Synthesis and pharmacological profile of serofendic acids A and B

Article abstract of DOI:10.1016/j.bmc.2007.07.037

We present efficient syntheses of serofendic acids A and B (SA-A and SA-B), novel neuroprotective substances isolated from fetal calf serum. Biological and pharmacological evaluation showed that SA-A and SA-B have potent protective action against glutamat

Full text of DOI:10.1016/j.bmc.2007.07.037

Bioorganic & Medicinal Chemistry 15 (2007) 7098–7107
Synthesis and pharmacological profile of serofendic acids A and B
Taro Terauchi,a,* Naoki Asai,a Takashi Doko,a Ryota Taguchi,a Osamu Takenaka,a
Hideki Sakurai,a Masahiro Yonaga,a Teiji Kimura,a Akiharu Kajiwara,a
Tetsuhiro Niidome,b Toshiaki Kume,b Akinori Akaikeb and Hachiro Sugimotob
aTsukuba Research Laboratories, Eisai Co., Ltd, Tsukuba-shi 300-2635, Japan
bDepartment of Pharmacology, Graduate School of Pharmaceutical Sciences, Kyoto University, Kyoto 606-8501, Japan
Received 26 June 2007; revised 11 July 2007; accepted 13 July 2007
Available online 22 August 2007
Abstract—We present efficient syntheses of serofendic acids A and B (SA-A and SA-B), novel neuroprotective substances isolated
from fetal calf serum. Biological and pharmacological evaluation showed that SA-A and SA-B have potent protective action against
glutamate-induced neurotoxicity, but do not interact directly with glutamate receptors. A pharmacokinetic study showed that they
have good oral bioavailability in rats. The results indicate that SA-A and SA-B are potential lead compounds for candidate drugs to
treat various neurological disorders.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
libraries, including natural products derived from mar-
ine sources or plants.7 We have sought to discover novel
neuroprotective substances from animal sources, based
on the hypothesis that animals may have intrinsic anti-
neurodegenerative mechanisms.
L-Glutamate (Glu) is a major excitatory neurotransmit-
ter in the central nervous system (CNS) and plays an
important role in neurological processes, including cog-
nition, learning, and memory.1 Excessive stimulation of
glutamate receptors, under pathophysiological condi-
tions, leads to neuronal damage and death. This Glu
neurotoxicity is associated with various neurological
disorders, including hypoxic–ischemic brain injury,2
Alzheimer’s disease,3 Huntington’s disease, and Parkin-
son’s disease.4 MK-801 (Dizocilpine) and phencyclidine,
typical N-methyl-D-aspartate receptor (NMDAR)
antagonists, showed potent neuroprotective effects in
various neurodegenerative models,5 but in clinical trials
they induced schizophrenia-like symptoms, which were
ascribed to the NMDAR-antagonistic activity itself.6
Therefore, substances which can prevent Glu neurotox-
icity without acting directly on NMDARs would be can-
didate drugs for the treatment of various neurological
and neurodegenerative diseases.
We previously reported serofendic acids A and B (SA-A
1 and SA-B 2, respectively) (Fig. 1), which were found in
a lipophilic fraction of fetal calf serum, as novel neuro-
protective substances.8 These compounds have a unique
structure, that is, an atisane-type diterpenoid (15-hydro-
xy-17-methylsulfinylatisane-19-oic acid) with a sulfoxide
group, the epimers of which correspond to SA-A and
SA-B. In order to clarify the pharmacological profile
of the serofendic acids and their derivatives, we required
a synthetic route to provide substantial amounts of the
compounds. Here, we report the synthesis and pharma-
cokinetic profile of SA-A and SA-B and their
derivatives.
Recently, many drug discovery programs have been
based on high-throughput screening of huge chemical
17
21
12
17
21
12
18
S
16
S
18
16
1
9
1
O
9
O
15
OH
15
OH
H
H
5
5
3
7
3
7
HO2C
H
H
HO2C
19
Keywords: Serofendic acid; Neuroprotective; Neurodegenerative dis-
ease; Neurotoxicity; Glutamate; Atisane.
20
19
20
1
2
*
Corresponding author. Tel.: +81 29 847 5843; fax: +81 29 847
4012; e-mail: t-terauchi@hhc.eisai.co.jp
Figure 1. Structure of serofendic acids A (1) and B (2).
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.07.037

T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
7099
2. Results and discussion
bond with the (COCl)2–DMSO–TEA reagent system.11
When this condition was examined, we obtained 13 with
a modestly improved selectivity (13:14 = 4.5:1, 95%
yield). Regarding E2 elimination, it is known that Hof-
mann elimination tends to be favored over Saytzeff elim-
ination when the leaving group is large or cationic.
Based on this insight, dehydration of 12 with Martin’s
dehydrating agent was conducted.12 Finally, we found
that preparation of the reagent and the dehydration
reaction could be conducted in one pot to obtain the ole-
fins with high selectivity (13:14 = 11.5:1) and high yield
(93%). Next, treatment of the olefins 13 and 14 with
SeO2 afforded 16, 17, and 18, together with unchanged
14. Each alcohol was isolated and the stereochemistries
of 16 or 17 were determined by means of NOESY exper-
iments (Fig. 2). Although we tried to get the diol 22 by
straightforward oxidation, hydroboration of 16 with the
same stereochemistry at the C15 position as those of the
serefendic acids gave a syn-diastereomer 19 as the only
product. Similarly, hydroboration of 17 afforded the
corresponding syn-diastereomer 20. The relative config-
uration at C15 and C16 of 19 or 20 was determined
based on the coupling constants for H15 and H16 in
the 1H NMR spectra. The results indicated that the bor-
ane reagent selectively approached from the opposite
side to the secondary alcohol at the C15 position. Since
the product obtained by the oxidation reaction did not
exhibit the natural configuration, inversion of the sec-
ondary alcohol on 20 by the Mitsunobu method
(HCO2H, DEAD, PPh3, and THF)13 was investigated.
However, the reaction did not proceed at all. Then, we
tried to invert the secondary alcohol of 20 by means of
an oxidation followed by reduction process. Selective
oxidation of the secondary alcohol of the diol 20 was
2.1. Chemistry
Our synthetic route is depicted in Schemes 1 and 2.9 The
diterpene skeleton of SA-A and SA-B was derived from
commercially available isosteviol (3) according to
Coates and Bertram’s method with some modifica-
tions.10 Isosteviol (3) was converted to 4 by Baeyer–Vil-
liger oxidation, followed by esterification and reduction
with lithium borohydride to provide the diol 6. Acetyla-
tion of the primary hydroxyl group of 6 gave 7. Coates
reported that dehydration with thionyl chloride–colli-
dine/CH2Cl2 afforded the desired olefin 8 and undesired
olefin 9 in 2:1 ratio. To increase the selectivity between 8
and 9, we tried to use only acid. Use of methanesulfonic
acid gave a complex reaction mixture that did not in-
clude 8 or 9, but use of TFA gave 8 in a 6:1 ratio over
9. Hydrolysis of the mixture of 8 and 9 with NaOH/
EtOH provided a primary alcohol. The C19 carbome-
thoxy group remained intact, presumably because of
high steric hindrance. Tosylation then gave a mixture
of 10 and 11. Solvolysis of the tosylate mixture, followed
by alkaline hydrolysis, gave the cyclized 12 and un-
changed 11. Dehydration of the tertiary alcohol 12 with
thionyl chloride/pyridine–CH2Cl2 afforded the exo-ole-
fin 13 and endo-olefin 14 in 1:1 ratio (76% yield).
Although regioselectivity was not obtained, both iso-
mers might lead to the desired product. Thus, we tried
to introduce the functional groups at the C15 and C16
positions of 13 and 14 by means of SeO2 oxidation fol-
lowed by hydroboration. Oxidation of 13 gave the cor-
responding oxidized products, while the reaction of 14
did not proceed at all. This result presumably reflects
steric hindrance at C15, which is the primary reaction
site for SeO2. Next, chemoselective dehydration of 12
to 13 was investigated. Satisfactory results were not
achieved with TFA or MsCl–collidine (Table 1). It has
been reported that 2-methylbicyclo[2.2.1]heptan-2-ol
was predominantly dehydrated on the exocyclic double
14
attempted. Use of NaBrO3–NaHSO3 gave the keto
alcohol 21 in high yield (80%), and this was reduced with
NaB(OAc)3H to provide the diol 22 in 73% yield. Selec-
tive tosylation of the resulting diol 22 followed by thi-
omethylation and simultaneous hydrolysis of the
methylester with NaSMe gave the sulfide 24 in 51%
OH
O
O
O
a
b
c
H
H
O
H
O
H
OH
HO2C
H
HO2C
OH
H
MeO2C
H
MeO2C
MeO2C
H
5
6
4
3
12
13
14
f
e
d
H
OTs
H
H
OAc
H
OAc
H
10: Δ
11:
MeO2C
MeO2C
H
7
12-13
13-14
12-13
8:
9:
Δ
Δ
13-14
Δ
17
17
16
16
15
OH
15
g
h
H
H
H
MeO2C
H
13
MeO2C
H
12
MeO2C
H
14
Scheme 1. Reagents and conditions: (a) H2O2 aq, H2SO4, AcOH, rt, 71 h, 93%; (b) i—(COCl)2, DMF, CH2Cl2, rt, 2.5 h; ii—TEA, MeOH, rt, 1 h,
92%; (c) LiBH4, THF, rt, 61 h, 72%; (d) Ac2O, pyridine, rt, 9 h, 97%; (e) TFA, CH2Cl2, rt, 18 h; (f) i—NaOH, EtOH, rt; ii—TsCl, pyridine, rt, 16 h;
(g) i—Na2CO3, HCO2H, rt, 16 h; ii—NaOH, EtOH, rt, 3 h, 61% from 7; (h) dehydrating condition.

