Absolute stereochemistry of fungal beauveriolide III and ACAT inhibitory activity of four stereoisomers

Article abstract of DOI:10.1021/jo0611667

(Chemical Equation Presented) Fungal beauveriolide III (BeauIII, 1b), a cyclodepsipeptide inhibiting acyl-CoA:cholesterol acyltransferase (ACAT) and showing antiatherogenic activity in mouse models, consists of L-Phe, L-Ala, D-allo-Ile, and 3-hydroxy-4-me

Full text of DOI:10.1021/jo0611667

Absolute Stereochemistry of Fungal Beauveriolide III and ACAT
Inhibitory Activity of Four Stereoisomers
Taichi Ohshiro,† Ichiji Namatame,† Kenichiro Nagai,† Takafumi Sekiguchi,‡ Takayuki Doi,‡
Takashi Takahashi,‡ Kazuaki Akasaka,| Lawrence L. Rudel,§ Hiroshi Tomoda,*,#, and
Satoshi Ohmura†,
Kitasato Institute for Life Sciences and Graduate School of Infection Control Sciences, Kitasato UniVersity,
5-9-1 Shirokane, Minato-ku, Tokyo 108-8641, Japan, Department of Applied Chemistry, Tokyo Institute of
Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo 152-8552, Japan, Graduate School of Life Sciences,
Tohoku UniVersity, 1-1 Tsutsumidori-Amamiyamachi, Aobaku, Sendai 981-8555, Japan, The Kitasato
Institute, 5-9-1 Shirokane, Minato-ku, Tokyo 108-8642, Japan, Department of Pathology, Atherosclerosis
Research Program, Wake Forest UniVersity School of Medicine, Winston-Salem, North Carolina, and
School of Pharmacy, Kitasato UniVersity, 5-9-1 Shirokane, Minato-ku, Tokyo 108-8641, Japan
tomodah@pharm.kitasato-u.ac.jp
ReceiVed June 7, 2006
Fungal beauveriolide III (BeauIII, 1b), a cyclodepsipeptide inhibiting acyl-CoA:cholesterol acyltransferase
(ACAT) and showing antiatherogenic activity in mouse models, consists of L-Phe, L-Ala, D-allo-Ile, and
3-hydroxy-4-methyloctanoic acid (HMA) moieties, but the stereochemistry of the HMA part has not
until now been fully defined. To determine it, four HMA stereoisomers were synthesized and labeled
with (S)-(+)-2-(anthracene-2,3-dicarboximido)-1-propyl trifluoromethane sulfonate (AP-OTf), a chiral
fluorescent reagent. The derivatives were separated by HPLC and compared with the natural HMA
derivative, which was thereby identified as (3S,4S)HMA in BeauIII. Furthermore, the four beauveriolide
III isomers ((3S,4S)BeauIII (23a), (3R,4R)BeauIII (23b), (3R,4S)BeauIII (23c), and (3S,4R)BeauIII (23d))
were synthesized, and it was shown that all the spectral data for 23a were identical with those for natural
1b. Isomers 23a and 23d showed potent inhibitory activity of lipid droplet accumulation in macrophages,
while the other two isomers caused weak inhibition. Thus, the 3S configuration of BeauIII is important
for this activity. Furthermore, 23a and 23d showed rather specific inhibition against the ACAT1 isozyme.
Introduction
to determine the absolute stereochemistry of natural products
in order to understand their biological function.
The chiral discrimination of natural products has received
considerable attention because the enantiomers usually have
different biological properties.1 Therefore, it is very important
Cyclodepsipeptides beauveriolides I and III to IX produced
by BeauVeria sp. FO-6979 were discovered from a screening
program for inhibitors of lipid droplet accumulation in mouse
macrophages.2,3 Among them, beauveriolides I (BeauI, 1a) and
III (BeauIII, 1b) (Figure 1) showed very potent and specific
inhibition. Studies on the mechanism of action revealed that
* Address correspondence to this author. Phone: +81-3-5791-6241. Fax:
+81-3-3444-6197.
† Kitasato University.
‡ Tokyo Institute of Technology.
| Tohoku University.
(2) Namatame, I.; Tomoda, H.; Arai, H.; Inoue, K.; Ohmura, S. J. Biochem.
1999, 125, 319-327.
The Kitasato Institute.
§ Wake Forest University School of Medicine.
# Kitasato University.
(3) (a) Namatame, I.; Tomoda, H.; Si, S.; Yamaguchi, Y.; Masuma, R.;
Ohmura. S. J. Antibiot. 1999, 52, 1-6. (b) Namatame, I.; Tomoda, H.; Tabata,
N.; Si, S.; Ohmura. S. J. Antibiot. 1999, 52, 7-12. (c) Matsuda, D.;
Namatame, I.; Tomoda, H.; Kobayashi, S.; Zocher, R.; Kleinkauf, H.;
Omura, S. J. Antibiot. 2004, 57, 1-9. (d) Mochizuki, K.; Ohmori, K.;
Tamura, H.; Shizuri, H.; Nishiyama, S.; Miyoshi, E.; Yamamura, S. Bull.
Chem. Soc. Jpn. 1993, 66, 3041-3046.
