., J
in Hz)
Position
2
5
6
7
10 (NH)
11
12
14, 18
20
21
22
23 (NH)
24 (OCH2-Ph)
Ar, OBn
N-CH3
Ar, mtpa
Mtpa-OCH3
a
(R)-MTPA
ester (167a)
δH (mult., J
in Hz)
a
a
a
a
a
a
3.72 (s)
7.99 (t, 6.0)
3.26 (m)
4.715 (t, 6.0)
7.54 (s)
4.02 (t, 5.0)
3.72 (s)
7.97 (t, 6.0)
3.25 (m)
4.710 (t, 6.0)
7.55 (s)
4.02 (t, 5.0)
a
a
a
a
a
a
5.259 (s)
5.254 (s)
a
a
2.79 (s)
7.46 (m)
7.49 (m)
3.49 (s)
2.80 (s)
7.46 (m)
7.50 (m)
3.49 (s)
ΔδSR
(ppm)
0.00
+0.02
+0.01
+0.005
-0.01
0.00
+0.005
-0.01
Not observed
151
(S)-MTPA
ester (166b)
δH (mult., J in
Hz)
(R)-MTPA
ester (167b)
δH (mult., J in
Hz)
a
a
a
a
a
a
3.72 (s)
7.97 (t, 6.0)
3.24 (m)
4.71 (t, 6.0)
7.543 (s)
4.02 (t, 5.0)
3.72 (s)
7.98 (t. 6.0)
3.25 (m)
4.71 (t, 6.0)
7.535 (s)
4.02 (t, 5.0)
a
a
a
a
5.257 (s)
5.261 (s)
a
a
2.80 (s)
7.46 (m)
7.49 (m)
3.49 (s)
2.79 (s)
7.46 (m)
7.49 (m)
3.49 (s)
ΔδSR
(ppm)
0.00
-0.01
-0.01
0.00
+0.008
0.00
-0.004
+0.01
7.3.3 Proposed Biosynthesis of pseudoceralidinone A (148) and aplysamin 7 (149)
Br
HO
tyrosine
decarboxylase
NH2
HO
tyramine
β-hydroxylase
HO
HO
bromoperoxidase
NH2
COOH
NH2
ii
i
161
HO
O
OH
N
O
OH
P
O O O
O
O
O
O
O
P
P
P
O OO OO O
N
O
S
N
NH2
COO
Adenosine-5'-diphosphate
(ADP)
Br
N
H2N
N
Br
O
N
OH
SAM
C3-alanylmethyltransferase
decarboxylase
N
N
O
N
H2 N
S-adenosylmethionine
(SAM)
HO
OH
Carbonylphosphate
N
NH2
iii
(HCO3-)
O
OH
Adenosine-5'-triphosphate
(ATP)
Br
OH
Br
O
H2N
O
SAM
methyltransferase
148
Br
H2O
NH
OH
v
O
HPO3-
168
HO
bromoperoxidase
NH2
HO
OH
O2
NADPH
N
HO
Br
i
HO
OH
SAM
methyltransferase
N
O
168
HO
Br
Br
vi
OH
O
O
OH
OH
NH2
iv
O
O
Br
OH
HO
NH2
NH2
Br
vii
viii
Scheme 7.7 A plausible biosynthesis of 148 and 149
The plausible biosynthesis of pseudoceralidinone A (148) and aplysamine 7
(149) is illustrated in Scheme 7.7. On the basis of biogenetic considerations, an amino
acid tyrosine (i) is known as a precusor of all bromotyrosine alkaloids.97 The presence
of a hydroxy group at Cβ of a 3,5-dibromotyramine-derived system in 148 and 149 is
due to the conversion of the tyrosine amino acid into octopamine (161) via a
decarboxylation followed by β-hydroxylation of tyramine (ii) as proposed by Alkema et
al.98 Dibromination of 161 gives a partial skeleton for these compounds. Containing a
structural motif, purpurealidin F (168), a known natural product isolated from the
sponge Psammaphysilla purpurea,92 may be an intermediate product on the biosynthetic
pathway to create 148 and 149. The N-alkylation and O-methylation of the 3,5dibromooctopamine (iii) in the generation of 168 can be explained by the reaction of iii
with S-adenosylmethionine (SAM) under the presence of “C3-alanyl-methyltransferase”
and methyltransferase which is a plausible mechanism suggested by Rogers et al..2 We
propose that the conversion of 168 into 148 involves the participation of carbonic
anhydrase99 which is a source of bicarbonate HCO3-. Bicarbonate is activated by the
presence of adenosine-5’-triphosphate (ATP) in cells to give carbonyl phosphate100,101
which is an intermediate transferring carboxyl group to 168 followed by formation of
the oxazolidinone 148. For compound 149, it is conceivable that compound 168 joins in
amide bond formation with an oxime unit (viii) that is derived separately by the
monobromination and oxime oxidation of tyrosine and then O-methylation with SAM.
152
149
7.4 Evaluation of drug-like properties
The result of drug-like evaluation indicated all compounds have logP in a druglike area ranging from -0.76 to 3.86 (Table 7.6). It can be seen that the
spirocyclohexadienylisoxazoline bromotyrosine compounds (134, 140-147) have higher
number of rotatable bonds than the simple and oxime bromotyrosine derivatives (139,
148-149 and 151-153). In total, eight out of sixteen isolated compounds pass the
requirements of bioavailability.
Table 7.6 Physico-chemical properties of all isolated compounds
Compounds
Formula
MW
logP
HBA
HBD
134
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
C15H21Br2N5O4
C11H16Br2NO
C15H21Br2N5O4
C16H19Br2N5O4
C16H23Br2N5O4
C31H30Br6N4O9
C24H26Br4N4O9
C24H26Br4N4O8
C25H28Br4N4O8
C31H30Br6N4O11
C14H18Br2N2O3
C23H28Br3N3O5
C5H6N2O2
C13H12Br2N2O5
C22H21Br4N3O8
C9H8Br2O3
495.17
338.06
495.17
505.16
509.19
1082.02
834.10
818.10
832.13
1114.01
422.11
666.20
126.11
436.05
775.03
323.97
-0.76
-0.73
-1.45
0.36
-0.35
3.86
-0.16
0.98
1.43
2.25
2.88
1.44
-0.23
2.33
2.29
2.99
8
1
8
7
8
11
11
10
10
13
3
7
3
3
8
3
5
1
6
4
5
4
5
4
4
6
1
3
2
2
4
1
PSA
142.05
20.23
153.05
134.85
142.05
169.53
180.53
160.30
160.30
209.99
50.80
103.62
61.69
85.89
147.94
46.53
NROT
7
3
9
6
8
12
9
9
10
12
6
12
0
4
8
3
Predicted
Bioavailability
9
9
9
9
9
9
9
9
7.5 Biological activity
Some compounds were evaluated against two cancer cell lines (HeLa and PC3)
and non-cancer cell line (NFF) (Table 7.7). Compound 145 displayed the most activity
against all three cell lines with IC50 values from 4.3 to 9.5 μM. A new compound,
aplysamine 7 (149), was the most active against PC3 cells (IC50 4.9 μM) and less active
against NFF and HeLa cells with IC50 values of 15.7 and 18.9 μM, respectively. On the
contrary, a new bromotyrosine alkaloid, pseudoceralidinone A (148) showed no
inhibition against three screened cell lines.
153
Table 7.7 Biological activity of some isolated compounds
Compound
141
143
145
147
148
149
152
Vincristine sulfate
IC50 ± SD (μM) or % Inhibition ± SD
HeLa
PC3
NFF
19.5±4.7%
66.1±4.8%
24.9±8.1%
41.1±8.6
31.9±7.8
17.9±9.1
6.6±2.9
4.3±1.2
9.5±3.7
40.3±3.4
73±12%
68±3%
44.2±14.3%
35.0±6.7%
28.7±2.5%
18.9±5.2
4.9±1.8
15.7±7.2
46.1±13.6%
60.4±19.9%
48.3±12.3%
0.0058±0.0009
0.0060±0.0005
0.0048±0.0015
7.6 References
(1)
Faulkner, D. J. Nat. Prod. Rep. 2000, 17, 7-55.
(2)
Rogers, E. W.; Molinski, T. F. J. Nat. Prod. 2007, 70, 1191-1194.
(3)
Peng, J.; Li, J.; Hamann, M. T. The Marine Bromotyrosine Derivatives - The
Alkaloids; Elsevier Inc., 2005; Vol. 61.
(4)
Konig, G. M.; Wright, A. D. Heterocycles 1993, 36, 1351-1358.
(5)
Gao, H. F.; Kelly, M.; Hamann, M. T. Tetrahedron 1999, 55, 9717-9726.
(6)
Kernan, M. R.; Cambie, R. C.; Bergquist, P. R. J. Nat. Prod. 1990, 53, 720-723.
(7)
Norte, M.; Fernandez, J. J. Tetrahedron Lett. 1987, 28, 1693-1696.
