The identification of naturally occurring labdane diterpenoid calcaratarin D as a potential anti-inflammatory agent

Article abstract of DOI:10.1016/j.ejmech.2019.04.023

In this study we report, for the first time, the synthesis of the natural product calcaratarin D via a stereo- and regio-selective aldol condensation with (S)-β-hydroxy-γ-butyrolactone as key steps. A concise synthetic route (under 10 steps) to a series of structurally related normal-labdane diterpenes was also developed and their anti-inflammatory activities were evaluated in an in vitro model of inflammation. The structure-activity relationships (SARs) pertaining to the labdane scaffold were elucidated and results suggest that an α-alkylidene-β-hydroxy-γ-butyrolactone system is necessary for potent activity in the labdanes. Our studies identified the natural product calcaratarin D (1) as a promising anti-inflammatory agent, which effectively modulates the production of pro-inflammatory mediators (e.g., TNF-α, IL-6, NO) at both transcriptional and translational levels. These inhibitory effects are likely to occur via the suppression of nuclear factor kappa B (NF-κB) activation by reducing the p65 nuclear translocation but not its phosphorylation or protein expression. Calcaratarin D exhibited significantly greater inhibition of NF-κB activation than andrographolide, a well-known NF-κB inhibitor from the labdane family, suggesting that a normal-configuration labdane ring or the absence of hydroxyl groups at C-3 and C-19 positions is favorable for potent NF-κB inhibition. We further investigated the effects of calcaratarin D on the upstream signalling pathways and found that the compound selectively suppressed the LPS-induced activation of PI3K/Akt pathway without affecting much of the MAPK (i.e., ERK, JNK, and p38) activation. These findings demonstrate that calcaratarin D exerts its anti-inflammatory effects via a selective Akt-NF-κB-mediated mechanism and potentially offers a new therapeutic strategy for the management of inflammatory diseases.

Full text of DOI:10.1016/j.ejmech.2019.04.023

Accepted Manuscript
The identification of naturally occurring labdane diterpenoid calcaratarin D as a
potential anti-inflammatory agent
Quy T.N. Tran, W.S. Fred Wong, Christina L.L. Chai
PII:
S0223-5234(19)30329-0
DOI:
https://doi.org/10.1016/j.ejmech.2019.04.023
EJMECH 11258
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 3 February 2019
Revised Date: 28 March 2019
Accepted Date: 10 April 2019
Please cite this article as: Q.T.N. Tran, W.S.F. Wong, C.L.L. Chai, The identification of naturally
occurring labdane diterpenoid calcaratarin D as a potential anti-inflammatory agent, European Journal of
Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.04.023.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
α-alkylidene-β-hydroxy-
γ-butyrolactone
Calcaratarin D (1)
1
1
5
4
O
Bio-evaluation
16
1
3
HO
β
O
Anti-inflammatory
activity
α
1
2
2
0
11
9
1
4
Natural
products
NF-κB
PI3K/Akt
1
7
2
3
1
5
0
8
7
6
H
H
1
8
19
Pro-inflammatory
gene expression
normal-labdane diterpene
Calcaratarin D (1)
TNF-α, IL-6, NO
scaffold

ACCEPTED MANUSCRIPT
The identification of naturally occurring labdane diterpenoid calcaratarin
D as a potential anti-inflammatory agent
a
b,c,d,e*
a,e*
Quy T.N. Tran, W.S. Fred Wong,
Christina L.L. Chai
a
Department of Pharmacy, Faculty of Science, National University of Singapore, Block S4A,
Level 3, 18 Science Drive 4, 117543, Singapore
b
Department of Pharmacology, Yong Loo Lin School of Medicine, National University Health
System, 117600, Singapore
c
Immunology Program, Life Science Institute, National University of Singapore, 117456,
Singapore
d
SHARE, Molecular Mechanisms of Inflammatory Disease Interdisciplinary Research Group,
Singapore
e
Drug Discovery and Optimization Platform, Medical Science Cluster, Yong Loo Lin School of
Medicine, National University Health System, Singapore
*
Corresponding author. E-mail address: phcwongf@nus.edu.sg, (W.S.F. Wong) or
phacllc@nus.edu.sg (C.L.L. Chai)
1

ACCEPTED MANUSCRIPT
Tran Thi Ngoc Quy
Department of Pharmacy, National University of Singapore
8 Science Drive 4, Singapore 117543
1
Tel: +65 9370 1602
Email: A0099426@u.nus.edu
Associate Professor Wong Wai-Shiu, Fred
Department of Pharmacology, NUS Yong Loo Lin School of Medicine
MD3 #04-01 Level 4, 16 Medical Drive Singapore 117600
Tel: +65 6516 3266
Fax: +65 6873 7690
Email: phcwongf@nus.edu.sg
Professor Christina Li Lin Chai
Department of Pharmacy, National University of Singapore
1
8 Science Drive 4, Singapore 117543
Tel: +65 6601 1061
Fax: +65 6779 1554
Email: phacllc@nus.edu.sg
2

ACCEPTED MANUSCRIPT
ABSTRACT
In this study we report, for the first time, the synthesis of the natural product calcaratarin D via a
stereo- and regio-selective aldol condensation with (S)-β-hydroxy-γ-butyrolactone as key steps.
A concise synthetic route (under 10 steps) to a series of structurally related normal-labdane
diterpenes was also developed and their anti-inflammatory activities were evaluated in an in vitro
model of inflammation. The structure-activity relationships (SARs) pertaining to the labdane
scaffold were elucidated and results suggest that an α-alkylidene-β-hydroxy-γ-butyrolactone
system is necessary for potent activity in the labdanes. Our studies identified the natural product
calcaratarin D (1) as a promising anti-inflammatory agent, which effectively modulates the
production of pro-inflammatory mediators (e.g., TNF-α, IL-6, NO) at both transcriptional and
translational levels. These inhibitory effects are likely to occur via the suppression of nuclear
factor kappa B (NF-κB) activation by reducing the p65 nuclear translocation but not its
phosphorylation or protein expression. Calcaratarin
D
exhibited significantly
greater inhibition of NF-κB activation than andrographolide, a well-known NF-κB inhibitor from
the labdane family, suggesting that a normal-configuration labdane ring or the absence of
hydroxyl groups at C-3 and C-19 positions is favorable for potent NF-κB inhibition. We further
investigated the effects of calcaratarin D on the upstream signalling pathways and found that the
compound selectively suppressed the LPS-induced activation of PI3K/Akt pathway without
affecting much of the MAPK (i.e., ERK, JNK, and p38) activation. These findings demonstrate
that calcaratarin D exerts its anti-inflammatory effects via a selective Akt-NF-κB-mediated
mechanism and potentially offers a new therapeutic strategy for the management of
inflammatory diseases.
3

ACCEPTED MANUSCRIPT
KEYWORDS: Labdane-type diterpene; nuclear factor kappa B; anti-inflammatory activity
ABBREVIATIONS
ERK (extracellular signal–regulated kinase), IL (interleukin), iNOS (inducible nitric oxide
synthase), JNK (Jun N-terminal kinase), LPS (lipopolysaccharide), MAPK (mitogen-activated
protein kinase), MCP-1 (monocyte chemoattractant protein-1), NF-κB (nuclear factor-kappa B),
NO (nitric oxide), NSAID (non-steroid anti-inflammatory drug), PI3K/Akt (phosphatidylinositol
3
-kinase/protein kinase B), TNF-α (tumor necrosis factor-alpha), SAR (structure-activity
relationship)
INTRODUCTION
The development of therapeutics to combat inflammation has long been an area of intense
research, as inflammation plays an important role in the progression of various diseases,
including heart attack, stroke, rheumatoid arthritis, inflammatory bowel disease, and cancer.[1-3]
Labdanes or labdane diterpenes are bioactive natural products with significant therapeutic
potential. They belong to a class of natural bicyclic diterpenes that are widely distributed in both
terrestrial and marine organisms.[4] Structurally, labdane diterpenes are divided into four
categories: normal-, ent-, syn-, and ent-syn-labdane; of which the normal-configuration
constitutes the majority of naturally occurring labdanes.[5, 6] A wide range of biological
activities has been uncovered for this class of compounds including anti-inflammatory,
antimicrobial, and anti-cancer activities.[7, 8] In particular, normal-labdanes with α,β-
unsaturated carbonyl systems are of great interest to drug development due to their potent
pharmacological properties and the high natural abundance in various members of the
4