7100
T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
OH
a
H
H
H
Ph(CF3)2C O Ph
Ph(CF3)2C O Ph
MeO2C
H
12
MeO2C
H
13
MeO2C
H
14
Martin's agent
F3C
OK
CF3
15
OH
b
OH
OH
+
+
H
H
H
MeO2C
H
MeO2C
H
MeO2C
H
18
16
17
c
d
OH
OH
OH
OH
OH
e
f
O
H
H
H
MeO2C
H
MeO2C
H
20
MeO2C
H
19
21
OH
OH
SMe
OH
OTs
OH
g
h
H
H
H
MeO2C
H
HO2C
H
MeO2C
H
i
j
22
23
24
SOMe
SOMe
OH
OH
H
H
MeO2C
H
HO2C
H
25
serofendic acid
Scheme 2. Reagents and conditions: (a) Ph–C(CF3)2–OK, (Ph)2S, Br2, CCl4, rt, 3 h then 12, CHCl3, rt, 3 h, 93%, 13:14 = 11.5:1; (b) SeO2, t-BuOOH,
CH2Cl2, rt, 15 h, 26% 16, 28% 17, 2.9% 18; (c) BH3–THF, THF, rt, 4 h then NaOH aq, H2O2 aq, rt, 2 h, 80%; (d) BH3–THF, THF, rt, 4 h then
NaOH aq, H2O2 aq, rt, 3 h, 91%; (e) NaBrO3, NaHSO3, CH3CN, H2O, rt, 3 h, 80%; (f) NaB(OAc)3H, AcOH, CH3CN, rt, 4 h, 73%; (g) TsCl,
DMAP, pyridine, rt, 17 h, 67%; (h) NaSMe, HMPA, 80 ꢁC, 43 h, 76%; i—Davis’s oxaziridine (2-benzenesulfonyl-3-phenyloxazirine), CHCl3, 0 ꢁC,
2 h, quant; (j) i—NaSMe, DMF, rt, 2 h; ii—Davis’s oxaziridine (2-benzenesulfonyl-3-phenyloxazirine), CHCl3, 0 ꢁC, 1 h, 73%, 2 step.
Table 1. Dehydration of 12
The corresponding ester derivatives of the serofendic
acids were also prepared.16 Since the ester group at C4
12 ! 13 14
13:14
Yield (%)
Condition (equiv.)
is resistant to hydrolysis at room temperature, the
methylester analogues of serefendic acid (25) were
obtained by thiomethylation at room temperature, fol-
lowed by oxidation of the sulfide group in the same
way as described for the synthesis of serofendic acid.
HPLC separation afforded 25A and 25B.
1 SOCl2 (4.5), pyridine–CH2Cl2, rt, 5 min
2 TFA (10), CH2Cl2, rt, 3 h
1:1
—
76
0
a
3 MsCl (3), collidine (10), CH2Cl2,
ꢀ60 ꢁC ! rt, 14 h
1.4:1
—
4 DMSO (8), (COCl)2 (4), TEA (12)
CH2Cl2, ꢀ70 ꢁC ! rt, 2 h
4.5:1
95
93
5 Martin’s agent (2), CH2Cl2, rt, 3 h
11.5:1
2.2. Neuroprotective activity
a Yield was not confirmed.
Neuroprotective effect of SA-A (1) or SA-B (2) was eval-
uated on primary neuronal cultures obtained from the
cerebral cortex of fetal rats (17–19 days of gestation).17
Neurotoxicity was induced by incubating the cells with
500 lM Glu for 10 min. Under this condition, MK-
801 (1 lM) completely blocked the Glu-induced neuro-
toxicity (data not shown). As shown in Figure 4,
incubation of the cells with SA-A (1) or SA-B (2) for
1 h before, 10 min during, and 24 h after Glu exposure
had a protective effect. In order to identify the mecha-
nism of action, we employed a panel assay with 76
receptors and ion channels, including NMDAR and
overall yield. Finally, oxidation of the sulfide group in
24 by using Davis’s oxaziridine15 quantitatively yielded
SA-A (1) and SA-B (2) as a 1:2 mixture. After separa-
tion of the diastereomers by HPLC, it was confirmed
that each synthetic isomer was consistent with the corre-
sponding natural isomer in physico-chemical properties
and biological activity.8 The absolute stereochemistry
was estimated by chiral derivatization and HPLC-MS
analysis.9a X-ray structure determination of SA-A (1)
or SA-B (2) (Fig. 3) confirmed the structures.

T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
7101
H
H
H
H
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
OH
H
H
H
H
H
H
H
16
17
Figure 2. Protons of 16 and 17 were assigned with COSY spectra, NOESY spectra, TOCSY spectra, HMQC spectra, and HMBC spectrum. The key
NOEs were observed in the NOESY spectra which indicate the C15 stereochemistry as shown.
Figure 3. Single crystal X-ray structure determination of serofendic acids A (1) and B (2).
8,18
100
oxidative stress was investigated.
The neuroprotec-
tive effect was found to be as potent as that against
Glu neurotoxicity. Further, it has been reported that
the serofendic acids protect cardiac myocytes against
oxidant-induced cell death,19 and prevent the loss of
mitochondrial membrane potential and the activation
of caspase-3 induced by Glu exposure in rat cultured
cortical cells.20 Since nitric oxide or oxidative stress acti-
vates caspase-321 and activated caspase-3 induces over-
release of Glu in synapses,22 the serofendic acids might
act at some point prior to over-activation of glutamate
receptors, for example, by protecting the functional
integrity of mitochondria. Detailed investigation of the
mechanism of action is ongoing.
**
**
80
60
40
20
0
2.3. Pharmacokinetics
sham
-
1
2
Since the serofendic acids showed attractive in vitro bio-
logical activity, an in vivo study was conducted. Table 2
shows the pharmacokinetic parameters of SA-A (1), SA-
B (2), and the methylester analogue 25 after intravenous
administration at a dose of 1 mg/kg in rats. The AUC
(area under the concentration–time curve) value of 2
was 1.4 times higher than that of 1. The stereochemistry
of the sulfoxide moiety thus influenced the pharmacoki-
netic profile. Although biological conversion of sulfox-
ide isomers might occur via a reduction–oxidation
process, in fact, interconversion between 1 and 2 was
not observed when either isomer was administered.
The methylester analogue 25, a mixture of isomers,
exhibited a relatively high clearance value and low
AUC value in a preliminary study, presumably reflecting
its greater lipophilicity. In addition, the serofendic acids
were detected as metabolites (AUC value: 0.08 lg h/ml).
compound (10 μM)
Glu (500 μM)
Figure 4. Protective effect of synthetic serofendic acids against Glu
neurotoxicity in cultured cortical neurons. Cultures were exposed to
Glu (500 lM) for 10 min and then incubated in glutamate-free medium
for 24 h. Cultures were treated with serofendic acid for 1 h before,
10 min during, and 24 h after Glu exposure. **P < 0.01, compared with
Glu alone.
a-amino-3-hydroxy-5-methylisoxazole-4-propionic acid
receptor (AMPAR). The serofendic acids had no effect
on any of them up to 10 lM. Next, the effect of the sero-
fendic acids on neurotoxicity induced by nitric oxide or

7102
T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
Table 2. Pharmacokinetic result in rats for intravenous administration of serofendic acids (1, 2) and methylester analogue (25) (1 mg/kg each)
Compound
T1/2 (h)
Vdss (l/kg)
Cl total (l/h/kg)
AUC (lg h/ml)
B/P 0.5 h
B/P 2 h
1 (1 mg/kg iv)
2 (1 mg/kg iv)
25 (1 mg/kg iv)
0.48
0.65
0.55
0.33
0.33
0.57
0.52
0.36
0.88
1.95
2.78
1.44
N.T
0.021
2.20
N.T
N.D
2.21
N.T, not tested; N.D, no detected.