(1) (a) Tomoda, H.; Kumagai, H.; Ogawa, Y.; Sunazuka, T.; Hashizume,
H.; Nagashima, H.; Ohmura, S. J. Org. Chem. 1997, 62, 2161-2165. (b)
Boruwa, J.; Kalita, B.; Barua, N. C.; Borah, J. C.; Mazumder, S.; Thakur,
D.; Gogoi, D. K.; Bora, T. C. Bioorg. Med. Chem. Lett. 2004, 14, 3571-
3574. (c) Morohashi, A.; Satake, M.; Nagai, H.; Oshima, Y.; Yasumoto, T.
Tetrahedron 2000, 56, 8995-9001.
10.1021/jo0611667 CCC: $33.50 © 2006 American Chemical Society
Published on Web 08/25/2006
J. Org. Chem. 2006, 71, 7643-7649
7643

Ohshiro et al.
Using this assay system, we evaluated the selectivity of BeauI
and BeauIII in the inhibition of ACAT isozymes. We demon-
strated that the stereochemistry of the HMA part of beauveri-
olides is important for the ability to strongly inhibit lipid droplet
accumulation in mouse macrophages and for the selectivity
toward the two ACAT isozymes.
Results
FIGURE 1. Structure of beauveriolides.
Absolute Stereochemistry of HMA in BeauI and BeauIII.
To elucidate the absolute stereochemistry of the HMA in natural
1a and 1b, four isomers for HMA were synthesized as shown
in Scheme 1.10 Asymmetric hydrogenation of (S)-3-oxo-4-
methyloctanoate 2 with RuCl2[(R)-binap]11 under 90 atm of
hydrogen at 45 °C provided â-hydroxyl ester 3 (94%, de 95%),
while use of RuCl2[(S)-binap] gave diastereomer 6 (94%, de
90%).12 Hydrolysis of the ethyl ester provided carboxylic acids
4 and 7 which were converted to methyl esters 5 and 8,
respectively. ent-4, ent-5, ent-7, and ent-8 were prepared by
using the same method from ethyl (R)-3-oxo-4-methyloctanoate.
they inhibit cholesteryl ester (CE) synthesis by blocking acyl-
CoA:cholesterol acyltransferase (ACAT) activity in macroph-
ages, leading to inhibition of lipid droplet accumulation. More
importantly, they proved orally active in atherogenic mouse
models, reducing the atherogenic lesions in the artery and heart.4
Therefore, they are expected to be novel agents for the
prevention and treatment of atherosclerosis.5
BeauI/BeauIII are 13-membered cyclodepsipeptides consisting
of L-Phe, L-Ala, D-Leu/D-allo-Ile and 3-hydroxy-4-methyloc-
tanoic acid (HMA), respectively. Mochizuki et al. previously
reported that the absolute stereochemistry of the HMA part is
The structures of 5 and 8 were confirmed by the 1H and 13
C
3S,4S. However, the [R]20 values appeared different between
NMR spectra, and the configurations were determined by
D
comparison of the optical rotations of 5 {[R]28 -38 (c 0.98,
natural-derived HMA (-19) and synthetic HMA (-44).3d On
the other hand, our group reported that the [R]20D value (-29)
of HMA prepared from BeauIII shows good agreement with
that (-27) of HMA from BeauI, and we therefore concluded
that the HMA stereochemistry in BeauIII is also 3S,4S.3b
D
CHCl3), lit. [R]20D -44 (c 0.98, CHCl3)} and 9 {[R]29D +18 (c
1.02, CHCl3), lit. [R]22 +23 (c 0.62, CHCl3)} with those
D
previously reported.3d
Next, the hydrolytic products of natural 1a and 1b and the
four HMA isomers 4, 9, ent-4, and ent-9 were treated with (S)-
(+)-2-(anthracene-2,3-dicarboximido)-1-propyl trifluoromethane
sulfonate ((S)-(+)-AP-OTf)13 to give HMA-(S)-(+)-AP deriva-
tives 6, 7, 11, and 12 (Scheme 1), which were analyzed by
HPLC with a silica gel column. The derivatives 7, 11, 12, and
6 were separated with retention times of 69, 71, 74, and 78
min, respectively (Figure 2). As shown in Figure 2B, the
derivatives from natural 1a- and 1b-derived HMA were found
to elute at 78 min, demonstrating that the absolute stereochem-
istry of HMA was 3S,4S.
Synthesis of Four Stereoismers in BeauIII. By using the
four HMA isomers, (3S,4S)BeauIII, (3R,4R)BeauIII, (3R,4S)-
BeauIII, and (3S,4R) were synthesized as summarized in
Schemes 2 and 3.10 In brief, after conversion of 4 to benzyl
ester 13, coupling of the secondary alcohol with Boc-D-allo-
Ile-OH using DCC in the presence of DMAP followed by
hydrogenolysis of the benzyl group produced ester unit 14
(Scheme 2). Synthesis of the four stereoisomers started with
attachment of the carboxylic acid in Fmoc-L-Ala-OH onto
2-chlorotrityl chloride resin 18 (Scheme 3). The Fmoc group
in 19 was removed with 20% piperidine in DMF, and subsequent
condensation with Fmoc-L-Phe-OH afforded dipeptide 20. After
subsequent deprotection, coupling of the resulting amine with
four ester units (14, 15, 16, and 17) with use of PyBrop resulted
in depsipeptides 21a-d, respectively. Concurrently, removal
of the Boc group and cleavage from the polymer support with
4 M HCl in 1,4-dioxane released the desired depsipeptides
22a-d in parallel. Finally, the cyclization of 22a-d with EDCI
However, the [R]20 values from Mochizuki’s report and our
D
report still showed a discrepancy. Therefore, in this study we
first defined the absolute stereochemistry by preparing the four
HMA stereoisomers.