(8)
Ciminiello, P.; Dellaversano, C.; Fattorusso, E.; Magno, S.; Carrano, L.; Pansini,
M. Tetrahedron 1996, 52, 9863-9868.
(9)
Saeki, B. M.; Granato, A. C.; Berlinck, R. G. S.; Magalhaes, A.; Schefer, A. B.;
Ferreira, A. G.; Pinheiro, U. S.; Hajdu, E. J. Nat. Prod. 2002, 65, 796-799.
(10)
Dambrosio, M.; Guerriero, A.; Pietra, F. Helv. Chim. Acta. 1984, 67, 1484-1492.
(11)
Ciminiello, P.; Costantino, V.; Fattorusso, E.; Magno, S.; Mangoni, A.; Pansini,
M. J. Nat. Prod. 1994, 57, 705-712.
(12)
Encarnacion, R. D.; Sandoval, E.; Malmstrom, J.; Christophersen, C. J. Nat.
Prod. 2000, 63, 874-875.
(13)
Fendert, T.; Wray, V.; Soest, R. W. M. V.; Proksch, P. Z. Naturforsch 1999,
54c, 246-252.
(14)
Acosta, A. L.; Rodriguez, A. D. J. Nat. Prod. 1992, 55, 1007-1012.
(15)
Capon, R. J.; MacLeod, J. K. Aust. J. Chem. 1987, 40, 341-346.
(16)
Shin, J.; Lee, H. S.; Seo, Y.; Rho, J. R.; Cho, K. W.; Paul, V. J. Tetrahedron
2000, 56, 9071-9077.
(17)
Tabudravu, J. N.; Jaspars, M. J. Nat. Prod. 2002, 65, 1798-1801.
154
(18)
Fulmor, W.; Lear, G. E. V.; Morton, G. O.; Mills, R. O. Tetrahedron Lett. 1970,
4551-4452.
(19)
Pettit, G. R.; Butler, M. S.; Williams, M. D.; Filiatrault, M. J.; Pettit, R. K. J.
Nat. Prod. 1996, 59, 927-934.
(20)
Lees, W. J.; Whitesides, G. M. J. Org. Chem. 1993, 58, 642-647.
(21)
Coll, J. C.; Kearns, P. S.; Rideout, J. A.; Sankar, V. J. Nat. Prod. 2002, 65, 753756.
(22)
Costantino, V.; Fattorusso, E.; Mangoni, A.; Pansini, M. J. Nat. Prod. 1994, 57,
1552-1556.
(23)
Park, Y.; Liu, Y.; Hong, J.; Lee, C.; Cho, H.; Kim, D. K.; Im, K. S.; Jung, J. H.
J. Nat. Prod. 2003, 66, 1495-1498.
(24)
Nicholas, G. M.; Newton, G. L.; Fahey, R. C.; Bewley, C. A. Org. Lett. 2001, 3,
1543-1545.
(25)
Longeon, A.; Guyot, M.; Vacelet, J. Experientia 1990, 46, 548-550.
(26)
Chang, C. W. J.; Weinheimer, A. J. Tetrahedron Lett. 1977, 18, 4005-4007.
(27)
Kobayashi, J.; Honma, K.; Sasaki, T.; Tsuda, M. Chem. Pharm. Bull. 1995, 43,
403-407.
(28)
Ciminiello, P.; Fattorusso, E.; Magno, S.; Pansini, M. J. Nat. Prod. 1995, 58,
689-696.
(29)
Kernan, M. R.; Cambie, R. C.; Bergquis, P. R. J. Nat. Prod. 1990, 53, 615-622.
(30)
Benharref, A.; Pais, M. J. Nat. Prod. 1996, 59, 177-180.
(31)
Kijjoa, A.; Watanadilok, R.; Sonchaeng, P.; Silva, A. M. S.; Eaton, G. Z.
Naturforsch 2001, 56c, 1116-1119.
(32)
Rodriguez, A. D.; Akee, R. K.; Scheuer, P. J. Tetrahedron Lett. 1987, 28, 49894992.
(33)
He, H.; Faulkner, J. D.; Shumsky, J. S.; Hong, K.; Clardy, J. J. Org. Chem.
1989, 54, 2511-2514.
(34)
Faulkner, D. J.; Ireland, C. M. Mar. Nat. Prod. Chem. 1977, 23.
(35)
Aydogmus, Z.; Ersoy, N.; Imre, S. Turk. J. Chem. 1999, 23, 339-344.
(36)
Venkateswarlu, Y.; Chavakula, R. J. Nat. Prod. 1995, 58, 1087-1088.
(37)
Krejcarek, G. E.; White, R. H.; Hager, L. P.; McClure, W. O.; Johnson, R. D.;
Rinehart, K. L.; McMillan, J. A.; Paul, I. C.; Shaw, P. D.; Brusca, R. C.
Tetrahedron Lett. 1975, 16, 507-510.
(38)
Sharma, G. M.; Vigand, B.; Burkholder, P. R. J. Org. Chem. 1970, 35, 28232826.
155
(39)
Mancini, I.; Guella, G.; Laboute, P.; Debitus, C.; Pietra, F. J. Chem. Soc. Perkin
Trans. 1 1993, 3121-3125.
(40)
Sharma, G. M.; Burkholder, P. R. Tetrahedron Lett. 1967, 8, 4147-4150.
(41)
Borders, B.; Morton, G. O.; Wetzel, E. R. Tetrahedron Lett. 1974, 15, 27092712.
(42)
Gunasekera, M.; Gunasekera, S. P. J. Nat. Prod. 1989, 52, 753-756.
(43)
Ciminiello, P.; Dellaversano, S.; Fattorusso, E.; Magno, S.; Pansini, M. J. Nat.
Prod. 2000, 63, 263-266.
(44)
Cosulich, D. B.; Lovell, F. M. Chem. Comm. 1971, 397-398.
(45)
Norte, M.; Rodriguez, M. L.; Fernández, J. J.; Eguren, L.; Estrade, D. M.
Tetrahedron 1988, 44, 4973-4980.
(46)
Minale, L.; Sodano, G.; Chan, W. R.; Chen, A. M. J. Chem. Soc., Chem. Comm.
1972, 674-675.
(47)
Mierzwa, R.; King, A.; Conover, M. A.; Tozzi, S.; Puar, M. S.; Patel, M.; Coval,
S. J.; Pomponi, S. A. J. Nat. Prod. 1994, 57, 175–177.
(48)
Jurek, J.; Yoshida, W. Y.; Scheuer, P. J.; Kelly-Borges, M. J. Nat. Prod. 1993,
56, 1609-1612.
(49)
Arabshahi, L.; Schmitz, F. J. J. Org. Chem. 1987, 52, 3584-3586.
(50)
Boehlow, T. R.; Harburn, J. J.; Spilling, C. D. J. Org. Chem. 2001, 66, 31113118.
(51)
Conte, V.; Furi, F. D.; Moro, S. Tetrahedron Lett. 1994, 35, 7429-7432.
(52)
Noda, H.; Niwa, M.; Yamamura, S. Tetrahedron Lett. 1981, 22, 3247-3248.
(53)
Kazlauskas, R.; Lidgard, R. O.; Murphy, P. T.; Wells, R. J. Tetrahedron Lett.
1980, 21, 2277-2280.
(54)
Venkateswarlu, Y.; Venkatesham, U.; Rao, M. R. J. Nat. Prod. 1999, 62, 893894.
(55)
Mcdonald, L. A.; Swersey, J. C.; Ireland, C. M.; Caroll, A. R.; Coll, J. C.;
Bowden, B. F.; Fairchild, C. R.; Cornell, L. Tetrahedron 1995, 51, 5237-5244.
(56)
Kasman, Y.; Groweiss, A.; Carmely, S.; Kinamoni, Z.; Czarkie, D.; Rotem, M.
Pure Appl. Chem. 1982, 54, 1995-2010.
(57)
Copp, B. R.; Ireland, C. M.; Barrows, L. R. J. Nat. Prod. 1992, 55, 822-823.
(58)
Ichiba, T.; Scheuer, P. J.; Borges, M. K. J. Org. Chem. 1993, 58, 4149-4150.
(59)
Liu, S.; Fu, X.; Schmitz, F. J.; Borges, M. K. J. Nat. Prod. 1997, 60, 614-615.
(60)
Tsukamoto, S.; Kato, H.; Hirota, H.; Fusetani, N. Tetrahedron 1996, 52, 81818186.
156
(61)
Roll, D. M.; Chang, C. W. J.; Scheuer, P. J.; Gray, G. A.; Shoolery, J. N.;
Matsumoto, G. K.; Duyne, G. D. V.; Clardy, J. J. Am. Chem. Soc. 1985, 107,
2916-2920.
(62)
Cimino, G.; Rosa, S. D.; Stefano, S. D.; Self, R.; Sodano, G. Tetrahedron Lett.
1983, 24, 3029-3032.
(63)
McMillan, J. A.; Paul, I. C.; Goo, Y. M.; Rinehart, K. L.; Krueger, W. C.;
Pschigoda, L. M. Tetrahedron Lett. 1981, 22, 39-42.
(64)
Compagnone, R. S.; Avila, R.; Suarez, A. I.; Abrams, O. V.; Rangel, H. R.;
Arvelo, F.; Pina, I. C.; Merentes, E. J. Nat. Prod. 1999, 62, 1443-1444.