Table 3. Pharmacokinetic result in rats for oral administration of serofendic acid B (2) (5 mg/kg)
Compound
Tmax (h)
T1/2 (h)
Cmax (lg/ml)
AUC (lg h/ml)
5.64
B.A. (%)
41
2 (5 mg/kg po)
0.33
3.90
2.05
Brain concentrations of the compounds were measured
at 0.5 h and 2 h after administration. While the methy-
lester analogue showed a high brain-to-plasma (B/P)
concentration ratio, serofendic acid (only isomer B
was measured) had a very low B/P value. We speculate
that the carboxylic group of serofendic acid might be
the main cause of the low permeability into the brain.
A pharmacokinetic study of serofendic acid B (2) in rats
after oral administration at a dose of 5 mg/kg (Table 3)
showed relatively good oral bioavailability (B.A.). On
the other hand, compound 25 was not detected in plas-
ma after oral administration at a dose of 5 mg/kg (data
not shown). These data indicate that the serofendic acids
and 25 may be ineffective in the clinical context. Work to
improve the brain penetration or oral bioavailability by
modification of the serofendic acids or 25 is ongoing.
ppm (d) from tetramethylsilane as an internal standard
and coupling constants are given in Hz (J). When
CD3OD was used as the NMR solvent, chemical shifts
were referred to the solvent peaks: dH 3.35 for CD2HOD
and dC 49.0 for CD3OD. Column chromatography was
performed on Fuji Silysia BW silica gel (200–400 mesh).
TLC was carried out on silica gel Merck 60F254. Re-
agents and solvents were purchased and used without
further purification. HPLC separation was performed
with a Biotage Parallex Flex or a MDA-0054 (Waters)
apparatus. Electrospray ionization (ESI) MS were re-
corded on a Thermo Fisher Scientific SSQ7000 mass
spectrometer. ESI-HRMS spectra were recorded with
a Q-TOF Ultima Global Alliance 2695 (Waters) and
API-III plus mass spectrometer (Applied Biosystems).
Specific rotations were measured on a JASCO DIP-
1000 digital polarimeter.
3. Conclusion
4.1.1. (1R,4S,5R,9S,10R,13S)-5,9,13-Trimethyl-15-oxo-
14-oxatetracyclo[11.3.1.01,10.04,9]heptadecane-5-carbox-
ylic acid (4). To a solution of isosteviol (30 g, 94.1 mmol)
in AcOH (315 ml) and H2SO4 (5.25 ml) was added drop-
wise a 30% hydrogen peroxide solution (105 ml) over
30 min at 0 ꢁC. The reaction mixture was then stirred
for 71 h at room temperature and cooled with a dry-
ice ethanol bath (ca. ꢀ40 ꢁC). Saturated aq Na2S2O3
was slowly added, while the temperature was kept below
20 ꢁC. The disappearance of peroxide was confirmed
with potassium iodide-starch paper, then the mixture
was diluted with water (2000 ml) and extracted with
Et2O (700 ml ·2). The organic layer was washed with
brine (300 ml), dried over MgSO4, and evaporated un-
der reduced pressure. The resulting solid was washed
with n-heptane/Et2O (4:1, v/v, 400 ml) and collected by
We have developed an efficient synthesis of serofendic
acids A and B, novel neuroprotective substances isolated
from fetal calf serum. Biological and pharmacological
evaluation indicated that the serofendic acids exhibited
a potent protective action against neurotoxicity induced
by glutamate, but this did not involve direct interaction
with glutamate receptors. A pharmacokinetic study
showed good oral bioavailability, but low concentration
in the brain. In contrast, the methylester analogues
exhibited high concentrations in the brain. Single crystal
X-ray diffraction studies of the serofendic acids sup-
ported the stereochemistry of the sulfoxide group indi-
cated by NMR spectroscopic analysis. SA-A and SA-B
appear to be potential lead compounds for candidate
drugs to treat various neurological disorders.
filtration to afford 4 (29.1 g, 93%) as a white solid. mp
26
268–270 ꢁC. ½aꢁ ꢀ50.90 (c 1.007, CHCl3). ESI-MS
D
1
m/z 357 (M+Na)+. H NMR (CDCl3): d 0.83–1.52 (m,
9H), 0.87 (s, 3H), 1.25 (s, 3H), 1.35 (s, 3H), 1.53–1.62
(m, 2H), 1.66–1.75 (m, 1H), 1.76–2.02 (m, 5H), 2.05
(d, J = 18.4 Hz, 1H), 2.18 (broad d, J = 13.6 Hz, 1H),
3.12 (dd, J = 18.8, 2.8 Hz, 1H). Anal. Calcd for
C20H30O4 (334.2144): C, 71.82; H, 9.04. Found: C,
71.62; H, 8.90.
4. Experimental
4.1. Chemistry
Melting points were taken with a Yanaco MP-S3. The
infrared absorption spectra were obtained on an FT/
IR-620 spectrometer (Jasco, Tokyo). The following
NMR spectra were recorded on a Varian Unity 400, a
Varian Unity INOVA 500, or a JEOL JNM-a600 spec-
4.1.2. Methyl (1R,4S,5R,9S,10R,13S)-5,9,13-Trimethyl-
15-oxo-14-oxatetracyclo [11.3.1.01,10.04,9] heptadecane-5-
carboxylate (5). Oxalyl chloride (22.4 ml, 261 mmol) was
added to a solution of 4 (29.1 g, 87 mmol) in CH2Cl2
trometer: H NMR, 13C NMR, COSY, NOESY, TOC-
1
SY, HMQC, and HMBC. Chemical shifts are given in

T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
7103
(426 ml) and DMF (2.8 ml) at 0 ꢁC. The mixture was at
room temperature for 2.5 h, then cooled to 0 ꢁC again,
and MeOH (71 ml) was added, followed by triethyl-
amine (60.6 ml, 436 mmol). The reaction mixture was
stirred for 1 h at room temperature, solvent was evapo-
rated under reduced pressure, and water (500 ml) was
added. The mixture was extracted with EtOAc
(400 ml ·2), and the organic layer was washed with
brine (250 ml), dried over MgSO4, and evaporated un-
der reduced pressure. The resulting solid was washed
(29.5 g, 74.8 mmol) in CH2Cl2 (300 ml) over 30 min at
0 ꢁC. The reaction mixture was stirred for 18 h at room
temperature and then evaporated to half the initial vol-
ume under reduced pressure. Water (500 ml) was added,
and the mixture was extracted with Et2O (300 ml ·2).
The organic layer was washed with water (200 ml), sat-
urated aq NaHCO3 (200 ml), and brine (200 ml), dried
over MgSO4, and evaporated under reduced pressure
to afford a mixture of 8 and 9 as a crude light yellow
oil (31.8 g, 100%). The ratio of 8 and 9 was determined
from the olefin proton NMR peak integration value,
and was found to be 6:1. HRESI(+)MS calcd for
with MeOH (300 ml) and collected by filtration to afford
26
D
5 (27.8 g, 92%) as a white solid. mp 206–207 ꢁC. ½aꢁ
1
C23H36NaO4 (M+Na)+ 399.2511; found: 399.2538. H
ꢀ40.90 (c 1.392, CHCl3). ESI-MS m/z 371 (M+Na)+.
1H NMR (CDCl3): d 0.76 (s, 3H), 0.81–0.91 (m, 1H),
0.94 (dd, J = 12.4, 3.0 Hz, 1H), 1.01 (td, J = 13.6,
4.4 Hz, 1H), 1.08 (dd, J = 11.8, 2.8 Hz, 1H), 1.18 (s,
3H), 1.19–1.32 (m, 2H), 1.35 (s, 3H), 1.35–1.50 (m,
3H), 1.52–1.63 (m, 2H), 1.65–1.94 (m, 5H), 1.95–2.03
(m, 1H), 2.04 (d, J = 18.8 Hz, 1H), 2.18 (broad d,
J = 13.6 Hz, 1H), 3.08 (dd, J = 18.8, 2.6 Hz, 1H), 3.63
(s, 3H). Anal. Calcd for C21H32O4 (348.2300): C,
72.38; H, 9.26. Found: C, 72.32; H, 9.17.
NMR (CDCl3): d 0.68 (s, 3H, for 9), 0.72 (s 3H, for
8), 0.78–2.24 (m, 23H), 1.17 (s, 3H, for 9), 1.18 (s 3H,
for 8), 2.04 (s, 3H), 3.65 (s, 3H), 3.86–3.98 (m, 1H),
4.0–4.14 (m, 1H), 5.08 (broad s, 1H, for 9), 5.35 (broad
s, 1H, for 8).