ACCEPTED MANUSCRIPT
Zingiberaceae family.[9-12] In this study, we evaluated a series of structurally-related normal-
labdane diterpenes including the natural products 1 – 10 and their semi-synthetic analogues for
potential anti-inflammatory properties. Crude plant extracts containing these compounds have
demonstrated anti-inflammatory effects in various experimental models.[9-12] However, the
anti-inflammatory activities of many of these labdane diterpenoids in their isolated and purified
forms have not been reported and structure-activity relationships (SARs) pertaining to the
normal-labdane scaffold have not yet been established. It is thus of interest to investigate the
potential of these compounds for anti-inflammatory drug development and to elucidate the
structural requirements for potent, yet safe anti-inflammatory agents based on the labdane
scaffold. Herein, we also report, for the first time, the synthesis of calcaratarin D (1), coronarin D
(
3) and its naturally occurring derivatives (4 – 6) from the commercially available (+)-sclareolide
10).
(
Figure 1. Stereochemistry of normal-labdane and the chemical structures of naturally occurring
labdane diterpenoids 1 – 10. (1) calcaratarin D; (2) vitexolide E; (3) coronarin D; (4) coronarin D
methyl ether; (5) coronarin D ethyl ether; (6) coronarin D acid; (7) galanolactone; (8) (E)-labda-
8
(17),12,14-trien-15(16)-olide; (9) villosin; (10) sclareolide. (2 column-fitting image)
5

ACCEPTED MANUSCRIPT
RESULTS AND DISCUSSION
Chemistry
Synthesis of calcaratarin D (1), vitexolide E (2), and bicyclohomofarnesal (12). A stereoselective
retrosynthetic route for calcaratarin D (1) was developed based on the principle of
diastereoselectivity in aldol reactions as detailed in scheme 1, starting with the preparation of
bicyclohomofarnesal 12 from commercially available (+)-sclareolide (10).[13, 14] In brief, aldol
condensation of aldehyde 12 with (S)-β-hydroxy-γ-butyrolactone was carried out with LDA as a
base, leading to the formation of anti-aldols 13 and 14 as major products. This process proceeded
via a trans-enolate intermediate and the selectivity observed can be explained by the
Zimmerman-Traxler chair-like transition state. A small amount of vitexolide E (2) was obtained
when the enolate of (S)-
β-hydroxy-γ-butyrolactone was formed at the higher temperature of -
3
0°C. [14-16] The absolute configuration at C-12 was assigned based on the reported
spectroscopic data of natural vitexolide E.[14, 15] The mixture of 13 and 14 were then subjected
to protective O-silylation with tert-butyldimethylsilyl chloride (TBSCl). This process is
regioselective and only the C-14 hydroxyl group of 14 reacted to form the intermediate 15.
Mesylation of 15 yielded a mixture of C-12-(R) and (S) epimers (16 and 17, respectively) in a
temperature dependent manner. At low temperatures (i.e., -78 °C), the reaction may have
proceeded through the formation of a stable carbocation resulting in racemization at C-12 and
both compounds 16 and 17 were formed, whereas, at higher temperatures (i.e., 0 °C), bond
formation proceeded rapidly and only compound 16 was obtained with the retention of the
configuration at C-12. In subsequent reactions, the isolated mesylates 16 and 17 were subjected
2
to elimination reaction using different bases (e.g, i-Pr EtN or DMAP) to afford Z-butyrolactone
(18) and E-butyrolactone (19), respectively. The removal of the silyl ether protecting group was
6

ACCEPTED MANUSCRIPT
achieved under acidic conditions (p-TsOH.H O in acetic acid) to provide calcaratarin D (1) and
2
its Z-isomer (20) in excellent yield.[17] Scheme 1 details the regioselectivity and
stereoselectivity of the above series of reactions.
Scheme 1. Synthesis of calcaratarin D (1) and vitexolide E (2)
7

ACCEPTED MANUSCRIPT
3 3
Reagents and conditions: (a) CH NHOCH .HCl, i-PrMgCl (2 M in THF), THF, 0 ºC – RT,
overnight, 84%; (b) SOCl , pyridine, DCM, -78 ºC, 1 h, 78%; (c) LiAlH , THF, 0 ºC, 2 h, 94%;
2
4
(d) (S)-(-)-β-hydroxy-γ-butyrolactone, LDA, HMPA/THF, -78 ºC – -30 ºC, overnight, 35% (13
and 14); or -30 ºC, 5% (2); (e) TBSCl, imidazole, THF, RT, overnight, 85%; (f) MsCl, Et N,
3
DCM, -78 ºC, 1h, 30% (16), and 21% (17); or 0 ºC, 1 h, quantitative yield (16); (g) (i) i-Pr EtN,
2
DCM, RT, overnight, 44% (18), and 41% (19); (ii) DMAP, DCM, RT, 1 h; (h) p-TsOH,
2
AcOH/H O/THF (7:2:1 v/v/v), RT, 2 days, 85% (1), and 77% (20). (2 column-fitting image)
Synthesis of coronarin D (
coronarin D acid ( ), galanolactone (
villosin ( ). Wittig-Horner olefination of 12 with diethylphosphono-2-butyrolactone 21 gave an
3
), coronarin D methyl ether (
4
), coronarin D ethyl ether (
5),
6
7
), (E)-labda-8(17),12,14-trien-15(16)-olide (
8
), and
9
isomeric mixture of minor Z-butyrolactone (22) and major E-butyrolactone (23) in a ratio of 3:2,
respectively (Scheme 2).[18] Hydrolysis of the major E-isomer (23) under basic conditions
afforded coronarin D acid (6) in almost quantitative yield. Acid 6 was then converted to the
hemi-acetal 3, namely coronarin D, under mild oxidation conditions. Pyridinium chlorochromate
(PCC) was the most effective oxidizing reagent for this oxidation with a yield of 60%, while the
use of Dess-Martin and Swern oxidation resulted in low-to-moderate yield. Coronarin D, a hemi-
acetal, was then converted to its acetal derivatives, coronarin D methyl ether (4) and ethyl ether
(5), by refluxing in the respective alcohol with a catalytic amount of p-toluenesulfonic acid
monohydrate (p-TsOH.H O), yielding 48% of 4 and 46% of 5. Synthetic compounds 3 – 6
2
1
13
exhibited H and C-NMR spectroscopic data in agreement with those reported for the natural
products derived from Hedychium coronarium.[19-21] Galanolactone (7) and its isomer (24)
were synthesized from 23 as previously reported.[22]
8