(65)
Okamoto, Y.; Ojika, M.; Kato, S.; Sakagami, Y. Tetrahedron 2000, 56, 58135818.
(66)
Kennedy, J. P.; Brogan, J. T.; Lindsley, C. W. J. Nat. Prod. 2008, 71, 17831786.
(67)
Swanson, D. M.; Wilson, S. J.; Boggs, J. D.; Xiao, W.; Apodaca, R.; Barbier, A.
J.; Lovenberg, T. W.; Carruthers, N. I. Bioorg. Med. Chem. Lett. 2006, 16, 897900.
(68)
Ross, S. A.; Weete, J. D.; Schinazi, R. F.; Wirtz, S. S.; Tharnish, P.; Scheuer, P.
J.; Hamann, M. T. J. Nat. Prod. 2000, 63, 501-503.
(69)
Kotoku, N.; Tsujita, H.; Hiramatsu, A.; Mori, C.; Koizumi, N.; Kobayashi, M.
Tetrahedron 2005, 61, 7211-7218.
(70)
Pick, N.; Rawat, M.; Arad, D.; Lan, J.; Fan, J.; Kend, A. S.; AvGay, Y. J. Med.
Microbiol. 2006, 55, 407-415.
(71)
Sepcic, K.; Mancini, I.; Vidic, I.; Franssanito, R.; Pietra, F.; Macek, P.; Turk, T.
J. Nat. Toxins 2001, 10, 181-191.
(72)
Nicholas, G. M.; Eckman, L. L.; Newton, G. L.; Fahey, R. C.; Ray, S.; Bewley,
C. A. Bioorg. Med. Chem. 2003, 11, 601-608.
(73)
Schoenfeld, R. C.; Conova, S.; Rittschof, D.; Ganem, B. Bioorg. Med. Chem.
Lett. 2002, 12, 823825.
(74)
Fu, X.; Schmitz, F. J. J. Nat. Prod. 1999, 62, 1072-1073.
(75)
Rodríguez, A. D.; Piña, I. C. J. Nat. Prod. 1993, 56, 907–914.
(76)
Moody, K.; Thomson, R. H.; Fattorusso, E.; Minale, L.; Sodano, G. J. Chem.
Soc., Perkin Trans. 1 1972, 18-24.
(77)
Gopichand, Y.; Schmitz, F. J. Tetrahedron Lett. 1979, 41, 3921-3924.
(78)
Rogers, E. W.; Oliveira, M. F. D.; Berlinck, R. G. S.; Konig, G. M.; Molinski, T.
F. J. Nat. Prod. 2005, 68, 891-896.
157
(79)
Kijjoa, A.; Bessa, J.; Wattanadilok, R.; Sawangwong, P.; Nascimento, M. S. J.;
Pedro, M.; Silva, A. M. S.; Eaton, G.; Soest, R. V.; Herz, W. Naturforsh, B
2005, 60, 904-908.
(80)
Kalaitzis, J. A.; Leone, P. D. A.; Hooper, J. N. A.; Quinn, R. J. Nat. Prod. Res.
2008, 22, 1257-1263.
(81)
Seco, J. M.; Quinoa, E.; Riguera, R. Chem. Rev. 2004, 104, 17-117.
(82)
Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512-519.
(83)
Sullivan, G. R.; Dale, J. A.; Mosher, H. S. J. Org. Chem. 1973, 38, 2143-2147.
(84)
Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991,
113, 4092-4096.
(85)
Ohtani, I.; Kusumi, T.; Ishisuka, M. O.; Kakisawa, H. Tetrahedron Lett. 1989,
30, 3147-3150.
(86)
Kusumi, T.; Fujita, Y.; Ohtani, I.; Kakisawa, H. Tetrahedron Lett. 1991, 32,
2923-2926.
(87)
Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Org. Chem. 1991, 56,
1296-1298.
(88)
Hoye, T. R.; Jeffrey, C. S.; Shao, F. Nature Protocols 2007, 2, 2451-2458.
(89)
Xynas, R.; Capon, R. J. Aust. J. Chem. 1989, 42, 1427-1433.
(90)
Tsuda, M.; Sakuma, Y.; Kobayashi, J. J. Nat. Prod. 2001, 64, 980-982.
(91)
Buchanan, M. S.; Carroll, A. R.; Wessling, D.; Jobling, M.; Avery, V. M.;
Davis, R. A.; Feng, Y.; Hooper, J. N. A.; Quinn, R. R. J. Nat. Prod. 2009, 72,
973-975.
(92)
Tilvi, S.; Rodrigues, C.; Naik, C. G.; Parameswaran, P. S.; Wahidhulla, S.
Tetrahedron 2004, 60, 10207-10215.
(93)
Hoshino, O.; Murakata, M.; Yamada, K. Bioorg. Med. Chem. Lett. 1992, 2,
1561-1562.
(94)
Kende, A. S.; Lan, J.; Fan, J. Tetrahedron Lett. 2004, 45, 133-135.
(95)
Rudkevich, D. M.; Chalmers, J. D. M.; Verboom, W.; Ungaro, R.; Jong, F. D.;
Reinhoudt, D. N. J. Am. Chem. Soc. 1995, 117, 6124–6125.
(96)
Reddy, S. M.; Srinivasulu, M.; Venkateswarlu, Y. Indian J. Chem. 2006, 45B,
2757-2762.
(97)
Dewick, P. M. Medicinal Natural Products: A Biosynthetic Approach; 3rd ed.;
John Wiley & Sons, 2009.
(98)
Alkema, M. J.; Ensor, M. H.; Ringstad, N.; Horvitz, H. R. Neuron 2005, 46,
247-260.
158
(99)
Chegwidden, W. R.; Carter, N. D.; Edwards, Y. H. The Carbonic Anhydrases:
New Horizons; Birkhauser: Berlin, 2000.
(100) Frey, P. A.; Northrop, D. B. Enzymatic mechanisms; IOS Press: Amsterdam,
1999.
(101) Metzler, D. E.; Metzler, C. M. Biochemistry: The Chemical Reactions of Living
Cells, Vol 1; Academic Press: New York, 2003.
159
160
Chapter 8
Conclusion
This thesis reports the chemical investigation of one Australian plant and six
Australian marine sponges whose prefractionated fractions showed cytotoxicity against
at least one of four cancer cell lines (A549, HeLa, LNCaP and PC3). Subsequent massguided isolation of these fractions led to the identification of forty-four secondary
metabolites, nine of which were assigned as new natural products. The biologically
active components isolated from seven samples were found to have the activity with the
Number of compounds
IC50 values ranging from 1nM to 10 μM towards the four test cancer cell lines.
45
40
35
30
25
20
15
10
5
0
MW
CLogP
HBD
HBA
PSA
NROT
Physico-chemical parameters
% count
Figure 8.1 Number of isolated compounds obeys Lipinski’s and Veber’s parameters
100
90
80
70
60
50
40
30
20
10
0
Lipinski
rule
Lipinski
and Veber
rules
Figure 8.2 Percentage of isolated compounds passing the Lipinski’s rule and combined
Lipinski’s and Veber’s rules
In silico physico-chemical properties were evaluated for all 44 isolated
compounds (Figure 8.1). The results showed that
− 21 compounds (47.7%) have MWs less than 500 Da
− 39 compounds (88.6%) have ClogPs less than 5
− 38 compounds (86.4%) have HBDs less than 5
− 38 compounds (86.4%) have HBAs less than 10
− 30 compounds (68.2%) have PSAs less than 140Å2
161
− 35 compounds (79.5%) have NROTs less than 10
Interestingly, 36 out of 44 compounds (81.8%) passed the Lipinski’s rule and 29
compounds (65.9%) showed no violation against the requirements of both Lipinski’s
and Veber’s rules (Figure 8.2). These results demonstrate the practicality of addressing
physico-chemical properties to front-load natural products in drug discovery campaign.
Furthermore, mass-guided isolation was shown to be effective to isolate the active
components.
The strategy of selecting a subset of 330 UV-active fractions instead of all 1155
fractions from 105 biota samples reduced the screening points by 3.5 times. The hit rate
of 5.8% obtained from screening and 20.5% of new natural products isolated from
bioactive fractions demonstrated that the selection of fractions with the predicted logP
values and also with detectable mass by UV absorptions is an effective procedure. This
approach may be particularly useful when screening campaigns use expensive or limited
reagents and a full library can not be screened.
162
Appendix 1
General Experimental Details
1.1 Acquisition of Spectroscopy Data
Optical rotation was obtained on a JASCO P-1020 polarimeter at 569 nm Na
source. Ultraviolet absorption spectra (UV) and infrared spectra (IR) were measured
with a Camspec M501 single beam scanning UV/VIS spectrophotometer and Bruker
Tensor 27 FTIR spectrometer, respectively. Circular dichroism spectra (CD) were
acquired with A JASCO J-715 Spectropolarimeter Circular Dichroism/Optical Rotatory
Dispersion. LRESIMS spectra were operated using Applied Biosystems, Mariner TOF
Biospectrometer. High resolution mass measurement (HRESIMS) was performed on a
Bruker Daltonics Apex III 4.7e Fourier transform mass spectrometer, fitted with an
Apollo API source. Nuclear magnetic resonance (NMR) spectra were recorded at 30oC
on a Varian INOVA 500 MHz spectrometer for 1D-NMR and a Varian INOVA 600
MHz spectrometer for 2D-NMR. The 1H and 13C chemical shift were referenced to the
solvent peaks.