4.1.6. Methyl (5b,8a,9b,10a,12a)-16-Hydroxyatisan-19-
oate (12). According to the procedure reported by
Coates, the crude mixture of 8 and 9 (34.3 g) was con-
verted to diastereomeric 12 (15.9 g, 61% from 7, white
solid), from which the diastereomers 12-A and 12-B
were slightly separated by column chromatography on
4.1.3. Methyl (5b,8a,9b,10a,13a)-13-Hydroxy-8-(2-
hydroxyethyl)-13-methylpodocarpan-15-oate (6). LiBH4
(6.52 g, 299 mmol) was added to a solution of 5 (17.4 g,
49.9 mmol) in THF (250 ml) at 0 ꢁC. The reaction mixture
was stirred for 61 h at room temperature, and then water
(300 ml) was added at 0 ꢁC. The mixture was stirred for
3 h at room temperature, then extracted with EtOAc
(300 ml ·2), and the organic layer was washed with brine
(200 ml), dried over MgSO4, and evaporated under re-
duced pressure. The resulting solid was washed with n-
silica gel with n-heptane/EtOAc (100:0 to 6:1, v/v) as
26
an eluent. 12-A: mp 139–141 ꢁC. ½aꢁ ꢀ51.76 (c 0.701,
D
CHCl3). ESI-MS m/z 357 (M+Na)+ 1H NMR (CDCl3):
d 0.77 (s, 3H), 0.78–1.90 (m, 19H), 1.17 (s, 3H), 1.28 (s,
3H), 1.92–2.04 (m, 1H), 2.16 (broad d, J = 12.4 Hz,
1H), 3.64 (s, 3H). Anal. Calcd for C21H34O3 (334.2507):
C, 75.41; H, 10.25. Found: C, 75.30; H, 10.14. 12-B: mp
26
148–150 ꢁC. ½aꢁ ꢀ36.73 (c 0.802, CHCl3). ESI-MS m/z
D
1
357 (M+Na)+. H NMR (CDCl3): d 0.78 (s, 3H), 0.79–
0.89 (m, 1H), 0.90–1.09 (m, 5H), 1.17 (s, 3H), 1.17–1.69
(m, 9H), 1.28 (s, 3H), 1.69–1.76 (m, 2H), 1.76–1.91 (m,
2H), 1.96–2.07 (m, 1H), 2.16 (broad d, J = 13.2 Hz,
1H), 3.64 (s, 3H). Anal. Calcd for C21H34O3 (334.2507):
C, 75.41; H, 10.25. Found: C, 75.35; H, 10.14.
heptane/Et2O (1:2, v/v, 200 ml) to afford 6 (12.7 g, 72%)
26
D
as a white solid. mp 213–214 ꢁC. ½aꢁ ꢀ26.63 (c 0.815,
CHCl3). ESI-MS m/z 375 (M+Na)+. 1H NMR (CDCl3):
d 0.68 (s, 3H), 0.74–0.94 (m, 4H), 1.01 (td, J = 13.6,
4.4 Hz, 1H), 1.09 (dd, J = 12.0, 3.2 Hz, 1H), 1.16 (s,
3H), 1.17 (s, 3H), 1.36 (td, J = 13.2, 4.4 Hz, 1H), 1.40–
1.52 (m, 2H), 1.53–1.94 (m, 8H), 1.98 (dd, J = 14.8, 2.4,
1H), 2.10–2.20 (m, 2H), 3.63 (s, 3H), 3.69–3.79 (m, 2H).
Anal. Calcd for C21H36O4 (352.2613): C, 71.56; H,
10.29. Found: C, 71.27; H, 10.18.
4.1.7. Methyl (5b,8a,9b,10a,12a)-Atis-16-en-19-oate
(13). Br2 (2.63 ml, 51.4 mmol) was added dropwise to
a solution of potassium 1,1,1,3,3,3-hexafluoro-2-phe-
nyl-2-propoxide (29 g, 103 mmol)12a and diphenyl sul-
fide (8.56 ml, 51.4 mmol) in CCl4 (200 ml) over 10 min
at room temperature. The reaction mixture was stirred
for 3 h at rt, then cooled with an ice-NaCl bath (ca.
ꢀ15 ꢁC), and a solution of diastereomeric 12 (8 g,
23.9 mmol) in CHCl3 (120 ml) was added dropwise for
15 min. The reaction mixture was stirred for 3 h at room
temperature and then evaporated under reduced pres-
sure. The mixture was diluted with Et2O (300 ml),
washed with 5 N aq NaOH (100 ml ·2), water
(100 ml), saturated aq NH4Cl (100 ml), and brine
(100 ml), dried over MgSO4, and evaporated under re-
duced pressure. The residue was purified by column
chromatography on silica gel with n-heptane/Et2O
(100:0 to 25:1, v/v) as an eluent to afford 13 (7.05 g,
93%) as a white solid. The resulting 13 contained 14 in
ca. 11.5:1 ratio, as determined from the olefin proton
NMR peak integration values. HRESI(+)MS calcd for
4.1.4. Methyl (5b,8a,9b,10a,13a)-8-(2-Acetoxyethyl)-13-
hydroxy-13-methylpodocarpan-15-oate (7). According to
the procedure reported by Coates, 6 (27.3 g, 77.4 mmol)
was converted to 7 (29.5 g, 97%, white solid). mp 145–
26
147 ꢁC. ½aꢁ ꢀ37.21 (c 0.088, CHCl3). ESI-MS m/z 417
D
1
(M+Na)+. H NMR (CDCl3): d 0.71 (s, 3H), 0.74–0.95
(m, 4H), 1.01 (td, J = 13.6, 4.4 Hz, 1H), 1.09 (dd,
J = 12.2, 2.6 Hz, 1H), 1.16 (s, 3H), 1.17 (s, 3H), 1.33
(td, J = 13.6, 4.4 Hz, 1H), 1.40–1.65 (m, 3H), 1.69–1.98
(m, 8H), 2.05 (s, 3H), 2.15 (broad d, J = 13.6 Hz, 1H),
2.26–2.36 (m, 1H), 3.65 (s, 3H), 4.03 (td, J = 10.0,
5.6 Hz, 1H), 4.34 (td, J = 10.0, 5.6 Hz, 1H). Anal. Calcd
for C23H38O5 (394.2719): C, 70.02; H, 9.71. Found: C,
69.97; H, 9.66.
4.1.5. Methyl (5b,8a,9b,10a)-8-(2-Acetoxyethyl)-13-
methylpodocarp-12-en-15-oate (8) and methyl (5b,8a,9b,
10a)-8-(2-acetoxyethyl)-13-methylpodocarp-13-en-15-oate
(9). TFA (58 ml) was added dropwise to a solution of 7
1
C21H32NaO2 (M+Na)+ 339.2300; found: 339.2343. H
NMR (CDCl3): d 0.78 (s, 3H), 0.84 (td, J = 13.6,

7104
T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
4.0 Hz, 1H), 0.93–1.16 (m, 5H), 1.18 (s, 3H), 1.34–1.65
(m, 7H), 1.67–2.00 (m, 5H), 2.04 (dt, J = 16.8, 2.0 Hz,
1H), 2.13–2.20 (m, 1H), 2.22 (quintet, J = 3.0 Hz, 1H),
3.65 (s, 3H), 4.57 (q, J = 2.0 Hz, 1H), 4.73 (q,
J = 2.0 Hz, 1H), 5.58 (broad t, J = 1.8 Hz, 1 H for 14).
evaporated under reduced pressure.The resulting solid
was washed with n-heptane/Et2O (2:1, v/v, 100 ml) and
collected by filtration to afford 19 (1.68 g, 80%) as a
26
white solid. mp 185–187 ꢁC. ½aꢁ ꢀ3.86 (c 0.609,
D
CHCl3). ESI-MS m/z 373 (M+Na)+. HRESI(+)MS
calcd for C21H34NaO4 (M+Na)+ 373.2355; found:
373.2355. 1H NMR (CD3OD): d 0.83 (s, 3H), 0.84–
0.99 (m, 2H), 0.99–1.15 (m, 4H), 1.17 (s, 3H), 1.34–
1.44 (m, 1H), 1.44–1.55 (m, 2H), 1.55–1.71 (m, 5H),
1.73–2.04 (m, 5H), 2.14 (broad d, J = 12.8 Hz, 1H),
3.40 (d, J = 9.2 Hz, 1H), 3.57 (dd, J = 10.4, 7.8 Hz,
1H), 3.63 (s, 3H), 3.88 (dd, J = 10.4, 7.6 Hz, 1H).
4.1.8. Methyl (5b,8a,9b,10a,12a,15b)-15-Hydroxyatis-
16-en-19-oate (16) and methyl (5b,8a,9b,10a,12a,15a)-
15-hydroxyatis-16-en-19-oate
(17)
and
methyl
(5b,8a,9b,10a,12a)-17-hydroxyatis-15-en-19-oate (18).
