Design, synthesis, and biological evaluation of beauveriolide analogues bearing photoreactive amino acids

Article abstract of DOI:10.1248/cpb.c16-00095

Beauveriolides I and III, which are naturally occurring cyclodepsipeptides, have been reported to bind to sterol O-acyltransferase (SOAT), inhibiting its ability to synthesize cholesteryl esters. To facilitate an analysis of the binding site(s) of these c

Full text of DOI:10.1248/cpb.c16-00095

754
Chem. Pharm. Bull. 64, 754–765 (2016)
Vol. 64, No. 7
Special Collection of Papers
This article is dedicated to Professor Satoshi Ōmura in celebration of his 2015 Nobel Prize.
Regular Article
Design, Synthesis, and Biological Evaluation of Beauveriolide Analogues
Bearing Photoreactive Amino Acids
Yuichi Masuda,a Kazumasa Aoyama,a Masahito Yoshida,a Keisuke Kobayashi,b Taichi Ohshiro,b
,a
Hiroshi Tomoda,b and Takayuki Doi*
a Graduate School of Pharmaceutical Sciences, Tohoku University; 6–3 Aza-Aoba, Aramaki, Aoba-ku,
b
Sendai 980–8578, Japan: and Graduate School of Pharmaceutical Sciences, Kitasato University;
5–9–1 Shirokane, Minato-ku, Tokyo 108–8641, Japan.
Received January 28, 2016; accepted February 25, 2016; advance publication released online March 26, 2016
Beauveriolides I and III, which are naturally occurring cyclodepsipeptides, have been reported to bind
to sterol O-acyltransferase (SOAT), inhibiting its ability to synthesize cholesteryl esters. To facilitate an
analysis of the binding site(s) of these compounds, we designed beauveriolide analogues 1a–d wherein the Leu
or D-allo-Ile residue was replaced by photoreactive amino acids possessing methyldiazirine or trifluorometh-
yldiazirine in the side chains. The methyldiazirine moiety was installed by reaction of methyl ketones with
liquid ammonia to provide imine intermediates, followed by treatment with hydroxylamine-O-sulfonic acid
to provide the diaziridines. Subsequent oxidation gave methyldiazirines. In contrast, trifluoromethyldiazirine
derivatives were prepared from trifluoromethyl ketones via the oxime intermediates, which were transformed
into diaziridines. Subsequent oxidation afforded trifluoromethyldiazirines. The synthesized photoreactive
amino acids 3a–d were coupled with 3-hydroxy-4-methyloctanoic acid 4 and dipeptide 5, followed by macro-
lactamization to provide beauveriolide analogues 1a–d. The SOAT inhibitory activities of 1a–d were found
to be as potent as those of beauveriolides I and III. Moreover, 1a–d inhibited SOAT1 selectively rather than
SOAT2, which was also consistent with the behavior of beauveriolides I and III.
Key words beauveriolide; sterol O-acyltransferase; photoaffinity labeling; diazirine
Sterol O-acyltransferase (SOAT, also known as acyl- microsomal fraction prepared from SOAT1-CHO or SOAT2-
CoA:cholesterol acyltransferase (ACAT)) catalyzes conversion CHO.10) SOAT1 and SOAT2 share extensive homology, but
of cholesterol and long-chain fatty-acyl-CoA to cholesteryl es- the topology of the protein domains is thought to be differ-
ters (CE).1) Since CE accumulation in plasma is closely related ent.1,5,8,9) These data implied that beauveriolides would bind
to artery plaque formation, SOAT inhibitors are expected to to a homology domain in SOAT1 and SOAT2, but also that
be therapeutic agents for atherosclerosis.2,3) Recent molecular the environment wherein the binding site is located might
biological studies have revealed the existence of two differ- be different for SOAT1 and SOAT2, e.g., inside or outside of
ent SOAT isozymes—SOAT1 and SOAT2—in mammals.4–7) the cell membrane. We have accomplished the total synthesis
SOAT1 is ubiquitously expressed in most tissues and cells, of beauveriolide III and determined the absolute configura-
including macrophages, and is related to foam cell formation. tion of the (3S,4S)-3-hydroxy-4-methyloctanoic acid (HMA)
On the other hand, SOAT2, which is expressed predominantly moiety.16) Our intensive structure–activity relationship (SAR)
in the liver and intestine, is related to food cholesterol absorp- study showed that structural modification of beauveriolide
tion and secretion of low-density lipoprotein (LDL). Since could change the selectivity against SOAT1 and SOAT2.16–21)
SOAT1 and SOAT2 function differently,8,9) isozyme selectivity Structural modification of beauveriolides might change not
is important for developing SOAT inhibitors. So far, a number only their affinity for each SOAT isozyme but also their dis-
of SOAT inhibitors have been reported and their isozyme se- tributions in the cell. The mechanism of isozyme selectivity
lectivities have been discussed.2,3,10–12)
remains unclear because the binding site of beauveriolide to
Beauveriolides I and III (Fig. 1A), isolated from a culture SOAT1 and SOAT2 has not been elucidated by high-resolution
broth of Beauveria sp. FO-6979, inhibit lipid droplet forma- (HR) structural analysis due to the structural complexity of
tion in mouse peritoneal macrophages.13,14) These cyclodep- SOAT, which is known to be a membrane protein.
sipeptides were found to inhibit CE synthesis by blocking
Photoaffinity labeling enables direct analysis of a target
SOAT activity.15) Ohshiro et al. reported that beauveriolides protein by photo-induced cross-linking between a ligand
I and III selectively inhibited SOAT1 in a cell-based assay and its binding protein.22) Photoaffinity labeling has been
using two Chinese hamster ovary (CHO) cell lines express- utilized to identify the target protein and binding site of low-
ing African green monkey SOAT1 (SOAT1-CHO) and SOAT2 molecular-weight ligands. Since the analysis of interactions
(SOAT2-CHO).10) However, beauveriolides I and III inhibit between peptide and protein is expected, many researchers
SOAT2 as well as SOAT1 in an enzyme assay using the have developed photoreactive amino acids bearing photore-
*To whom correspondence should be addressed. e-mail: doi_taka@mail.pharm.tohoku.ac.jp
© 2016 The Pharmaceutical Society of Japan

Vol. 64, No. 7 (2016)
Chem. Pharm. Bull.
755
Fig. 1. (A) Beauveriolides and Their Analogues with Methyl- or Trifluoromethyl-Diazirines in the Side Chains of D-Leu or D-allo-Ile; (B) Photoreac-
tive Amino Acids with Alkyldiazirines
active groups such as phenylazide,23) benzophenone,24) and
diazirine25–30) in their side chains to probe target molecules
and the binding-site structure of bioactive peptides and pro-
teins. Diazirine is frequently used as a useful photoreactive
group owing to its practical advantages22,31): relatively high
stability under ambient light and various synthetic reaction
conditions, quick photolysis under long-wavelength ultraviolet
light (350–355nm), and compact structure with few effects on
the chemical properties of the ligand molecules. Suchanek et
al. have developed photo-methionine and -leucine (Fig. 1B),
which possess alkyl methyldiazirines in their side chains and
exhibit chemical and biological features similar to those of
methionine and leucine.27) Several researchers have reported
the successful application of photo-methionine and -leucine in
capturing the target molecules of bioactive peptides,32,33) and
in the structural analysis of a β-hairpin-forming peptide.34)
It has been reported that photo-excited diazirines are first
transformed into carbenes or diazoisomers.25,35–37) As carbenes
are highly reactive species with extremely short lifetimes,38,39)
they can form intermolecular cross-links with the first bio-
Chart 1. Retrosynthetic Analysis of 1a–d
molecule that they encounter. In contrast, diazoisomers of
diazirines, which have reduced reactivity and much longer inhibitory activities of the synthesized analogues were also
lifetimes, can diffuse from an initial site of generation to other investigated.
sites in biomolecules. It is a critical problem that photo-excited
diazirines, carbenes, and diazoisomers can induce intramolec-
Results and Discussion
ular 1,2-rearrangements of a hydrogen atom, resulting in the
Design of Beauveriolide Analogues Bearing Photoreac-
production of alkenes.40,41) In comparison to methyldiazirines, tive Amino Acids Our SAR studies of beauveriolides16–21)
trifluoromethyldiazirines are less prone to the rearrange- indicated that the HMA moiety is indispensable for SOAT
ments,41,42) the absence of which could increase the efficiency inhibitory activity. The phenylalanine residue was also found
of photoaffinity labeling experiments. The electron-with- to be important for binding activity and SOAT selectivity.
drawing effect of the trifluoromethyl group can stabilize the Since the L-Ala residue is not critical for this activity, it would
carbenes, preventing them from undergoing rearrangements. not face the target protein and would be left for the incorpora-
Moreover, fluorine does not migrate as readily as hydrogen tion of detecting groups, such as a biotin tag.21) In contrast,
owing to the greater strength of the C–F bond compared the D-Leu and D-allo-Ile residues in beauveriolides I and III,
with C–H. Despite many reports on aryl(trifluoromethyl)- respectively, are related to the activity, but small modifications
diazirines,31) there is little precedent for alkyl(trifluoromethyl)- of the side chains are allowed while maintaining the activity.
diazirines, which possesses a compact structure that is less Overall, we have designed four kinds of beauveriolide ana-
likely to undergo rearrangement. Recently, Franks and col- logues with various photoreactive amino acids in place of the
leagues reported the synthesis of the photoaffinity probe of D-Leu and D-allo-Ile residues (1a–d) (Fig. 1A).
propofol bearing dimethylene(trifluoromethyl)diazirine.43)
Retrosynthesis of 1a–d The synthesis of 1a–d can be
Herein, we have demonstrated the synthesis of beauveri- performed in a manner similar to that used in our total syn-
olide analogues bearing a methyldiazirine or trifluorometh- thesis of beauveriolide III16) (Chart 1). In principle, 1a–d can
yldiazirine moiety in their side chains. Synthetic methods be synthesized by macrolactamization of the linear compound
for the preparation of trifluoromethyl photo-methionine and 2, which can be obtained by sequential couplings of 3–5.
trifluoromethyl photo-leucine (Fig. 1B) are described, along
Synthesis of 3a–d First, we conducted the synthesis of the
with those of photo-methionine and -leucine. The photochemi- photoreactive amino acids 3a–d. Vila-Perelló et al. reported
cal properties of the photoreactive amino acids and the SOAT an efficient synthesis of L-photo-methionine from a L-glutamic

