Isolation and confirmation of the proposed cleistanthol biogentic link from Croton insularis

Article abstract of DOI:10.1021/ol103083m

The proposed cleistanthol biosynthetic intermediate en route to spruceanol, and other related family members, was isolated for the first time from Croton insularis, confirming the Jacobs-Reynolds hypothesis. Anticancer evaluation of the new isolates and their aerial oxidation products is also reported.(Figure Presented)

Full text of DOI:10.1021/ol103083m

ORGANIC
LETTERS
2
011
Vol. 13, No. 5
032–1035
Isolation and Confirmation of the
Proposed Cleistanthol Biogentic Link
from Croton insularis
1
†
Lidia A. Maslovskaya, Andrei I. Savchenko, Victoria A. Gordon,
‡
§
§
Paul W. Reddell, Carly J. Pierce, Peter G. Parsons, and Craig M. Williams*
†
†
,‡
School of Chemistry and Molecular Biosciences, University of Queensland, Brisbane
072, Queensland, Australia, Queensland Institute of Medical Research, P.O. Royal
4
Brisbane Hospital, Brisbane 4029, Queensland, Australia, and EcoBiotics Limited,
P.O. Box 1, Yungaburra, 4884 Queensland, Australia
c.williams3@uq.edu.au
Received December 21, 2010
ABSTRACT
The proposed cleistanthol biosynthetic intermediate en route to spruceanol, and other related family members, was isolated for the first time from
Croton insularis, confirming the Jacobs-Reynolds hypothesis. Anticancer evaluation of the new isolates and their aerial oxidation products is
also reported.
2
known members of the rare cleistanthane series of diter-
In 2001, Jacobs and Reynolds isolated from Jatropha
divaricata two pimarane-type diterpenes (1 and 2) of which
2
3
penes [spruceanol (3) and cleistanthol (4)] were isolated
4
1
was a new skeletal type (Scheme 1). Considering two
5
together with 1 and 2, Jacobs and Reynolds surmised that
a close biogenetic link most likely exisited between the
three sketal types (i.e., 1-3).
This notion was further supported by the fact that
†
University of Queensland.
Queensland Institute of Medical Research.
EcoBiotics Limited.
‡
§
(
1) Denton, R. W.; Harding, W. W.; Anderson, C. I.; Jacobs, H.;
McLean, S.; Reynolds, W. F. J. Nat. Prod. 2001, 64, 829–831.
2) Recent family members and sources of isolation: (a) Torres-
Mendoza, D.; Gonz ꢀa lez, L. D. U.; Ortega-Barrı
a, E.; Coley, P. D.;
Kursar, T. A.; Capson, T. L.; McPhail, K.; Cubilla-Rios, L. J. Nat. Prod.
pimarane derivatives have been converted by acid catalysis
6
to the cleistanthane system previously.
(
´
2
1
004, 67, 1711–1715. (b) Craveiro, A. A.; Silveira, E. R. Phytochemistry
982, 21, 2571–2574. (c) Kaufman, T. S.; Mischine, M. P.; Gonzalez-
(3) Gunasekera, S. P.; Cordell, G. A.; Farnsworth, N. R. J. Nat.
Prod. 1979, 42, 658–662.
Sierra, M.; Ruveda, E. A. Can. J. Chem. 1987, 65, 2024–2026. (d) Du,
Z.-Z.; Shen, Y.-M. Helv. Chim. Acta 2006, 89, 2841–2845. (e) Pinto,
A. C.; Patitucci, M. L.; Zocher, D. H. T.; Kelecom, A. Phytochemistry
(4) (a) McGarry, E. J.; Pegel, K. H.; Phillips, L.; Waight, E. S.
J. Chem. Soc., Chem. Commun. 1969, 1074. (b) McGarry, E. J.; Pegel,
K. H.; Phillips, L.; Waight, E. S. J. Chem. Soc. C 1971, 904–909.
(5) Pimarane and cleistanthane family members have been isolated
together on other occasions; see: (a) Riehl, C. A. S.; Pinto, A. C.
Phytochemistry 2000, 53, 917–919. (b) Pinto, A. C.; Peixoto, E. M.;
Fiorani, N. G. M. Phytochemistry 1984, 23, 1293–1296. (c) Pinto, A. C.;
Patitucci, M. L.; Da Silva, R. S.; Queiroz, P. P. S.; Kelecom, A.
Tetrahedron 1983, 39, 3351–3354. (d) Candy, H. A.; Pakshong, J. M.;
Pegel, K. H. J. Chem. Soc. C 1970, 2536–2538.
1985, 24, 2345–2347. (f) Pinto, A. C.; Antunes, O. A. C.; Rezende, C. M.;
Correia, C. R. D. Phytochemistry 1995, 38, 1269–1271. (g) Pinchin, R.;
Mayer, L. M. U.; Pinto, A. C. Phytochemistry 1978, 17, 1671–1672.
(h) Sutthivaiyakit, S.; Nakorn, N. N.; Kraus, W.; Sutthivaiyakit, P.
Tetrahedron 2003, 59, 9991–9995. (i) Yadav, D.; Tiwari, N.; Gupta,
M. M. Phytochem. Lett. 2010, 3, 143–147. (j) Sakai, T.; Nakagawa, Y.
Phytochemistry 1988, 27, 3769–3779.
1
0.1021/ol103083m r 2011 American Chemical Society
Published on Web 02/04/2011