ACAT, an ER membrane protein that is the target of
beauveriolides, is responsible for the esterification of cholesterol
for storage inside cells. ACAT has been recognized as a target
of inhibition for a new type of antiatherosclerotic agent. Thus
far, many companies have attempted to develop synthetic ACAT
inhibitors. However, almost none have been successfully
developed. Recent molecular biological studies revealed the
existence of two different isozymes, ACAT1 and ACAT2.6,7
ACAT1 is ubiquitously expressed in tissues and cells including
macrophages, while ACAT2 is expressed predominantly in the
liver (hepatocytes) and intestine.6-8 Therefore, for the develop-
ment of a new antiatherosclerotic agent, it is important to
understand the selectivity of an inhibitor against the two ACAT
isozymes.9 Recently, Rudel’s group established sophisticated
cell-based assays to evaluate ACAT1 and ACAT2 activities.9a
(4) Namatame, I.; Tomoda, H.; Ishibashi, S.; Ohmura, S. Proc. Natl. Acad.
Sci. U.S.A. 2004, 101, 737-742.
(5) Chem. Eng. News 2004, 18, 15.
(6) Chang, C. C.; Huh, H. Y.; Cadigan, K. M.; Chang, T. Y. J. Biol.
Chem. 1993, 268, 20747-20755.
(7) (a) Anderson, R. A.; Joyce, C.; Davis, M.; Reagan, J. W.; Clark, M.;
Shelness, G. S.; Rudel, L. L. J. Biol. Chem. 1998, 273, 26747-26754. (b)
Cases, S.; Novak, S.; Zheng, Y. W.; Myers, H. M.; Lear, S. R.; Sande, E.;
Welch, C. B.; Lusis, A. J.; Spencer, T. A.; Krause, B. R.; Erickson, S. K.;
Farese, R. V., Jr. J. Biol. Chem. 1998, 273, 26755-26764. (c) Oelkers, P.;
Behari, A.; Cromley, D.; Billheimer, J. T.; Sturley, S. L. J. Biol. Chem.
1998, 273, 26765-26771.
(8) Parini, P.; Davis, M.; Lada, A. T.; Erickson, S. K.; Wright, T. L.;
Gustafsson, U.; Sahlin, S.; Einarsson, C.; Eriksson, M.; Angelin, B.;
Tomoda, H.; Ohmura, S.; Willingham, M. C.; Rudel, L. L. Circulation 2004,
110, 2017-2023.
(9) (a) Lada, A. T.; Davis, M.; Kent, C.; Chapman, J.; Tomoda, H.;
Ohmura, S.; Rudel, L. L. J. Lipid. Res. 2004, 45, 378-386. (b) Giovannoni,
M. P.; Piaz, V. D.; Vergelli, C.; Barlocco, D. Mini ReV. Med. Chem. 2003,
3, 576-584.
(10) Nagai, K.; Doi, T.; Sekiguchi, T.; Namatame, I.; Sunazuka, T.;
Tomoda, H.; Ohmura, S.; Takahashi, T. J. Comb. Chem. 2006, 8, 103-109
(11) Noyori, R.; Ohkuma, T.; Kitamura, M.; Takaya, H.; Sayo, N.;
Kumobayashi, H.; Akutagawa, S. J. Am. Chem. Soc. 1987, 109, 5856-
5858.
(12) Diastereoselectivity was determined by H NMR analysis.
(13) Akasaka, K.; Ohrui, H.; Meguro, H.; Umestu, T. Anal. Sci. 1997,
13, 461-466.
1
7644 J. Org. Chem., Vol. 71, No. 20, 2006

Absolute Stereochemistry of Fungal BeauVeriolide III
SCHEME 1. Synthesis of Four Isomers of HMAa
a Reagents and conditions: (a) RuCl2[(R)-binap], EtOH, H2, 90 atm, 45 °C, 40 h; (b) RuCl2[(S)-binap], EtOH, H2, 90 atm, 45 °C, 40 h; (c) 1 M LiOH,
THF, rt, 2.5 h; (d) cat. H2SO4, MeOH, CH2Cl2 40 °C, 15 h; (e) (S)-(+)-AP-OTf, TEAC, MeCN, rt, 30 min.
23c, and at 4.94 ppm for both 23a and 23d. In addition, the
proton signals corresponding to 4-H were observed at 2.03 ppm
for 23a, which was close to those for 1b, while they were
observed at 2.13 ppm for 23d. As is clearly shown, all the proton
signals for 23a were found to be identical with those for
naturally occurring 1b. Furthermore, other spectral data (13C
NMR, UV) were also identical between 23a and 1b. Thus, all
the absolute stereochemistry of 1b was finally defined as that
of 23a.