756
Chem. Pharm. Bull.
Vol. 64, No. 7 (2016)
Chart 2. Synthesis of 3a and b
Chart 3. Synthesis of 3c and d
acid derivative.32) We followed their synthetic route for the ing ketone 8b by the methyl Grignard reagent. Our investiga-
preparation of tert-butoxycarbonyl (Boc)-D-photo-methionine tion of the methylation conditions found that treatment of the
(3a) (Chart 2). Briefly, Boc-D-glutamic acid α-tert-butyl ester Weinreb amide 7b with methyllithium in tetrahydrofuran at
(6a) (the preparation method for this compound is summa- −78°C gave the desired compound 8b in 90% yield. The re-
rized in Chart S1) was transformed into the corresponding sultant ketone 8b was then transformed into 3b in the same
ε-Weinreb amide 7a, which was then converted into methyl manner as that used for the synthesis of 3a.
ketone 8a by treatment with a methyl Grignard reagent. The
Next, we attempted the synthesis of trifluoromethyl-photo-
methyldiazirine moiety was then installed by reaction of methionine 3c (Chart 3). The light-sensitive diazirine moiety
the ketone with liquid ammonia, followed by treatment with should be formed at a late stage in the synthetic protocol, thus
hydroxylamine-O-sulfonic acid to give diaziridine 9a. Sub- we utilized the trifluoromethyl ketone 13c as a key intermedi-
sequent oxidation afforded the desired diazirine 10a in 44% ate. Ketone 13c was prepared from aldehyde 11c (the prepara-
overall yield. Removal of the Boc and t-butyl groups, followed tion method for this compound is summarized in Chart S2)
by Boc protection of the resulting amine gave Boc-D-photo- by nucleophilic addition of the trifluoromethyl group utilizing
methionine (3a). Since the synthesis of L-photo-leucine from Ruppert–Prakash reagent (TMS-CF3),44) followed by oxida-
an L-aspartic acid derivative has not been reported, we at- tion of the resulting alcohol 12c with Dess–Martin periodin-
tempted to synthesize Boc-D-photo-leucine (3b) in the same ate (DMP),45) in 75% yield in three steps. In this synthesis,
manner as that used for 3a (Chart 2). However, the δ-Weinreb protection of the α-amino group with bis-Boc groups was
amide 7b was not completely converted into the correspond- adopted because mono-Boc protected aldehydes readily cy-

Vol. 64, No. 7 (2016)
Chem. Pharm. Bull.
757
Fig. 2. LC-MS Analysis of the Photolysis of 17a (A–C) and 17c (D–F) in MeOH Irradiated with 352nm Blacklight (Output, 6W; UV Output, 0.6W;
Distance, 5cm)
Irradiation times were 0min (A, D), 15min (B, E), and 90min (C, F). The analytical conditions are as follows: column, X Bridge™ C1,8 (3.5µm, 4.6×75mm); gradient
method, 10–95% of B (0.00–4.00min), 95% of B (4.00–11.0min), 95–10% of B (11.0–11.1min), 10% of B (11.1–15.0min) (A: 0.1% HCOOH/H2O, B: 0.1% HCOOH/MeOH);
flow rate, 1.1mL/min; UV 214nm. Chemical structures of the observed peaks were estimated by m/z of the molecular ion (MH+).
clize to the aminal, which is then often reluctant to undergo that the absorbance maximum between 300 and 400nm is
further reactions.46) Preparation of the trifluoromethyldiazirine 347nm for 17a, and 317nm for 17c (Fig. S1). This could be a
moiety in 16c was unsuccessful using the previous reaction reason for 17c being hard to decompose under 352nm black
conditions, including the treatment of the ketone 13c with light. These data indicate that trifluoromethyldiazirine is more
liquid ammonia and hydroxylamine-O-sulfonic acid followed efficient for cross linking, but needs 300–330nm wavelength
by oxidation. Therefore, we converted 13c to tosyloxime 14c; UV light.
treatment with liquid ammonia then furnished diaziridine 15c.
Total Synthesis of 1a–d With the desired photoreactive
Due to the instability of 15c under the purification conditions, amino acids 3a–d in hand, we incorporated them into beau-
the compound was subsequently oxidized with iodine in the veriolide analogues (Chart 4). After allyl ester protection of
presence of triethylamine47) to afford the desired trifluoro- the carboxyl group in 18, which was prepared as previously re-
methyldiazirine 16c. Finally, removal of the Boc and t-butyl ported,16) the resultant alcohol 4 was coupled with 3a–d using
groups and subsequent Boc protection furnished 3c in 68% N,N′-diisopropylcarbodiimide (DIC) and 4-(dimethylamino)-
yield over two steps. In a fashion similar to the synthesis of pyridine (DMAP) to afford esters 19a–d in 81–90% yields.
3c, the synthesis of trifluoromethyl-photo-leucine 3d was also After removal of the allyl groups in 19a–d and the benzy-
accomplished from the aldehyde 11d (Chart 3).
loxycarbonyl (Cbz) group in 5 (the preparation method for
Photochemical Properties The photolysis of the methyl- this compound is described in Supplementary materials), the
and trifluoromethyldiazirines was investigated. To quantitate resultant carboxylic acids and amine were coupled by treat-
the amounts of substrates and products based on UV absorp- ment with N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide
tion, we attached a benzyl ester to 3a and c to afford 17a and (EDCI), 1-hydroxybenzotriazole (HOBt), and diisopropyl-
c, respectively (Fig. 2, Chart S3). The amino acid derivatives ethylamine (DIEA) to afford the linear compounds 20a–d.
in MeOH, cooled in an ice bath, were irradiated with 352nm Finally, macrolactamization was successfully conducted with
black light at 6W (UV output: 0.6W) from a distance of 5cm, EDCI for 20a and b (R=CH3) or 1-[bis(dimethylamino)-
and the resultant solution was analyzed by LC-MS (Fig. 2). methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexaflu-
Although methyldiazirine 17a rapidly decomposed, a large orophosphate) (HATU)48) for 20c and d (R=CF3) under high-
quantity of alkenes was produced. In contrast, photolysis of dilution conditions to afford the desired macrocycles 1a–d in
trifluoromethyldiazirine 17c generated much less alkenes, and moderate yields. Considerable amount of a dimeric product
approximately half of the consumed trifluoromethyldiazirine (ca. 25%) was observed only in the formation of 1c. No seri-
was cross-linked with MeOH. However, the photolysis of tri- ous epimerization at the α-position of the alanine residue in
fluoromethyldiazirine 17c was sluggish compared with meth- the C-terminus was observed by LC-MS analysis.16–18,21) The
yldiazirine 17a (Fig. 2). The UV absorbance spectra showed structures of 1a–d were unambiguously determined by instru-

758
Chem. Pharm. Bull.
Vol. 64, No. 7 (2016)
Chart 4. Total Syntheses of 1a–d
Table 1. SOAT Inhibitory Activities of 1a–da)
Inhibition of CE syntheses by 1a–d was also investigated
by an enzyme assay. After addition of the compounds and
[14C]oleoyl-CoA to a microsomal fraction prepared from
SOAT1-CHO or SOAT2-CHO, the generated [14C]CE was
measured, and the IC50 values were calculated (Table 1). Com-
pounds 1a–d were found to inhibit CE synthesis catalyzed by
both SOAT1 and SOAT2. This indicates that photoaffinity la-
beling experiments are applicable not only to SOAT1 but also
SOAT2 using microsomal fractions.
IC50 (µM) for cholesteryl ester synthesis
Cell-based assay Enzyme assay
SOAT1-CHO SOAT2-CHO SOAT1 SOAT2
Compound
1a
0.98
2.7
>20
>20
>20
>20
20
0.53
1.4
3.0
1.1
2.2
3.0
3.4
9.5
2.6
7.3
1.9
3.0
1b
1c
1.9
1d
2.1
Beauveriolide I9)
Beauveriolide III9)
0.60
0.90
Conclusion
>20
To facilitate an analysis of the binding site in SOAT pro-
teins, we have designed beauveriolide analogues wherein the
D-Leu or D-allo-Ile residue was replaced by photoreactive
amino acids possessing methyldiazirine (1a, b) or trifluoro-
methyldiazirine (1c, d) in the side chain (Fig. 1). The meth-
a) Inhibitions of triacylglycerol and phospholipid syntheses were not observed
even at a concentration of 20µM of 1a–d in cell-based assay.
mental analyses, including 2D NMR.
SOAT Inhibitory Activity The SOAT inhibitory ac- yldiazirine moiety was installed by the reaction of methyl
tivities of 1a–d were estimated by a cell-based assay using ketones 8a and b with liquid ammonia to give an imine
SOAT1-CHO and SOAT2-CHO. Compounds 1a–d and [14C]- intermediate, followed by treatment with hydroxylamine-O-
oleic acid were added to each cell culture, and the resultant sulfonic acid to provide the diaziridines 9a and b. Subsequent
radioactivities of [14C]CE, [14C]triacylglycerol (TG), and [14C]- oxidation gave the desired methyldiazirines 10a and b. In con-
phospholipid (PL) were measured. Compounds 1a–d inhibited trast, the trifluoromethyldiazirines were prepared from trifluo-
production of CE in a dose dependent manner, but did not romethyl ketones 13c and d via oxime intermediates 14c and
inhibit the formation of TG and PL, even at a concentration d, which were transformed into diaziridines 15c and d. Subse-
of 20µM. This suggests that 1a–d selectively inhibit CE syn- quent oxidation afforded the desired trifluoromethyldiazirines
thesis in the CHO cells. The IC50 values for CE synthesis are 16c and d. Despite slow photolysis, rearrangement to yield al-
summarized in Table 1. SOAT1 inhibitory activities of 1a–d kene products was not prominent in the alkyl(trifluoromethyl)-
are as potent as those of beauveriolides I and III, whereas CE diazirines, suggesting that such compounds could be effective
synthesis by SOAT2-CHO was not inhibited by 1a–d at 20µM. for photo-affinity labeling. Sequential couplings of 3–5, fol-
The SOAT inhibitory activity and isozyme selectivity of 1a–d lowed by macrolactamization gave the desired macrocycles
were almost identical to those of beauveriolides I and III. 1a–d. The SOAT inhibitory activities of 1a–d were found to
These results strongly suggest that the alterations made to the be as potent as those of beauveriolides I and III. In addition,
side chain of the Leu or Ile residue in beauveriolides, which 1a–d inhibited SOAT1 selectively rather than SOAT2, which
resulted in incorporation of alkyl methyldiazirine or trifluoro- was also consistent with the behavior of beauveriolides I and
methyldiazirine, do not affect their SOAT inhibitory activity III (Table 1).
and isozyme selectivity.