1.2 Isolation of Natural Products
All solvents used for extraction and chromatography were Omnisolv, Merck
HPLC grade and the H2O used was Millipore Milli-Q PF filtered. HPLC separations
were carried out on a Waters 600 pump fitted with a 996 photodiode array detector and
Gilson FC204 fraction collector. Separation and purification were performed by using
Hypersil Betasil C18 columns (150 x 21.2 mm, 5 μm), Hypersil BDS C18 columns (250
x 10 mm, 5 μm) and Betasil C18 columns (150 x 10 mm, 5 μm). LC/MS analysis was
controlled by the MassLynx 4.1 software using a Phenomenex Luna C18 column (3μm,
4.6 x 50 mm).
1.3 Physico-chemical Parameter Calculation
The Instant JChem 2.5.2 software was employed to calculate physico-chemical
properties of all isolated compounds.
1.4 Biological screening
Alamar Blue was purchased from Invitrogen, USA. Reference drugs used in the
assay were paclitaxel (Sigma, USA) and vincristine sulphate (Sigma, USA). Falcon
163
black clear 384 TC microtitre plates (BD Biosciences, USA) were used and wells were
read at excitation 535 nm, emission 590 nm on a Victor II Wallac plate reader
(PerkinElmer, USA). Ten-point concentration response curves were then analysed using
non-linear regression and IC50 values determined by using GraphPad Prism version 5.0.
Procedure of cytotoxicity assay
Human embryonic kidney (HEK 293), human neonatal foreskin fibroblasts
(non-cancer cells, NFF) and human lung adenocarcinoma cells (A549) were grown in
media – DMEM supplemented with 10% Foetal Bovine Serum (FBS), 100 Unit/ml
Penicillin and 100ug/mL Streptomycin. Cervical adenocarcinoma cells (HeLa), prostate
adenocarcinoma cells (LNCaP and PC3) were grown in media – RPMI supplemented
with 10% FBS, 100 Unit/ml Penicillin and 100ug/mL Streptomycin. Cells were grown
under 5% CO2 in a humidified environment at 37oC. Compound toxicity was measured
after 72 hour incubation using an Alamar Blue proliferation assay. Forty-five
microlitres of media containing 1000 cells were added to a 384 well microtitre plate
(Falcon black clear 384 TC microtitre plates). Plates were incubated overnight at 37°C
5% CO2 and 80% humidity to allow cells to adhere. Stock concentrations of pure
compounds were diluted 1 in 10 in media. Five microlitres of diluted compound were
added to the cells to give a total volume 50μl. Each concentration in media was tested in
triplicate. Cells and compounds were then incubated in 72 hours at 37°C 5% CO2 and
80% humidity. Cell proliferation was measured with the addition of 10 ml of a 60%
Alamar blue solution in media to each well of the microtitre plate to give a final
concentration of 10% Alamar blue. The plates were incubated at 37°C 5% CO2 and 80%
humidity within 24 hours. The fluorescence of each well was read at excitation 535 nm
and emission 590 nm on the Victor II Wallac plate reader (PerkinElmer, USA). Tenpoint concentration response curves were then analysed using non-linear regression and
IC50 values determined by using GraphPad Prism 5. Vincristine sulphate was used
during each screening as a positive control compound.
164
Appendix 2
Chapter 3: Experimental
Sigmosceptrellin A (27)
White powder; (+)-LRESIMS m/z 393; 1H,
O
O
O
OH
13
C-NMR data and
optical rotation value were identical with those reported in the
literature (references 2 and 5 in chapter 3)
H
Diacarperoxide A (28)
Colorless oil; (+)-LRESIMS m/z 353; 1H,
O
O
13
C-NMR data and
optical rotation value were identical with those reported in the
O
O O
literature (reference 9 in chapter 3)
Methyl diacarnoate A (29)
Colorless oil; (+)-LRESIMS m/z 353; 1H, 13C-NMR data and
O
optical rotation value were identical with those reported in
O
O
O O
the literature (references 14 and 15 in chapter 3)
Motuporamine C (42)
Colorless oil; (+)-LRESIMS m/z 324; 1H and
N
N
H
NH2
13
C-
NMR data were identical with those reported in the
literature (references 44-50 in chapter 3)
Dehydrocyclostellettamine A (43)
Shallow yellowish gum; UV (MeOH) λmax (logε) 268
N
(3.8); IR (film) νmax 2934, 2855, 1631 and 1202 cm-1;
N
NMR data are summarised in Table 3.3; (+)HRESIMS m/z 244.2066 (calcd for [M]2+ m/z (z=2)
244.2060, Δ 2.5 ppm).
165
Appendix 3
Chapter 4: Experimental
(6aR)-normecambroline (55)
Light brown amorphous solid; [α ]D -284.4o (c 0.05, C2H5OH); CD
25
O
(MeOH) λmax (Δε) 275.0 (5.5), 238 (-18.8), 207 (9.2) nm; UV
NH
O
H
(MeOH) λmax (logε) 309 (4.6), 275 (4.7), 265 (sh) (4.7), 240 (4.7)
HO
nm; IR (film) νmax 3386, 2925, 1679, 1460, 1203; 1H (600 MHz) and
13
C (150 MHz) NMR data are summarized in Table 4.1; (+)
HRESIMS m/z 282.1135 ([M+H]+) (calcd (+)m/z 282.1125, Δ
3.5ppm).
(6aR)-roemerine (56)
Light yellow amorphous solid; (+) LRESIMS m/z 280 ([M+H]+);
O
N
O
H
CD (MeOH) λmax (Δε) 270 (7.2), 231 (-20.6) and 208 (13.1) nm;
1
H,
13
C-NMR data and optical rotation value were identical with
those reported in the literature (reference 26 in chapter 4)
(6S,6aR)-roemerine-Nα-oxide (57)
Light yellow amorphous solid; [α]D25-13.9o (c 0.04, C2H5OH); CD
O
OH
O
N
H
(MeOH) λmax (Δε) 270 (5.9), 231 (-18.4) and 207 (12.5) nm; 1H
(600 MHz) and 13C (150 MHz) NMR data are summarized in Table
4.2. (+) HRESIMS m/z 296.1296 ([M+H]+) (calculated 296.1281,
Δ5.0 ppm);
(6R,6aR)-roemerine-Nβ-oxide (58)
Light yellow amorphous solid; [α]D25-17.4o (c 0.04, C2H5OH); CD
O
O
N
OH
H
(MeOH) λmax (Δε) 271 (6.8), 232 (-19.8) and 208 (15.9) nm; 1H
(600 MHz) and 13C (150 MHz) NMR data are summarized in Table
4.3; (+) HRESIMS m/z 296.1295 ([M+H]+) (calculated 296.1281,
Δ4.7 ppm)
166
(6aS)-actinodaphnine (59)
Light brown amorphous solid; (+) LRESIMS m/z 312 ([M+H]+); 1H,
O
NH
O
13
C-NMR data and optical rotation value were identical with those
H
reported in the literature (reference 27 in chapter 4)
O
OH
(6aS)-laurolitsine (60)
Colorless amorphous solid; (+) LRESIMS m/z 314 ([M+H]+); 1H,
HO
NH
13
O
C-NMR data and optical rotation value were identical with those
H
reported in the literature (references 28 and 29 in chapter 4)
O
OH
(6aS)-boldine (61)
Colorless amorphous solid; (+) LRESIMS m/z 328 ([M+H]+); 1H,
HO
N
O
H
13
C-NMR data and optical rotation value were identical with those
reported in the literature (reference 30 in chapter 4)
O
OH
(1S)-norjuziphine (62)
Colorless amorphous solid; (+) LRESIMS m/z 286 ([M+H]+);
1
NH
O
H
H, 13C-NMR data and optical rotation value were identical with
those reported in the literature (references 31 and 32 in chapter
OH
OH
4)
(1S)-juziphine (63)
Colorless amorphous solid; (+) LRESIMS m/z 286 ([M+H]+);
1
N
O
H
OH
H,
13
C-NMR data and optical rotation value were identical
with those reported in the literature (reference 32 in chapter 4)
OH
167
Appendix 4
Chapter 5: Experimental
Neamphamide B (86)
Colourless amorphous powder; [α]24D -5.5 (c 0.08, MeOH); UV (MeOH) λmax
(logε) 233 (3.8), 274 (3.1) nm; IR (film) νmax 3315, 1740, 1659, 1510 cm-1; 1H (600
MHz) and 13C (150 MHz) NMR data are summarized in Table 5.4; (+) HRESIMS m/z
1574.8935 ([M+H]+) (calcd (+) m/z 1574.8901, Δ 2.2 ppm).
