Design and synthesis of simple, yet potent and selective non-ring-A pyripyropene A-based inhibitors of acyl-coenzyme A: Cholesterol acyltransferase 2 (ACAT2)

Article abstract of DOI:10.1039/c5ob02019k

A series of pyripyropene A-based compounds were designed and synthesized by opening the upper section of the A-ring, which significantly simplifies the structure and synthesis from commercially available starting materials. Representative compound (-)-3 exhibited potent activity against ACAT2 and greater selectivity for ACAT2 than for ACAT1.

Full text of DOI:10.1039/c5ob02019k

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Design and synthesis of simple, yet potent and
selective non-ring-A pyripyropene A-based
inhibitors of acyl-coenzyme A: cholesterol
acyltransferase 2 (ACAT2)†
Cite this: Org. Biomol. Chem., 2016,
14, 747
Yang Zhan,‡a Xiao-Wei Zhang,‡b Ying Xiong,b Bo-Liang Li*b and Fa-Jun Nan*a
Received 28th September 2015,
Accepted 9th November 2015
A series of pyripyropene A-based compounds were designed and synthesized by opening the upper
section of the A-ring, which significantly simplifies the structure and synthesis from commercially available
starting materials. Representative compound (−)-3 exhibited potent activity against ACAT2 and greater
selectivity for ACAT2 than for ACAT1.
DOI: 10.1039/c5ob02019k
www.rsc.org/obc
ACAT2 and esterification of excess oxysterols for secretion to
avoid cytotoxicity, is established in a subset of HCCs for tumor
Introduction
Acyl-coenzyme A: cholesterol acyltransferase (ACAT) is the growth. ACAT2 is also an attractive target for the treatment of
exclusive intracellular enzyme that catalyzes the formation of HCCs, and selective ACAT2 inhibitors will be potential
cholesteryl esters from free cholesterol and long-chain fatty anti-HCC candidates with novel modes of action compared
acyl-CoA. The ACAT family includes two members, ACAT1 and with current chemotherapeutics.
ACAT2. ACAT1 contributes significantly to macrophage-derived
To date, natural product pyripyropene A (PPPA) and its
foam cell formation in atherosclerosis. ACAT1 is expressed analogues are the only chemical type ACAT2-specific
ubiquitously in almost all human tissues, and the steryl esters inhibitors.9 PPPAs were isolated from the fungal strain
catalyzed by ACAT1 incorporate into cellular lipid droplets A. fumigatus FO-1289710 and characterized by Omura et al. in
exclusively. Whereas, ACAT2 is responsible for absorption of detailed structure–activity relationship (SAR) studies.11–16
dietary cholesterol from the intestine and lipoprotein PPPAs exhibited efficacy in an atherosclerosis animal model.17
production by the liver, the steryl esters catalyzed by ACAT2 However, the acquisition of PPPA from either natural sources10
can incorporate into both intracellular lipid droplets and or chemical synthesis18 is difficult.
extracellular lipoproteins.1–3 Therefore, ACAT is regarded as a
According to the total synthesis of PPPA reported18,19 by
potential therapeutic target for cardiovascular diseases.4–6 Atsuki et al., eleven steps are required to construct the A-ring,
However, synthetic ACAT inhibitors failed to show efficacy or and these steps account for the majority of the synthesis pro-
cause severe side effects in clinical trials. In 2006, Fazio et al. cedure. Satoshi et al. reported structure–activity relationships
observed7 that inhibition of ACAT1 activity led to cytotoxicity (SARs) of PPPA modifications through simple replacement of
toward macrophages and vascular cells, whereas ACAT2 small substituents.11–16 Their results suggested that the
inhibition decreased the synthesis of cholesteryl esters and pyridine pyrone moiety, free hydroxyl and three acetyl
provided therapeutic benefits. More recently, Lu et al. segments are essential for potent activity. However, A-ring,
reported8 that by investigating tissue samples from hepato- which is the most difficult portion of the molecule to
cellular carcinoma (HCC) patients and HCC cell lines, a synthesize, does not appear to be essential for maintaining
specific cholesterol metabolic pathway, involving induction of ACAT2 activity. We analyzed these key pharmacophores in
complicated natural products and analogues and rationally
designed some simplified PPPA-based compounds by opening
the upper portion of ring A (red part in Fig. 1) and maintaining
aState Key Laboratory of Drug Research, The National Center for Drug Screening,
Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai,
the remaining acetyl esters. As shown in Fig. 1, to prove our
initial concept, we selected a method to prepare a racemic
mixture (1) as the first target molecule.