In the course of our ongoing search for natural products
7
,8
displaying anticancer activity, we investigated Croton
insularis (Baillon), a plant endemic to northeast Australian
rain forests, driven by previous accounts of biological
screening on this species.
We report herein isolation of the proposed biogenetic
Scheme 1. Relationship between the Proposed Biogentic Link 5
and or 7) to Pimaranes (1 and 2) and the Cleistanthane
Ditepenes (i.e., 3 and 6). Conversion of 5 into Spruceanol 3,
Confirming the Biogentic Link, and the Peroxy Derivative 8
(
9
link pimarane 5, its corresponding keto derivative 7,
b,10
together with spruceanol (3) and sonderianol (6).
2
1 13
Table 1. H and C NMR Data for Compound 5 Recorded in
CDCl at 500 MHz
3
1
13
position
H, σ (ppm) multiplicity
J, Hz
C, σ (ppm)
1
1
2
2
3
4
5
6
6
7
7
8
9
1
a
b
a
b
1.54
1H (dd)
1H (dt)
1H (m)
1H (m)
1H (dd)
13.20, 3.91
13.20, 3.42
35.8
1.99
1.68-1.73
1.73-1.78
3.25
27.5
11.74, 4.40
8.31, 3.91
78.2
39.5
47.7
19.1
1.21
1H (dd)
1H (m)
1H (m)
1H (m)
1H (d)
a
b
a
b
1.64-1.68
1.78-1.83
2.43-2.52
2.74
29.2
16.14
129.3
166.9
38.8
0
1
1
5.81
1H (s)
117.7
204.3
52.6
1
1
1
1
1
1
1
1
1
2
2
3
4
5.91
5.73
5.00
5.03
1.26
1.02
0.88
1.14
1H (s)
1H (dd)
1H (d)
1H (d)
3H (s)
3H (s)
3H (s)
3H (s)
139.2
139.9
114.1
5
17.12, 10.27
10.27
6a
6b
7
17.12
23.4
27.9
15.4
23.0
8
9
0
þ
HRESIMS analysis of 5 ([M þ Na] , m/z 323.1982), in
1
conjunction with a well deconvoluted C NMR spectrum,
3
1
established a molecular formula of C H O . The H
2
0
28 2
NMR (Table 1) spectrum displayed characteristic reso-
nances for the vinyl moiety, two olefinic protons (δ 5.8
H
With this knowledge in hand, the ent-3β-hydroxypi-
mara-8(14),9,15-trien-12-one 5 was proposed as the most
probable biosynthetic intermediate. Further justification
for 5 was outlined with plausible biosynthetic routes to
skeletons 1-3 (Scheme 1). For example, 5 could concei-
vably undergo aromatization via a 1,2-shift of the vinyl
moiety contained in the C-ring to give spruceanol (3);
epoxidation of the 8,14 double bond and subsequent
rearrangement would lead to 1, and either a [1,3]- or
and 5.9), one oxygenated methine (δ 3.2), isolated axial
H
ring protons (δ 2.7, 2.5 and 2.0), and four methyl groups.
H
13
The C NMR (Table 1) spectra revealed all 20 carbons,
which in combination with DEPT measurements were
found to be four methyl groups, five methylenes (one
olefinic), five methines (three olefinic, one oxygenated,
and one saturated), and six quaternary centers (two olefinic,
(
8) Dong, L.; Schill, H.; Grange, R. L.; Porzelle, A.; Johns, J. P.;
[2,3]-sigmatropic shift of the vinyl group from C13 to C8
arrives at 2 (Scheme 1).
Parsons, P. G.; Gordon, V. A.; Reddell, P. W.; Williams, C. M Chem.;
Eur. J. 2009, 15, 11307–11318.
(
9) (a) Setzer, M. C.; Setzer, W. N.; Jackes, B. R.; Gentry, G. A.;
Moriarity, D. M. Pharm. Biol. 2001, 39, 67–78. (b) Graikou, K.;
Aligiannis, N.; Skaltsounis, A.-L.; Chinou, I.; Michel, S.; Tillequin,
F.; Litaudon, M. J. Nat. Prod. 2004, 67, 685–688. (c) Graikou, K.;
Aligiannis, N.; Chinou, I.; Skaltsounis, A.-L.; Tillequin, F.; Litaudon,
M. Helv. Chim. Acta 2005, 88, 2654–2660 and references cited therein.
(10) Itokawa, H.; Ichihara, Y.; Takeya, K.; Morita, H.; Motidome,
M. Phytochemistry 1991, 30 (12), 4071–4073.
(
6) (a) Ellestad, G. A.; Kunstmann, M. P.; Morton, G. O. J. Chem.
Soc., Chem. Commun. 1973, 312–313. (b) Ceccherelli, P.; Curini, M.;
Coccia, R.; Cagnoli, N. J. Chem. Soc., Perkin Trans. 1 1984, 589–592.
(
7) Dong, L.; Gordon, V. A.; Grange, R. L.; Johns, J.; Parsons, P. G.;
Porzelle, A.; Reddell, P.; Schill, H.; Williams, C. M. J. Am. Chem. Soc.
008, 130, 15262–15263.
2
Org. Lett., Vol. 13, No. 5, 2011
1033