ACCEPTED MANUSCRIPT
The unreacted 13, recovered from the above silyl protection (Scheme 2), was acetylated to form
diacetylated 25. This intermediate was subjected to a temperature-dependent (i.e., 0 °C and RT)
regioselective elimination with DMAP to afford 8 and 26 (at 0 °C) and 9 (at RT),
respectively.[23]
Scheme 2. Synthesis of coronarin D (3), coronarin D methyl ether (4), coronarin D ethyl ether
(5), coronarin D acid (6), galanolactone (7), (E)-labda-8(17),12,14-trien-15(16)-olide (8), and
villosin (9)
Reagents and conditions: (a) α-Bromo-γ-butyrolactone, triethylphosphite, 130 ºC, 5 h; (b)
phosphonate 21, NaH (60% oil dispersion), THF, 0 ºC – RT, overnight, 40% (22), and 57% (23);
(c) KOH, EtOH/THF (1:10 v/v), RT, overnight, 99%; (d) (i) PCC, DCM, RT, 2 h, 55%; or (ii)
9

ACCEPTED MANUSCRIPT
2 3
DMP, DCM, RT, 1 h, 10%; or (iii), DMSO, (COCl) , Et N, -78 ºC – RT, 1 h, 29%; (e)
appropriate alcohol, cat. p-TsOH.H O, reflux, overnight, 48% (4), 46% (5); (f) m-CPBA, DCM,
2
2
h, 20% (7), 64% (24); (g) Ac O, cat. DMAP, 0 ºC, 1 h, 88%; (h) DMAP (3 equiv.), DCM, 0
2
ºC, 1 h, 53% (8); or RT, 1h, 81% (9). (2 column-fitting image)
Biological evaluation
Anti-inflammatory and cytotoxicity profiling of labdane diterpene compounds. Tumor necrosis
factor-alpha (TNF-α) and interleukin-6 (IL-6) are key pro-inflammatory cytokines involved in
systemic inflammation and are secreted by activated macrophages under inflammatory
conditions. In our preliminary screening, the synthesized natural products and their synthetic
analogs were tested for their inhibitory effects on LPS-induced TNF-α and IL-6 production in
murine macrophage RAW264.7 cells with dexamethasone (1 µM) as the positive control. The
cytotoxicity profiles of these compounds with or without LPS stimulation were also determined.
All compounds were prepared in DMSO as a 100 mM stock solution and tested for both assays
at four different concentrations (e.g., 0.1, 1, 5, and 20 µM). Figure 2 summarizes the cytotoxic
effects of the test compounds on non-stimulated RAW264.7 cells. Figure 3 displays the results of
anti-inflammatory screening of test compounds, where the percentage of TNF-α (Figure 3A) and
IL-6 (Figure 3B) level was normalized against the total percentage of viable LPS-stimulated
RAW264.7 cells at a particular concentration.
0
.1 µM 103.0
108.9
110.6
107.7
91.9
2
102.2
92.7
86.3
79.0
3
101.8
102.3
101.0
100.3
4
102.0
100.1
94.9
96.3
5
102.5
99.2
96.6
87.6
6
100.7
97.0
90.9
80.4
7
102.0
97.7
90.8
82.6
8
103.8
105.3
102.7
59.0
9
99.3
96.6
97.2
95.9
10
100.5
101.2
101.2
91.9
107.5
106.2
107.1
98.3
100.2
95.5
81.7
47.5
20
103.3
100.6
95.4
85.0
22
103.0
107.3
102.4
100.0
23
97.8
97.6
95.1
85.7
24
1
00%
1
5
µM
µM
95.6
82.7
53.7
1
8
0%
0%
6
20 µM
12
13
10

ACCEPTED MANUSCRIPT
Figure 2. Percent cell viability of non-stimulated RAW264.7 cells under increasing
concentrations of test compounds visualized on a heat map. Data are presented as mean of three
separate experiments with triplicate samples. The linear scale is set from white (0%) to grey
(100%). White squares represent higher cytotoxic effects. (2 column-fitting image)
(A)
0
.1 µM
99.3
63.4
19.2
1.7
112.0
101.5
79.3
33.4
2
100.2
89.6
81.9
71.1
3
104.7
77.7
44.7
27.5
4
101.6
86.8
65.3
48.1
5
104.7
80.7
60.7
50.2
6
114.8
88.6
48.4
20.5
7
94.1
84.6
53.6
26.6
8
80.4
86.6
74.6
133.5
9
104.8
101.0
79.9
63.9
10
104.0
91.7
73.7
48.5
12
111.1
108.6
109.5
86.8
98.7
71.6
17.5
4.8
99.8
81.8
69.9
48.0
22
104.4
94.6
73.1
52.6
23
100.3
88.3
55.6
33.1
24
1
00%
1
5
µM
µM
50%
2
0 µM
1
13
20
(B)
0
.1 µM
99.2
68.1
40.0
10.8
1
100.6
84.0
75.6
47.1
2
99.1
82.4
64.0
59.4
3
101.0
79.1
70.0
59.9
4
102.3
100.3
95.9
80.9
5
106.0
88.3
79.9
63.5
6
101.6
79.7
35.2
22.3
7
99.0
99.6
85.9
61.5
8
98.4
87.6
101.5
170.1
9
94.6
100.3
93.8
89.7
10
100.0
95.0
93.0
72.8
12
104.6
101.7
104.3
71.8
100.7
76.9
32.2
10.1
20
101.5
94.1
85.6
64.8
22
103.0
93.9
83.9
76.3
23
102.3
92.3
81.4
63.9
24
150%
1
5
µM
µM
1
00%
0%
5
2
0 µM
13
Figure 3. Percent normalized TNF-α (A) and IL-6 (B) levels visualized on a heat map. Total
amounts of TNF-α or IL-6 are normalized to the total number of viable LPS-stimulated
RAW264.7 cells under increasing concentrations of test compounds. Data are presented as mean
of three separate experiments with triplicate samples. The linear scale is set from white (0%) to
grey (100%). White squares represent stronger TNF-α or IL-6 reduction. (2 column-fitting
image)
In terms of anti-inflammatory activities, compounds 1 and 20 are the most active in the normal-
labdane series. Interestingly, these two compounds share a common structural feature, which is
11

ACCEPTED MANUSCRIPT
an α-alkylidene-β-hydroxy-γ-butyrolactone system. The presence of a hydroxyl group at the β-
position is important and is likely to contribute significantly to the anti-inflammatory activity, as
removal of this group in compounds 22 and 23 resulted in much weaker activities. Similarly,
shifting the OH group from β-position in 1 to
γ-position in coronarin D (3) led to a drastic drop
in activities. Although an intact α-alkylidene-β-hydroxy-
γ-butyrolactone system appears to be
crucial for anti-inflammatory activity, it is also associated with higher cytotoxicity.
Another key SAR observation is that an α,β-unsaturated carbonyl system is required for potent
activity, as saturation of this system led to a loss in activity (1 versus 13). Similarly, (+)-
sclareolide (10) or aldehyde 12 with no α,β-unsaturated system showed reduced activity. These
observations confirm the importance of the α,β-unsaturated system for potent anti-inflammatory
activity. The location of the double bond is also critical as compounds 1 and 20 with unsaturation
at C12-C13 exhibited stronger activities than that with unsaturation at C13-C14 (see compound
2
2
). The anti-inflammatory and cytotoxic effects of configurational isomers 1 (E-) and 20 (Z-), or
1
2, 13
2 (Z-) and 23 (E-) suggest that the configuration of ∆
double bond does not affect activity
and cytotoxicity.
The inhibitory effects of coronarin D (3) and its acetal derivatives (4 and 5) vary with the length
of the O-alkyl chain substituent. Methyl coronarin D ether (4) was the most active amongst all
three against TNF-α release, and is as potent as the parent compound 3 in IL-6 inhibition.
Compound 6, an acid derivative of 3, showed greater TNF-α inhibitory activity, but possessed a
similar IL-6 inhibitory profile as compared to 3. There was no consistent trend in TNF-α and IL-
6
inhibition observed for this class of compounds.
12