Effect of BeauIII Stereoisomers on Lipid Droplet Ac-
cumulation in Macrophages. In this macrophage assay,
massive amounts of lipid droplets were accumulated in the
cytosol when macrophages were incubated with liposomes, and
FIGURE 2. Chromatographic profiles of isolation of (S)-(+)-AP
derivatives of four HMA isomers by preparative HPLC: (A) synthetic
were observed microscopically after oil red O staining (Figure
4E). Under the assay conditions, 23a and 23d inhibited lipid
droplet accumulation in a dose-dependent fashion (2-20 µM,
data not shown). As shown in Figure 4A,D, the size and the
number of lipid droplets in macrophages were reduced in the
presence of 23a or 23d (6 µM). By contrast, 23b and 23c
showed no inhibition of lipid droplet accumulation even at 20
µM (Figure 4B,C). No cytotoxic effects were observed on
macrophages upon addition of up to 20 µM of the four BeauIII
stereoisomers.
HMA, (a) (3R,4R)HMA (ent-4), (b) (3R,4S)HMA (9), (c) (3S,4R)HMA
(ent-9), (d) (3S,4S)HMA (4), (B) HMA of BeauIII (1b). Column,
PEGASIL Silica 120-5, 4.6 × 500 mm; hexanes-ethyl acetate (42:8);
flow rate, 0.7 mL/min; fluorometer, excitation 298 nm, emission 462
nm.
SCHEME 2. Synthesis of Four Stereoisomers of HMAa
Effect of BeauIII Stereoisomers on CE and TG Synthesis
by Macrophages. Under the conditions of lipid droplet ac-
cumulation in mouse macrophages, about 40% of exogenously
added [14C]oleic acid was incorporated into [14C]CE and [14C]-
triacylglycerol (TG), which are the main constituents of lipid
droplets in the cytosol.2 As shown in Figure 5A, 23a selectively
inhibited [14C]CE synthesis in a dose-dependent fashion with
an IC50 value of 1.19 µM. Isomer 23d inhibited not only [14C]-
CE synthesis (IC50, 0.69 µM) but also [14C]TG synthesis (IC50,
14.0 µM) although the TG inhibition was much weaker than
the CE synthesis (Figure 5D). By contrast, 23b and 23c showed
weak inhibition of [14C]CE synthesis at the highest concentration
(50 µM) (Figure 5B,C). These biochemical data are consistent
with the morphological data of the lipid droplet accumulation
in macrophages (Figure 4).
a Reagents and conditions: (i) Cs2CO3, BnBr, DMF, rt, 12 h; (j) Boc-
D-allo-Ile-OH, DCC, DMAP, CH2Cl2, 0 °C to room temperature, 22 h; (k)
10% Pd/C, H2, EtOH, rt, 12 h.
and i-Pr2NEt under high dilution conditions proceeded to
produce four stereoisomers, (3S,4S)BeauIII (23a, 11%), (3R,4R)-
BeauIII (23b, 10%), (3R,4S)BeauIII (23c, 17%), and (3S,4R)-
1
BeauIII (23d, 12%), respectively. The H NMR spectra of the
isomers are shown in Figure 3. The proton signals corresponding
Selectivity in Inhibition of ACAT Isozymes by BeauIII
to 3-H were observed at 5.18 ppm for 23b and 5.15 ppm for
Stereoisomers. Previously, we reported that BeauI and BeauIII
J. Org. Chem, Vol. 71, No. 20, 2006 7645

Ohshiro et al.
SCHEME 3. Synthesis of Beauveriolide III Isomers with Solid Supporta
a Reagents and conditions: (l) Fmoc-L-Ala-OH, i-Pr2NEt, CH2Cl2, rt, 2 h; (m) 20% piperidine/DMF, rt, 1 h; (n) Fmoc-L-Phe-OH, DIPCI, HOBt, CH2Cl2/
DMF (4/1), rt, 2 h; (o) 14, 15, 16 or 17, PyBrop, i-Pr2NEt, CH2Cl2/DMF (4/1), rt, 1.5 h; (p) 4M HCl/1,4-dioxane, rt, 2 h; (q) EDCI‚HCl, i-Pr2NEt, CH2Cl2,
rt, 2 h, 51%.
FIGURE 3. 1H NMR spectra of (a) (3S,4S)BeauIII (23a), (b) (3R,4R)BeauIII (23b), (c) (3R,4S)BeauIII (23c), (d) (3S,4R)BeauIII (23d), and (e)
naturally occurring beauIII (1b).
inhibited ACAT activity in enzyme-based assays using mi-
crosomes prepared from mouse macrophages (ACAT1 expres-
sion) and livers (ACAT2 expression), suggesting that they
inhibited both the ACAT1 and ACAT2 isozymes.4 To define
the selectivity of the inhibition of four BeauIII isomers for the
isozymes, a cell-based assay using ACAT1- and ACAT2-CHO
cells was used.9a
As shown in Figure 6, 23a and 23d showed rather specific
inhibition against ACAT1 compared to ACAT2. The selectivity
was about 4, as shown by comparison of the IC50 values
(Table 1). On the other hand, 23b and 23c showed almost no
effect on CE synthesis at 10 µM in ACAT1- or ACAT2-CHO
cells, although 23c gave weak inhibition of both isozymes at
20 µM.