O
H2N
*
NH
*
*
O
OH
N
O
NH
H
N
H2N
*
O
O
*
NH
OH
*
O
O
H
N
*
O
HN
NH
O
O
O
OH
O
OH
N
H
O
O
HN
H2N
OH
*
N
HN
NH2
O
O
HN
H2N
NH
Neamphamide C (87)
Colourless amorphous powder; [α]24D -8.1 (c 0.08, MeOH); UV (MeOH) λmax
(logε) 233 (3.8), 275 (3.1) nm; IR (film) νmax 3315, 1740, 1660, 1512 cm-1; 1H (600
MHz) and 13C (150 MHz) NMR data are summarized in Table 5.5; (+) HRESIMS m/z
1575.8726 ([M+H]+) (calcd 1575.8741, Δ -0.9 ppm).
O
H2 N
*
NH
*
*
O
OH
N
O
H
N
H2 N
NH
*
O
O
NH
*
OH
*
OH
O
O
H
N
*
O
O
OH
N
H
HN
O
HN
H2 N
NH
O
O
O
OH
*
O
N
HN
OH
O
O
HN
H2N
NH
Neamphamide D (88)
Colourless amorphous powder; [α]24D -18.9 (c 0.08, MeOH); UV (MeOH) λmax (logε)
233 (3.7), 275 (3.0) nm; IR (film) νmax 3318, 1739, 1659, 1518 cm-1; 1H (600 MHz) and
13
C (150 MHz) NMR data are summarized in Table 5.5; (+) HRESIMS m/z 1588.9052
([M+H]+) (calculated 1588.9057, Δ -0.3 ppm).
168
O
H2 N
*
NH
*
*
O
OH
N
O
H
N
H2 N
NH
*
O
O
NH
*
OH
*
OH
O
O
H
N
*
O
O
OH
N
H
HN
O
HN
H2 N
NH
O
O
O
OH
*
O
N
HN
NH2
O
O
HN
H2N
NH
Peptide Hydrolysis
Peptide samples (200μg) were dissolved in degassed 6N HCl (500μl) and heated
at 120oC for 16h. The solvent was removed under dry nitrogen and the resulting
material was subjected to further derivatization for stereochemical assignment.
LC/MS Analysis of Marfey’s Derivatives
A portion of the hydrolysate mixture (100μg) or the amino acid standard was
added a solution of L-FDAA 1% (w/w) in acetone and 100μl of a 1N NaHCO3
solution. The vial was heated at 50oC for 3h and the contents were neutralized with
0.2N HCl (50μl) after cooling to room temperature. An aliquot of the L-FDAA
derivative was dried under dry nitrogen, diluted in DMSO and loaded on a
Phenomenex Luna column (C18, 3μm, 2.0 mm x 150 mm) using a linear gradient
from 100% water (0.1% formic acid) to 100% acetonitrile (0.1% formic acid) in 50
minutes and an isocratic at 100% acetonitrile (0.1% formic acid) in the next ten
minutes. FDAA derivatives were detected by absorption at 340nm and assignment
was secured by ion-selective monitoring.
Retention times of authentic FDAA-amino acids are given in parenthesis: L-Arg
(16.18), D-Arg (16.32), L-Asn (20.71), D-Asn (21.08), L-Hpr (27.03), D-Hpr (26.35),
L-Leu (28.95), D-Leu (30.09), L-NMeGln (25.16), D-NMeGln (25.89), L-NMeGlu
(25.71), D-NMeGlu (26.52), L-Thr (21.68), D-Thr (23.27), L-aThr (21.24) and D-aThr
(22.54).
The hydrolysate of neamphamides B (2) contained D-Arg (16.35), L-Asn
(20.74), L-Hpr (27.08), L-Leu (28.93), L-NMeGln (25.19) and D-aThr (22.58).
The hydrolysate of neamphamides C (3) contained D-Arg (16.31), L-Asn
(20.73), L-Hpr (27.03), L-Leu (28.94), L-NMeGlu (25.75) and D-aThr (22.55).
The hydrolysate of neamphamides D (4) contained D-Arg (16.34), L-Asn
(20.75), L-Hpr (27.06), L-Leu (28.95), L-NMeGln (25.20) and D-aThr (22.59).
169
ESI-FTICR-MSn Analysis
Mass spectral data was obtained in the positive ion mode on a Bruker Apex III
4.7 Tesla, which was equipped with an ESI Apollo source. Samples were directly
infused by a Cole-Parmer syringe pump with a flow rate of 2 μL per minute. The end
plate or counter electrode voltage was biased at 3900 V and the capillary voltage at
4400 V relative to the ESI needle. N2 gas was used as nebulizing gas with a pressure of
50 psi and as counter-current drying N2 gas with a flow of 50 L/min. The drying gas
temperature was maintained at 125ºC. The capillary exit voltage was tuned at 120 V.
ESI mass spectra were recorded in the mass range m/z 50-3000 Da. SORI-CID was used
for fragmentation in FTMSn experiments. Data acquisition and processing were
performed using Xmass software.
Parameters for MS2 were correlated sweep pulse length, 1000 μs; correlated
sweep attenuation, 21.4 dB; ejection safety belt, 0 Hz; user pulse length, 40000 μs; ion
activation pulse length, 250000 μs; ion activation attenuation, 42.0 dB; frequency offset
from activation mass, 500 Hz; user delay length, 10 s.
Parameters for MS3 were correlated sweep pulse length, 600 μs; correlated
sweep attenuation, 42.5 dB; ejection safety belt, 0 Hz; user pulse length, 40000 μs; ion
activation pulse length, 250000 μs; ion activation attenuation, 42.0 dB; frequency offset
from activation mass, 500 Hz; user delay length, 10 s.
170
Appendix 5
Chapter 6: Experimental
Hemiasterlin (89)
Colourless amorphous powder; (+) LRESIMS m/z
O
O
527 ([M+H]+);
N
N
H
NH
1
H,
13
C-NMR data and optical
OH
rotation value were identical with those reported in
O
N
the literature (reference 1 in chapter 6)
Hemiasterlin A (91)
Colourless amorphous powder; (+) LRESIMS m/z
O
537 ([M+H]+); 1H,
O
N
N
H
NH
OH
O
N
H
13
C-NMR data and optical
rotation value were identical with those reported
in the literature (reference 3 in chapter 6)
Milnamide A (90)
Colourless amorphous powder; (+) LRESIMS m/z
O
539 ([M+H]+);
O
N
N
H
N
1
H,
13
C-NMR data and optical
OH
rotation value were identical with those reported in
O
N
the literature (reference 3 in chapter 6)
Milnamide C (94)
Colourless amorphous powder; (+) LRESIMS m/z
O
O
553 ([M+H]+); 1H,
N
N
H
N
13
C-NMR data and optical
OH
rotation value were identical with those reported
O
N
in the literature (reference 5 in chapter 6)
O
Milnamide D (95)
Light yellow powder; (+) LRESIMS m/z 537
O
O
N
N
H
N
N
OH
O
([M+H]+); 1H,
13
C-NMR data and optical rotation
value were identical with those reported in the
literature (reference 8 in chapter 6)
171
Geodiamolide D (101)
Colourless amorphous powder; (+) LRESIMS m/z 628
I
OH
13
C-NMR data and optical rotation value were
identical with those reported in the literature (reference 16 in
H
N
N
chapter 6)
O
O
([M+H]+); 1H,
O
O
O
NH
Geodiamolide E (102)
Colourless amorphous powder; (+) LRESIMS m/z 580
Br
OH
13
C-NMR data and optical rotation value were
identical with those reported in the literature (reference 16 in
H
N
N
chapter 6)
O
O
([M+H]+); 1H,
O
O
O
NH
Geodiamolide E (103)
Colourless amorphous powder; (+) LRESIMS m/z 536
Cl
OH
13
C-NMR data and optical rotation value were
identical with those reported in the literature (reference 16 in
H
N
N
chapter 6)
O
O
([M+H]+); 1H,
O
O
O
NH
Jaspamide (115)
White amorphous powder; (+) LRESIMS m/z 709 ([M+H]+);
HO
Br
NH
H,
13
C-NMR data and optical rotation value were identical
with those reported in the literature (reference 19 in chapter
H
N
N
6)
O
O
1
O
O
O
NH
Minamide E (116)
Colourless amorphous powder; [α]25D +10.8 (c 0.02,
O
O
MeOH); CD (MeOH) λmax (Δε) 222 (-5.5), 235 (-
N
N
H
N
N
H
OH
O
3.5), 270 (+0.3) nm; UV (MeOH) λmax (logε) 227
(4.3), 284 (3.5), 290 (3.4) nm; IR (film) νmax 3398,
2960, 1682, 1650, 1480 cm-1; 1H (600 MHz) and 13C
(150 MHz) NMR data are summarized in Table 6.2;
172
(+) HRESIMS m/z 525.3418 ([M+H]+) (calcd
525.3435, Δ -3.3 ppm).