To a suspension of SeO2 (246 mg, 2.22 mmol) in CH2Cl2
was added 70% aq tert-butyl hydroperoxide (9.21 ml,
66.6 mmol) at room temperature. The mixture was stir-
red for 15 min at room temperature, then cooled to 0 ꢁC,
and 13 (7.01 g, 22.2 mmol) was added. Stirring was con-
tinued for 15 h at room temperature, then the mixture
was cooled to 0 ꢁC, and saturated aq Na2S2O3
(200 ml) was added. The whole was extracted with
EtOAc (200 ml ·2), and the organic layer was washed
with brine (200 ml), dried over MgSO4, and evaporated
under reduced pressure. The residue was purified by col-
umn chromatography on silica gel with n-heptane/
EtOAc (20:1 to 8:1, v/v) to afford 16 (1.9 g, 26%, white
4.1.10. Methyl (5b,8a,9b,10a,12a,15a)-15,17-Dihydr-
oxyatisan-19-oate (20). Diborane (1.1 M in THF,
16.7 ml, 18.4 mmol) was added to a solution of 17
(2.04 g, 6.14 mmol) in THF (60 ml) at 0 ꢁC. The reaction
mixture was stirred for 4 h at room temperature and
then cooled to 0 ꢁC again. Water (10 ml), 5 N aq NaOH
(15 ml), and 30% aq H2O2 (20 ml) were added, and the
mixture was stirred for 3 h at room temperature. Satu-
rated aq Na2S2O3 (100 ml) was added, the mixture was
extracted with EtOAc (100 ml ·2), and the organic layer
was washed with brine (100 ml), dried over MgSO4, and
evaporated under reduced pressure. The residue was
purified by column chromatography on silica gel with
n-heptane/EtOAc (8:1 to 3:2, v/v) as an eluent to afford
solid), 17 (2.04 g, 28%, white solid), and 18 (217 mg,
26
3%, white solid). Compound 16: mp 194–195 ꢁC. ½aꢁ
D
ꢀ43.54 (c 0.790, CHCl3). ESI-MS m/z 355 (M+Na)+.
1H NMR (CDCl3): d 0.81 (s, 3H), 0.81–1.09 (m, 4H),
1.14–1.21 (m, 1H), 1.18 (s, 3H), 1.32–1.45 (m, 3H),
1.45–1.60 (m, 4H), 1.64–1.89 (m, 4H), 1.97–2.07 (m,
1H), 2.12–2.20 (m, 1H), 2.27–2.33 (m, 1H), 3.59 (dt,
J = 5.6, 2.0 Hz, 1H), 3.65 (s, 3H), 5.00 (broad t,
J = 1.6 Hz, 1H), 5.06 (broad t, J = 1.6 Hz, 1H). Anal.
Calcd for C21H32O3 (332.2351): C, 75.86; H, 9.70.
20 (1.95 g, 91%) as a white solid. mp 155–156 ꢁC.
26
½aꢁ ꢀ 92:67 (c 0.319, CHCl3). ESI-MS m/z 373
D
(M+Na)+. HRESI(+)MS calcd for C21H34NaO4
(M+Na)+ 373.2355; found: 373.2358. 1H NMR
(CD3OD): d 0.85 (s, 3H), 0.93 (td, J = 13.6, 4.0 Hz,
1H), 0.98–1.11 (m, 4H), 1.17 (s, 3H), 1.21–1.33 (m,
1H), 1.35–1.54 (m, 4H), 1.54–1.93 (m, 8H), 1.94–2.04
(m, 1H), 2.14 (broad d, J = 12.8 Hz, 1H), 3.42 (dd,
J = 9.6, 1.6 Hz, 1H), 3.56 (dd, J = 11.2, 7.6 Hz, 1H),
3.63 (s, 3H), 3.90 (dd, J = 10.8, 7.6 Hz, 1H).
Found: C, 75.74; H, 9.66. Compound 17: mp 134–
26
135 ꢁC. ½aꢁ ꢀ57.39 (c 0.905, CHCl3). ESI-MS m/z 355
D
(M+Na)+. 1H NMR (CDCl3): d 0.80–0.91 (m, 1H),
0.83 (s, 3H), 0.93–1.15 (m, 4H), 1.19 (s, 3H), 1.33–1.66
(m, 7H), 1.66–1.90 (m, 5H), 2.14–2.21 (m, 1H), 2.26–
2.31 (m, 1H), 3.58–3.62 (m, 1H), 3.65 (s, 3H), 4.98
(broad t, J = 1.6 Hz, 1H), 5.05 (broad t, J = 1.6 Hz,
1H). Anal. Calcd for C21H32O3 (332.2351): C, 75.86;
4.1.11. Methyl (5b,8a,9b,10a,12a)-17-Hydroxy-15-oxoa-
tisan-19-oate (21). A solution of NaHSO3 (1.39 g,
13.3 mmol) in H2O (13.4 ml) was added dropwise to a
solution of 20 (1.95 g, 5.56 mmol) and NaBrO3 (2.01 g,
13.3 mmol) in CH3CN (22.4 ml)–H2O (6.7 ml) for
10 min at 0 ꢁC. The reaction mixture was stirred for
3 h at room temperature, and saturated aq Na2S2O3
(30 ml) was added. The mixture was extracted with
EtOAc (80 ml ·2), and the organic layer was washed
with brine (80 ml), dried over MgSO4, and evaporated
under reduced pressure. The residue was purified by col-
umn chromatography on silica gel with n-heptane/
H, 9.70. Found: C, 75.89; H, 9.66. Compound 18: mp
26
97–99 ꢁC. ½aꢁ ꢀ61.62 (c 0.472, CHCl3). ESI-MS m/z
D
355 (M+Na)+. HRESI(+)MS calcd for C21H32NaO3
(M+Na)+ 355.2249; found: 355.2202. 1H NMR
(CDCl3): d 0.74–0.91 (m, 2H), 0.80 (s, 3H), 0.95–1.11
(m, 3H), 1.15–1.64 (m, 7H), 1.19 (s, 3H), 1.71–1.93 (m,
4H), 2.00 (dt, J = 9.6, 3.2 Hz, 1H), 2.11–2.20 (m, 1H),
2.42–2.48 (m, 1H), 3.65 (s, 3H), 4.13 (d, J = 1.6 Hz,
2H), 5.83 (s, 1H).
EtOAc (10:1 to 2:1, v/v) as an eluent to afford 21
29
(1.55 g, 80%) as a white solid. mp 151 ꢁC. ½aꢁ ꢀ22.13
D
(c 0.592, CH3OH). ESI-MS m/z 371 (M+Na)+. 1H
NMR (CD3OD): d 0.87–0.99 (m, 1H), 0.93 (s, 3H),
1.00–1.16 (m, 3H), 1.18 (s, 3H), 1.27–1.46 (m, 3H),
1.51–1.96 (m, 9H), 2.10–2.31 (m, 3H), 2.31–2.38 (m,
1H), 3.59 (dd, J = 10.8, 9.6 Hz, 1H), 3.65 (s, 3H), 3.87
(dd, J = 10.8, 5.0 Hz, 1H). Anal. Calcd for C21H32O4
(348.2300): C, 72.38; H, 9.26. Found: C, 72.29; H, 9.23.
4.1.9. Methyl (5b,8a,9b,10a,12a,15b,16b)-15,17-Dihydr-
oxyatisan-19-oate (19). Diborane (1 M in THF, 18.1 ml,
18.1 mmol) was added to a solution of 16 (2 g,
6.02 mmol) in THF (60 ml) at 0 ꢁC. The reaction mix-
ture was stirred for 4 h at room temperature and then
cooled to 0 ꢁC again. Water (9 ml), 2 N aq NaOH
(20 ml), and 30% aq H2O2 (18 ml) were added, and the
mixture was stirred for 2 h at room temperature. Satu-
rated aq Na2S2O3 (100 ml) was added, the mixture was
extracted with EtOAc (100 ml ·2), and the organic layer
was washed with brine (100 ml), dried over MgSO4, and
4.1.12. Methyl (5b,8a,9b,10a,12a,15b)-15,17-Dihydr-
oxyatisan-19-oate (22). AcOH (2.55 ml, 44.5 mmol)
was added to a solution of NaB(OAc)3H (4.72 g,

T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
7105
22.3 mmol) in CH3CN (45 ml) at 0 ꢁC. Then, a solution
of 21 (1.55 g, 4.45 mmol) in CH3CN (10 ml) was added,
and the mixture was stirred for 4 h at 0 ꢁC. H2O (150 ml)
was added, the mixture was extracted with EtOAc
(2· 80 ml), and the organic layer was washed with brine
(50 ml), dried over MgSO4, and evaporated under re-
duced pressure. The residue was purified by column
chromatography on silica gel with n-heptane/EtOAc
Calcd for C21H34O3S (366.22287): C, 68.81; H, 9.35.
Found: C, 68.52; H, 9.32.
4.1.15. Serofendic acid A (1) and serofendic acid B (2).
Davis’s oxaziridine (2-benzenesulfonyl-3-phenyloxazi-
rine) (856 mg, 3.28 mmol) was added to a solution of
24 (923 mg, 2.52 mmol) in CHCl3 (25.2 ml) at 0 ꢁC.