Vol. 64, No. 7 (2016)
Chem. Pharm. Bull.
759
for 3h. The reaction mixture was poured into 1M aqueous
Experimental
General Techniques All commercially available reagents HCl and the aqueous phase was extracted three times with
were used as received. Dry THF and CH2Cl2 (Kanto Chemi- CH2Cl2. The organic layers were dried over Na2SO4, filtered,
cal Co., Tokyo, Japan) were obtained by passing through and concentrated in vacuo. The residue was chromatographed
activated alumina column with commercially available pre- on silica gel (hexane–ethyl acetate=7:3) to afford Weinreb
1
dried, oxygen-free formulations. All solution-phase reactions amide 7a (5.33g, 15.4mmol, 88%) as a colorless oil. H-NMR
were monitored by thin-layer chromatography (TLC) carried (600MHz, CDCl3) δ: 5.18 (1H, d, J=7.6Hz), 4.19 (1H, m), 3.68
out on 0.2mm E. Merck silica gel plates (60F-254) with UV (3H, s), 3.18 (3H, s), 2.54 (1H, m), 2.50 (1H, m), 2.15 (1H, m),
light, visualized by p-anisaldehyde H2SO4–ethanol solution, 1.93 (1H, m), 1.47 (9H, s), 1.44 (9H, s). 13C-NMR (150MHz,
phosphomolybdic acid–ethanol solution, or ninhydrin–acetic CDCl3) δ: 173.6, 171.6, 155.5, 81.9, 79.5, 61.2, 53.7, 32.2, 28.3,
acid–1-butanol solution. Column chromatography was carried 28.03, 27.96, 27.6. IR (neat) cm−1: 3340, 2978, 2936, 1794,
out with silica gel 60N (Kanto Chemical Co., 100–210µm). 1716, 1667, 1512, 1456, 1392, 1367, 1313, 1251, 1157, 1051,
[1-14C]Oleic acid was purchased from PerkinElmer, Inc. Life 1026, 848. HR-ESI-MS m/z: 369.1989 (Calcd for C16H30N2O6Na
and Analytical Sciences (U.S.A.). [14C]Oleoyl-CoA was pur- [M+Na]+: 369.1996). [α]D24 −5.40 (c=0.950, CHCl3).
chased from GE Healthcare Bio-Science (U.S.A.). Fetal bo-
tert-Butyl (R)-2-{[(tert-Butoxy)carbonyl]amino}-3-[meth-
vine serum (FBS) was bought from HyClone (U.S.A.). Ham’s oxy(methyl)carbamoyl]propanoate (7b) 7b was prepared
F12 was obtained from Sigma-Aldrich (U.S.A.). Geneticin from 6b (8.42g, 29.1mmol) in the same manner to the syn-
1
(G-418 sulfate) and MEM vitamin solution were purchased thesis of 7a. Yield 90% (a colorless oil). H-NMR (400MHz,
from Wako Pure Chemical Industries, Ltd. (Japan). Penicillin CDCl3) δ: 5.66 (1H, d, J=8.4Hz), 4.47–4.43 (1H, m), 3.68 (3H,
(10000 units/mL)/streptomycin (10000 units/mL) solution was s), 3.16 (3H, s), 3.15–3.11 (1H, m), 2.92–2.83 (1H, m), 1.46
acquired from Invitrogen (U.S.A.). Plastic microplates (48- (9H, s), 1.44 (9H, s). 13C-NMR (100MHz, CDCl3) δ: 171.7,
well) were purchased from Asahi Techno Glass (Japan).
170.4, 155.6, 81.6, 79.4, 61.1, 50.3, 34.6, 31.9, 28.1, 27.8. IR
1H-NMR spectra and 13C-NMR spectra were recorded on (neat) cm−1: 3431, 3361, 2978, 2936, 1718, 1662, 1496, 1391,
1
a JEOL JNM-AL400 (400MHz for H) or a JEOL ECA-600 1368, 1252, 1157, 1023, 850. HR-ESI-MS m/z: 355.1832 (Calcd
(600MHz for 1H) spectrometer in the indicated solvent. for C15H28N2O6Na [M+Na]+: 355.1840). [α]D24 −19.8 (c=1.50,
1
Chemical shifts (δ) for H-NMR spectra are referenced to CHCl3).
signals for internal tetramethylsilane (0ppm) and residual
tert-Butyl
(R)-2-{[(tert-Butoxy)carbonyl]amino}-5-oxo-
non-deuterated solvents (chloroform 7.26ppm; methanol-d4 hexanoate (8a) To a solution of Weinreb amide 7a (6.00g,
3.30ppm). Chemical shifts (δ) for 13C-NMR spectra are refer- 17.3mmol, 1.00 equiv.) in toluene (57mL) at –78°C under an
enced to signals for residual deuterated solvents (chloroform-d argon atmosphere was added methyl magnesium bromide
77.0ppm, methanol-d4 49.0ppm). Multiplicities are reported by (43.3mL, 1M solution in hexanes, 43.3mmol, 2.50 equiv.) over
the following abbreviations: s (singlet), d (doublet), t (triplet), q 30min. The reaction was allowed to warm up to –5°C over 3h,
(quartet), quint. (quintet), m (multiplet), dd (double doublet), dt quenched with 1M aqueous HCl, and the aqueous layer was
(double triplet), brs (broad singlet). Coupling constants (J) are extracted three times with ethyl acetate. The organic layers
represented in hertz (Hz).
were dried over Na2SO4, filtered, and concentrated in vacuo.
MS and HR-MS were measured on JEOL JMS-DX303 (for The residue was chromatographed on silica gel (hexane–ethyl
electron ionization (EI)), MS-AX500 (for FAB), SYNAPT acetate=8:1) to afford ketone 8a (5.21g, 17.3mmol, quant.) as
1
G2 HDMS instruments (for electrospray ionization (ESI)). a colorless oil. H-NMR (400MHz, CDCl3) δ: 5.05 (1H, brs),
IR spectra were recorded on a JASCO FTIR-4100. Only the 4.15 (1H, brs), 2.62–2.44 (2H, m), 2.15 (3H, s), 2.12–2.00 (1H,
strongest and/or structurally important absorptions are report- m), 1.89–1.79 (1H, m), 1.46 (9H, s), 1.44 (9H, s). 13C-NMR
ed in wavenumbers (cm−1). Optical rotations were measured (100MHz, CDCl3) δ: 207.6, 171.5, 155.4, 82.1, 79.7, 53.4, 39.4,
on a JASCO P-1010 polarimeter at 589nm. Melting points 29.9, 28.3, 28.0, 26.8. IR (neat) cm−1: 3361, 2979, 2933, 1716,
were measured on a RFS-10 melting point apparatus (Round 1514, 1367, 1252, 1156, 1056, 848. HR-ESI-MS m/z: 324.1777
Science Inc.) and are uncorrected.
Analytical reversed-phase HPLC (RP-HPLC) was per- (c=1.00, CHCl3).
formed on a Waters LC/MS system consisting of a Waters 600 tert-Butyl (R)-2-{[(tert-Butoxy)carbonyl]amino}-4-oxo-
(Calcd for C15H27NO5Na [M+Na]+: 324.1781). [α]2D0 −4.00
HPLC System equipped with a Waters 2996 Photodiode Array pentanoate (8b) To a solution of Weinreb amide 7b (3.00g,
Detector and a Waters Micromass ZQ 2000 unit. Preparative 9.03mmol, 1.00 equiv.) in tetrahydrofuran (THF) (90mL) at
RP-HPLC was carried out on a Waters 1525EF HPLC System –78°C under an argon atmosphere was added methyl lithium
equipped with a Waters 2489 UV/Visible detector (monitoring (18.1mL, 1M solution in hexanes, 18.1mmol, 2.00 equiv.)
at 214, 254nm).
over 30min and the mixture was stirred at room temperature
tert-Butyl (R)-2-{[(tert-Butoxy)carbonyl]amino}-4-[meth- for 2h. The reaction mixture was quenched with 1M aque-
oxy(methyl)carbamoyl]butanoate (7a) To a solution of ous HCl, and the aqueous layer was extracted with ethyl
carboxylic acid 6a (5.30g, 17.5mmol, 1.00 equiv.) in dry acetate. The organic layers were pooled, dried over Na2SO4,
CH2Cl2 (84mL) at 0°C was added N-methylmorpholine filtered, and concentrated in vacuo. The residue was chroma-
(NMM) (4.80mL, 43.8mmol, 2.50 equiv.) followed by isobutyl tographed on silica gel (hexane–ethyl acetate=8:1) to afford
1
chloroformate (2.90mL, 21.9mmol, 1.25 equiv.). The mixture ketone 8b (2.34g, 8.13mmol, 90%) as a colorless oil. H-NMR
was stirred for 15min, and then N,O-dimethylhydroxylamine (400MHz, CDCl3) δ: 5.44 (1H, d, J=6.8Hz), 4.36 (1H, brs),
hydrochloride (2.05g, 21.0mmol, 1.20 equiv.) was added 3.10 (1H, dd, J=17.8, 4.0Hz), 2.90 (1H, dd, J=17.8, 4.4Hz),
portionwise. The reaction was then allowed to warm up to 2.16 (3H, s), 1.44 (18H, s). 13C-NMR (100MHz, CDCl3) δ:
room temperature and stirred under an argon atmosphere 206.5, 170.3, 155.6, 82.1, 79.7, 50.1, 45.6, 29.9, 28.3, 27.8. IR