indeed the stated biogentic link to diterpenes 1-3(Scheme 1).
Whena solution of5 indeuteratedchloroformwasallowed
to slowly evaporate in air three products could be identi-
fied. An inseparable mixture of exo- and endoperoxides 8
and 9 (9 notshown) alongwiththe keyproduct, spruceanol
(3), confirmed beyound doubt the Jacobs-Reynolds hy-
pothesis.
Scheme 2. Norrish Type 1 Photodegradation of Compound 7
Giving Derivatives 11, 12 (via Path A) and 13 (via Path B)
a
Figure 1. COSY and HMBC correlation diagram.
1
1
three saturated, and one carbonyl carbon). Key H- H-
COSY, HSQC, and HMBC correlations (see Figure 1), in
1
conjunction with the ACD/laboratories software and
1
1
3
12
reported C chemical shifts for similar diterpenes, estab-
lished the position of the hydroxyl at C3. This in turn
initiated the connectivity of the A ring and angular methyl,
both of which instate C10 and C9. H14 (δ 5.9) and H11
H
(
quaternary olefinic carbons (C8 and C9). Subsequent
δH 5.8) ascribed the position of the ajacent carbonyl and
1
1
correlations in the H- H-COSY, HSQC, and HMBC
build the remaining C ring olefin and quaternary carbon
(
δ 52.6) containing the methyl (δ 23.4) and vinyl moiety.
C C
The B ring methylene at C7 (δ 29.2) and H5 (δ 1.2) pin
C
H
points the second methylene carbon C6 (δ 19.1).
C
a
Double-headed arrows indicate NOSEY correlations.
The elucidation of the C3 keto 7 derivative was straight-
forward when compared to the spectra obtained for that of
5. Exposure of 7 to ambient laboratory light in CDCl3
unexpectedlyaffordedthree Norrish type1photodegraded
products via intermediate 10. Bicyclic derivatives 11and 12
Figure 2. NOESY correlation diagram for compound 5.
(1:1 mixture) via pathway A and the cyclopropane 13 via
pathway B (Scheme 2).
The elucidation of 11 and 12 was based on the fact that it
NOESY experiments (Figure 2) confirmed the 1,3-di-
axial syn relationship of H3 and H5 and the trans-decalin
system in compound 5. Only weak correlations between
the C13 methyl and H7R and H7β could be detected so the
stereochemistry at this position is assigned based on that
argued by Jacobs and Reynolds.
With the isolation of 5 now established, chemical con-
version was sought to provide definitive proof that 5 is
1
3
is a decomposition product of 7. The C NMR chemical
shifts of rings A and B atoms did not change drastically.
1
3
However, according to C NMR and HSQC correlations,
1 and 12 only have two double bonds with three quatern-
1
2
1
ary carbons (i.e., C8, C9, C14), one CH sp carbon (C13),
and possess eight ring and double bond equivalents as seen
3
in 7. The key feature was two new sp CH carbons C11
(46.6) and C15 (55.3), clearly indicating the formation of
an additional ring instead of double bond. Correlations
involving, but not between, H11, H15, and methyl group
C17 in the H- H-COSY and the HMBC were key for
building the additional unsaturated ring and showed the
(
11) ACD/H and C NMR Predictor; ACD/Structure Elucidator,
version 10.0, Advanced Chemistry Development, Inc., Toronto, ON,
Canada, 2007; www.acdlabs.com.
(
1
1
12) Seca, A. M. L.; Pinto, D. C. G. A.; Silva, A. M. S. Nat. Prod.
Commun. 2008, 3, 399–412.
1
034
Org. Lett., Vol. 13, No. 5, 2011