7646 J. Org. Chem., Vol. 71, No. 20, 2006

Absolute Stereochemistry of Fungal BeauVeriolide III
FIGURE 4. Effect of BeauIII stereoisomers on lipid droplet accumulation in macrophages: (A) (3S,4S)BeauIII (23a), 6 µM (B) (3R,4R)BeauIII
(23b), 6 µM, (C) (3R,4S)BeauIII (23c), 6 µM, (D) (3S,4R)BeauIII (23d), 6 µM, and (E) control.
FIGURE 5. Effect of BeauIII stereoisomers on CE and TG synthesis by macrophages [14C]CE (9) and [14C]TG (b): (A) (3S,4S)BeauIII (23a),
(B) (3R,4R)BeauIII (23c) and (3S,4R)BeauIII (23d), (C) (3R,4S)BeauIII (23c,) and (D) (3S,4R)BeauIII (23d).
FIGURE 6. Effect of beauIII stereoisomers on CE synthesis by ACAT1- or ACAT2-CHO cells. [14C]CE in ACAT1-CHO cells (9) or
ACAT2-CHO cells (0): (A) (3S,4S)BeauIII (23a), (B) (3R,4R)BeauIII (23b), (C) (3R,4S)BeauIII (23c), and (D) (3S,4R)BeauIII (23d).
analogue library by combinatorial chemistry.10 According to the
same strategy (Scheme 3), the four BeauIII stereoisomers were
TABLE 1. IC50 Values of BeauIII Stereoismers in Cell-Based
Assays
IC50 (µM)
synthesized by using the four HMA isomers as one of the
building units. As a result, all the spectral data, including H
1
BeauIII isomer
macrophages
ACAT1
ACAT2
NMR (Figure 3), of synthetic (3S,4S)BeauIII (23a) were found
to be consistent with those of natural BeauIII (1b), corroborating
the absolute stereochemistry of BeauIII.
(3S,4S)BeauIII (23a ) 1b)
(3R,4R)BeauIII (23b)
(3R,4S)BeauIII (23c)
(3S,4R)BeauIII (23d)
1.19
>44.2
>52.1
0.69
5.5
>20
>20
2.7
>20
>20
19.0
10.7
Fungal BeauIII was originally isolated as an inhibitor of lipid
droplet accumulation in mouse macrophages.3a,b First, the four
BeauIII stereoisomers were evaluated in this macrophage assay.
Not only 23a but 23d inhibited lipid droplet accumulation
(Figure 4), while the other isomers showed very weak or almost
no inhibition even at higher concentration (20 µM). The
microscopic observations were supported by the biochemical
data (Figure 5); 23a and 23d inhibited CE synthesis strongly,
but the other two isomers showed very weak inhibition at the
highest concentration (50 µM). These results indicate that the
3S HMA configuration is responsible for the strong inhibition
of lipid droplet accumulation in macrophages. Interestingly, 23d
was found to inhibit not only CE synthesis but also TG
synthesis, although the TG inhibition was 20 times weaker than
the CE inhibition, as shown by comparison of the IC50 values.
Accordingly, inhibition of lipid droplet synthesis by 23d
appeared more effective than that by 23a (Figure 4A,D). The
Discussion
Several research groups, including our group, reported that
BeauI (1a) and BeauIII (1b) are composed of two L-amino acids,
one D-amino acid, and HMA,3 but the absolute stereochemistry
of HMA was not fully defined. In this study, the four
stereoisomers of HMA were synthesized and the (S)-(+)-AP
derivatives were separated by HPLC and compared with the
derivatives from natural HMA (Figure 2). This method proved
effective to determine the absolute stereochemistry of the
hydroxy fatty acid at the picomole level.13,14 Thus, we defined
clearly the stereochemistry of HMA in BeauI and BeauIII as
3S,4S. Furthermore, we succeeded in preparing a beauveriolide
(14) Akasaka, K.; Ohrui, H. Biosci. Biotechnol. Biochem. 2004, 68, 153-
158.
J. Org. Chem, Vol. 71, No. 20, 2006 7647

Ohshiro et al.
Analysis of (S)-(+)-AP Derivatives of HMA by HPLC. To
determine the HMA configuration, (S)-(+)-AP derivatives of HMA
were analyzed by HPLC (column, PEGASIL Silica 120-5 (4.6 ×
500 mm); solvent, hexane/ethyl acetate (42:8); excitation, 298 nm,
emission, 462 nm; flow rate, 0.7 mL/min; column temperature, 40
°C). HPLC was carried out with use of the SSC flow system 3100
(Senshu). Fluorometry was carried out with a JASCO FP-210
spectrofluorometer (JASCO).
Synthesis of BeauIII Stereoisomers. Synthesis of BeauIII
stereoisomers was carried out by the method described previously.10
Details of the synthesis are available from the Supporting Informa-
tion.
difference might be due to the fact that 23d has multiple
inhibition targets in macrophages.