ACCEPTED MANUSCRIPT
8
, 17
The effects of the terminal double bond at ∆
on activity and cytotoxicity were examined by
comparing epoxides 7 and 24 with the parent compound 23. Both 7 and 24 were more active than
8
, 17
2
3, suggesting that a terminal epoxide ring is preferable to a terminal ∆
double bond for anti-
inflammatory activity.
To further examine the importance of an extended α,β-unsaturated carbonyl system, the effects
of villosin (9) and its positional isomer 8 were evaluated. Interestingly, treatment with 9 induced
the release of both TNF-α and IL-6 in a dose-dependent manner, while 8 showed moderate
inhibitory activities. Compound 8, which has an extended α,β-unsaturated carbonyl system at
C12 to C15, was found to possess a better activity profile as compared to the normal conjugated
system in 3 – 6, 22 – 23, but weaker than 1. In contrast, shifting this conjugated system to C-11
to C-14 position in 9 further induced the inflammatory responses and cytotoxicity in LPS-
stimulated RAW264.7 cells. The mechanism by which 9 induced the production of pro-
inflammatory cytokines remains unknown, despite a report on its anti-inflammatory
property.[11, 23]
Calcaratarin D (1) on the production of nitric oxide in LPS-stimulated RAW264.7 cells. Results
from the above SAR study have identified the natural product calcaratarin D as the most active
compound amongst other normal-labdanes. Hence, further studies on its anti-inflammatory
activities were carried out. Nitric oxide (NO) is a cell-signaling molecule that produces anti-
inflammatory effects under normal physiological conditions. However, NO acts as a pro-
inflammatory mediator when over-produced in abnormal situations, and is involved in the
13

ACCEPTED MANUSCRIPT
pathogenesis of many inflammatory disorders.[24] In view of this, the inhibitory effects of 1 on
LPS-induced NO generation were examined. LPS stimulation of RAW264.7 macrophages
resulted in a significant increase in nitrite concentration in the cell supernatant. Pre-treatment
with 1 strongly suppressed the NO accumulation in a concentration-dependent manner with an
IC value of 3.91 ± 1.12 µM (Figure 4A).
5
0
Calcaratarin D (1) on pro-inflammatory cytokine expressions in LPS-stimulated RAW264.7
cells. The inducible nitric oxide synthase (iNOS) is the major enzyme responsible for the
overproduction of NO under pathological conditions. IL-1β is an important cytokine that
mediates the expression of various genes involved in inflammation and tissue injury, while
MCP-1 is one of the key chemokines that regulate the migration and infiltration of
monocytes/macrophages. In view of the biological relevance of these molecules for
inflammatory responses, the effects of compound 1 on the mRNA expression of the pro-
inflammatory mediators (e.g., TNF-α, IL-6, IL-1β, MCP-1, and iNOS) were examined in LPS-
stimulated RAW264.7 cells in comparison with dexamethasone. As shown in Figure 4B-F, LPS
treatment significantly up-regulated the gene expression levels of TNF-α, IL-6, IL-1β, MCP-1
and iNOS after 4 h stimulation. However, the mRNA expression levels of these mediators were
significantly abated in a concentration-dependent manner by compound 1 (at 1, 5 and 20 µM).
14

ACCEPTED MANUSCRIPT
6
4
0
0
5
000
A
1
B
C
###
00
4000
###
3000
2000
1000
0
*
*
*
5
0
IC50 = 3.91 ± 1.12 µM
20
0
*
*
0
Ctrl Media Veh DEX
1
5
20
Ctrl Media Veh DEX
1
5
20
-
2
-1
0
1
2
LPS (25 ng/mL)
DMSO (0.02 %)
-
+
+
-
+
+
+
+
+
+
+
+
+
+
-
+
+
-
+
+
+
+
+
+
+
+
+
+
LPS (25 ng/mL)
DMSO (0.02 %)
Log [compound 1] (µM))
Compound 1 (µM)
Compound 1 (µM)
2
2
1
1
5000
0000
5000
0000
200
80
#
##
D
E
F
#
##
#
##
1
50
00
60
40
20
0
1
*
*
*
*
***
50
5000
***
**
0
0
Ctrl Media Veh DEX
1
5
20
Ctrl Media Veh DEX
1
5
20
Ctrl Media Veh DEX
1
5
20
LPS (25 ng/mL)
DMSO (0.02 %)
-
+
+
-
+
+
+
+
+
+
+
+
+
+
LPS (25 ng/mL)
DMSO (0.02 %)
-
+
+
-
+
+
+
+
+
+
+
+
+
+
LPS (25 ng/mL)
DMSO (0.02 %)
-
+
+
-
+
+
+
+
+
+
+
+
+
+
Compound 1 (µM)
Compound 1 (µM)
Compound 1 (µM)
Figure 4. Effects of compound 1 on the LPS-induced NO production and pro-inflammatory gene
expression in RAW264.7 cells. The NO concentration-dependent inhibition curve of compound
1
(A). Gene expression of TNF-α (B), IL-6 (C), IL-1β (D), MCP-1 (E) and iNOS(F) after
stimulation with LPS (25 ng/mL) for 4 h in the presence of increasing concentrations of
compound 1 (1 to 20 M), dexamethasone (DEX; 1 M) or DMSO vehicle control. Total RNA
was isolated and analyzed by real-time PCR. The amount of mRNA was normalized to -actin
µ
µ
β
expression and was presented as relative expression levels compared to the control cells. Data
are presented as mean ± SEM (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001 indicate statistically
#
##
significant differences compared with 0.02% DMSO-treated cells.
p < 0.001 indicates
statistically significant difference compared with the untreated control. (2 column-fitting image)
15

ACCEPTED MANUSCRIPT
Calcaratarin D (
1
) on NF-
κ
B transcriptional activity. NF-κB is a master switch for pro-
inflammatory gene transcription, including genes encoding for TNF-α, IL-6, IL-1β, MCP-1, and
iNOS.[25] Andrographolide, an ent-labdane, is one of the most widely reported NF-κB inhibitors
from the family of labdane diterpenoids. In view of this, the effect of compound 1 on the
transcriptional activity of NF-κB was examined in comparison with andrographolide. A NF-κB-
SEAP reporter gene assay was performed in RAW-Blue™ cells. LPS stimulation elicited a 6-
fold increase in NF-κB transcriptional activity as compared to the untreated control (p < 0.001,
Figure 5A). Pretreatment with 1 effectively reversed LPS-induced NF-κB transcriptional activity
in a concentration-dependent manner. Unexpectedly, the labdane diterpene andrographolide
failed to inhibit NF-κB activity in the range of concentrations tested (e.g., 0.1, 1, 5 and 20
µM).[26] These results implicate that calcaratarin D is a more potent NF-κB inhibitor than the
notable andrographolide, and a normal-labdane stereochemistry or the absence of hydroxyl
groups at C-3 and C-19 positions is favorable for potent NF-κB inhibition.
Phosphorylation and nuclear translocation of the p65 subunit play an important role in NF-κB
function. As shown in Figure 5B, compound 1 did not affect the levels of both phosphorylated
and total p65 subunit. Next, we evaluated the effects of compound 1 on the LPS-induced p65
nuclear translocation. The nuclear p65 NF-κB level increased significantly after LPS stimulation
and was blocked by pre-treatment with 1 at 20 µM (Figure 5C and 5D). These results indicate
that compound 1 might exert its NF-κB inhibitory activity by suppressing the nuclear
translocation of p65 NF-kB but not its phosphorylation nor expression.
16