Table 2. Inhibition of Growth of Human Cell Lines in Culture
ED50 (μg/mL)
cell line
origin
3
6
5
7
11
12
13
NFF
normal fibroblasts
cervical carcinoma
colon cancer
23
3.4
12
18
5.1
13
23
38
20
50
25
50
43
28
60
>100
20
70
53
55
60
28
90
HeLa
13
13
16
16
15
22
2.7
4.8
19
HT 29
MCF-7
MM96L
K562
35
breast cancer
melanoma
>100
45
2.8
17
7.4
23
0.6
0.47
leukemia
>100
double-bond migration to positions 8 and 9. The elucida-
tionof 13beganwiththe analysisofthe HMBC correlation
between the C20 methyl indicating the position of an sp
tion does not adequately account for the potency of 7
because 5 oxidized in air to the endoperoxide 8 at a similar
rate yet was less active. Although 1 and 2 have not under-
2
1
4
atom at position 9. Additional HMBC cross peaks involv-
ing, but not between, H11 and methyl C17 determined the
position of the carbonyl on C12 adjacent the cyclopropane
ring. The regiochemistry of derivatives 11-13 was ob-
tained from NOESY spectra (see the Supporting Informa-
tion), which shows specific correlation H12-H1 observed
only for 12 and H12-H20 found only for 11. Compound
gone biological evaluation, both spruceanol (3) and
sonderianol (6) have previously been evaluated. Sprucea-
nol (3) displayed marginal activity (ED 3.2 μg/mL)
5
0
3
against the P-388 test system in vitro, whereas sonderianol
(6) displayed cytotoxicity against leukemia L1210 (and
P388) cells in mice (100% growth inhibition at a concen-
tration of 50 μg/mL) and liver cancer HEPAlclc7 cells in
1
5
1
3 has clear NOESY correlations H15-H20, H15-H13,
mice (IC50 0.403 mmol/L).
and H16-H13 corresponding to the structure depicted
In conclusion, the Jacobs-Reynolds proposed biogenetic
link, pimarane 5, between pimarane and cleistanthane
diterpenes has been isolated, elucidated, and chemically
proven.
(
Scheme 2).
Biological evaluation of 3, 5-7, and 11-13 was per-
formed against a panel of human tumor cell lines (Table
2). Clonogenic-type assays for comparing the growth of
treated cells showed some inhibitory activity for 3, 6, and 5
in a normal cell type (NFF) and in a range of tumor cell lines
(Table 2). The results from 7, however, were notable because
Acknowledgment. We thank EcoBiotics Ltd. and
the University of Queensland for financial support.
Dr. Lynette Lambert from the Centre of Advanced
Imaging at the University of Queensland for assistance
with NMR measurements.
of a 5- to 20-fold potency increase in four of the five tumor
cell lines compared with the other compounds and selectivity
against tumor cells compared with normal fibroblasts.
The MM96L line is sensitive to killing by oxidants
Supporting Information Available. Experimental pro-
1
13
cedures and copies of H, C and 2D NMR spectra. This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
3
released during autoxidation of catechols, but autoxida-
(13) Parsons, P. G.; Morrison, L. E. Cancer Res. 1982, 42, 3783–3788.
(14) Similar systems have been evaluated; see: He, F.; Pu, J.-X.;
Huang, S.-X.; Xiao, W.-L.; Yang, L.-B.; Li, X.-N.; Zhao, Y.; Ding, J.;
Xu, C.-H.; Sun, H.-D. Helv. Chim. Acta 2008, 91, 2139–2147.
(15) Grace, M. H.; Jin, Y.; Wilson, G. R.; Coates, R. M. Phytochem-
istry 2006, 67, 1708–1715.
Org. Lett., Vol. 13, No. 5, 2011
1035