The synthesis of 1 is summarized in Scheme 1. The hydro-
xycarvone ( )-6,20 which was prepared from (R)-carvone, was
sulfonated with triflic anhydride to produce enol triflate ( )-7
201203, China. E-mail: fjnan@simm.ac.cn
bState Key Laboratory of Molecular Biology, Institute of Biochemistry and Cell
Biology, Shanghai Institutes for Bio-logical Sciences, Chinese Academy of Sciences,
Shanghai, 200031, China. E-mail: blli@sibs.ac.cn
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5ob02019k
‡These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 747–751 | 747

View Article Online
Paper
Organic & Biomolecular Chemistry
NBD22-steryl ester fluorescence assay indicated that com-
pound 1 exhibited activity comparable to PPPA, with an IC50
value of 0.155 μM, whereas the IC50 of PPPA was 0.185 μM
under the same assay conditions. We also test the selectivity
index (SI) by using the cholesterol oxidase assay with an
Amplex red cholesterol assay kit, compound 1 exhibited high
selectivity to ACAT2 isoform, with a selectivity index (SI) of 733
(ACAT1-IC50: 107.7 μM; ACAT2-IC50: 0.147 μM), which was also
comparable to that of PPPA (Table 1).
Fig. 1 Rational design of simplified pyripyropene A-based compounds
via opening the upper portion of the A-ring.
These results encouraged us to perform detailed SAR
studies for these simplified PPPA-based ACAT2 inhibitors.
Because racemic 1 was obtained as a mixture of four isomers,
acquisition of the single enantiomer was different from the
procedure for the racemic mixture (1). We changed several
steps to obtain chiral aldehyde 11 from natural (R)-carvone or
(S)-carvone, followed by a similar procedure to afford the
single isomers.
The synthesis of (+)-2 and (+)-3 is outlined in Scheme 2.
The known allylic alcohol (+)-19,24 which was prepared from
commercially available (R)-carvone, was protected as a TBS
ether to form product (+)-20. The nitrile (+)-20 was reduced
with DIBAL-H to give aldehyde (+)-11 in 84% yield.25 The sub-
sequent five steps were the same as those used in the synthetic
method shown in Scheme 1. Prevost dihydroxylation of (+)-16
resulted in (+)-17A and (+)-17B as a 1 : 1 mixture of diastereo-
mers at the newly formed stereocenter C10, and these diaster-
eomers were isolated and identified by 2d-NMR (Fig. 2). As
shown in Fig. 2, on the basis of the evidence of the methyl
NOESY cross-correlation with both methylene (H4 and H8),
and the configural-superiority effect, we obtained two stable
Newman forms. The NOE between methylene (H11) and
methylene (H4 or H8) protons was consistent with (S)-(+)-17A
followed by carbonylation using a procedure described by
Stille21 to produce the methyl ester ( )-8. Stereoselective Luche
reduction22 provided ( )-9 quantitatively, which was protected
as a TBS ether to form product ( )-10. Coupling reaction of the
5-iodo-1,3-dioxin-4-one derivative 12 with ( )-11, which was
generated from reduction with DIBALH of the ester ( )-10 fol-
lowed by oxidation with DMP, afforded intermediate ( )-13
under conditions reported by Knochel et al.23 Subsequent
Dess–Martin oxidation formed ( )-14 in 45% yield over two
steps. Deprotection of ( )-14 followed by cyclization was rea-
lized under methanolysis conditions, producing ( )-15. Enoli-
zation of ( )-15 by LHMDS, γ-H acylation with nicotinoyl
chloride, and treatment with an additional portion of LHMDS
by intramolecular cyclization formed ( )-16 from ( )-15 in one
pot. Prevost dihydroxylation of ( )-16 furnished ( )-17 as a
1 : 1 mixture of diastereomers at the newly formed stereocenter
C10. Protecting group manipulations of ( )-17 by two steps
resulted in ( )-18, which was subjected to Luche reduction to
give 1 as the final product. Synthetic racemic mixture 1 was
obtained as a mixture of four isomers. ACAT2 inhibition in the
Scheme 1 Preparation of simplified pyripyropene A-like racemic mixture 1.