ACCEPTED MANUSCRIPT
1
1
50
00
A
**
B
LPS (25 ng/mL)
DMSO (0.02 %)
Compound 1 (µM)
-
+
-
+
-
-
+
+
-
+
+
5
+
+
+
+
*
10 20
***
p-p65
p65
50
***
***
0
GAPDH
Ctrl Veh DEX 0.1
1
+
+
5
+
+
20 0.1
1
+
+
5
+
+
20
LPS (25 ng/mL)
DMSO (0.02 %)
-
+
+
+
+
+
+
+
+
+
+
+
+
+
1.5
Andrographolide
µM)
Calcaratarin D
(
(µM)
1
0
0
.0
C
D
LPS (25 ng/mL)
-
-
+
-
+
-
.5 ***
1
(20 µM)
Nuclear p65
Nuclear TBP
Cytoplasmic p65
GAPDH
.0
Ctrl Media Veh
5
10 20
LPS (25 ng/mL)
DMSO (0.02 %)
-
+
-
+
+
+
+
+
+
+
+
+
Compound
1 (µM)
Figure 5. Effects of compound 1 on LPS-induced NF-κB activation. (A) NF-κB transcriptional
TM
activity in RAW-Blue cells using a reporter gene assay. Cells were pretreated with compound
1
(0.1 to 20 µM), dexamethasone (DEX; 1 µM) or with DMSO as vehicle for 1 h before
stimulated with LPS (25 ng/mL) for 24 h. The supernatants were analyzed for SEAP activity
with LPS-treated cells inin 0.02% DMSO counted as 100% SEAP activity. Data are presented as
mean ± SEM (n = 4). (B) Immunoblots and corresponding quantifications of phosphorylated and
total p65 levels in whole cell lysate in response to increasing concentrations of compound 1 (5 to
2
0
µ
M), or vehicle control after stimulation with LPS (25 ng/mL) for 4 h. (C) Immunoblots of
p65 in cytoplasmic and nuclear fractions in response to compound 1 (20 M), or vehicle control
µ
after stimulation with LPS (25 ng/mL) for 4 h. TBP and GAPDH were used as nuclear and
cytoplasmic protein loading controls, respectively. (D) Quantification of nuclear p65 levels.
17

ACCEPTED MANUSCRIPT
Band intensity was quantified using Image Lab software. Data are presented as mean ± SEM (n
=
4). *p < 0.05, **p < 0.01, ***p < 0.001 indicate statistically significant difference as compared
with the 0.02% DMSO-treated cells. (2 column-fitting image)
Calcaratarin D (1) on other LPS-induced signaling pathways. The crosstalk between the
upstream signalling pathways (e.g., MAPK and PI3K) and the downstream NF-κB pathway has
been well-documented.[27] Hence, the effects of calcaratarin D (1) on these pathways were
investigated. As shown in Figure 6A, LPS treatment induced marked phosphorylation of ERK,
JNK, and p38 MAPK in RAW264.7 cells, and pre-treatment with 1 did not significantly alter
LPS-induced phosphorylation of these kinases. In contrast, remarkably, compound 1 suppressed
LPS-induced phosphorylation of Akt in a concentration-dependent manner (Figure 6B). Our
findings suggest that calcaratarin D may act on the upstream inhibition of PI3K/Akt pathway
leading to downstream inhibition of NF-κB transcriptional activity. On the other hand,
andrographolide showed broader inhibition on all the pathways tested (supplementary
information). In fact, the multi-targeting properties of andrographolide have been well
documented.[28] These findings further suggest that the selectivity of the labdanes can be fine-
tuned by selecting an appropriate stereochemistry of the decalin ring or by manipulating the
substituents present at C-3 and C-19 positions.
18

ACCEPTED MANUSCRIPT
A
B
LPS (25 ng/mL)
DMSO (0.02 %)
Compound 1 (µM)
-
+
-
+
-
-
+
+
-
+
+
5
+
+
+
+
LPS (25 ng/mL)
DMSO (0.02 %)
Compound 1 (µM)
-
+
-
+
-
-
+
+
-
+
+
5
+
+
+
+
10 20
10 20
p-AKT
p-ERK1/2
AKT
ERK1/2
GAPDH
1
1
0
0
.5
.0
.5
.0
p-JNK/SAPK
JNK/SAPK
p-p38
***
***
p38
***
GAPDH
***
Ctrl Media Veh
5
10 20
LPS (25 ng/mL)
DMSO (0.02 %)
-
+
+
-
+
+
+
+
+
+
+
+
Compound
1
(µM)
Figure 6. Effects of compound 1 on LPS-induced activation of major signaling pathways. (A)
Immunoblots of ERK1/2, SAPK/JNK, and p38 phosphorylation, and (B) immunoblots and
quantitative analysis of Akt phosphorylation in response to LPS (25 ng/mL) for 4 h in the
presence of increasing concentrtions of compound 1 (5 to 20 µM), or vehicle control. GAPDH
was used as total protein loading control. Band intensity was quantified using Image Lab
software. Data are presented as mean ± SEM (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001
indicate statistically significant difference compared with the 0.02% DMSO-treated cells. (2
column-fitting image)
CONCLUSIONS
In this study, methods for preparing a series of bioactive normal-labdane diterpenes were
reported, including the first reported synthesis of the natural product calcaratarin D (1) via a
stereo- and regioselective route. The ability of normal-labdane diterpenes to inhibit the
19

ACCEPTED MANUSCRIPT
production of pro-inflammatory cytokines (e.g., TNF-α, IL-6, NO) shows great promise as anti-
inflammatory agents. Calcaratarin D (1) and its isomer 20 were identified as the most potent
compounds amongst all the labdanes tested. The structure-activity relationship (SAR) study
suggests that an
α-alkylidene-β-hydroxy-γ-butyrolactone system is necessary for potent anti-
inflammatory activity in the labdanes.
The natural product calcaratarin D was selected for further studies of its pharmacological
mechanisms. It was found that the compound effectively blocked the production of pro-
inflammatory mediators (e.g., TNF-α, IL-6, NO) at both transcriptional and translational levels
via the suppression of NF-κB activation. Calcaratarin D (a normal-labdane) demonstrated
superior NF-κB inhibitory activities than andrographolide, a well-known ent-labdane NF-κB
inhibitor from the labdane family. This finding suggests that a normal-labdane stereochemistry
or a system with the absence of hydroxyl groups at C-3 and C-19 positions is favorable for
potent NF-κB inhibition. Further mechanistic studies revealed that the compound selectively
suppressed the activation of PI3K/Akt pathway without affecting much of the MAPK (i.e., ERK,
JNK, and p38) activation – a mechanism distinct from that of the multi-targeting
andrographolide. Structurally, calcaratarin D contains a single polar functional group at one end
and a hydropholic scaffold, while both ends of andrographolide are composed of polar functional
groups. This bipolarity of andrographolide has been suggested to lead to interactions with many
inter- and intracellular constituents, giving rise to its multi-targeting properties. Thus it is
possible that the selectivity of calcaratarin D is due to its unipolar structural feature. Altogether,
our study demonstrates the therapeutic potential of calcaratarin D as an effective anti-
inflammatory agent and the involvement of a selective Akt/NF-κB-mediated mechanism
20