Our previous experiments on the mechanism of action
suggested that natural BeauIII (1b) inhibited both ACAT1 and
ACAT2 isozymes in an enzyme-based assay using microsomes
prepared from mouse macrophages and livers.4 Second, we
confirmed the selectivity of the four BeauIII stereoisomers for
the inhibition of ACAT isozymes in sophisticated cell-based
assays using ACAT1- and ACAT2-CHO cells. As shown in
Figure 6, 23a and 23d inhibited ACAT1 activity strongly, while
the other isomers, 23b and 23c, showed weak or no inhibition
of ACAT1 activity. The results are consistent with those in
macrophage assays. These findings are reasonable because
mouse macrophages predominantly express ACAT1.6 Further-
more, 23a and 23d also showed inhibition against ACAT2
activity, although the potency was several times weaker than
that against ACAT1 (Table 1). Thus, we demonstrated that the
3S sterochemistry of HMA in BeauIII is responsible for potent
ACAT inhibition and that BeauIII causes rather specific
inhibition against ACAT1 compared to ACAT2.
Selected Spectral Data of BeauIII Stereoisomers: BeauIII
(natural product, 1b). HRFABMS: calcd for C27H42N3O5 [M +
H]+ 488.3124, found 488.3127. 1H NMR (400 MHz, CDCl3:CD3-
OD 4:1): δ (ppm) 7.23 (ddd, J ) 8.0, 7.0, 1.0 Hz, 2H), 7.0.18
(ddd, J ) 7.0, 1.5 Hz, 1H), 7.14 (ddd, J ) 8.0, 1.5, 1.0 Hz, 2H),
4.93 (ddd, J ) 10.0, 4.5, 4.0 Hz, 1H), 4.27 (d, J ) 10.0 Hz, 1H),
4.19 (dd, J ) 8.5, 7.5 Hz, 1H), 3.80 (q, J ) 7.0 Hz, 1H), 3.04 (dd,
J ) 13.5, 7.5 Hz, 1H), 2.95 (dd, J ) 13.5, 8.5 Hz, 1H), 2.46 (dd,
J ) 14.0, 4.5 Hz, 1H), 2.38 (dd, J ) 14.0, 10.0 Hz, 1H), 2.04 (m,
1H), 1.67 (m, 1H), 1.36 (m, 1H), 1.32 (m, 1H), 1.32 (m, 1H), 1.22
(d, J ) 7.0 Hz, 3H), 1.21 (m, 2H), 1.20 (m, 2H), 1.11 (m, 1H),
0.99 (m, 1H), 0.88 (d, J ) 7.0 Hz, 3H), 0.86 (t, J ) 7.0 Hz, 3H),
0.85 (d, J ) 6.0 Hz, 3H), 0.84 (t, J ) 7.0 Hz, 3H). 13C NMR (100
MHz, CDCl3:CD3OD 4:1): δ (ppm) 172.4, 171.8 (×2), 169.7,
136.7, 129.3 (×2), 128.8 (×2), 127.2, 76.8, 59.6, 57.2, 49.7, 37.4,
36.0, 35.9, 35.8, 30.7, 29.8, 26.1, 23.2, 15.7, 15.1, 14.8, 14.1, 11.1.
(3S,4S)BeauIII (23a). HRFABMS: calcd for C27H42N3O5 [M
In this study, we determined that the absolute stereochemistry
of HMA in beauveriolide III is 3S,4S and that the 3S config-
uration in beauveriolide is responsible for potent inhibition of
lipid droplet accumulation in macrophages and of ACAT
activity.
1
Experimental Section
+ H]+ 488.3124, found 488.3117. H NMR (300 MHz, CDCl3:
CD3OD 4:1): δ (ppm) 7.28-7.11 (5H, m), 4.94 (1H, dt, J ) 10.0,
4.5 Hz), 4.26 (1H, d, J ) 10.0 Hz), 4.17 (1H, dd, J ) 8.5, 7.5 Hz),
3.82 (1H, q, J ) 7.0 Hz), 3.02 (1H, dd, J ) 13.5, 8.5 Hz), 2.94
(1H, dd, J ) 13.5, 7.5 Hz), 2.46 (1H, dd, J ) 14.0, 5.0 Hz), 2.38
(1H, dd, J ) 14.0, 10.0 Hz), 2.03 (1H, m), 1.67 (1H, m), 1.44-
1.26 (3H, m), 1.22 (3H, d, J ) 7.0 Hz), 1.21-1.13 (5H, m), 1.00
(1H, m), 0.87 (3H, t, J ) 7.0 Hz), 0.86 (3H, d, J ) 7.0 Hz), 0.85
(3H, d, J ) 7.0 Hz), 0.83 (3H, t, J ) 7.0 Hz). 13C NMR (75.4
MHz, CDCl3:CD3OD 4:1): δ (ppm) 172.1, 171.5, 171.4, 169.4,
136.5, 129.0 (×2), 128.5 (×2), 126.9, 76.6, 59.4, 57.0, 49.3, 37.2,
35.9, 35.7 (×2), 30.6, 29.7, 26.0, 23.1, 15.7, 15.0, 14.8, 14.1, 11.1.