The reaction mixture was stirred for 2 h at 0 ꢁC, Me2S
(1 ml) was added, and the whole was evaporated under
reduced pressure. The residue was purified by column
chromatography on silica gel with n-heptane/EtOAc
(2:1, v/v) and CH2Cl2/MeOH (15:1 to 10:1, v/v) as elu-
ents to yield serofendic acids A and B (1.01 g, 105%)
as a white solid. The resulting isomer mixture (1 g)
was purified by HPLC with a Daicel Chiralpak AD-H
column and an n-hexane/EtOH (30%) isocratic solvent
system to yield serofendic acid A (1) (393 mg, front
peak, white solid) and serofendic acid B (2) (544 mg,
back peak, white solid). Single crystals of the two com-
(5:1 to 1:2, v/v) to afford 22 (1.14 g, 73%) as a white so-
26
lid. mp 163–165 ꢁC. ½aꢁ ꢀ74.56 (c 0.542, CHCl3). ESI-
D
MS m/z 373 (M+Na)+. 1H NMR (CD3OD): d 0.64–0.76
(m, 1H), 0.83 (s, 3H), 0.90–1.12 (m, 4H), 1.17 (s, 3H),
1.23–1.34 (m, 1H), 1.34–1.45 (m, 2H), 1.46–1.89 (m,
10H), 1.94 (ddd, J = 14.8, 11.2, 3.6 Hz, 1H), 2.16 (broad
d, J = 13.4 Hz, 1H), 2.63 (d, J = 4.4 Hz, 1H), 3.42 (t,
J = 10.6 Hz, 1H), 3.60–3.67 (m, 1H), 3.63 (s, 3H). Anal.
Calcd for C21H34O4 (350.2457): C, 71.96; H, 9.78.
Found: C, 71.68; H, 9.69.
4.1.13. Methyl (5b,8a,9b,10a,12a,15b)-15-Hydroxy-17-
{[(4-methylphenyl)sulfonyl]oxy}atisan-19-oate (23). TsCl
(1.09 g, 5.71 mmol) and 4-dimethylaminopyridine
(417 mg, 3.43 mmol) were added to a solution of 22
(400 mg, 1.14 mmol) in pyridine (11 ml) at 0 ꢁC. This
mixture was stirred for 17 h at room temperature, then
H2O (80 ml) was added. The mixture was extracted with
EtOAc (50 ml ·2), and the organic layer was washed
with brine (50 ml), dried over MgSO4, and evaporated
under reduced pressure. The residue was purified by col-
umn chromatography on silica gel with n-heptane/
pounds were obtained by crystallization from n-hexane/
29
EtOAc. Serofendic acid A: mp 214–216 ꢁC. ½aꢁ ꢀ122.9
D
(c 1.280, CH3OH). HRESI(+)MS: (M+H)+ m/z
383.2256 (C21H34O4S, D 0.0 mmu). ESI(+)MS/MS: m/z
383, 365, 347, 319, 301, 283, 255. IR (KBr, cmꢀ1):
3432 (mO–H), 2927 (mC–H), 1698 (mC@O). 1H NMR
(600 MHz, CD3OD): d 2.99 (1H, dd, J = 13.0, 9.3 Hz,
H17a), 2.97 (1H, d, J = 3.7 Hz, H15), 2.87 (1H, dd,
J = 13.0, 6.1 Hz, H17b), 2.72 (3H, s, H21), 2.17 (1H,
brd, J = 14.2 Hz, H3a), 2.06 (1H, ddd, J = 14.2, 11.7,
2.9 Hz, H14a), 1.96 (1H, m, H16), 1.93 (1H, m, H2b),
1.90 (1H, m, H6a), 1.83 (1H, m, H6b), 1.74 (1H, ddd,
J = 13.2, 13.2, 4.5 Hz, H7b), 1.71 (1H, m, H12), 1.68
(1H, m, H13a), 1.66 (1H, m, H1a), 1.63 (1H, m,
H11b), 1.61 (1H, m, H9), 1.45 (1H, m, H11a), 1.42
(1H, m, H2a), 1.42 (1H, m, H13b), 1.24 (3H, s, H18),
1.11 (1H, J = ddd, 13.2, 2.9, 2.9 Hz, H7a), 1.07 (1H,
m, H3b), 1.07 (1H, m, H5), 1.00 (1H, m, H1b), 0.99
(3H, s, H20), 0.87 (1H, ddd, J = 14.2, 12.0, 6.6 Hz,
H14b). 13C NMR (150 MHz, CD3OD): d 181.7 (C19),
81.3 (C15), 60.9 (C17), 57.8 (C5), 44.7 (C4), 44.5
(C16), 42.8 (C9), 41.2(C1), 39.3 (C3), 39.2 (C10), 38.8
(C21), 37.7 (C8), 34.3 (C7), 32.2 (C12), 30.0 (C11),
29.5 (C18), 28.2 (C14), 22.2 (C13), 20.9 (C6), 20.0
(C2), 13.4 (C20). Anal. Calcd for C21H34O4S
(382.2177): C, 65.93; H, 8.96; O, 16.73; S, 8.38. Found:
EtOAc (8:1 to 3:1, v/v) as an eluent to yield 23
29
D
(386 mg, 67%) as a white solid. mp 155 ꢁC. ½aꢁ
ꢀ42.57 (c 0.237, CH3OH). ESI-MS m/z 527 (M+Na)+.
1H NMR (CDCl3): d 0.59–0.69 (m, 1H), 0.77 (s, 3H),
0.92 (td, J = 13.2, 4.0 Hz, 1H), 0.96–1.10 (m, 3H), 1.16
(s, 3H), 1.20–1.45 (m, 4H), 1.45–1.87 (m, 9H), 1.72 (d,
J = 4.4 Hz, 1H) 1.93 (ddd, J = 14.4, 11.2, 3.2 Hz, 1H),
2.15 (broad d, J = 13.6 Hz, 1H), 2.45 (s, 3H), 2.80
(t, J = 4.2 Hz, 1H), 3.63 (s, 3H), 3.95 (t, J = 9.4 Hz,
1H), 4.05 (dd, J = 9.4, 7.4 Hz, 1H), 7.34 (d, J = 8.4 Hz,
2H), 7.79 (d, J = 8.4 Hz, 2H). Anal. Calcd for
C28H40O6S (504.2545): C, 66.64; H, 7.99. Found: C,
66.50; H, 7.93.
4.1.14. Methyl (5b,8a,9b,10a,12a,15b)-15-Hydroxy-17-
(methylsulfanyl)atisan-19-oate (24). NaSMe (1.16 g,
16.6 mmol) was added to a solution of 23 (1.66 g,
3.29 mmol) in HMPA (16.5 ml) at room temperature.
The mixture was stirred for 43 h at 80 ꢁC, then cooled
to room temperature, and saturated aq NH4Cl
(100 ml) was added. The mixture was extracted with
EtOAc (100 ml ·2), and the organic layer was washed
with brine (100 ml), dried over MgSO4, and evaporated
under reduced pressure. The residue was purified by col-
umn chromatography on silica gel with n-heptane/
EtOAc (5:1 to 2:1, v/v) as an eluent to afford 23
C, 65.64; H, 8.99; O, 16.81; S, 8.34. Serofendic acid
29
B: mp 229–230 ꢁC. ½aꢁ ꢀ13.73 (c 1.408, CH3OH).
D
HRESI(+)MS: (M+H)+ m/z 383.2261 (C21H34O4S, D
+0.5 mmu). ESI(+)MS/MS: m/z 383, 365, 347, 319,
301, 283, 255. IR (KBr, cmꢀ1): 3420(mO–H), 2927
1
(mC-H), 1699 (mC@O). H NMR (500 MHz, CD3OD): d
3.04 (1H, dd, J = 13.2, 6.8, H17a), 2.95 (1H, d, J =
4.4 Hz, H15), 2.91 (1H, dd, J = 13.2, 9.0 Hz, H17b),
2.73 (3H, s, H21), 2.18 (brd, J = 13.2 Hz, H3a), 2.08
(1H, ddd, J = 14.0, 11.7, 3.0 Hz, H14a), 1.96 (1H, m,
H2b), 1.92 (1H, m, H16), 1.90 (1H, m, H6a), 1.83
(1H, m, H6b), 1.79 (1H, m, H12), 1.73 (1H, ddd,
J = 13.2, 13.2, 4.4 Hz, H7b), 1.66 (1H, m, H1a), 1.66
(1H, m, H13a), 1.63 (1H, m, H11b), 1.62 (1H, m, H9),
1.46 (1H, m, H11a), 1.44 (1H, m, H13b), 1.42 (1H, m,
H2a), 1.23 (3H, s, H18), 1.10 (1H, ddd, J = 13.2, 2.9,
2.9 Hz, H7a), 1.06 (1H, m, H3b), 1.06 (1H, m, H5),
1.00 (1H, m, H1b), 0.99 (3H, s, H2O), 0.86 (1H, ddd,
(923 mg, 76%) as a colorless solid. mp 145–147 ꢁC.