748 | Org. Biomol. Chem., 2016, 14, 747–751
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 1 Effect of inhibitors on cholesteryl ester (CE) synthesis and selectivity of inhibitors toward ACAT isoforms
IC50 SD (μM)
Selectivity index
(IC50 for ACAT1b/IC50
for ACAT2b)
Compound
ACAT2a
ACAT2b
ACAT1b
Pyripyropene A
1
(+)-2
(+)-3
(+)-4
0.185 0.007
0.155 0.010
0.743 0.098
1.272 0.149
1.789 0.084
3.059 0.222
1.700 0.062
0.077 0.006
2.456 0.161
0.416 0.052
0.198 0.016
0.147 0.018
—
—
—
>20c
107.7
—
—
—
—
—
56.66
—
—
>200c
733
—
—
—
—
—
1035
—
—
d
(+)-5
—
—
(−)-2
(−)-3
(−)-4
(−)-5
0.055 0.007
—
—
a By using the NBD22-steryl ester fluorescence assay. b By using the cholesterol oxidase assay. IC50 values were means SD of three replications.
c From Ref. 29. d Not test.
Fig. 2 Identification of the configuration at stereocenter C10
.
and H13 (Scheme 2: NOESY). The synthetic method of (−)-2
and (−)-3 is similar to the method shown in Scheme 2. The
known allylic alcohol (−)-1924 was prepared from commercially
available (S)-carvone.
Four single isomers, (+)-2, (+)-3, (−)-2 and (−)-3, were tested
for ACAT2 inhibition. Surprisingly, compounds (+)-2 and (+)-3,
which have the same core-structure stereochemistry as the
natural product PPPA, showed weak inhibitory activity of
ACAT2, with IC50 values of 0.743 and 1.272 μM, respectively.
However, (−)-3 was the most potent ACAT2 inhibitor among all
of the investigated isomers, with an IC50 value of 0.077 μM in
NBD22-steryl ester fluorescence assay and a much higher SI
value of 1035 (ACAT1-IC50: 56.66 μM; ACAT2-IC50: 0.055 μM, by
assaying steryl esters) (Table 1). Isomer (−)-2 also exhibited
weak inhibitory activity of ACAT2, with an IC50 value of
1.700 μM. The latter two isomers actually exhibit opposite
stereochemistry in the B, C ring compared with the natural
product PPPA.
Scheme 2 Preparation of one single enantiomer (+)-2, (+)-3 and (−)-2,
(−)-3.
and (R)-(+)-17B at stereocenter C10. The following three steps
imitated the route of the racemic mixture to access (+)-2 and
(+)-3 as the final product. The relative configuration of the
stereocenters was determined by irradiation of the methyl-H14
Scheme 3 Preparation of (+)-4 and (+)-5.
Org. Biomol. Chem., 2016, 14, 747–751 | 749
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 The structure of eight isomers: (+)/(−)-2–5.
By comparing the stereochemistry of four single isomers with pared with PPPA, which suggests that the SAR profiles of these
PPPA, we observed that the relative stereochemistry at the C13 simplified analogues differ from the profile of the natural
position of the most potent compound, (−)-3, also differed from product. These non-ring-A analogues are novel compounds
that at the same position of the natural product. These results that can be easily assembled from naturally abundant carvone
suggested that, in non-A-ring analogs, an exact structure–activity in several steps and in high yield. More detailed SARs are
relationship should be further assigned. Therefore, we required to generate more drug-like and selective ACAT2
synthesized (+)/(−)-4 and (+)/(−)-5 as four corresponding C13 inhibitors with potential therapeutic use in atherosclerosis
epi-isomers of (+)/(−)-2 and (+)/(−)-3, respectively.
and HCC. Further studies assessing the SARs of these com-
The synthesis of (+)-4 and (+)-5 is outlined in Scheme 3. pounds are in progress.
Starting from the intermediate (+)-18A or (+)-18B, respectively,
the final asymmetric hydroxyl was introduced by reduction of
the enone using (S)-Me-CBS-oxazaborolidine,26 which provided
Acknowledgements
allylic alcohol (+)-4 or (+)-5 in 50% yield. The corresponding
enantiomers (−)-4 and (−)-5 were obtained using a similar
procedure starting from (−)-18A and (−)-18B, respectively.