Hemiasterlin D (30)
Colourless amorphous solid; [α]24D -52.3 (c 0.02,
O
O
N
N
H
HN
OH
(3.3) nm; IR (film) νmax 3410, 2965, 1689, 1659,
O
N
O
HO
MeOH); UV (MeOH) λmax (logε) 225 (4.3), 270
1418 cm-1; 1H (600 MHz) and
O
HN
13
C (150 MHz)
OH
N
O
NMR data are summarized in Table 6.4; (+)
HRESIMS
m/z
853.5470
([M+H]+)
(calcd
853.5434, Δ 4.2 ppm).
Peptide Hydrolysis
Peptide samples (150μg) were dissolved in degassed 6N HCl (500μl) and heated
at 120oC for 8h. The solvent was removed under dry nitrogen and the resulting material
was subjected to further derivatization for stereochemical assignment.
LC/MS Analysis of Marfey’s Derivatives
The hydrolysate mixture or the amino acid standard was added a solution of LFDAA 1% (w/w) in acetone and 100μl of a 1N NaHCO3 solution. The vial was heated
at 50oC for 3h and the contents were neutralized with 0.2N HCl (50μl) after cooling to
room temperature. An aliquot of the L-FDAA derivative was dried under dry nitrogen,
diluted in DMSO and loaded on a Phenomenex Luna column (C18, 3μm, 2.0 mm x 150
mm) using a linear gradient from 100% water (0.1% formic acid) to 100% acetonitrile
(0.1% formic acid) in 50 minutes and an isocratic at 100% acetonitrile (0.1% formic
acid) in the next ten minutes. FDAA derivatives were detected by absorption at 340nm
and assignment was secured by ion-selective monitoring.
Retention times of authentic FDAA-amino acids are given in parenthesis: L-tertLeu (33.49) and D-tert-Leu (35.12). The hydrolysate of hemiasterlin D (1) and
milnamide E (2) contained L-tert-Leu eluted at 33.54 min and 33.51 min, respectively.
173
Appendix 6
Chapter 7: Experimental
Purealidin L (134)
Light yellow amorphous powder; (+) LRESIMS m/z 494,
OCH3
Br
Br
496, 498 (1:2:1); 1H,
HO
O
N
H
C-NMR data and optical rotation
value were identical with those reported in the literature
NH
H
N
N
13
NH2
(reference 27 in chapter 7)
O
N,N,N-trimethyl-3,5-dibromotyrosine (139)
Light yellow amorphous powder; (+) LRESIMS m/z 336, 338, 340 (1:2:1);
OH
Br
Br
1
H and
13
C-NMR data were identical with those reported in the literature
(reference 74 in chapter 7)
N
Purealidin O (140)
Light yellow amorphous powder; (+) LRESIMS m/z 494,
OCH3
Br
Br
HO
O
496, 498 (1:2:1); 1H and
H
N
NH2
N
H
N
13
C-NMR data were identical
with those reported in the literature (reference 27 in
chapter 7)
NH
HO
Aerophobins 2 (141)
Yellow amorphous powder; (+) LRESIMS m/z 504, 506, 508
OCH3
Br
Br
NH2
HO
O
N
H
N
N
NH
(1:2:1); 1H,
13
C-NMR data and optical rotation value were
identical with those reported in the literature (reference 62 in
chapter 7)
O
Aplysinamisine 2 (142)
Light yellow amorphous powder; (+) LRESIMS m/z
OCH3
Br
Br
508, 510, 512 (1:2:1); 1H,
13
C-NMR data and optical
HO
O
H
N
N
O
H
N
NH2
NH
rotation value were identical with those reported in the
literature (reference 75 in chapter 7)
174
11,19-dideoxyfistularin 3 (143)
Br
Light yellow amorphous powder;
OCH3
(-) LRESIMS m/z 1080 (cluster);
Br
Br
H3CO
H
N
Br
O
O
1
OH
N
O
Br
O
OH
N
H
N
H,
13
C-NMR data and optical
rotation value were identical with
O
Br
those reported in the literature
(reference 29 in chapter 7)
11-hydroxyaerothionin (144)
Light yellow amorphous powder; (-) LRESIMS
OCH3
Br
m/z 830, 832, 834, 836, 838; 1H, 13C-NMR data
Br
HO
O
N
O
OH
and optical rotation value were identical with
Br
O
H
N
those reported in the literature (reference 29 in
O
N
H
N O
HO
chapter 7)
Br
Aerothionin (145)
Light yellow amorphous powder; (-) LRESIMS
OCH3
Br
m/z 814, 816, 818, 820, 822; 1H, 13C-NMR data
Br
Br
HO
and optical rotation value were identical with
O
O
H
N
N
O
those reported in the literature (reference 76 in
N
H
O
N O
Br
chapter 7)
HO
Homoaerothionin (146)
Light yellow amorphous powder; (-) LRESIMS
OCH3
OCH3
Br
Br
Br
HO
O
m/z 828, 830, 832, 834, 836; 1H,
OH
and optical rotation value were identical with
O
H
N
H
N
N
Br
N
C-NMR data
those reported in the literature (reference 76 in
O
O
13
chapter 7)
Fistularin 3 (147)
Br
OCH3
Br
Br
OH
H3CO
H
N
Br
O
O
OH
N
O
Br
O
OH
N
N
H
LRESIMS m/z 1114 (cluster); 1H,
13
C-NMR data and optical rotation
value were identical with those
O
OH
Light yellow amorphous powder; (-)
Br
175
reported in the literature (reference
77 in chapter 7)
Pseudoceralidinone A (148)
Colourless amorphous solid; [α]25D +3.5 (c 0.1, MeOH); UV
Br
N
O
Br
(MeOH) λmax (logε) 257 (3.4), 362 (2.7) nm; IR (film) νmax
NH
3444, 1748, 1682, 1203 cm-1; 1H (600 MHz) and
O
O
13
C (150
MHz) NMR data are summarized in Table 7.3; (+) HRESIMS
m/z 420.9746 (calcd (+) m/z 420.9757, Δ -2.6 ppm).
Aplysamine 7 (149)
Colourless amorphous solid; [α]24D +8.1 (c
Br
N
0.08, MeOH); UV (MeOH) λmax (logε) 282
O
O
Br
Br
(3.5) nm; IR (film) νmax 3405, 1673, 1204 cm-
O
1
N
H
OH
N
OH
; 1H (600 MHz) and
13
C (150 MHz) NMR
data are summarized in Table 2; (+) HRESIMS
m/z 663.9662 (calcd (+)-m/z 663.9652, Δ 1.5
ppm).
3-methylmaleimide-5-oxime (150)
Light yellow amorphous powder; (+) LRESIMS m/z 126; 1H and
N
N
H
O
OH
13
C-
NMR data were identical with those reported in the literature (reference 79
in chapter 7)
5-[3,5-dibromo-4-[(2-oxo-5-oxazolidinyl)methoxy]phenyl]-2-oxazolidinone (151)
HN
Light yellow amorphous powder; (+) LRESIMS m/z 435,
Br
O
O
437, 439 (1:2:1); 1H, 13C-NMR data and optical rotation value
O
Br
NH
O
were identical with those reported in the literature (reference
O
41 in chapter 7).
Fistularin 2 (152)
Light yellow amorphous powder; (+) LRESIMS m/z
OCH3
Br
Br
772, 774, 776, 778, 780; 1H,
HO
O
C-NMR data and
optical rotation value were identical with those
OH
H
N
N
13
Br
reported in the literature (reference 77 in chapter 7)
O
O
Br
NH
O
O
176
(3,5-dibromo-2-hydroxy-4-methoxyphenyl)-acetic acid (153)
OCH3
Br
Br
Colourless amorphous powder; (-) LRESIMS m/z 320, 322, 324, 326, 328;
1
OH
H and
13
C-NMR data were identical with those reported in the literature
(reference 12 in chapter 7)
O
Hydrolysis and Boc-protection of Pseudoceralidinone A. A solution of 148 (7.0 mg,
0.016 mmol) in HCl 6N (1.5 mL) was heated at 140oC in 10 min under microwave
irradiation. The reaction mass was dried in vacuo before it was basified by NaOH 10%
(1 mL) and added Boc2O (5 mg, 0.02 mmol) with stirring at r.t in 30 min. The solvent
was removed in vacuo and the residue was subjected on RP-HPLC column to give 154
(3.8 mg, 46% yield in two steps).
Compound 154: colourless amorphous solid; [α]25D +3.1 (c 0.1, MeOH); 1H-NMR
(600MHz, DMSO-d6) δ 7.51 (2H, s), 6.67 (1H, t, J=5.4 Hz), 4.54 (1H, t, J=6.0 Hz),
3.95 (2H, t, J=6.6 Hz), 3.08 (2H, m), 2.41 (2H, t, J=6.6 Hz), 2.14 (6H, s), 1.91 (2H, m),
1.32 (9H, s); 13C-NMR (150MHz, DMSO-d6) δ 155.7 (C), 151.5 (C), 143.2 (C), 130.3
(2 x CH), 117.3 (2 x C), 77.7 (C), 71.5 (CH2), 69.8 (CH), 55.4 (CH2), 47.0 (CH2), 44.9
(2 x N-CH3), 27.8 (3 x CH3), 27.4 (CH2); (+)-LRESIMS m/z 495.0, 497.0, 499.0
Preparation of MTPA Esters of 154 (155 and 156). (R) or (S)-MTPA-Cl (2μL,
0.01mmol) was added to 154 (0.5mg, 0.001mmol) in anhydrous pyridine (150μL) and
stirred at room temperature. Reactions were monitored by LC/MS and stopped after
24h. An aliquot was then dried under dry nitrogen. The residue was partitioned with the
solvent system 1:1 (H2O-CH2Cl2). The CH2Cl2 fraction was evaporated to dryness to
yield the Mosher ester. The 1H and COSY NMR were performed on the Mosher esters
(155 and 156) to obtain the δ
S
and δ R values, which were used to determined the
absolute stereochemistry at C-5.