760
Chem. Pharm. Bull.
Vol. 64, No. 7 (2016)
(neat) cm−1: 3364, 2979, 2933, 1719, 1499, 1368, 1252, 1156, extracted with ethyl acetate. The combined organic layer was
1052, 847. HR-ESI-MS m/z: 310.1620 (Calcd for C14H25NO5Na washed with saturated aqueous NaHCO3 and brine, dried over
[M+Na]+: 310.1625). [α]D23 −13.8 (c=0.95, CHCl3).
tert-Butyl
methyl-3H-diazirin-3-yl)butanoate (10a)
MgSO4, filtered and concentrated in vacuo. The residue was
(R)-2-{[(tert-Butoxy)carbonyl]amino}-4-(3- chromatographed on silica gel (hexane–ethyl acetate=1:1) to
solution of afford carboxylic acid 3a (692mg, 2.69mmol, 71% in 2 steps)
A
ketone 8a (3.30g, 10.9mmol, 1.00 equiv.) in liquid am- as a slightly yellow oil. 1H-NMR (400MHz, CDCl3, mix-
monia (24mL) was stirred under reflux (dry ice condenser) ture of rotamers) δ: 10.4 (1H, brs), 6.80 (0.4×1H, brs), 5.07
for 5h. Then, the reaction flask was cooled to −50°C and (0.6×1H, brs), 4.29 (0.6×1H, brs), 4.12 (0.4×1H, brs), 1.77
hydroxylamine-O-sulfonic acid (1.43g, 12.7mmol, 1.15 equiv.) (1H, m), 1.49 (1H, m), 1.45 (10H, m), 1.25 (1H, m), 1.03 (3H,
dissolved in dry methanol (6.3mL) was added over 30min. m). 13C-NMR (100MHz, CDCl3, mixture of rotamers) δ: 176.5,
The reaction was then allowed to stir under reflux for 10h 175.6, 156.9, 155.5, 82.0, 80.4, 53.7, 52.7, 30.4, 30.0, 28.2, 27.1,
until ammonia was evaporated. The resulting slurry was fil- 25.2, 19.6, 14.1. IR (neat) cm−1: 3324, 3107, 2979, 2932, 1719,
tered and the filter cake was washed with several portions of 1514, 1454, 1394, 1369, 1252, 1164, 1051, 1027, 855, 779. HR-
methanol. The filtrate was concentrated until ammonia was FAB-MS 258.1461 (Calcd for C11H20N3O4 [MH]+: 258.1454).
completely evaporated.
The residue was then diluted with methanol (38mL) and
[α]2D7 −16.8 (c=1.00, CHCl3).
(R)-2-{[(tert-Butoxy)carbonyl]amino}-3-(3-methyl-3H-
cooled to 0°C. NEt3 (1.52mL, 10.9mmol, 1.00 equiv.) was diazirin-3-yl)propanoic Acid (3b) 3b was prepared from
added followed by portionwise addition of iodine until a Boc-D-photo-Leu-OtBu (10b) (230mg, 0.769mmol) in the
brown color persisted. After being stirred at room tempera- same manner to the synthesis of 3a. Yield 78% in 2 steps (a
1
ture for 2h, the reaction mixture was quenched with saturated slightly yellow oil). H-NMR (400MHz, CDCl3, mixture of
aqueous Na2S2O3 and the aqueous layer was extracted three rotamers) δ: 9.12 (1H, brs), 6.59 (0.3×1H, brs), 5.10 (0.7×1H,
times with ethyl acetate. The combined organic layers were brs), 4.38 (0.7×1H, brs), 4.12 (0.3×1H, brs), 2.06 (0.7×1H, dd,
washed with brine, dried over MgSO4, filtered, and concen- J=15.0, 4.6Hz), 1.89–1.87 (0.3×1H, m), 1.73–1.68 (0.3×1H,
trated in vacuo. The residue was chromatographed on silica m), 1.61 (0.7×1H, dd, J=15.0, 8.8Hz), 1.48 (9H, s), 1.10 (3H,
gel (hexane–ethyl acetate=10:1) to afford methyldiazirine 10a s). 13C-NMR (100MHz, CDCl3) δ: 176.0, 155.4, 80.7, 50.2,
1
(1.50g, 4.80mmol, 44% in 2 steps) as white solids. H-NMR 37.5, 28.3, 23.7, 19.7. IR (neat) cm−1: 3323, 2966, 2928, 2857,
(400MHz, CDCl3, mixture of rotamers) δ: 5.04 (1H, m), 4.06 1717, 1508, 1455, 1394, 1368, 1254, 1163, 1051, 1025, 878. HR-
(1H, brs), 1.62 (1H, m), 1.39–1.37 (19H, s) 1.29–1.19 (2H, m), FAB-MS m/z: 244.1302 (Calcd for C14H25N3O4Na [M+Na]+:
0.943-0.936 (3H, m). 13C-NMR (100MHz, CDCl3) δ: 171.2, 244.1297). [α]D24 −15 (c=0.47, CHCl3).
155.1, 81.9, 79.5, 53.2, 30.2, 28.1, 27.8, 27.2, 25.1, 19.5. IR
tert-Butyl (R)-2-{[Bis(tert-Butoxy)carbonyl]amino}-6,6,6-
(neat) cm−1: 3362, 2979, 2932, 2870, 1716, 1581, 1501, 1455, trifluoro-5-oxohexanoate (13c) To a solution of aldehyde
1368, 1253, 1160, 1049, 847. HR-ESI-MS m/z: 336.1882 (Calcd 11c (350mg, 0.903mmol, 1.00 equiv.) in dry THF (4.5mL)
for C15H27N3O4Na [M+Na]+: 336.1894). [α]D28 −15.7 (c=1.00, was added tetramethylsilane, or trimethylsilyl (TMS)-CF3
CHCl3).
tert-Butyl
(266 µL, 1.80mmol, 2.00 equiv.) and tetrabutylammonium
(R)-2-{[(tert-Butoxy)carbonyl]amino}-3-(3- fluoride (TBAF) (90µL, 1.0M in THF, 90.3µmol, 0.10 equiv.)
methyl-3H-diazirin-3-yl)propanoate (10b) 10b was pre- at 0°C under an argon atmosphere. After being stirred at room
pared from 8b (2.40g, 8.35mmol) in the same manner to temperature for 3h, the reaction mixture was poured into 1M
the synthesis of 10a. Yield 35% in 2 steps (white solids). aqueous HCl at 0°C and the mixture was stirred for 2h. The
1H-NMR (400MHz, CDCl3, mixture of rotamers) δ: 5.09 (1H, aqueous layer was extracted with ethyl acetate. The combined
d, J=6.8Hz), 4.26 (1H, m), 1.82 (1H, dd, J=14.6, 5.6Hz), 1.61 organic layer was washed with 1M aqueous HCl and brine,
(1H, dd, J=14.6, 6.8Hz), 1.55 (0.1×9H, s), 1.53 (0.1×9H, s), dried over MgSO4, filtered and concentrated in vacuo. The
1.49 (0.9×9H, s), 1.47 (0.9×9H, s), 1.09 (3H, s). 13C-NMR residue was used for the next reaction without further purifica-
(100MHz, CDCl3, mixture of rotamers) δ: 170.7, 155.0, 146.7, tion.
85.1, 82.5, 79.9, 67.0, 50.9, 38.2, 29.6, 28.3, 27.8, 27.3, 23.8,
To a solution of the above residue in CH2Cl2 (9mL) was
19.7, 14.1. IR (neat) cm−1: 3364, 2980, 2933, 1716, 1505, 1456, added NaHCO3 (567mg, 6.75mmol, 7.50 equiv.) and Dess-
1368, 1251, 1156, 1066, 847. HR-ESI-MS m/z: 322.1736 (Calcd Martin periodinane (572mg, 1.35mmol, 1.50 equiv.) at 0°C
for C14H25N3O4Na [M+Na]+: 322.1737). [α]D24 −13.1 (c=1.00, under an argon atmosphere and the mixture was stirred at
CHCl3).
room temperature for 3h. The reaction mixture was poured
(R)-2-{[(tert-Butoxy)carbonyl]amino}-4-(3-methyl-3H- into saturated aqueous Na2S2O3 and the aqueous layer was
diazirin-3-yl)butanoic Acid (3a) A solution of Boc-D- extracted with CH2Cl2. The combined organic layer was
photo-Met-OtBu (10a) (1.19g, 3.79mmol, 1.00 equiv.) in THF washed with brine, dried over MgSO4, filtered and concen-
(118mL) and 8M aqueous HCl (118mL) was stirred at room trated in vacuo. The residue was chromatographed on silica
temperature for 16h. The suspension was concentrated in gel (hexane–ethyl acetate=8:1) to afford trifluoromethyl ke-
vacuo. The residue was used for the next reaction without tone 13c (307mg, 0.677mmol, 75% in 2 steps) as white solids.
further purification.
1H-NMR (400MHz, CDCl3) δ: 4.74 (1H, dd, J=9.2, 5.6Hz),
To a solution of the above residue in a mixture of 1,4-diox- 2.89 (1H, m), 2.77 (1H, m), 2.48 (1H, m), 2.19 (1H, m), 1.50
ane (3.8mL) and H2O (3.8mL) was added di-tert-butyl dicar- (18H, s), 1.45 (9H, s). 13C-NMR (100MHz, CDCl3) δ: 190.6 (q,
bonate (910mg, 4.17mmol, 1.10 equiv.) and Na2CO3 (603mg, J=36Hz), 168.9, 152.4, 115.5 (q, J=293Hz), 83.3, 81.7, 57.6,
5.69mmol, 1.50 equiv.) at 0°C and the mixture was stirred at 33.2, 28.0, 27.9, 22.1. IR (neat) cm−1: 3431, 2981, 2936, 1739,
room temperature for 12h. The reaction mixture was acidi- 1699, 1480, 1458, 1369, 1256, 1149, 1007, 849. HR-ESI-MS
fied with 1M aqueous HCl to pH 4 and the aqueous layer was m/z: 478.2001 (Calcd for C20H32F3NO7Na [M+Na]+: 478.2023).