ACCEPTED MANUSCRIPT
underlying its actions. Given the multiplicity of protein targets that are commonly associated
with electrophilic natural products and consequently off-target side effects, the selectivity of
calcaratarin D is unexpected and would allow the development of future therapeutics that present
potent, and more specific activity.
EXPERIMENTAL SECTION
General. All reagents and solvents (of analytical and HPLC grades) were used as received from
commercial sources unless otherwise stated. Analytical thin layer chromatography (TLC) was
performed on Merck’s pre-coated silica gel plates and visualized under UV light (254 nm) or by
staining with basic KMnO . Purification of compounds was done using flash chromatography on
4
a glass column using Merck silica gel 60 (230−400 mesh). Melting points were determined on a
1
MPA100 automated melting point apparatus from Stanford Research Systems (SRS). H and
1
3
C-NMR spectra were obtained at 400 MHz and 100 MHz respectively on a Bruker Ultra Shield
spectrometer. The chemical shifts are recorded in parts per million (ppm) on the δ scale using
residual protio-solvent signals as internal reference. MS spectra were performed with ESI or
APCI modes on a Shimadzu single quadrupole liquid chromatograph−mass spectrometer
(
LCMS-2020). High-resolution mass spectrometry was carried out using a Bruker micrOTOF-Q
II 10269 mass spectrometer. The purity of the synthesized compounds was determined by
analytical high performance liquid chromatography (HPLC) on a Shimadzu Ultra Fast Liquid
Chromatograph Prominence series (LC-20AD) system at 219 or 254 nm. The gradient systems
for HPLC separation were composed of either acetonitrile in water (10 – 90% or 50 – 90% v/v)
or methanol in water (60 − 90% v/v).
21

ACCEPTED MANUSCRIPT
Chemistry
[(1R,4aS,8aS)
methoxy N-methylacetamide (11). To a stirred solution of (+)-sclareolide (2000.0 mg, 7.98
2



2
hydroxy
2,5,5,8a
tetramethyl
decahydronaphthalen
1
yl]
N



mmol, 1.0 equiv.) and NHMe(OMe).HCl (1550.0 mg, 15.89 mmol, 2 equiv.) in dry THF (20
mL) at 0 ºC was added i-PrMgCl (2.0 M in THF, 16 mL, 32.00 mmol, 4 equiv.) dropwise within
1
0 min. The reaction mixture was stirred at 0 ºC for 2 h and at room temperature overnight. After
cooling to 0 ºC, aqueous solution of NH Cl (50 mL) was added slowly and carefully until an
evolution of a large amount of gas was observed. The reaction mixture was allowed to warm to
room temperature, diluted with EA, and washed successively with H O, and brine, dried over
anhydrous Na SO , filtered, evaporated and purified by silica gel column chromatography to
4
2
2
4
afford the title product as a white solid (2080 mg, 84 %). 11 showed identical spectroscopic data
to those reported. [29]
2


[(1S,4aS,8aS)
12). The compound was synthesized following a procedure described in the literature.[30]
3R,4S) [(1S) [(1S,4aS,8aS) 5,5,8a trimethyl methylidene
yl] hydroxyethyl] hydroxyoxolan
2
one (13). To a stirred

5,5,8a
trimethyl
2
methylidene
decahydronaphthalen
1
yl]acetaldehyde
(
(


3




2








2




decahydronaphthalen


1




1




4


solution of diisopropylamine (220 µL, 1.570 mmol, 3.65 equiv.) in dry THF (2 mL) was added
n-BuLi (1.6 M in hexane, 480 µL, 0.768 mmol, 1.8 equiv.) dropwise at 0 °C under an inert (N2)
atmosphere, and stirred for ca 30 min. The resulting LDA solution was cooled to –78 °C, and
was added (S)-(-)-β-hydroxy-γ-butyrolactone (70 mg, 0.686 mmol, 1.6 equiv.) in dry THF (0.6
mL). After stirring for another 30 min, a suspension of white solid was formed and a solution of
aldehyde 12 (100 mg, 0.427 mmol, 1.0 equiv.) in dry THF (1 mL) and HMPA (0.9 mL) was
added dropwise. The reaction mixture was allowed to warm to –30 °C and stirred overnight
22

ACCEPTED MANUSCRIPT
before quenching with saturated aqueous NaHCO
3
(5 mL). The resulting mixture were diluted
with EA, and washed successively with H O, and brine, dried over anhydrous Na SO , filtered,
2
2
4
evaporated and purified by silica gel column chromatography to afford a mixture of 13 and 14 in
a 2:3 ratio (50 mg, 35%). The isolated title product was obtained as a white solid. Mp 139 – 141
1
°
C. H NMR (400 MHz, CDCl ) δ 4.92 (s, 1H), 4.77 (m, 1H), 4.70 (s, 1H), 4.47 (m, 1H), 4.15
3
1
3
(
m, 1H), 4.05 (m, 1H), 2.60 – 1.01 (m, 17H), 0.88 (s, 3H), 0.80 (s, 3H), 0.70 (s, 3H); C NMR
100 MHz, CDCl ) δ 176.5 (C-16), 150.1 (C-8), 107.3 (C-17), 73.3 (C-15), 71.4 (C-14), 68.8 (C-
2), 55.6 (C-9), 54.2 (C-13), 53.7 (C-5), 42.0 (C-3), 40.1 (C-10), 39.0 (C-1), 38.3 (C-7), 33.6 (C-
(
3
1
4
), 33.5 (C-18), 29.1 (C-11), 24.4 (C-6), 21.7 (C-19), 19.3 (C-2), 14.5 (C-20). APCI-MS (m/z)
+
for C H O [M + H] 337.2301; found 337.2.
2
0
32
4
(
3S,4S)


3


[(1R)


2


[(1S,4aS,8aS)


5,5,8a


trimethyl

2
methylidene

decahydronaphthalen


1


yl] hydroxyethyl]


1




4


hydroxyoxolan
2
one (14). Obtained
from the same reaction mixture as 13 after column chromatography. The isolated title product
1
was obtained as a white solid. H NMR (400 MHz, CDCl ) δ 4.86 (d, J = 1.6 Hz, 1H), 4.60 (m,
3
1
0
1
H), 4.54 – 4.42 (m, 2H), 4.09 – 4.00 (m, 2H), 2.65 – 1.01 (m, 17H), 0.88 (s, 3H), 0.81 (s, 3H),
1
3
.69 (s, 3H); C NMR (100 MHz, CDCl
3
) δ 175.8 (C-16), 148.6 (C-8), 106.6 (C-17), 72.9 (C-
5), 70.3 (C-15), 68.7 (C-12), 55.5 (C-9), 54.4 (C-13), 51.9 (C-5), 42.1 (C-3), 39.4 (C-10), 39.0
(C-1), 38.3 (C-7), 33.6 (C-4), 33.6 (C-18), 28.9 (C-11), 24.4 (C-6), 21.7 (C-19), 19.3 (C-2), 14.6
+
(
C-20). APCI-MS (m/z) for C H O [M + H] 337.2301; found 337.2.
2
0
32
4
3
1
1




[(1R)


2


[(1S,4aS,8aS)


5,5,8a


trimethyl
2
methylidene
decahydronaphthalen
1
yl]

hydroxyethyl]


2,5
dihydrofuran
2
one (2). Obtained from the same reaction mixture as
3 and 14 after column chromatography, when the solution of aldehyde 12 (300 mg, 1.281
mmol, 1.0 equiv.) in dry THF was added at –30 °C. The title product was isolated as an
23

ACCEPTED MANUSCRIPT
1
amorphous white solid (20 mg, 5 %). H NMR (400 MHz, CDCl
3
) δ 7.31 (m, 1H), 4.89 (d, J =
1
1
1
.5 Hz, 1H), 4.82 (m, 2H), 4.71 (d, J = 1.5 Hz, 1H), 4.55 (m, 1H), 2.41 (m, 2H), 2.13 – 0.96 (m,
1
3
3H), 0.88 (s, 3H), 0.80 (s, 3H), 0.68 (s, 3H); C NMR (100 MHz, CDCl ) δ 173.1 (C-16),
3
48.3 (C-8), 144.1 (C-14), 137.6 (C-13), 107.1 (C-17), 70.5 (C-15), 65.8 (C-12), 55.5 (C-5), 51.9
(C-9), 42.1 (C-3), 39.3 (C-10), 39.0 (C-1), 38.3 (C-7), 33.6 (C-4), 30.3 (C-11), 24.3 (C-6), 21.7
+
(
C-19), 19.3 (C-2), 14.6 (C-20). APCI-TOF-HRMS (m/z) for C H O [M + H] 319.2268;
2
0
30
3
found 319.2271.[14, 15]
(
3S,4S)
decahydronaphthalen
15). To a mixed solution of 13 and 14 in a 2:3 ratio (168.0 mg, 0.500 mmol, 1.0 equiv.) in dry