(3R,4R)BeauIII (23b). HRFABMS: calcd for C27H42N3O5 [M
Materials. Reagents and solvents were purchased from com-
mercial sources and used without further purification. BeauI and
BeauIII were purified from the culture broth of BeauVeria sp. FO-
6979 as reported.3a 2-Chlorotrityl chloride resin 18 (100-200 mesh,
1% cross-linked, 1.25 mmol/g; batch, U1125032) and MicroKan
microreactors were purchased from IRORI (now Biotage). The resin
in the MicroKan microreactor was swelled in solvent for 1 h before
use. CHO cells expressing African Green monkey ACAT1 (ACAT1-
CHO) and ACAT2 (ACAT2-CHO) were established previously by
Lada et al.9a ACAT1- or ACAT2- CHO cells were cultured by the
method described previously.9a In brief, both cell lines were
maintained at 37 °C in 5% CO2 in Ham’s F-12 medium supple-
mented with MEM vitamins, Geneticin (G418) (300 µg/mL), and
10% heat inactivated fetal bovine serum (FBS) (hereafter referred
to as medium A).
1H NMR (270 MHz) and 13C NMR (67.9 MHz) spectra were
recorded on a JEOL JNM-EX270 instrument with CDCl3. HR-
FABMS were measured with a JEOL JMS-AX505 HA mass
spectrometer.
Synthesis of Four HMA Stereoisomers. Synthesis of the four
HMA stereoisomers was carried out by the method described
previously.10 Details of the synthesis are available from the
Supporting Information.
1
+ H]+ 488.3124, found 488.3108. H NMR (300 MHz, CDCl3:
CD3OD 4:1): δ (ppm) 7.30-7.14 (5H, m), 5.18 (1H, dd, J ) 11.0,
5.0, 4.0 Hz), 4.45 (1H, dd, J ) 9.0, 5.5 Hz), 4.44 (1H, q, J ) 7.0
Hz), 4.39 (1H, d, J ) 6.0 Hz), 3.06 (1H, dd, J ) 14.0, 5.5 Hz),
2.84 (1H, dd, J ) 14.0, 9.0 Hz), 2.44 (1H, dd, J ) 15.5, 11.0 Hz),
2.35 (1H, dd, J ) 15.5, 4.0 Hz), 1.94 (1H, m), 1.77 (1H, m), 1.39-
1.26 (3H, m), 1.25 (3H, d, J ) 7.0 Hz), 1.24-1.05 (5H, m), 1.00
(1H, m), 0.88 (3H, t, J ) 7.0 Hz), 0.84 (3H, d, J ) 7.0 Hz), 0.83
(3H, t, J ) 7.0 Hz), 0.82 (3H, d, J ) 7.0 Hz). 13C NMR (75.4
MHz, CDCl3:CD3OD 4:1): δ (ppm) 172.1, 171.8, 171.7, 170.3,
136.0, 128.9 (×2), 128.7 (×2), 127.2, 76.0, 56.7, 55.9, 48.3, 36.6,
36.1 (×2), 35.9, 31.3, 29.5, 26.6, 22.9, 15.7, 15.4, 14.4, 14.0, 11.6.
(3R,4S)BeauIII (23c). HRFABMS: calcd for C27H42N3O5 [M
Hydrolysis of BeauIII. BeauIII (1 mg, 2 µmol) was degraded
in a gas phase of 6 N HCl (500 µL) at 150 °C for 1 h, using a
PICO TAG work station (Waters). The degradation products were
dissolved in toluene (100 µL).
1
+ H]+ 488.3124, found 488.3119; H NMR (300 MHz, CDCl3:
CD3OD 4:1): δ (ppm) 7.29-7.14 (5H, m), 5.15 (1H, dd, J ) 11.0,
5.0, 4.0 Hz), 4.48 (1H, dd, J ) 9.0, 5.5 Hz), 4.46 (1H, q, J ) 7.0
Hz), 4.43 (1H, d, J ) 6.0 Hz), 3.07 (1H, dd, J ) 14.0, 5.5 Hz),
2.84 (1H, dd, J ) 14.0, 9.0 Hz), 2.42 (1H, dd, J ) 15.5, 11.0 Hz),
2.31 (1H, dd, J ) 15.5, 4.0 Hz), 2.00-1.89 (2H, m), 1.38-1.26
(3H, m), 1.25 (3H, d, J ) 7.0 Hz), 1.24-1.04 (5H, m), 1.00 (1H,
m), 0.88 (3H, t, J ) 7.0 Hz), 0.84 (3H, d, J ) 7.0 Hz), 0.82 (3H,
d, J ) 7.0 Hz), 0.81 (3H, t, J ) 7.0 Hz); 13C NMR (75.4 MHz,
CDCl3:CD3OD 4:1) δ (ppm) 172.3, 172.0 (×2), 170.6, 136.2, 129.0
(×2), 128.9 (×2), 127.3, 76.1, 56.6, 55.6, 48.2, 36.6, 36.0, 35.3,
34.8, 32.8, 29.4, 26.6, 22.9, 15.6, 14.4, 13.9 (×2), 11.7.