Vol. 64, No. 7 (2016)
Chem. Pharm. Bull.
761
1
[α]D24 +0.92 (c=1.00, CHCl3).
23% in 4 steps (white solids). H-NMR (600MHz, CDCl3)
tert-Butyl (R)-2-{[Bis(tert-butoxy)carbonyl]amino}-5,5,5- δ: 4.74 (1H, t, J=7.3Hz), 2.43 (2H, d, J=7.6Hz), 1.52 (18H,
trifluoro-4-oxopentanoate (13d) 13d was prepared from s), 1.43 (9H, s). 13C-NMR (150MHz, CDCl3) δ: 168.1, 151.9,
aldehyde 11d (200mg, 0.536mmol) in the same manner to the 122.3 (q, J=273Hz), 83.5, 82.4, 54.0, 28.0, 27.8, 27.1, 25.8 (q,
1
synthesis of 13c. Yield 75% in 2 steps (white solids). H-NMR J=40Hz). IR (neat) cm−1: 2982, 2933, 1739, 1703, 1394, 1383,
(600MHz, CDCl3) δ: 5.44 (1H, dd, J=7.7, 5.2Hz), 3.76 (1H, 1369, 1281, 1236, 1154, 849. HR-ESI-MS m/z: 476.1967 (Calcd
dd, J=18.6, 7.8Hz), 2.95 (1H, dd, J=18.6, 5.2Hz), 1.51 (18H, for C19H30F3N3O6Na [M+Na]+: 476.1979). [α]D28 +27.8 (c=1.00,
s), 1.44 (9H, s). 13C-NMR (150MHz, CDCl3) δ: 188.5 (q, CHCl3).
J=36Hz), 167.9, 152.0, 115.4 (q, J=291Hz), 83.7, 82.7, 54.0,
( R) -2 - {[(t e r t - B u t ox y) c a r b o n y l ]a m i n o} - 4 - [3 -
37.7, 28.0, 27.8. IR (neat) cm−1: 2982, 2937, 1766, 1739, 1701, (trifluoromethyl)-3H-diazirin-3-yl]butanoic Acid (3c)
A
1480, 1458, 1395, 1369, 1258, 1172, 1149, 846. HR-ESI-MS solution of 16c (56.0mg, 0.121mmol, 1.00 equiv.) in 4M HCl/
m/z: 464.1847 (Calcd for C19H30F3NO7Na [M+Na]+: 464.1867). dioxane (6.6mL) was stirred at room temperature for 20h. The
[α]2D6 +23.4 (c=1.00, CHCl3).
suspension was concentrated in vacuo. The residue was used
tert-Butyl (R)-2-{[Bis(tert-butoxy)carbonyl]amino}-4-[3- for the next reaction without further purification.
(trifluoromethyl)-3H-diazirin-3-yl]butanoate (16c) To To a solution of the above residue in a mixture of 1,4-di-
a
solution of trifluoromethyl ketone 13c (240mg, 0.527mmol, oxane (0.28mL) and H2O (0.28mL) was added di-tert-butyl
1.00 equiv.) in dry pyridine (1mL) and dry ethanol (0.5mL) dicarbonate (30.0mg, 0.133mmol, 1.10 equiv.) and Na2CO3
was added HONH2·HCl (40.3mg, 0.58mmol, 1.10 equiv.) (19.0mg, 0.182mmol, 1.50 equiv.) at 0°C and the mixture was
under an argon atmosphere and the mixture was stirred at stirred at room temperature for 12h. The reaction mixture
60°C for 14h. The suspension was concentrated in vacuo. The was acidified with 1M aqueous HCl and the aqueous layer was
residue was used for the next reaction without further purifica- extracted with ethyl acetate. The combined organic layer was
tion.
washed with saturated aqueous NaHCO3 and brine, dried over
To a solution of the above residue in dry CH2Cl2 (1mL) was MgSO4, filtered and concentrated in vacuo. The residue was
added NEt3 (190 µL, 1.37mmol, 2.60 equiv.), tosyl chloride chromatographed on silica gel (hexane–ethyl acetate=3:1) to
(TsCl) (120mg, 0.632mmol, 1.20 equiv.) and DMAP (3.20mg, afford 3c (25.7mg, 0.0820mmol, 68% in 2 steps) as white sol-
1
26.3µmol, 0.05 equiv.) at 0°C under an argon atmosphere and ids. H-NMR (400MHz, CDCl3, mixture of rotamers) δ: 10.55
the mixture was stirred at room temperature for 2h. The reac- (1H, brs), 7.03 (0.5×1H, brs), 5.08 (0.5×1H, d, J=7.3Hz),
tion mixture was poured into H2O and the aqueous layer was 4.29 (0.5×1H, m), 4.14 (0.5×1H, m), 1.87–1.77 (4H, m), 1.47
extracted with ethyl acetate. The combined organic layer was (0.5×9H, s), 1.45 (0.5×9H, s); 13C-NMR (100MHz, CDCl3,
washed with brine, dried over MgSO4, filtered and concen- mixture of rotamers) δ: 176.0, 174.8, 157.0, 155.6, 122.3 (q,
trated in vacuo. The residue was used for the next reaction J=275Hz), 82.6, 80.8, 53.3, 52.4, 28.2, 27.2 (q, J=40Hz), 26.3,
without further purification.
26.2, 22.5, 22.0. IR (neat) cm−1: 3323, 2982, 2936, 1719, 1513,
A solution of the above residue in liquid ammonia (10mL) 1456, 1370, 1157, 1056, 849. HR-FAB-MS m/z: 312.1163 (Calcd
and dry CH2Cl2 (1mL) was stirred under reflux equipped with for C11H17F3N3O7 [M+H]+: 312.1171). [α]D23 −5.01 (c=1.00,
a dry ice condenser for 5h until the ammonia was evaporated. CHCl3).
Resulting slurry was filtered and the filter cake was washed
( R) -2 - {[(t e r t - B u t ox y) c a r b o n y l ]a m i n o} - 3 - [3 -
with several portions of methanol. The combined filtrate was (trifluoromethyl)-3H-diazirin-3-yl]propanoic Acid (3d) 3d
concentrated in vacuo and the residue was used for the next was prepared from 16d (100mg, 0.221mmol) in the same
reaction without further purification.
manner to the synthesis of 3c. Yield 53% in 2 steps (white
1
A solution of the above residue in dry MeOH (1.8mL) was solids). H-NMR (400MHz, CDCl3, mixture of rotamers) δ:
added NEt3 (110 µL, 0.760mmol, 1.50 equiv.) and I2 (147mg, 9.66 (1H, brs), 6.56 (0.3×1H, brs), 5.08 (0.7×1H, s), 4.21–4.10
0.580mmol, 1.10 equiv.) at 0°C under an argon atmosphere (1H, m), 2.47–2.37 (1H, m), 2.21–2.11 (1H, m), 1.47 (9H, s).
and the mixture was stirred at room temperature for 2h. The 13C-NMR (100MHz, CDCl3, mixture of rotamers) δ: 174.7
reaction mixture was poured into saturated aqueous Na2S2O3 (major) and 174.0 (minor), 156.5 (minor) and 155.2 (major),
and the aqueous layer was extracted with ethyl acetate. The 129.9 (minor, q, J=211Hz) and 122.0 (major, q, J=274Hz),
combined organic layer was washed with brine, dried over 83.0 (minor) and 81.1 (major), 49.9 (minor) and 49.0 (major),
MgSO4, filtered and concentrated in vacuo. The residue was 30.1 (minor) and 29.7 (major), 28.7 (minor) and 28.2 (major),
chromatographed on silica gel (hexane–ethyl acetate=30:1) 25.6 (q, J=41Hz). IR (neat) cm−1: 3334, 2982, 2927, 2853,
to afford trifluoromethyldiazirine 16c (128.1mg, 0.274mmol, 1718, 1701, 1523, 1395, 1370, 1319, 1195, 1156, 1078. HR-FAB-
1
56% in 4 steps) as white solids. H-NMR (400MHz, CDCl3) MS 320.0828 (Calcd for C10H14F3N3O4Na [M+Na]+: 320.0834).
δ: 4.53 (1H, dd, J=8.3, 5.2Hz), 1.89 (1H, m), 1.72–1.59 (3H, [α]D23 −6.73 (c=0.415, CHCl3).
m), 1.43 (18H, s), 1.37 (9H, s). 13C-NMR (100MHz, CDCl3) δ:
Allyl (3S,4S)-3-Hydroxy-4-methyloctanoic Acid (4) To
168.8, 152.3, 122.4 (q, J=272Hz), 83.2, 81.7, 57.8, 27.9, 27.8, a solution of carboxylic acid 18 (2.00g, 11.5mmol, 1.00 equiv.)
27.2 (q, J=40Hz), 23.3, 23.1. IR (neat) cm−1: 2982, 2935, 1741, in MeOH (4.1mL) was added a solution of cesium carbon-
1701, 1458, 1369, 1269, 1155, 849. HR-ESI-MS m/z: 490.2125 ate (1.87g, 5.74mmol, 0.50 equiv.) in distilled water (4.1mL)
(Calcd for C20H32F3N3O6Na [M+Na]+: 490.2135). [α]D24 +0.58 at 0°C and the mixture was stirred at room temperature for
(c=1.50, CHCl3).
40min. The reaction mixture was concentrated in vacuo and
tert-Butyl (R)-2-{[Bis(tert-butoxy)carbonyl]amino}-3-[3- diluted with N,N-dimethylformamide (DMF) (57mL). To
(trifluoromethyl)-3H-diazirin-3-yl]propanoate (16d) 16d the solution was added allyl bromide (1.03mL, 12.0mmol,
was prepared from trifluoromethyl ketone 13d (300mg, 1.05 equiv.) at 0°C and the mixture was stirred at the same
0.680mmol) in the same manner to the synthesis of 16c. Yield temperature for 1h under an argon atmosphere. The reaction

762
Chem. Pharm. Bull.
Vol. 64, No. 7 (2016)
mixture was quenched with saturated aqueous NaHCO3 and (100MHz, CDCl3, mixture of rotamers) δ: 170.9 (major) and
diluted with ethyl acetate. The organic layer was separated 170.7 (minor), 170.0 (minor) and 169.9 (major), 155.1, 131.7,
and the aqueous layer was extracted with ethyl acetate. The 122.2 (q, J=274Hz), 118.5 (minor) and 118.4 (major), 79.9,
combined organic layer was washed with H2O, brine, dried 74.7, 65.31 (minor) and 65.28 (major), 52.5 (minor) and 52.3
over MgSO4 and filtered. The filtrate was concentrated in (major), 36.3 (major) and 36.2 (minor), 35.9, 32.1 (major) and
vacuo, and then the resulting residue was purified by column 31.8 (minor), 29.1, 28.0, 27.1 (q, J=40Hz, major) and 27.0 (q,
chromatography on silica gel (hexane–ethyl acetate=15:1) J=40Hz, minor), 26.2 (minor) and 26.1 (major), 22.5, 22.1
to afford the allyl (3S,4S)-3-hydroxy-4-methyloctanoic acid (minor) and 21.9 (major), 14.3 (major) and 14.1 (minor), 13.73
(4) (2.12g, 9.89mmol, 87%) as a colorless oil. 1H-NMR (major) and 13.72 (minor). IR (neat) cm−1: 3370, 2966, 2934,
(400MHz, CDCl3) δ: 5.92 (1H, ddt, J=17.2, 10.4, 5.6Hz), 2874, 2862, 1744, 1735, 1719, 1508, 1500, 1456, 1392, 1367,
5.33 (1H, d, J=17.2Hz), 5.25 (1H, d, J=10.4Hz), 4.62 (2H, 1346, 1301, 1276, 1252, 1159, 1104, 1052, 1026, 990, 934. HR-
d, J=5.6Hz), 3.95 (1H, brs), 2.79 (1H, s), 2.49 (2H, m), 1.50 ESI-MS m/z: 530.2432 (Calcd for C23H36F3N3O6Na [M+Na]+:
(2H, m), 1.30 (4H, m), 1.14 (1H, m), 0.90 (6H, m). 13C-NMR 530.2448). [α]2D5 −62 (c=0.90, CHCl3).
(100MHz, CDCl3) δ: 173.0, 131.9, 118.5, 71.2, 65.3, 38.8, 38.0,
Ester Unit 19d (n=1, R=CF3) 19d was prepared from
32.3, 29.40, 22.9, 14.2, 14.0. IR (neat) cm−1: 3459, 2958, 2930, 3d (29.0mg, 97.6µmol, 1.10 equiv.) and 4 (19.0mg, 88.7µmol,
2873, 2860, 1737, 1380, 1276, 1173, 987cm−1. HR-ESI-MS 1.00 equiv.) in the same manner to the synthesis of 19a. Yield
m/z: 237.1458 (Calcd for C12H22O3Na [M+Na]+: 237.1461). [α]D24 80% (a colorless oil). H-NMR (400MHz, CDCl3, mixture
1
−55.2 (c=1.50, CHCl3).
of rotamers) δ: 5.91 (1H, dt, J=16.8, 10.4, 5.6Hz), 5.35–5.25
Ester Unit 19a (n=2, R=Me) To a solution of 3a (154mg, (3H, m), 5.05 (1H, d, J=6.8Hz), 4.57 (2H, d, J=6.0Hz),
0.600mmol, 1.30 equiv.) and 4 (100mg, 0.470mmol, 1.00 4.14 (1H, m), 2.66 (1H, dd, J=16.0, 8.1Hz), 2.57 (1H, dd,
equiv.) in dry CH2Cl2 (4mL) was added DMAP (5.7mg, J=16.0, 4.4Hz), 2.35 (1H, dd, J=16.2, 4.1Hz), 2.09 (1H, dd,
47 µmol, 0.10 equiv.) and DIC (218µL, 1.40mmol, 3.00 equiv.) J=16.2, 8.8Hz), 1.46 (9H, s), 1.30–1.11 (7H, m), 0.90 (6H,
at 0°C under an argon atmosphere and the mixture was stirred m). 13C-NMR (100MHz, CDCl3, mixture of rotamers) δ:
at the same temperature for 10h. The reaction mixture was 170.2 (major) and 170.1 (minor), 170.0, 154.8, 131.8, 122.0 (q,
diluted with CH2Cl2, washed with 1M aqueous HCl, saturated J=275Hz), 118.8, 80.5, 75.4 (minor) and 75.4 (major), 65.56
aqueous NaHCO3 and brine, dried over MgSO4, filtered and (minor) and 65.54 (major), 48.8, 36.3, 36.2, 32.1, 29.2, 29.0,
concentrated in vacuo. The residue was chromatographed 28.0, 25.7 (q, J=40Hz), 22.7, 14.4, 14.0. IR (neat) cm−1: 3373,
on silica gel (hexane–ethyl acetate=5:1) to afford ester unit 2963, 2932, 2876, 2862, 1742, 1728, 1722, 1717, 1507, 1368,
1
19a (188mg, 0.414mmol, 88%) as a colorless oil. H-NMR 1286, 1204, 1158, 1051. HR-ESI-MS m/z: 516.2285 (Calcd for
(400MHz, CDCl3) δ: 5.92 (1H, ddt, J=16.8, 10.0, 5.8Hz), C22H34F3N3O6Na [M+Na]+: 516.2292). [α]2D2 −13.2 (c=0.475,
5.34 (1H, dd, J=17.2, 1.6Hz), 5.26 (2H, m), 4.98 (1H, brs), CHCl3).
4.59 (2H, d, J=5.8Hz), 4.22 (1H, brs), 2.65–2.52 (2H, m),
Depsipeptide 20a (n=2, R=Me) To a solution of ester
1.73–1.65 (2H, m), 1.44 (9H, s), 1.42–1.30 (9H, m), 1.01 (3H, unit 19a (227mg, 0.500mmol, 1.00 equiv.) and morpholine
s), 0.92–0.87 (6H, m). 13C-NMR (100MHz, CDCl3) δ: 171.3, (109 µL, 1.25mmol, 2.50 equiv.) in dry THF (15mL) was
169.9, 155.0, 131.7, 118.3, 79.5, 74.4, 65.2, 52.6, 36.1, 36.0, 31.7, added a catalytic amount of Pd(PPh3)4 (57.0mg, 50µmol,
29.9, 29.0, 28.7, 26.8, 25.0, 22.8, 19.4, 14.3, 13.8. IR (neat) 0.10 equiv.) at room temperature under an argon atmosphere.
cm−1: 3373, 2962, 2932, 2861, 1741, 1718, 1507, 1454, 1367, After being stirred at the same temperature for 1h, the reac-
1276, 1250, 1164, 1022. HR-ESI-MS m/z: 476.2713 (Calcd tion mixture was diluted with ethyl acetate, washed with 1M
for C23H39N3O6Na [M+Na]+: 476.2731). [α]D21 −47.7 (c=1.50, aqueous HCl and brine, dried over MgSO4, filtered and con-
CHCl3).
centrated in vacuo. The residue was used for the next reaction
Ester Unit 19b (n=1, R=Me) 19b was prepared from 3b without further purification.
(44mg, 0.182mmol, 1.30 equiv.) and 4 (30.0mg, 0.140mmol,
To a solution of dipeptide 5 (260mg, 0.610mmol, 1.20
1.00 equiv.) in the same manner to the synthesis of 19a. equiv.) in MeOH (11mL) was added 10% Pd/C (52.0mg,
1
Yield 81% (a colorless oil). H-NMR (400MHz, CDCl3) δ: 20wt%) under an argon atmosphere and the reaction mixture
5.90 (1H, ddt, J=16.8, 10.4, 5.8Hz), 5.32 (1H, d, J=16.8Hz), was purged with hydrogen three times. The reaction mixture
5.26 (2H, m), 5.05 (1H, d, J=7.2Hz), 4.56 (2H, d, J=5.8Hz), was stirred at room temperature for 2h under a hydrogen
4.34 (1H, m), 2.73–2.55 (2H, m), 1.89 (1H, dd, J=5.6Hz), 1.75 atmosphere. The suspension was filtered through a pad of
(1H, m), 1.47 (9H, s), 1.31–1.26 (7H, m), 1.08 (3H, s), 0.91 Celite® and the filtrate was concentrated in vacuo. The residue
(6H, m). 13C-NMR (100MHz, CDCl3) δ: 171.1, 170.2, 155.0, was used for the next reaction without further purification.
131.8, 118.7, 80.1, 75.0, 65.5, 50.3, 38.0, 36.3, 36.2, 32.1, 29.2,
To a solution of the carboxylic acid and amine in CH2Cl2
28.3, 23.8, 22.7, 19.6, 14.4, 14.0. IR (neat) cm−1: 3380, 2961, (6mL) was added HOBt (101mg, 0.75mmol, 1.50 equiv.),
2931, 2874, 1741, 1718, 1507, 1457, 1367, 1279, 1248, 1165, DIEA (235µL, 1.35mmol, 2.70 equiv.) and EDCI·HCl (125mg,
1023. HR-ESI-MS m/z: 462.2562 (Calcd for C22H37N3O6Na 0.65mmol, 1.30 equiv.) at 0°C under an argon atmosphere and
[M+Na]+: 462.2575). [α]D24 −17.3 (c=1.00, CHCl3).
the mixture was stirred at the same temperature for 10h. The
Ester Unit 19c (n=2, R=CF3) 19c was prepared from reaction mixture was poured into 1M aqueous HCl and the
3c (342mg, 1.10mmol, 1.10 equiv.) and 4 (214mg, 1.00mmol, aqueous layer was extracted with ethyl acetate. The combined
1.00 equiv.) in the same manner to the synthesis of 19a. Yield organic layer was washed with saturated aqueous NaHCO3,
1
90% (a colorless oil). H-NMR (400MHz, CDCl3, mixture of brine, dried over MgSO4, filtered and concentrated in vacuo.
rotamers) δ: 5.86–5.74 (1H, m), 5.25–5.07 (4H, m), 4.47 (2H, The residue was chromatographed on silica gel (hexane–ethyl
dd, J=8.5, 7.1Hz), 4.13 (1H, s), 2.57–2.43 (2H, m), 1.77–1.56 acetate=3:1) to afford depsipeptide 20a (265mg, 0.385mmol,
1
(4H, m), 1.34 (9H, s), 1.20 (7H, m), 0.80 (6H, m); 13C-NMR 77% from 19a) as a colorless oil. H-NMR (400MHz, CDCl3)