3


[(1R)


2


[(1S,4aS,8aS)


5,5,8a


trimethyl

2
methylidene



1


yl] 1-hydroxyethyl]




4

[(tert-butyldimethylsilyl)oxy]oxolan

2
one
(
THF (5 mL), was added imidazole (135 mg, 1.982 mmol, ~ 4 equiv.) and TBSCl (265 mg, 1.758
mmol, ~ 3.5 equiv.). The reaction mixture was stirred at room temperature under an inert (N2)
atmosphere for 2 days before quenching with saturated aqueous NaHCO (20 mL), diluted with
3
EA, and washed successively with H O, and brine, dried over anhydrous Na SO , filtered,
2
2
4
evaporated and purified by silica gel column chromatography to afford the title product as a
1
white solid (140 mg, quantitative yield). Mp 142 – 144 °C. H NMR (400 MHz, CDCl
3
) δ 4.84
(d, J = 1.5 Hz, 1H), 4.49 (m, 1H), 4.44 – 4.35 (m, 2H), 3.98 – 3.85 (m, 2H), 2.56 – 2.36 (m, 3H),
2
3
3
.06 – 1.02 (m, 13H), 0.88 (s, 9H), 0.87 (s, 3H), 0.80 (s, 3H), 0.68 (s, 3H), 0.09 (s, 3H), 0.07 (s,
1
3
H); C NMR (100 MHz, CDCl ) δ 176.2, 148.8, 106.2, 73.4, 71.1, 68.5, 55.4, 55.0, 52.0, 42.1,
3
9.3, 39.0, 38.2, 33.6, 33.6, 29.7, 25.6, 24.4, 21.7, 19.3, 18.0, 14.6, -4.4, -4.9. APCI-MS (m/z)
+
for C H O Si [M + H] 451.3165; found 451.3.
2
6
46
4
(
1R)


2


[(1S,4aS,8aS)


5,5,8a


trimethyl


2


methylidene decahydronaphthalen



1
yl]
1

[
(3R,4S)


4


[(tert butyldimethylsilyl)oxy]




2


oxooxolan
3
yl]ethyl methanesulfonate (16).
24

ACCEPTED MANUSCRIPT
To a solution of compound 15 (140 mg, 0.311 mmol, 1 equiv.) in dry DCM (2 mL) was added
Et N (200 µL, 1.435 mmol, ~ 4.5 equiv.) under an inert (N ) atmosphere. After cooling to –78
3
2
ºC, MsCl (100 µL, 1.291 mmol, ~ 4.0 equiv.) was added dropwise at same temperature. The
reaction mixture was stirred for 1h before quenching with saturated aqueous NaHCO (20 mL),
3
diluted with EA, and washed successively with H O, and brine, dried over anhydrous Na SO ,
2
2
4
filtered, evaporated and purified by silica gel column chromatography to afford the title product
1
as a colorless gel (50 mg, 30 %). H NMR (400 MHz, CDCl
3
) δ 5.25 (m, 1H), 4.95 (s, 1H), 4.70
(m, 1H), 4.48 – 4.39 (m, 2H), 3.95 (m, 1H), 3.23 (s, 3H), 3.07 (m, 1H), 2.48 – 1.02 (m, 14H),
1
3
0
.90 (s, 9H), 0.88 (s, 3H), 0.80 (s, 3H), 0.69 (s, 3H), 0.13 (s, 3H), 0.10 (s, 3H); C NMR (100
MHz, CDCl ) δ 172.6, 147.8, 107.3, 81.5, 72.8, 69.6, 68.9, 55.3, 53.4, 51.6, 42.4, 41.9, 39.6,
3
3
8.7, 37.9, 33.6, 33.5, 25.6, 24.3, 21.7, 19.3, 17.7, 14.6, -4.5, -4.9. APCI-MS (m/z) for
-
C H O SSi [M + H] 527.2941; found 527.2.
27
48
6
(1S)


2

[(1S,4aS,8aS)


5,5,8a


trimethyl


2

methylidene


decahydronaphthalen
1
yl]

1
[(3R,4S) [(tert butyldimethylsilyl)oxy]


4






2

oxooxolan
3

yl]ethyl methanesulfonate (17).
Obtained from the same reaction mixture as 16 after column chromatography. The title product
was isolated as a colorless gel (35 mg, 21 %). It is noteworthy that only compound 16 was
1
obtained in almost quantitative yield when MsCl was added to the reaction mixture at 0 °C. H
NMR (400 MHz, CDCl ) δ 5.10 (m, 1H), 4.93 (s, 1H), 4.71 (m, 1H), 4.50 (s, 1H), 4.42 (m, 1H),
3
3
.99 (m, 1H), 3.02 (s, 3H), 3.00 (m, 1H), 2.48 – 0.96 (m, 14H), 0.89 (s, 9H), 0.88 (s, 3H), 0.81
13
(
s, 3H), 0.69 (s, 3H), 0.13 (s, 3H), 0.10 (s, 3H); C NMR (100 MHz, CDCl ) δ 173.1, 147.7,
3
1
2
5
07.3, 78.0, 73.7, 70.0, 55.8, 54.0, 52.1, 42.0, 39.5, 39.1, 38.3, 38.1, 33.6, 33.5, 27.0, 25.6, 24.3,
-
1.7, 19.3, 17.8, 14.6, -4.7, -4.8. APCI-MS (m/z) for C27
27.2.
H
48
O
6
SSi [M + H] 527.2941; found
25

ACCEPTED MANUSCRIPT
(
3Z,4S)


3


{2

[(1S,4aS,8aS)


5,5,8a


trimethyl


2


methylidene
decahydronaphthalen
1

yl]ethylidene}


4

[(tert-butyldimethylsilyl)oxy]oxolan
2

one (18). To a solution of 16 (45
mg, 0.085 mmol, 1.0 equiv.) in dry DCM (2 mL) was added DIEA (600 µL, 3.446 mmol, ~ 40
equiv.) under an inert (N ) atmosphere and stirred for overnight. The reaction mixture was
2
diluted with DCM and washed successively with saturated aqueous NH Cl, H O, and brine, dried
4
2
over anhydrous Na SO , filtered, evaporated and purified by silica gel column chromatography
2
4
1
to afford the title product as a deliquescent white solid (16 mg, 44 %). H NMR (400 MHz,
CDCl ) δ 6.38 (m, 1H), 4.83 (s, 1H), 4.77 (m, 1H), 4.50 (s, 1H), 4.35 (m, 1H), 4.01 (m, 1H), 3.02
m, 1H), 2.70 (m, 1H), 2.39 (m, 1H), 2.04 – 1.32 (m, 11H), 1.25 (s, 9H), 1.11 (s, 3H), 0.81 (s,
3
(
1
3
3
1
1
H), 0.74 (s, 3H), 0.08 – 0.05 (m, 6H); C NMR (100 MHz, CDCl ) δ 168.9, 149.5, 148.0,
3
27.5, 107.8, 73.5, 69.9, 57.1, 55.5, 42.2, 39.7, 39.1, 38.0, 33.6, 33.6, 25.6, 24.2, 23.3, 22.7, 21.7,
+
9.3, 14.4, -4.5, -4.5. APCI-MS (m/z) for C H O Si [M + H] 433.3060; found 433.3.
2
6
44
3
(
3E,4S)
yl]ethylidene}
.057 mmol, 1.0 equiv.) analogously to compound 18. The title product was isolated as a