(S)-MTPA ester of pseudoceralidinone A (155): 1H-NMR (500MHz, DMSO-d6) δ
7.62 (2H, s, H-7 and H-11), 7.49 (2H, m, MTPA-ArH), 7.46 (3H, m, MTPA-ArH), 7.13
(1H, t, J=5.5 Hz, NH-3), 5.96 (1H, t, J=6.0 Hz, H-5), 4.01 (2H, t, J=6.0 Hz, H-13), 3.49
(3H, s, MTPA-OCH3), 3.45 (1H, m, H-4b), 3.32 (1H, m, H-4a), 3.31 (2H, m, H-15),
2.79 (6H, s, 2 x N-CH3), 2.24 (2H, m, H-14), 1.32 (9H, s, Boc-H); (+)-LRESIMS m/z
711.1, 713.1, 715.1
177
(R)-MTPA ester of pseudoceralidinone A (156): 1H-NMR (500MHz, DMSO-d6) δ
7.63 (2H, s, H-7 and H-11), 7.50 (2H, m, MTPA-ArH), 7.46 (3H, m, MTPA-ArH), 7.12
(1H, t, J=5.5 Hz, NH-3), 5.95 (1H, t, J=6.0 Hz, H-5), 4.01 (2H, t, J=6.0 Hz, H-13), 3.49
(3H, s, MTPA-OCH3), 3.44 (1H, m, H-4b), 3.30 (1H, m, H-4a), 3.33 (2H, m, H-15),
2.83 (6H, s, 2 x N-CH3), 2.24 (2H, m, H-14), 1.32 (9H, s, Boc-H); (+)-LRESIMS m/z
711.1, 713.1, 715.1
3-bromo-O-methyltyrosine (158): To a cooled solution (5oC) of O-methyl-L-tyrosine
(157, 261mg, 1.3 mmol) in glacial acetic acid (8.0 mL, 0.14 mol), Br2 (0.12 mL, 2.3
mmol) was added and stirred for 3h at r.t. After completion of the reaction, the reaction
was quenched with saturated Na2S2O3 solution and the solvent was removed under
vacuum pressure. The reaction mass was then extracted with ethyl acetate. The organic
layer was dried prior to being purified by RP-HPLC column to give 158 as a colorless
amorphous solid (285 mg, 80%); IR (film) νmax 3411, 1625, 1255, 1024 cm-1, ; 1H-NMR
(600MHz, DMSO-d6) δ 8.31 (3H, br.s), 7.47 (1H, d, J=1.8 Hz), 7.23 (1H, dd, J=9.0, 1.8
Hz), 7.07 (1H, d, J=8.4 Hz), 4.17 (1H, br.s), 3.83 (3H, s), 3.05 (2H, m);
13
C-NMR
(150MHz, DMSO-d6) δ 170.3 (C), 154.7 (C), 133.8 (CH), 130.1 (CH), 128.5 (C), 112.7
(CH), 110.6 (C), 56.2 (OCH3), 53.1 (CH), 34.5 (CH2); (+)-LRESIMS m/z 274.0, 276.0
3-bromo-4-methoxyphenylpyruvic acid (159): A solution of 158 (161 mg, 0.6 mmol)
in (CF3CO)2O (3.0 mL, 21.2 mmol) was heated at 90oC for 18h. The solvent was
removed under reduced pressure. The residue was again dissolved in 70% aqueous TFA
and allowed to stand 16h at r.t. The product was chromatographed on RP-HPLC column
to yield 159 (74 mg, 46%); IR (film) νmax 3410, 1650, 1254, 1054 cm-1; 1H-NMR
(600MHz, DMSO-d6) δ 9.28 (1H, br.s), 8.10 (1H, d, J=1.8 Hz), 7.66 (1H, dd, J=8.4, 1.8
Hz), 7.10 (1H, d, J=8.4 Hz), 3.85 (3H, s); 13C-NMR (150MHz, DMSO-d6) δ 166.2 (C),
154.3 (C), 141.1 (C), 133.2 (CH), 130.2 (CH), 129.2 (C), 112.5 (CH), 110.5 (C), 108.2
(CH), 56.2 (OCH3); (-)-LRESIMS m/z 271.0, 273.0
2-(benzyloxyimino)-3-(3-bromo-4-methoxyphenyl)propanoic acid (160): To a
sulution of 159 (33 mg, 0.12 mmol) in EtOH (2 mL), O-benxylhydroxylamine (56 mg,
0.35 mmol) was added and refluxed for 4h. The crude product was purified by RPHPLC to afford 160 (29 mg, 64%); IR (film) νmax 3420, 2940, 1625, 1599, 1541, 1397,
178
1255, 1022, 806 cm -1; 1H-NMR (600MHz, DMSO-d6) δ 7.42 (1H, d, J=1.8 Hz); 7.32
(2H, d, J=7.8 Hz), 7.29 (1H, d, J=7.2 Hz), 7.25 (2H, d, J=7.2 Hz), 7.19 (1H, dd, J=8.4,
1.8 Hz), 6.96 (1H, d, J=8.4 Hz), 5.05 (2H, s), 3.79 (3H, s), 3.72 (2H, s);
13
C-NMR
(150MHz, DMSO-d6) δ 164.7 (C), 158.2 (C), 153.5 (C), 138.2 (C), 133.1 (CH), 131.5
(C), 129.6 (CH), 128.2 (2 x CH), 127.54 (2 x CH), 127.50 (CH), 112.3 (CH), 110.0 (C),
74.8 (CH2), 56.2 (OCH3), 30.8 (CH2); (+)-LRESIMS m/z 378.0, 380.0
tert-butyl-2-(3,5-dibromo-4-hydroxyphenyl)-2-hydroxyethylcarbamate (162): To a
solution of racemic octopamine (384 mg, 2.0 mmol) in distilled water (3 mL), 6N HCl
(3 mL) was added and the mixture was cooled to 5oC. Bromine (0.35 mL, 6.8 mmol)
was then injected into the stirred solution. The reaction mass was dried in vacuo before
it was basified by NaOH 10% (5 mL) and added Boc2O (500 mg, 2.2 mmol) with
stirring at r.t in 30 min. The solvent was removed in vacuo and the residue was
subjected on RP-HPLC column to give 162 (654 mg, 80% yield in two steps); IR (film)
νmax 3411, 2977, 2932, 1693, 1468, 1250, 1170, 1055 cm-1; 1H-NMR (600MHz,
DMSO-d6) δ 7.10 (2H, s), 6.64 (1H, t, J=5.4 Hz), 4.29 (1H, dd, J=7.2, 5.4 Hz), 3.00
(1H, m), 2.92 (1H, m), 1.36 (9H, s); 13C-NMR (150MHz, DMSO-d6) δ 160.8 (C), 155.7
(C), 128.8 (2 x CH), 123.9 (C), 114.4 (2 x C), 77.5 (C), 70.7 (CH), 48.3 (CH2), 28.3 (3 x
CH3); (+)-LRESIMS m/z 410.0, 412.0, 414.0
tert-butyl-2-(3,5-dibromo-4-(3-(dimethylamino)propoxy)phenyl)-2-hydroxyethyl
carbamate (163): To a mixture of compound 162 (327 mg, 0.8 mmol), K2CO3 (0.90 g,
6.5 mmol) and KI (1.08 g, 6.5 mmol) in dry 50% acetone – 50 % acetonitrile (5.0 mL),
3-dimethylamino-1-propyl chloride hydrochloride (406 mg, 2.5 mmol) was added and
refluxed for 16h. The solvent was evaporated in vacuo and extracted with ethyl acetate.