Vol. 64, No. 7 (2016)
Chem. Pharm. Bull.
763
δ: 7.32–7.20 (5H, m), 6.83 (1H, d, J=6.8Hz), 6.67 (1H, d, J=7.3Hz), 4.17 (1H, m), 3.11 (1H, dd, J=13.9, 6.8Hz), 3.04
J=7.2Hz), 5.26 (1H, d, J=7.2Hz), 5.15 (1H, m), 4.70 (1H, (1H, dd, J=13.9, 7.3Hz), 2.42 (2H, m), 1.83–1.67 (2H, m), 1.44
q, J=7.2Hz), 4.37–4.30 (1H, dq, J=7.2, 6.8Hz), 4.17 (1H, (18H, s), 1.32 (3H, d, J=7.3Hz), 1.30–1.18 (7H, m), 0.88 (6H,
m), 3.09 (2H, d, J=7.2Hz), 2.44 (2H, m), 1.71–1.64 (2H, m), m). 13C-NMR (100MHz, CDCl3, mixture of rotamers) δ: 171.8
1.44 (9H, s), 1.32 (9H, s), 1.43–1.20 (12H, m), 0.99 (3H, s), (major) and 171.5 (minor), 170.5 (major) and 170.4 (minor),
0.89–0.86 (6H, m). 13C-NMR (100MHz, CDCl3) δ: 171.6, 170.2, 169.5, 155.0, 136.4, 129.2 (major) and 129.1 (minor),
171.4, 170.3, 169.4, 155.3, 136.5, 129.1, 128.4, 126.7, 81.7, 79.7, 128.6, 127.0, 122.1 (J=273Hz), 82.1, 80.3, 75.9, 54.6 (minor)
75.1, 54.4, 52.9, 48.7, 38.3, 38.1, 36.2, 32.0, 30.3, 29.1, 28.2, and 54.4 (major), 48.8, 38.4, 38.1, 36.4, 32.1 (major) and 31.9
27.8, 26.6, 25.1, 22.6, 19.5, 18.1, 14.2, 13.9. HR-ESI-MS m/z: (minor), 29.7, 29.1, 28.2, 27.9, 25.8 (J=38Hz), 22.7, 18.3, 18.1,
710.4091 (Calcd for C36H57N5O8Na [M+Na]+: 710.4099). IR 14.4 (major) and 14.3 (minor), 14.1 (minor) and 14.0 (major).
(neat) cm−1: 3289, 2977, 2931, 2859, 1735, 1719, 1654, 1647, IR (neat) cm−1: 3285, 3067, 2963, 2931, 2858, 1738, 1733, 1645,
1551, 1454, 1367, 1158, 1048, 1024, 755. [α]2D2 −17 (c=0.50, 1557, 1456, 1368, 1284, 1154, 1051, 750. HR-ESI-MS m/z:
CHCl3).
750.3651 (Calcd for C35H52F3N5O8Na [M+Na]+: 750.3660). [α]2D2
Depsipeptide 20b (n=1, R=Me) 20b was prepared from −10.3 (c=1.00, CHCl3).
ester unit 19b (37.1mg, 81.6µmol) and dipeptide 5 (41.8mg,
Beauveriolide Analogue 1a (n=2, R=Me) To a solution
97.9 µmol) in the same manner to the synthesis of 20a. Yield of depsipeptide 20a (140mg, 0.204mmol, 1.00 equiv.) in 4M
1
77% from 19b (a colorless oil). H-NMR (600MHz, CDCl3) HCl/dioxane (13mL) was stirred at room temperature for 16h
δ: 7.30–7.20 (5H, m), 6.45 (1H, d, J=7.2Hz), 6.37 (1H, d, under an argon atmosphere. The suspension was concentrated
J=6.6Hz), 5.25 (1H, d, J=8.4Hz), 5.14–5.11 (1H, m), 4.61 in vacuo. The residue was used for the next reaction without
(1H, dt, J=7.8, 7.2Hz), 4.35–4.30 (1H, quint., J=7.8Hz), 4.25 further purification.
(1H, m), 3.07 (2H, m), 2.50 (2H, dd, J=15.0, 8.4Hz), 1.88–1.81
To a solution of the above residue in dry CH2Cl2 (210mL)
(1H, m), 1.68 (1H, m), 1.58 (1H, m), 1.46 (9H, s), 1.44 (9H, was added DIEA (142µL, 0.816mmol, 4.00 equiv.) and the
s), 1.32–1.18 (9H, m), 1.08 (3H, s), 0.87 (6H, m). 13C-NMR mixture was stirred at 0°C at 20min under an argon at-
(150MHz, CDCl3) δ: 171.7, 171.5, 170.1, 169.7, 155.4, 136.6, mosphere. To the reaction mixture was added EDCI·HCl
129.3, 128.7, 127.0, 82.0, 80.2, 75.7, 54.6, 50.5, 48.8, 38.9, 38.1, (117.3mg, 0.612mmol, 3.00 equiv.) at 0°C and the mixture
37.5, 36.5, 32.3, 29.2, 28.4, 28.0, 23.9, 22.8, 19.7, 18.3, 14.3, was stirred at the same temperature for 8h and concentrated
14.0. IR (neat) cm−1: 3287, 2977, 2931, 2859, 1740, 1719, 1644, in vacuo. The residue was diluted with organic solvent (ethyl
1551, 1455, 1392, 1368, 1248, 1222, 1163. HR-ESI-MS m/z: acetate–acetonitrile=4:1), washed with 1M aqueous HCl,
696.3921 (Calcd for C35H55N3O8Na [M+Na]+: 696.3943). [α]2D6 saturated aqueous NaHCO3, brine, dried over MgSO4, filtered
−14.2 (c=1.05, CHCl3).
and concentrated in vacuo. The residue was chromatographed
Depsipeptide 20c (n=2, R=CF3) 20c was prepared on silica gel (CHCl3–MeOH=9:1) and further purified by
from ester unit 19c (300mg, 0.591mmol) and dipeptide reversed-phase HPLC (column, YMC-Pack R&D ODS-A
5 (302mg, 0.71mmol) in the same manner to the synthe- 20mm×150mm; solvent, H2O–MeOH=25:75 to 10:90 lin-
1
sis of 20a. Yield 63% from 19c (a colorless oil). H-NMR ear gradient (0.0−15.0 min), H2O–MeOH=10:90 isocratic
(400MHz, CDCl3, mixture of rotamers) δ: 7.31–7.19 (5H, m), (15.0−20.0min); flow rate, 12.0mL/min) to afford beauveri-
6.36 (1H, d, J=7.2Hz), 6.28 (1H, d, J=7.2Hz), 5.22 (1H, d, olide analog 1a (46.1mg, 89.7µmol, 44% in 2 steps) as white
1
J=8.4Hz), 5.17–5.13 (1H, dt, J=8.0, 4.4Hz), 4.65–4.58 (1H, solids. H-NMR (600MHz, CDCl3–CD3OD=1:1) δ: 7.30–7.18
m), 4.36–4.29 (1H, quint, J=7.2Hz), 4.16 (1H, m), 3.10 (1H, (5H, m), 4.97 (1H, ddd, J=19.8, 10.8, 4.8Hz), 4.47 (1H, t,
dd, J=13.8, 6.8Hz), 3.04 (1H, dd, J=13.8, 7.2Hz), 2.50–2.37 J=7.8Hz), 4.27 (1H, m), 3.88 (1H, q, J=7.2Hz), 3.10 (1H, m),
(2H, m), 1.86–1.67 (4H, m), 1.44 (18H, s), 1.33–1.25 (10H, 2.99 (1H, m), 2.50 (2H, m), 2.11 (1H, m), 1.63–1.27 (13H, m),
m), 0.88 (6H, m). 13C-NMR (150MHz, CDCl3, mixture of 1.01 (3H, s), 0.93–0.89 (6H, m). 13C-NMR (150MHz, CDCl3–
rotamers) δ: 171.54 (major) and 171.45 (minor), 171.31 (major) CD3OD=1:1) δ: 171.6, 171.4, 171.3, 168.7, 136.1, 128.7, 128.2,
and 171.26 (minor), 170.2, 169.31 (major) and 169.27 (minor), 126.6, 76.3, 56.4, 53.2, 49.2, 36.1, 35.4, 35.3, 30.6, 30.3, 29.3,
155.4, 136.47 (minor) and 136.42 (major), 129.2, 128.5, 126.9, 28.9, 26.4, 22.5, 19.0, 15.0, 14.4, 13.4. IR (neat) cm−1: 3380,
122.2 (q, J=275Hz), 81.97 (major) and 81.96 (minor), 80.1, 3294, 2958, 2929, 2857, 1724, 1683, 1640, 1537, 1450, 1371,
75.9, 75.41 and 75.35 (1:1), 54.5, 52.7 and 52.6 (1:1), 48.79 1254, 1149, 1006. HR-ESI-MS m/z: Found 536.2837 (Calcd
and 48.75 (1:1), 38.3, 36.4 and 35.9 (1:1), 32.2 (major) and for C27H39N5O5Na [M+Na]+: 536.2843). [α]D24 −41 (c=0.25,
32.1 (minor), 29.2 (major) and 29.1 (minor), 28.2, 27.8, 27.2 (q, CHCl3–MeOH=1:1).
J=40.6Hz), 25.9 (major) and 25.8 (minor), 22.7, 22.3 (minor)
Beauveriolide Analogue 1b (n=1, R=Me) 1b was pre-
and 22.2 (major), 18.4 (minor) and 18.2 (major), 14.40 (minor) pared from depsipeptide 20b (26mg, 38.6µmol) in the same
and 14.3 (major), 13.9. IR (neat) cm−1: 3291, 2978, 2932, 2875, manner to the synthesis of 1a. Yield 41% in 2 steps (white
1
1740, 1719, 1645, 1551, 1456, 1368, 1155. HR-ESI-MS m/z: solids). H-NMR (600MHz, CDCl3–CD3OD=1:1) δ: 7.24 (5H,
764.3802 (Calcd for C36H54F3N5O8Na [M+Na]+: 764.3817). m), 4.98–4.95 (1H, m), 4.70 (1H, dd, J=8.4, 7.2Hz), 4.25 (1H,
[α]D23 −23.0 (c=0.50, CHCl3).
t, J=8.4Hz), 3.89 (1H, m), 3.08 (1H, dd, J=13.2, 8.4Hz), 2.97
Depsipeptide 20d (n=1, R=CF3) 20d was prepared from (1H, dd, J=13.2, 8.4Hz), 2.51 (1H, m), 2.44 (1H, dd, J=13.9,
ester unit 19d (26.1mg, 52.7µmol) and dipeptide 5 (27mg, 9.5Hz), 2.11 (1H, m), 1.73 (1H, dd, J=14.8, 7.2Hz) 1.63 (1H,
63.2 µmol) in the same manner to the synthesis of 20a. dd, J=14.8, 8.4Hz), 1.21 (6H, m), 1.08 (3H, s), 0.89 (6H, m).
1
Yield 58% from 19d (a colorless oil). H-NMR (400MHz, 13C-NMR (150MHz, CDCl3–CD3OD=1:1) δ: 171.8, 171.3,
CDCl3, mixture of rotamers) δ: 7.31–7.19 (5H, m), 6.36 (1H, 171.1, 168.3, 136.1, 128.8, 128.3, 126.7, 56.4, 50.1, 49.5, 37.2,
d, J=7.8Hz), 6.28 (1H, d, J=7.3Hz), 5.22 (1H, d, J=7.8Hz), 35.9, 35.5, 35.5, 30.6, 29.4, 28.9, 23.2, 22.6, 18.8, 15.1, 14.5,
5.15 (1H, dt, J=8.3, 4.9Hz), 4.62 (1H, m), 4.33 (1H, quint, 13.6. IR (neat) cm−1: 3380, 3298, 3063, 2959, 2930, 2858, 1726,