26
½aꢁ
ꢀ125.3 (c 0.255, CHCl3). ESI-MS m/z 389
D
(M+Na)+. 1H NMR (CDCl3): d 0.69–0.81 (m, 1H),
0.92 (s, 3H), 0.92–1.18 (m, 4H), 1.25 (s, 3H), 1.25–1.46
(m, 4H), 1.46–1.75 (m, 6H), 1.76–1.92 (m, 3H), 1.99
(td, J = 11.4, 3.0 Hz, 1H), 2.11–2.20 (m, 1H), 2.14 (s,
3H), 2.53 (d, J = 7.6 Hz), 2.95 (d, J = 4.0 Hz, 1H). Anal.

7106
T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
J = 14.0, 12.2, 6.5 Hz, H14b). 13C NMR (125 MHz,
CD3OD): d 181.0 (C19), 81.2 (C15), 59.8 (C17), 57.9
(C5), 44.8 (C4), 44.3 (C16), 42.9 (C9), 41.3 (C1), 39.4
(C3), 39.2 (C10), 38.3 (C21), 37.7 (C8), 34.3 (C7), 30.8
(C12), 29.9 (C11), 29.6 (C18), 28.4 (C14), 21.9 (C13),
21.0 (C6), 20.1 (C2), 13.5 (C20). Anal. Calcd for
C21H34O4S: C, 65.93; H, 8.96; O, 16.73; S, 8.38. Found:
C, 65.73; H, 8.99; O, 16.85; S, 8.28.
3H), 1.48–1.91 (m, 10H), 1.94–2.04 (m, 1H), 2.14 (broad
d, J = 13.6 Hz, 1H), 2.68 (s, 3H), 2.87 (dd, J = 13.2,
9.4 Hz, 1H), 2.91 (d, J = 4.4 Hz, 1H), 2.99 (dd,
J = 13.2, 6.8 Hz, 1H), 3.63 (s, 3H). Anal. Calcd for
C22H36O4S (396.2334): C, 66.63; H, 9.15. Found: C,
66.38; H, 9.02.
4.2. Single crystal X-ray structure determination
4.1.16. Serofendic acid methylester (25). NaSMe
(55.5 mg, 0.792 mmol) was added to a solution of 23
(200 mg, 0.396 mmol) in DMF (4 ml) at room tempera-
ture. The reaction mixture was stirred for 2 h at room
temperature, then H2O (40 ml) was added. The mixture
was extracted with EtOAc (30 ml ·2), and the organic
layer was washed with brine (30 ml), dried over MgSO4,
and evaporated under reduced pressure. The residue was
purified by column chromatography on silica gel with
n-heptane/EtOAc (10:1 to 5:1, v/v) to afford methyl
Single crystal X-ray diffraction data were collected with
a R-AXIS RAPID II (Rigaku), using CuKa radiation
(k = 1.54187 A). Crystallographic data, excluding struc-
ture factors, were deposited with the Cambridge Crys-
tallographic Data Center as supplementary Publication
numbers CCDC 647840 and 647841 for compounds 1
and 2. These data can be obtained free of charge from
the Cambridge Crystallographic Data Center via
www.ccdc.cam..ac.uk/data_request/cif.
˚
(5b,8a,9b,10a,12a,
nyl)atisan-18-oate (121 mg, 80%) as a white solid. mp
15b)-15-hydroxy-17-(methylsulfa-
4.3. Neuroprotective activity
26
107–108 ꢁC. ½aꢁ ꢀ101.7 (c 0.200, CHCl3). ESI-MS
Eagle’s minimal essential salt medium (Eagle’s MEM)
was purchased from Nissui Pharmaceutical (Tokyo, Ja-
pan). Fetal bovine serum and horse serum were pur-
chased from JRH Biosciences (Lenexa, KS, USA).
Poly(ethyleneimine) solution and cytosine b-D-arabino-
furanoside hydrochloride were from Sigma–Aldrich
Corp. (St. Louis, MO, USA). Trypan blue was from
Wako Pure Chemical Industries, Ltd (Osaka, Japan).
Sodium pentobarbital was from Dainippon Pharmaceu-
tical, Co, Ltd (Osaka, Japan). All other drugs were ob-
tained from Nacalai Tesque (Kyoto, Japan).
D
1
m/z 403 (M+Na)+. H NMR (CDCl3): d 0.69–0.80 (m,
1H), 0.80 (s, 3H), 0.85–1.14 (m, 4H), 1.18 (s, 3H),
1.23–1.46 (m, 4H), 1.46–1.88 (m, 9H), 1.91 (d,
J = 4.0 Hz, 1H), 1.92–2.01 (m, 1H), 2.12–2.19 (m, 1H),
2.14 (s, 3H), 2.52 (s, 1H), 2.53 (d, J = 1.2 Hz, 1H),
2.94 (t, J = 4.0 Hz, 1H), 3.64 (s, 3H). Anal. Calcd for
C22H36O3S (380.2385): C, 69.43; H, 9.53. Found: C,
69.31; H, 9.39.
Davis’s oxaziridine (2-benzenesulfonyl-3-phenyloxazi-
rine) (91.6 mg, 0.35 mmol) was added to a solution of
methyl
(5b,8a,9b,10a,12a,15b)-15-hydroxy-17-(meth-
Primary neuronal cultures were obtained from the cere-
bral cortex of fetal rats (17–19 days of gestation) accord-
ing to the procedures described previously.17 Briefly,
pregnant Wistar rats (Nihon SLC, Shizuoka, Japan)
were anesthetized with sodium pentobarbital, and the
cerebral cortex was rapidly removed bilaterally and
placed in ice-cold Hanks’ solution (137 mM NaCl,
5.4 mM KCl, 3.4 mM Na2HPO4, 0.5 mM KH2PO4, 5.6
mM D-(+)-glucose, and 2.4 mM Hepes, pH 7.2). The
cerebral cortex was mechanically dissociated with a scal-
pel and by trituration with a Pasteur pipette. Dissoci-
ated cells were filtered through a stainless steel mesh
with a pore size of 150 lm and centrifuged at 200g for
3 min. The supernatant was removed, and cells were
resuspended in complete medium (Eagle’s MEM supple-
mented with 10% heat-inactivated fetal bovine serum, 2
mM L-glutamine, 11 mM D-(+)-glucose, 24 mM NaH-
CO3, and 10 mM Hepes). Finally, single cell suspensions
were plated onto 0.1% polyethyleneimine-coated glass
coverslips, which were placed in 60 mm dishes
(5.1 · 106 cells per dish). Cultures were maintained in
the above-mentioned complete medium (1–8 days after
plating) or Eagle’s MEM supplemented with 10%
heat-inactivated horse serum, 2 mM L-glutamine,
11 mM D-(+)-glucose, 24 mM NaHCO3, and 10 mM
Hepes (9–15 days after plating). Cultures were incubated
at 37 ꢁC in a humidified 5% CO2 atmosphere. To pre-
vent proliferation of non-neuronal cells, 10 lM cytosine
b-D-arabinofuranoside hydrochloride was added after 6
days of plating. The culture medium was changed every
ylsulfanyl)atisan-18-oate (111 mg, 0.292 mmol) in
CHCl3 (2 ml) at 0 ꢁC. The reaction mixture was stirred
for 1 h at 0 ꢁC, Me2S (0.1 ml) was added, and the mix-
ture was evaporated under reduced pressure. The resi-
due was purified by column chromatography on silica
gel with n-heptane/EtOAc (2:1, v/v) and EtOAc/MeOH
(15/1, v/v) to yield serofendic acid methylester (25)
(105 mg, 91%) as a white solid.
The resulting isomeric mixture of serofendic acid methy-
lester (60 mg) was purified by HPLC with a YMC pack
ProC18 column and an H2O/CH3CN (37%) isocratic
solvent system to afford serofendic acid methylester A
(25-A) (15.1 mg, front peak, white solid) and serofendic
acid methylester B (25-B) (24.3 mg, back peak, white so-
lid). The stereochemistry of the sulfoxide group of 25-A
and 25-B has not been determined. Compound 25-A: mp
26
D
164–166 ꢁC. ½aꢁ ꢀ132.0 (c 0.283, CHCl3). ESI-MS m/z
419 (M+Na)+. 1H NMR (CD3OD): d 0.76–0.88 (m, 1H),
0.83 (s, 3H), 0.89–1.11 (m, 4H), 1.17 (s, 3H), 1.25–1.44
(m, 3H), 1.48–1.88 (m, 9H), 1.88–1.94 (m, 1H), 1.94–
2.04 (m, 1H), 2.14 (broad d, J = 13.6 Hz, 1H), 2.67 (s,
3H), 2.81 (dd, J = 13.2, 6.0 Hz, 1H), 2.89–2.98 (m,
2H), 3.63 (s, 3H). Anal. Calcd for C22H36O4S
(396.2334): C, 66.63; H, 9.15. Found: C, 66.48; H,
26
D
9.19. Compound 25-B: mp 197–199 ꢁC. ½aꢁ ꢀ24.98 (c
0.304, CHCl3). ESI-MS m/z 419 (M+Na)+.1H NMR
(CD3OD): d 0.76–0.88 (m, 1H), 0.83 (s, 3H), 0.90–1.01
(m, 1H), 1.01–1.11 (m, 3H), 1.17 (s, 3H), 1.27–1.45 (m,

T. Terauchi et al. / Bioorg. Med. Chem. 15 (2007) 7098–7107
7107
3. (a) Koh, J. Y.; Yang, L. L.; Cotamn, C. W. Brain Res.