3


{2


[(1S,4aS,8aS)


5,5,8a


trimethyl


2


methylidene


decahydronaphthalen

1



4

[(tert butyldimethylsilyl)oxy]oxolan




2


one (19)
.
Prepared from 17 (30 mg,
0
deliquescent white solid (10 mg, 41 %). It is noteworthy that, when DMAP (4 equiv.) was used
as a base for the elimination (1 h), both 18 and 19 were obtained from either 16 or 17 in different
1
ratios. H NMR (400 MHz, CDCl ) δ 6.80 (m, 1H), 5.06 (m, 1H), 4.78 (d, J = 1.6 Hz, 1H), 4.42
3
(m, 1H), 4.23 (d, J = 1.5 Hz, 1H), 4.08 (m, 1H), 2.57 (m, 1H), 2.49 – 2.29 (m, 2H), 2.09 – 1.89
(m, 2H), 1.80 – 1.04 (m, 9H), 0.91 (s, 9H), 0.89 (s, 3H), 0.82 (s, 3H), 0.72 (s, 3H), 0.17 (s, 3H),
1
3
0
5
.12 (s, 3H); C NMR (100 MHz, CDCl ) δ 170.0, 149.0, 148.5, 127.4, 107.5, 73.9, 67.3, 56.4,
3
5.3, 42.0, 39.4, 39.4, 37.7, 33.5, 29.7, 25.6, 25.3, 24.1, 21.7, 19.3, 17.9, 14.6, -4.1, -4.7. APCI-
+
MS (m/z) for C26
H
44
O
3
Si [M + H] 433.3060; found 433.3.
26

ACCEPTED MANUSCRIPT
(
3Z,4S)
yl]ethylidene}
mmol, 1 equiv.) in AcOH/H O/THF (v/v/v = 7:2:1, 3 mL) was added p-TsOH.H O (35 mg,


3


{2


[(1S,4aS,8aS)


5,5,8a


trimethyl

2
methylidene
decahydronaphthalen
1



4

hydroxyoxolan

2
one (20). To a solution of compound 18 (53 mg, 0.122
2
2
0
.184 mmol, 1.5 equiv.). The reaction mixture was stirred for 1 day before quenching with
saturated aqueous NaHCO (2 mL), diluted with EA, and washed successively with H O, and
3
2
brine, dried over anhydrous Na SO , filtered, evaporated and purified by silica gel column
2
4
1
chromatography to afford the title product as an amorphous white solid (30 mg, 77 %). H NMR
400 MHz, CDCl ) δ 6.56 (m, 1H), 4.83 (d, J = 1.5 Hz, 1H), 4.78 (m, 1H), 4.51 (d, J = 1.5 Hz,
H), 4.39 (m, 1H), 4.17 (m, 1H), 3.02 – 2.76 (m, 2H), 2.39 (m, 1H), 2.18 – 1.02 (m, 12H), 0.87
(
3
1
1
3
(
s, 3H), 0.81 (s, 3H), 0.74 (s, 3H); C NMR (100 MHz, CDCl ) δ 168.8 (C-16), 151.3 (C-12),
3
1
48.4 (C-8), 127.7 (C-13), 107.8 (C-17), 73.3 (C-15), 69.5 (C-14), 57.2 (C-9), 55.5 (C-5), 42.1
(C-3), 39.7 (C-10), 39.1 (C-1), 38.0 (C-7), 33.6 (C-4), 33.6 (C-18), 24.2 (C-11), 23.4 (C-6), 21.7
+
(
C-19), 19.3 (C-2), 14.4 (C-20). APCI-TOF-HRMS (m/z) for C H O [M + H] 319.2268;
2
0
30
3
found 319.2261.
(
3E,4S)


3


{2


[(1S,4aS,8aS) trimethyl


5,5,8a



2
methylidene
decahydronaphthalen
1

yl]ethylidene}


4

hydroxyoxolan
2
one (1). Prepared from 19 (56 mg, 0.129 mmol, 1.0
equiv.) analogously to compound 20. The title product was isolated as an amorphous white solid
1
(
35 mg, 85 %). H NMR (400 MHz, CDCl ) δ 6.95 (m, 1H), 5.06 (m, 1H), 4.82 (d, J = 1.5 Hz,
3
1
2
H), 4.45 (m, 1H), 4.35 (d, J = 1.5 Hz, 1H), 4.25 (m, 1H), 2.67 (m, 1H), 2.51 – 2.34 (m, 2H),
.19 (m, 1H), 2.08 – 1.88 (m, 2H), 1.79 – 1.68 (m, 2H), 1.57 – 1.06 (m, 7H), 0.89 (s, 3H), 0.82
1
3
(
s, 3H), 0.74 (s, 3H); C NMR (100 MHz, CDCl ) δ 169.9 (C-16), 150.0 (C-12), 148.2 (C-8),
3
1
1
27.9 (C-13), 107.5 (C-17), 74.2 (C-15), 66.5 (C-14), 56.5 (C-9), 55.4 (C-5), 42.0 (C-3), 39.7 (C-
0), 39.2 (C-1), 37.8 (C-7), 33.6 (C-4), 33.6 (C-18), 25.2 (C-11), 24.1 (C-6), 21.7 (C-19), 19.3
27

ACCEPTED MANUSCRIPT
+
(
C-2), 14.4 (C-20). APCI-TOF-HRMS (m/z) for C20
19.2263.[17]
Diethyl (2 oxooxolane
600.0 mg, 3.64 mmol) and triethylphosphite (950 µL, 5.54 mmol, ~ 1.5 equiv.) was refluxed
30 3
H O [M + H] 319.2268; found
3



3
carbonyl)phosphonite (21). A solution of α-Bromo-γ-butyrolactone
(
under neat condition at 130 °C for 5 h. The excess of triethylphosphite was removed under
vacumm oven at 40 °C. The product was used without further purification.
(
3Z)
yl]ethylidene}oxolan
procedure described in the literature.[18]
3E) {2 [(1S,4aS,8aS) 5,5,8a trimethyl
yl]ethylidene}oxolan one (23). Mp 55 – 58 °C. The compound was synthesized following a
procedure described in the literature.[18]


3


{2


[(1S,4aS,8aS)

5,5,8a
trimethyl
2
methylidene
decahydronaphthalen
1


2

one (22). Mp 75 – 77 °C. The compound was synthesized following a
(


3









2
methylidene
decahydronaphthalen
1


2

(
2E) 5,5,8a


4


[(1S,4aS,8aS)



trimethyl
2
methylidene
decahydronaphthalen
1
yl]
2

(2


hydroxyethyl)but
2
enoic acid (6). To a stirred solution of 23 (200.0 mg, 0.661 mmol, 1.0
equiv.) in EtOH/THF (1/10 v/v, 5 mL) at room temperature was added KOH (1 M, 3.3 mL, 3.3
mmol, 5.0 equiv.). The resulting mixture was stirred for overnight and the organic solvent was
removed before acidifying with concentrated HCl to pH ≈ 2. The resulting suspension was
extracted with EA and the combined organic layer was washed successively with saturated
aqueous H O, and brine, dried over anhydrous Na SO , filtered, evaporated and purified by silica
2
2
4
gel column chromatography to afford the title product as a white solid (210 mg, 99 %). Mp 138 –
42 °C. 6 showed identical spectroscopic data to those reported.[10]
3E) {2 [(1S,4aS,8aS) 5,5,8a trimethyl methylidene decahydronaphthalen
yl]ethylidene} hydroxyoxolan one (3). To a solution of 6 (100.0 mg, 0.312 mmol, 1.0
1
(


3










2





1



5


2

28