764
Chem. Pharm. Bull.
Vol. 64, No. 7 (2016)
1682, 1640, 1536, 1375, 1330, 1246, 1002. HR-ESI-MS m/z: CHCl3–MeOH=1:1).
522.2679 (Calcd for C26H37N5O5Na [M+Na]+: 522.2687). [α]D24
−25.0 (c=0.150, CHCl3–MeOH=1:1).
Photolysis 17a and c in MeOH (1mg/mL) cooled in an
ice bath were irradiated with black light using Toshiba FL-
Beauveriolide Analogue 1c (n=2, R=CF3) To a solu- 6BL-B black light fluorescent lamp (peak wavelength 352nm;
tion of depsipeptide 20c (20mg, 27.0µmol, 1.00 equiv.) in 4M output 6W; UV output 0.6W) from a distance of 5cm. Ten
HCl/dioxane (2mL) was stirred at room temperature for 16h microliter of the resultant solution was injected to a LC-MS
under an argon atmosphere. The suspension was concentrated system. The analytical conditions are as follows: column, X
in vacuo. The residue was used for the next reaction without Bridge™ C18 (3.5µm, 4.6×75mm); gradient method, 10−95%
further purification.
of B (0.00–4.00min), 95% of B (4.00–11.0min), 95–10% of B
To a solution of the above residue in dry CH2Cl2 (27mL) (11.0–11.1min), 10% of B (11.1–15.0min) (A: 0.1% HCOOH–
was added DIEA (28.2µL, 162.0µmol, 6.00 equiv.) and the H2O, B: 0.1% HCOOH–MeOH); flow rate, 1.1mL/min; UV
mixture was stirred at 0°C at 20min under an argon atmo- 214 nm.
sphere. To the reaction mixture was added HATU (30.8mg,
Culture of SOAT1- and SOAT2-CHO Cells SOAT1- and
81.0 µmol, 3.00 equiv.) at 0°C and the mixture was stirred SOAT2-CHO cells expressing African Green monkey SOAT1
at the same temperature for 5h and concentrated in vacuo. and SOAT2,10) respectively, which were kind gifts from Dr.
The residue was diluted with organic solvent (ethyl acetate– Rudel L. L. (Wake Forest University, U.S.A.), were maintained
acetonitrile=4:1), washed with 1M aqueous HCl, saturated at 37°C in 5% CO2 in Ham’s F-12 medium supplemented with
aqueous NaHCO3, brine, dried over MgSO4, filtered and MEM vitamins, geneticin (300µg/mL) and 10% heat inacti-
concentrated in vacuo. The residue was chromatographed vated FBS (hereafter referred to as medium A).
on silica gel (CHCl3–MeOH=9:1) and further purified by
Cell-Based Assay for SOAT Inhibitory Activity
reversed-phase HPLC (column, YMC-Pack R&D ODS-A SOAT1- or SOAT2-CHO cells (1.25×105 cells in 250µL of
20mm×150mm; solvent, H2O–MeOH=25:75 to 10:90 lin- medium A) were cultured in a 48-well plastic microplate and
ear gradient (0.0−15.0 min), H2O–MeOH=10:90 isocratic allowed to recover overnight at 37°C in 5% CO2. The assays
(15.0−20.0min); flow rate, 12.0mL/min) to afford beauveri- were performed using cells at least 80% confluent. Following
olide analog 1c (5.19mg, 9.10µmol, 34% in 2 steps) as white overnight recovery, 2.5µL of a sample (methanol solution)
solids. 1H-NMR (600MHz, CDCl3–CD3OD=1:1, mixture and 5µL of [1-14C]oleic acid (1nmol, 1.85 KBq, 10% ethanol/
of rotamers) δ: 7.24 (5H, m), 4.94 (1H, m), 4.48 (1H, m), phosphate buffered saline (PBS) solution) were added to each
4.30 (1H, t, J=7.2Hz), 3.80 (1H, m), 3.05 (1H, m), 2.95 (1H, culture at 37°C in 5% CO2. After a 6h incubation, the me-
m), 2.59 (1H, m), 2.45 (1H, m), 2.06 (1H, m), 1.81–1.75 (2H, dium was removed, and the cells in each well were washed
m), 1.61 (1H, m), 1.53–1.28 (7H, m), 1.22 (3H, d, J=6.2Hz), twice with PBS. The cells were lysed by adding 0.25mL of
0.94–0.90 (6H, m). 13C-NMR (600MHz, CDCl3–CD3OD=1:1, 10 mM Tris–HCl (pH 7.5) containing 0.1% (w/v) SDS, cellu-
mixture of rotamers) δ: 173.3, 173.3, 172.7, 169.4, 137.5, lar lipids were extracted by the method of Bligh and Dyer.49)
130.1 and 129.9 (1:1), 129.4 (minor) and 129.2 (major), 127.8 After the organic phase had been concentrated, the total lipids
and 127.6 (1:1), 123.4 (q, J=273Hz), 78.1 (minor) and 77.9 were separated on a thin layer chromatography (TLC) plate
(major), 57.5 and 57.5 (1:1), 54.1 (major) and 54.0 (minor), (silica gel F254, 0.5mm thick, Merck, Germany) and the ra-
50.6 (minor) and 50.5 (major), 37.1, 36.8, 36.6, 31.8, 30.5 dioactivities of [14C]CE, [14C]TG, and [14C]PL were analyzed
(minor) and 30.3 (major), 28.0 (q, J=40Hz), 26.6, 23.7, 23.3, by a bioimaging analyzer (FLA-7000, FUJIFILM, Japan). In
15.9, 15.3 (minor) and 15.2 (major), 14.3. IR (neat) cm−1: 3296, this cell-based assay, [14C]CE was produced by the reaction
2919, 2850, 1727, 1683, 1640, 1537, 1454, 1420, 1375, 1322, of SOAT1 or SOAT2. SOAT inhibitory activity (%) is defined
1273, 1236, 1157, 1093. HR-ESI-MS m/z: 590.2555 (Calcd for as (1-[14C]CE-drug/[14C]CE-control)×100. The IC50 value is
C27H36F3N5O5Na [M+Na]+: 590.2561). [α]D24 −48.4 (c=0.100, defined as the drug concentration causing 50% inhibition of a
CHCl3–MeOH=1:1).
biological activity.
Preparation of Microsomes from SOAT1- or SOAT2-
Beauveriolide Analogue 1d (n=1, R=CF3) 1d was pre-
pared from depsipeptide 20d (20mg, 27.5µmol in the same CHO Cells SOAT1- or SOAT2-CHO cells (2×108 cells)
manner to the synthesis of 1c. Yield 37% in 2 steps (white were homogenized in 10mL of cold buffered sucrose solu-
1
solids). H-NMR (600MHz, CDCl3–CD3OD=1:1, mixture of tion (pH 7.2) containing 100mM sucrose, 50mM KCl, 40mM
rotamers) δ: 7.20–7.10 (5H, m), 4.77 (1H, m), 4.39–4.34 (1H, KH2PO4, 30mM ethylenediaminetetraacetic acid (EDTA) and
m), 4.22–4.16 (1H, m), 3.97 (1H, m), 2.94 (1H, m), 2.86–2.80 complete protease inhibitor cocktail (Roche, U.S.A.) (hereafter
(1H, m), 2.48 (1H, m), 2.33 (1H, m), 1.37 (2H, m), 1.23–1.17 referred to as buffer A) in a Teflon homogenizer. The micro-
(7H, m), 1.15 (3H, d, J=6.8Hz), 0.85–0.75 (6H, m). 13C-NMR somal fraction was pelleted by centrifugation at 100000×g for
(150MHz, CDCl3–CD3OD=1:1, mixture of rotamers) δ: 1h at 4°C, resuspended in the same buffer at a concentration
174.4, 173.3 (major) and 173.0 (minor), 172.6 (major) and 172.5 of 5mg protein/mL and stored at –80°C until use.
(minor), 167.9 (minor) and 167.8 (major), 137.3 (major) and
Enzyme Assay for SOAT Inhibitory Activity SOAT1
137.1 (minor), 129.8 (minor) and 129.7 (major), 129.2, 127.6, and SOAT2 activities were determined by using microsomes
122.9 (J=271Hz), 57.3 (minor) and 56.1 (major), 52.1(minor) prepared as described above as the enzyme source. Briefly, an
and 50.6 (major), 37.7 37.3, 37.0, 31.9 (major) and 31.8 (minor), assay mixture containing 2.5mg/mL bovine serum albumin
30.3, 30.0 (J=40Hz), 29.7, 28.7 (minor) and 28.6 (major), 23.5, (BSA) in buffer A and [1-14C]oleoyl-CoA (20µM, 3.7kBq)
16.2, 15.6, 15.1, 14.2. IR (neat) cm−1: 3308, 2958, 2925, 2854, together with a test sample (added as a 10µL methanol solu-
1730, 1690, 1675, 1641, 1534, 1454, 1378, 1319, 1288, 1259, tion), and the SOAT1 or SOAT2 microsomal fraction (150
1153, 1092, 1029, 803. HR-ESI-MS m/z: 576.2390 (Calcd for or 10µg of protein, respectively) in a total volume of 200µL
C26H34F3N5O5Na [M+Na]+: 576.2404). [α]D23 −11 (c=0.080, were incubated at 37°C for 5min. The reaction was started by