The organic layer was concentrated and purified by RP-HPLC column to obtain a
racemate of 163 (198 mg, 50%); IR (film) νmax 3336, 2974, 2823, 1698, 1456, 1253,
1168, 1042 cm-1; 1H-NMR (500MHz, DMSO-d6) δ 7.52 (2H, s), 6.76 (1H, br.s), 4.55
(1H, t, J=6.0 Hz), 3.95 (2H, t, J=6.5 Hz), 4.00 (2H, t, J=6.5 Hz), 2.41 (2H, t, J=7.0 Hz),
2.14 (6H, s), 1.91 (2H, m), 1.32 (9H, s);
13
C-NMR (125MHz, DMSO-d6) δ 155.5 (C),
151.3 (C), 142.7 (C), 130.4 (2 x CH), 117.0 (2 x C), 77.6 (C), 71.7 (CH2), 70.0 (CH),
55.6 (CH2), 47.3 (CH2), 45.1 (2 x N-CH3), 28.1 (3 x CH3), 27.7 (CH2); (+)-LRESIMS
m/z 495.0, 497.0, 499.0
179
2-amino-1-(3,5-dibromo-4-(3-(dimethylamino)propoxy)phenyl)ethanol
(164):
Compound 164 (158 mg, 0.3 mmol) was taken in 2 mL of 50% DCM – 50% TFA and
stirred for 30 min. After the solvent was removed, the crude product was purified by
RP-HPLC column to get free amine 164 (114 mg, 90%); IR (film) νmax 3408, 3033,
2739, 1676, 1202, 1133 cm-1; 1H-NMR (600MHz, DMSO-d6) δ 8.10 (2H, br.s), 7.68
(2H, s), 4.82 (1H, dd, J=9.0, 3.0 Hz), 4.00 (2H, t, J=6.6 Hz), 3.36 (2H, t, J=7.2 Hz),
3.10 (1H, d, J=12.0 Hz), 2.89 (1H, t, J=11.6 Hz), 2.84 (6H, s), 2.19 (2H, m); 13C-NMR
(150MHz, DMSO-d6) δ 151.4 (C), 141.4 (C), 130.5 (2 x CH), 117.6 (2 x C), 70.3
(CH2), 67.6 (CH), 54.3 (CH2), 45.1 (CH2), 42.3 (2 x N-CH3), 24.8 (CH2); (+)-LRESIMS
m/z 395.0, 397.0, 399.0
2-(benzyloxyimino)-3-(3-bromo-4-methoxyphenyl)-N-(2-(3,5-dibromo-4-(3(dimethylamino)propoxy)phenyl)-2-hydroxyethyl)propanamide (165): To a solution
of 164 (30 mg, 0.076 mmol) in dry dimethylformamide (3.5 mL), Nhydroxybenzotriazole (HOBt, 20 mg, 0.15 mmol) was added and the reaction mixture
was stirred for 15 min at r.t. The reaction mixture was then cooled to 0oC and N-(3Dimethylaminopropyl)-N′-ethylcarbodiimide (EDCI, 28.7 mg, 0.15 mmol) was added
and stirring continued for 30 min at 0oC. To this mixture, compound 160 (28 mg, 0.076
mmol) was then added and stirred for 2h at r.t. The crude product was concentrated in
vacuo and separated on RP-HPLC column (MeOH, H2O, 0.1% TFA) to get a racemic
mixture of 165 (34 mg, 60%); IR (film) νmax 3395, 2940, 1673, 1496, 1204 cm-1; 1HNMR (600MHz, DMSO-d6) δ 9.47 (1H, br.s), 7.96 (1H, t, J=6.0 Hz), 7.57 (2H, s), 7.36
(3H, m), 7.33 (2H, d, J=7.2 Hz), 7.08 (1H, d, J=8.4 Hz), 6.97 (1H, d, J=8.4 Hz), 5.25
(2H, s), 4.67 (1H, t, J=6.0 Hz), 3.98 (2H, t, J=6.0 Hz), 3.80 (3H, s), 3.72 (2H, s), 3.36
(1H, m), 3.35 (2H, m), 3.27 (1H, m), 2.84 (6H, d, J=4.2 Hz), 2.16 (2H, m); 13C-NMR
(150MHz, DMSO-d6) δ 162.2 (C), 154.0 (C), 152.2 (C), 150.8 (C), 143.1 (C), 136.8
(C), 133.1 (CH), 130.4 (2 x CH), 129.6 (C), 129.2 (CH), 128.5 (2 x CH), 128.1 (CH),
128.0 (2 x CH), 117.2 (2 x C), 112.6 (CH), 110.3 (CH), 76.6 (CH2), 70.2 (CH2), 69.4
(CH), 56.2 (OCH3), 54.4 (CH2), 46.3 (CH2), 42.4 (N-CH3), 28.6 (CH2), 24.8 (CH2); (+)LRESIMS m/z 754.0, 756.0, 758.0, 760.0
Isolation of two enantiomers in a racemic oxime-protected aplysamin 7 (165):
Compound 165 was further purified by HPLC on a chiral HPLC column (Phenomenex,
Lux 5μm, Amylose-2, 250 x 4.6 mm) with an isocratic condition of 30% acetonitrile
180
(0.1% TFA)-70% water (0.1% TFA) in 15 minutes, flow rate 0.8 ml/min. Two
enantiomers (165a and 165b) were eluted at 7.5 min and 9.8 min, respectively.
Compound 165a: colourless amorphous solid; [α]24D +5.2 (c 0.08, MeOH); 1H-NMR
(600MHz, DMSO-d6) δ 9.47 (1H, br.s), 7.96 (1H, t, J=6.0 Hz), 7.57 (2H, s), 7.36 (3H,
m), 7.33 (2H, d, J=7.2 Hz), 7.08 (1H, d, J=8.4 Hz), 6.97 (1H, d, J=8.4 Hz), 5.25 (2H, s),
4.67 (1H, t, J=6.0 Hz), 3.98 (2H, t, J=6.0 Hz), 3.80 (3H, s), 3.72 (2H, s), 3.36 (1H, m),
3.35 (2H, m), 3.27 (1H, m), 2.84 (6H, d, J=4.2 Hz), 2.16 (2H, m); (+)-LRESIMS m/z
754.0, 756.0, 758.0, 760.0
Compound 165b: colourless amorphous solid; [α]24D -6.7 (c 0.08, MeOH); 1H-NMR
(600MHz, DMSO-d6) δ 9.47 (1H, br.s), 7.96 (1H, t, J=6.0 Hz), 7.57 (2H, s), 7.36 (3H,
m), 7.33 (2H, d, J=7.2 Hz), 7.08 (1H, d, J=8.4 Hz), 6.97 (1H, d, J=8.4 Hz), 5.25 (2H, s),
4.67 (1H, t, J=6.0 Hz), 3.98 (2H, t, J=6.0 Hz), 3.80 (3H, s), 3.72 (2H, s), 3.36 (1H, m),
3.35 (2H, m), 3.27 (1H, m), 2.84 (6H, d, J=4.2 Hz), 2.16 (2H, m); (+)-LRESIMS m/z
754.0, 756.0, 758.0, 760.0
Preparation of MTPA Esters of 165a (166a and 167a). (R) or (S)-MTPA-Cl (2μL,
0.01mmol) was added to 165a (0.5mg, 0.001mmol) in anhydrous pyridine (150μL) and
stirred at room temperature. Reactions were monitored by LC/MS and stopped after
24h. An aliquot was then dried under dry nitrogen. The residue was partitioned with the
solvent system 1:1 (H2O-CH2Cl2). The CH2Cl2 fraction was evaporated to dryness to
yield the Mosher ester. The 1H and COSY NMR were performed on the Mosher esters
(166a and 167a) to obtain the δ S and δ R values, which were used to determined the
absolute stereochemistry at C-5.
(S)-MTPA ester of 165a (166a): 1H-NMR (600MHz, DMSO-d6) δ 7.99 (1H, t, J=6.0
Hz, NH-10), 7.541 (2H, s, H-14 and H-18), 5.259 (2H, s, H-24), 4.715 (1H, t, J=6.0 Hz,
H-12), 3.26 (1H, m, H-11), 2.79 (6H, s, N-CH3); (+)-LRESIMS m/z 970.0, 972.0, 974.0,
976.0
(R)-MTPA ester of 165a (167a): 1H-NMR (600MHz, DMSO-d6) δ 7.97 (1H, t, J=6.0
Hz, NH-10), 7.551 (2H, s, H-14 and H-18), 5.254 (2H, s, H-24), 4.710 (1H, t, J=6.0 Hz,
181
H-12), 3.25 (1H, m, H-11), 2.80 (6H, s, N-CH3) ; (+)-LRESIMS m/z 970.0, 972.0,
974.0, 976.0
Preparation of MTPA Esters of 165b (166b and 167b). Carried out as same
procedure with MTPA Esters of 165a.
(S)-MTPA ester of 165b (166b): 1H-NMR (600MHz, DMSO-d6) δ 7.97 (1H, t, J=6.0
Hz, NH-10), 7.543 (2H, s, H-14 and H-18), 5.257 (2H, s, H-24), 4.71 (1H, t, J=6.0 Hz,
H-12), 3.24 (1H, m, H-11), 2.80 (6H, s, N-CH3) ; (+)-LRESIMS m/z 970.0, 972.0,
974.0, 976.0
(R)-MTPA ester of 165b (167b): 1H-NMR (600MHz, DMSO-d6) δ 7.98 (1H, t, J=6.0
Hz, NH-10), 7.535 (2H, s, H-14 and H-18), 5.261 (2H, s, H-24), 4.71 (1H, t, J=6.0 Hz,
H-12), 3.25 (1H, m, H-11), 2.79 (6H, s, N-CH3) ; (+)-LRESIMS m/z 970.0, 972.0,
974.0, 976.0
182