1990, 553, 315; (b) Le, W. D.; Colom, L. V.; Xie, W. J.;
Smith, R. G.; Alexianu, M.; Appel, S. Brain Res. 1995,
686, 49.
4. Sawada, H.; Shimohama, S.; Tamura, Y.; Kawamura,
T.; Akaike, A.; Kimura, J. J. Neurosci. Res. 1996, 686,
49.
two days. Mature cells after 11–15 days in vitro were
used in all experiments. Animals were treated in accor-
dance with the guidelines of the Kyoto University ani-
mal experimentation committee and the guidelines of
the Japanese Pharmacological Society.
Cell viability was assessed by means of trypan blue
exclusion assay using Hoffman modulation contrast
microscopy, as described in our previous report.17a On
the day of the experiment, culture medium was replaced
with drug-containing medium at 37 ꢁC. After incuba-
tion, cultures were immediately stained with 1.5% try-
pan blue for 10 min at room temperature, fixed with
isotonic formalin, and then rinsed with physiological
saline. Cells stained with trypan blue were regarded as
non-viable. The viability of the cultures was calculated
as the percentage of the number of unstained cells (via-
ble cells) relative to the total number of cells counted
(viable cells plus non-viable cells). Over 200 cells per
coverslip were counted. In each experiment, cells on five
coverslips were counted to obtain means SEM of the
cell viability.
5. (a) John, W. M.; Faye, S. S.; Michael, V. J. Brain Res.
1989, 490, 33; (b) Clineschmidt, B. V.; Martin, G. E.;
Bunting, P. R. Drug Dev. Res. 1982, 2, 123; (c) Clinesch-
midt, B. V.; Martin, G. E.; Bunting, P. R.; Papp, N. L.
Drug Dev. Res. 1982, 2, 135; (d) Gill, R.; Foster, A. C.;
Woodruff, G. N. J. Neurosci. 1987, 7, 3343.
6. (a) Showalter, C. V.; Thornton, W. E. Am. J. Psychiatry
1977, 134, 1234; (b) Javitt, D. C.; Zukin, S. R. Am. J.
Psychiatry 1991, 148, 1301; (c) Willetts, J.; Balster, R. L.;
Leander, J. D. Trend Pharmacol. Sci. 1990, 11, 423; (d)
Tricklebank, M. D.; Singh, L.; Oles, R. J.; Preston, C.;
Iversen, S. D. Eur. J. Pharmacol. 1989, 167, 127; (e) Koek,
W.; Woods, J. H.; Winger, G. D. J. Pharmacol. Exp. Ther.
1988, 245, 969.
7. (a) Macarron, R. Drug Discov. Today 2006, 11, 277; (b)
Butler, M. S. J. Nat. Prod. 2004, 67, 2141; (c) Wilson, R.
M.; Danishefsky, S. J. J. Org. Chem. 2006, 71, 8329 .
8. Kume, T.; Asai, N.; Nishikawa, H.; Mano, N.; Terauchi,
T.; Taguchi, R.; Shirakawa, H.; Osakada, F.; Mori, H.;
Asakawa, N.; Yonaga, M.; Nishizawa, Y.; Sugimoto, H.;
Shimohama, S.; Katsuki, H.; Kaneko, S.; Akaike, A. Proc.
Natl. Acad. Sci. U.S.A. 2002, 99, 3288.
9. (a) Terauchi, T.; Asai, N.; Yonaga, M.; Kume, T.; Akaike,
A.; Sugimoto, H. Tetrahedron Lett. 2002, 43, 3625; (b)
Toyota, M.; Asano, T.; Ihara, M. Org. Lett. 2005, 7, 3929.
10. Coates, R. M.; Bertram, E. F. J. Org. Chem. 1971, 36,
2625.
Data were expressed as means SEM. The statistical
significance of differences between groups was deter-
mined by one-way analysis of variance (ANOVA) fol-
lowed by Dunnett’s two-tailed test using the InStat
program (Graph Pad Software, San Diego, CA, USA).
Statistical significance was defined as a probability value
of less than 5%.
11. Gleiter, R.; Herb, T.; Hofmann, J. Synlett 1996, 987.
12. (a) Martin, J. C.; Arhart, R. J.; Franz, J. A.; Perozzi, E. F.;
Kaplan, L. J. Org. Synth. 1988, VI, 163; (b) Arhart, R. J.;
Martin, J. C. J. Am. Chem. Soc. 1972, 94, 5003.
13. Mitsunobu, O. Synthesis 1981, 1.
4.4. Pharmacokinetic study
Fasted SD rats (n = 3, 290–260 g, obtained from Nihon
SLC) were used for the study. The dosing solution con-
taining the test compound (1 mg/ml) was administered
(dosing solution: 30% DMSO/5% glucose for iv; 0.5%
methyl cellulose for po). Blood samples were collected
at designated time points: 0.25, 0.5, 1, 2, 4, 6, and 8 h.
Brain samples were collected at designated time points:
0.5 and 2 h. MeOH/HClO4 (500:1, 0.2 ml) solution was
added to plasma (0.1 ml) or brain homogenate (20%
saline) (0.1 ml), centrifuged (12,000 rpm, 5 min), and
filtered. Then the samples were examined by using LC-
MS methodology (MS: SIM positive 406; column:
Develosil C30 2 · 150; column temperature: 40 ꢁC; mo-
bile phase: CH3CN/H2O = 30/70 containing 0.02%
HCO2H).
14. Sakaguchi, S.; Kikuchi, D.; Ishii, Y. Bull. Chem. Soc. Jpn.
1977, 70, 2561.
15. Davis, F. A.; Lal, S. G. J. Org. Chem. 1988, 53, 5004.
16. Terauchi, T.; Doko, T.; Yonaga, M.; Kajiwara, A.;
Niidome, T.; Taguchi, R.; Kume, T.; Akaike, A.;
Sugimoto, H. Bioorg. Med. Chem. Lett. 2006, 16,
5080.
17. (a) Nishikawa, H.; Hashino, A.; Kume, T.; Katsuki, H.;
Kaneko, S.; Akaike, A. Eur. J. Pharmacol. 2000, 404, 41;
(b) Kume, T.; Nishikawa, H.; Taguchi, R.; Hashino, A.;
Katsuki, H.; Kaneko, S.; Minami, M.; Satoh, M.; Akaike,
A. Eur. J. Pharmacol. 2002, 455, 91.
18. Osakada, F.; Kawato, Y.; Kume, T.; Katsuki, H.;
Sugimoto, H.; Akaike, A. . J. Pharmacol. Exp. Ther.
2004, 311, 51.
19. Takeda, T.; Akao, M.; Matsumoto-Ida, M.; Kato, M.;
Takenaka, H.; Kihara, Y.; Kume, T.; Akaike, A.; Kita, T.
J. Am. Coll. Cardiol. 2006, 47, 1882.
Acknowledgments
20. Kume, T.; Taguchi, R.; Katsuki, H.; Akao, M.; Sugimoto,
H.; Kaneko, S.; Akaike, A. Eur. J. Pharmacol. 2006, 542,
69.
21. (a) Takahashi, A.; Masuda, A.; Sun, M.; Victoria, E. C.;
Herman, B. Brain Res. Bull. 2004, 62, 497; (b) Blaise, G.
A.; Gauvin, D.; Gangal, M.; Authier, S. Toxicology 2005,
208, 177; (c) Uchiyama, T.; Otani, H.; Okada, T.;
Ninomiya, H.; Kido, M.; Imamura, H.; Nogi, S.; Kobay-
ashi, Y. J. Mol. Cell. Cardiol. 2002, 34, 1049.
We are grateful to Mrs. S. Sasaki of this laboratory for
the single crystal X-ray crystallographic measurement.
We are also grateful to Mrs. M. Gomibuchi of this lab-
oratory for the HRMS measurement.
References and notes
1. Collingridge, G. L.; Laster, R. A. Pharmacol. Rev. 1989,
41, 143.
2. Choi, D. W. Trends Neurosci. 1988, 11, 465.
22. William, B. H.; Gibb, S. L.; Williams, E. O.; Pasinelli, P.;
Brown, R. H., Jr.; Trotti, D. J. Biol. Chem. 2006, 281,
14076.