Vol. 64, No. 7 (2016)
Chem. Pharm. Bull.
765
adding [1-14C]oleoyl-CoA, and stopped by adding 1.2mL of
CHCl3–MeOH=2:1. The produced [14C]CE was extracted by
the method of Bligh and Dyer.49) After the organic solvent was
removed by evaporation, lipids was separated on a TLC plate
and the radioactivity of [14C]CE was measured as described
above.
18) Nagai K., Doi T., Ohshiro T., Sunazuka T., Tomoda H., Takahashi
T., Ōmura S., Bioorg. Med. Chem. Lett., 18, 4397–4400 (2008).
19) Ohshiro T., Matsuda D., Nagai K., Doi T., Sunazuka T., Takahashi
T., Rudel L. L., Ōmura S., Tomoda H., Chem. Pharm. Bull., 57,
377–381 (2009).
20) Tomoda H., Doi T., Acc. Chem. Res., 41, 32–39 (2008).
21) Doi T., Muraoka T., Ohshiro T., Matsuda D., Yoshida M., Takahashi
T., Ōmura S., Tomoda H., Bioorg. Med. Chem. Lett., 22, 696–699
(2012).
22) Sadakane Y., Hatanaka Y., Anal. Sci., 22, 209–218 (2006).
23) Schwyzer R., Caviezel M., Helv. Chim. Acta, 54, 1395–1400 (1971).
24) Kauer J. C., Erickson-Viitanen S., Wolfe H. R. Jr., DeGrado W. F.,
J. Biol. Chem., 261, 10695–10700 (1986).
Acknowledgment This study was supported by Grant-in-
Aid for Scientific Research (B) (No. 26282208) from the Min-
istry of Education, Culture, Sports, Science and Technology
(MEXT) of Japan.
25) Nassal M., J. Am. Chem. Soc., 106, 7540–7545 (1984).
26) Shih L. B., Bayley H., Anal. Biochem., 144, 132–141 (1985).
27) Suchanek M., Radzikowska A., Thiele C., Nat. Methods, 2, 261–268
(2005).
28) Srinivas N., Jetter P., Ueberbacher B. J., Werneburg M., Zerbe K.,
Steinmann J., Van der Meijden B., Bernardini F., Lederer A., Dias
R. L. A., Misson P. E., Henze H., Zumbrunn J., Gombert F. O.,
Obrecht D., Hunziker P., Schauer S., Ziegler U., Kach A., Eberl L.,
Riedel K., DeMarco S. J., Robinson J. A., Science, 327, 1010–1013
(2010).
29) Van der Meijden B., Robinson J. A., ARKIVOC, 6, 130–136 (2011).
30) Wartmann T., Lindel T., Eur. J. Org. Chem., 2013, 1649–1652
(2013).
31) Dubinsky L., Krom B. P., Meijler M. M., Bioorg. Med. Chem., 20,
554–570 (2012).
Conflict of Interest The authors declare no conflict of
interest.
Supplementary Materials The online version of this ar-
ticle contains supplementary materials. They include Fig. S1,
Charts S1–S3, supplementary experimental section, and copies
1
of 1-D H- and 13C-NMR spectra.
References
1) Chang T. Y., Li B. L., Chang C. C., Urano Y., Am. J. Physiol. Endo-
crinol. Metab., 297, E1–E9 (2009).
2) Sliskovic D. R., White A. D., Trends Pharmacol. Sci., 12, 194–199
(1991).
3) Tomoda H., Ōmura S., Pharmacol. Ther., 115, 375–389 (2007).
4) Chang C. C., Huh H. Y., Cadigan K. M., Chang T. Y., J. Biol.
Chem., 268, 20747–20755 (1993).
32) Vila-Perelló M., Pratt M. R., Tulin F., Muir T. W., J. Am. Chem.
Soc., 129, 8068–8069 (2007).
33) MacKinnon A. L., Garrison J. L., Hegde R. S., Taunton J., J. Am.
Chem. Soc., 129, 14560–14561 (2007).
34) Kölbel K., Ihling C. H., Sinz A., Angew. Chem. Int. Ed., 51, 12602–
12605 (2012).
35) Brunner J., Senn H., Richards F. M., J. Biol. Chem., 255, 3313–3318
(1980).
36) Erni B., Khorana H. G., J. Am. Chem. Soc., 102, 3888–3896 (1980).
37) Zhang Y., Burdzinski G., Kubicki J., Platz M. S., J. Am. Chem. Soc.,
130, 16134–16135 (2008).
38) Ford F., Yuzawa T., Platz M. S., Matzinger S., Fülscher M., J. Am.
Chem. Soc., 120, 4430–4438 (1998).
39) Toscano J. P., Platz M. S., Nikolaev V., J. Am. Chem. Soc., 117,
4712–4713 (1995).
40) Sulzbach H. M., Platz M. S., Schaefer H. F., Hadad C. M., J. Am.
Chem. Soc., 119, 5682–5689 (1997).
41) Das J., Chem. Rev., 111, 4405–4417 (2011).
5) Anderson R. A., Joyce C., Davis M., Reagan J. W., Clark M.,
Shelness G. S., Rudel L. L., J. Biol. Chem., 273, 26747–26754
(1998).
6) 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., 273, 26755–26764 (1998).
7) Oelkers P., Behari A., Cromley D., Billheimer J. T., Sturley S. L., J.
Biol. Chem., 273, 26765–26771 (1998).
8) Rudel L. L., Lee R. G., Cockman T. L., Curr. Opin. Lipidol., 12,
121–127 (2001).
9) Das A., Davis M. A., Rudel L. L., J. Lipid Res., 49, 1770–1781
(2008).
10) Ohshiro T., Rudel L. L., Ōmura S., Tomoda H., J. Antibiot. (Tokyo),
60, 43–51 (2007).
11) Ohshiro T., Tomoda H., Future Med. Chem., 3, 2039–2061 (2011).
12) Ohshiro T., Tomoda H., Expert Opin. Ther. Pat., 25, 145–158
(2015).
42) Admasu A., Gudmundsdottir A. D., Platz M. S., Watt D. S., Kwiat-
kowski S., Crocker P. J., J. Chem. Soc., Perkin Trans., 2, 1093–1100
(1998).
13) Namatame I., Tomoda H., Si S., Yamaguchi Y., Masuma R., Ōmura
S., J. Antibiot. (Tokyo), 52, 1–6 (1999).
43) Yip G. M. S., Chen Z.-W., Edge C. J., Smith E. H., Dickinson R.,
Hohenester E., Townsend R. R., Fuchs K., Sieghart W., Evers A. S.,
Franks N. P., Nat. Chem. Biol., 9, 715–720 (2013).
44) Prakash G. K. S., Yudin A. K., Chem. Rev., 97, 757–786 (1997).
45) Linderman R. J., Graves D. M., J. Org. Chem., 54, 661–668 (1989).
46) Roberts J. L., Chan C., Tetrahedron Lett., 43, 7679–7682 (2002).
47) Church R. F. R., Weiss M. J., J. Org. Chem., 35, 2465–2471 (1970).
48) Carpino L. A., J. Am. Chem. Soc., 115, 4397–4398 (1993).
49) Bligh E. G., Dyer W. J., Can. J. Biochem. Physiol., 37, 911–917
(1959).
14) Namatame I., Tomoda H., Tabata N., Si S., Ōmura S., J. Antibiot.
(Tokyo), 52, 7–12 (1999).
15) Namatame I., Tomoda H., Ishibashi S., Ōmura S., Proc. Natl. Acad.
Sci. U.S.A., 101, 737–742 (2004).
16) Ohshiro T., Namatame I., Nagai K., Sekiguchi T., Doi T., Takahashi
T., Akasaka K., Rudel L. L., Tomoda H., Ōmura S., J. Org. Chem.,
71, 7643–7649 (2006).
17) Nagai K., Doi T., Sekiguchi T., Namatame I., Sunazuka T., Tomoda
H., Ōmura S., Takahashi T., J. Comb. Chem., 8, 103–109 (2006).