Derivatization of Four HMA Stereoisomers by (S)-(+)-2-
(Anthracene-2,3-dicarboximido)-1-propyl Trifluoromethane Sul-
fonate (AP-OTf). The degradation products of beauIII 1b (800
nmol) and the four HMA stereoisomers (4, 9, ent-4, ent-9, 188
nmol) were reacted with (S)-(+)-2-(anthracene-2,3-dicarboximido)-
1-propyl trifluoromethane sulfonate (AP-OTf)13 and tetraethylam-
monium carbonate (TEAC) in dry CH3CN for 30 min at room
temperature. The reaction mixture were purified by preparative
TLC. Active bands were collected and extracted with CHCl3-CH3-
OH (10:1). The extracts were concentrated in vacuo to yield (S)-
(+)-AP derivatives 6, 7, 11, and 12 (375 pmol).
(3S,4R)BeauIII (23d). HRFABMS: calcd for C27H42N3O5 [M
1
+ H]+ 488.3124, found 488.3122. H NMR (300 MHz, CDCl3:
7648 J. Org. Chem., Vol. 71, No. 20, 2006

Absolute Stereochemistry of Fungal BeauVeriolide III
CD3OD 4:1): δ (ppm) 7.28-7.12 (5H, m), 6.71 (1H, d, J ) 9.5
Hz), 4.94 (1H, ddd, J ) 9.0, 5.5, 5.0 Hz), 4.28 (1H, t, J ) 9.5 Hz),
4.19 (1H, dd, J ) 8.5, 7.5 Hz), 3.85 (1H, q, J ) 7.0 Hz), 3.05 (1H,
dd, J ) 13.5, 8.5 Hz), 2.94 (1H, dd, J ) 13.5, 7.5 Hz), 2.40 (1H,
dd, J ) 13.5, 5.0 Hz), 2.37 (1H, dd, J ) 13.5, 9.0 Hz), 2.13 (1H,
m), 1.67 (1H, m), 1.39-1.26 (3H, m), 1.23 (3H, d, J ) 7.0 Hz),
1.21-1.02 (5H, m), 1.00 (1H, m), 0.89 (3H, d, J ) 7.0 Hz), 0.85
(3H, d, J ) 7.0 Hz), 0.85-0.84 (6H, m); 13C NMR (75.4 MHz,
CDCl3:CD3OD 4:1): δ (ppm) 172.0, 171.5, 171.4, 169.3, 136.4,
129.0 (x 2), 128.5 (x 2), 126.9, 75.9, 59.4, 57.0, 49.3, 37.4, 35.8,
35.3, 35.1, 32.8, 29.3, 26.0, 23.0, 15.0, 14.8, 14.0, 13.5, 11.0.
Assay for [14C]Neutral Lipid Synthesis by Mouse Peritoneal
Macrophages. The assay for [14C]CE and [14C]TG synthesis from
[14C]oleic acid in mouse peritoneal macrophages was carried out
by the method described previously.2 Details of the assay are
available from the Supporting Information.
were washed two times with PBS. The cells were lysed by adding
0.25 mL of 10 mM Tris-HCl (pH 7.5) containing 0.1% (w/v)
sodium dodecyl sulfate, and the cellular lipids were extracted by
the method of Bligh and Dyer.15 The organic solvent was reduced
by centrifugation under vacuum, and [14C]CE was separated on a
thin-layer chromatography (TLC) plate (silica gel) and analyzed
with a bioimaging analyzer (BAS 2000, Fuji Film).
Assay for Cell Viability. The numbers of viable macrophages,
ACAT1-CHO cells, and ACAT2-CHO cells were measured in the
presence of various inhibitors with use of alamar Blue.
Acknowledgment. This work was supported in part by
grants from Scientific Research on Priority Areas 16073215
(H.T.), Scientific Research (B) 18390008 (H.T.), Specific
Research (S) 14103013 (T.T.), and the 21st Century COE
Program, Ministry of Education, Culture, Sports, Science and
Technology, Japan, and from the Hoh-ansha Foundation, Japan
(H.T.). T.O. is supported by Research Fellowships of the Japan
Society for the Promotion of Science for Young Scientists.
Cellular Neutral Lipid Staining. Cellular neutral lipid staining
was carried out by the method described previously.2 Details of
the assay are available from the Supporting Information.
Assay for [14C]CE Synthesis in ACAT1- or ACAT2-CHO
Cells. ACAT1- or ACAT2-CHO cells (1.25 × 105 cells/0.25 mL
medium A) were plated in a 48-well plastic microplate and allowed
to recover overnight at 37 °C in 5% CO2. The assay was done
with cells that were at least 80% confluent. Following the overnight
recovery, 2.5 µL of a sample (methanol solution) and 5 µL of [14C]-
oleic acid (1 nmol, 1.85 KBq, 10% ethanol/PBS solution) were
added to each culture at 37 °C in 5% CO2. Following a 6 h
incubation, the medium was removed, and the cells in each well
Supporting Information Available: Materials, synthesis pro-
cedures, characterization of compounds, and biological assay
procedures. This material is available free of charge via the Internet
at http://pubs.acs.org.
JO0611667
(15) Bligh, E. G.; Dyer. W. Can. J. Biochem. Physiol. 1959, 37, 911-
917.
J. Org. Chem, Vol. 71, No. 20, 2006 7649