We next determined the ACAT2 inhibition of four isomers
(+)/(−)-4–5 using NBD22-steryl ester fluorescence assay. IC50
results showed that they all have weaker ACAT2 inhibitory
activity compared with PPPA and (−)-3. Surprisingly,
compound (−)-5, which is C13-epi (−)-3, showed 5 fold
decreased activity, indicating that non-ring-A compounds
exhibit different SARs toward ACAT2 (Fig. 3).
This work was supported by grants from the National Science
and Technology Major Projects for Major New Drugs Inno-
vation and Development (2012ZX09304011, 2013ZX09507002),
the National Natural Science Foundation of China (81321092,
31470802), and the Ministry of Science and Technology of
China (no. 2011CB910900).
Notes and references
Recently, Nagamitsu et al. reported27,28 several non-ring-A
PPPA-based analogues, which shared the same core-structure of
the B, C ring of the PPPA, showed weak activity of ACAT2 inhi-
bition and selectivity. Our results will possibly explain why their
non-ring-A compounds are weak or inactive, because all of the
compounds they reported have the same stereochemistry in the
B, C ring with the natural product PPPA.
1 K. K. Buhman, M. Accad and S. Novak, Nat. Med., 2000, 6,
1341–1347.
2 J. J. Repa, K. K. Buhman and R. V. Farese, Hepatology, 2004,
40, 1088–1097.
3 R. G. Lee, R. Shah and J. K. Sawyer, J. Lipid Res., 2005, 46,
1205–1212.
4 A. Miyazaki, K. T. anome and T. Watanabe, Curr. Drug
Targets: Cardiovasc. & Haematol. Disord., 2005, 5, 463–469.
5 D. R. Sliskovic, J. A. Picard and B. R. Krause, Prog. Med.
Chem., 2002, 39, 121–171.
Conclusions
In conclusion, eight new simplified non-ring-A PPPA-based
analogues were designed and synthesized in this article.
Among these analogues, compound (−)-3 exhibited more
potent ACAT2 inhibition and higher selectivity toward ACAT2
compared with the natural product PPPA. Compound (−)-3
also exhibited opposite stereochemistry in the B, C ring com-
6 T. Y. Chang, B. L. Li, C. C. Chang and Y. Urano,
Am. J. Physiol., 2009, 297, E1–E9.
7 S. Fazio and M. Linton, N. Engl. J. Med., 2006, 354, 1307–
1309.
8 M. Lu, X. H. Hu, Q. Li, Y. Xiong, G. J. Hu, J. J. Xu,
X. N. Zhao, X. X. Wei, C. C. Y. Chang, Y. K. Liu, F. J. Nan,
750 | Org. Biomol. Chem., 2016, 14, 747–751
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
J. Li, T. Y. Chang, B. L. Song and B. L. Li, J. Mol. Cell. Biol., 18 A. Odani, K. Ishihara, M. Ohtawa, H. Tomoda, S. Omura
2013, 5, 404–415. and T. Nagamitsu, Tetrahedron, 2011, 67, 8195–8203.
9 T. Ohshiro and H. Tomoda, Future Med. Chem., 2011, 3, 19 T. Nagamitsu, T. Sunazuka, R. Obata, H. Tomoda,
2039–2061.
H. Tanaka, Y. Harigaya, S. Omura and A. B. Smith, J. Org.
Chem., 1995, 60, 8126–8127.
20 J. W. Huffman and G. F. Hillenbrand, Tetrahedron, 1981,
37, 269–274.
21 W. J. Scott, G. T. Crisp and J. K. Stille, J. Am. Chem. Soc.,
1984, 106, 4630–4632.
10 S. Omura, H. Tomoda, Y. K. Kim and H. Nishida, J. Anti-
biot., 1993, 46, 1168–1169.
11 R. Obata, T. Sunazuka, Y. Kato, H. Tomoda, Y. Harigaya
and S. Omura, J. Antibiot., 1996, 49, 1149–1156.
12 R. Obata, T. Sunazuka, Z. R. Li, Z. M. Tian, Y. Harigaya,
N. Tabata, H. Tomoda and S. Omura, J. Antibiot., 1996, 49, 22 J. L. Luche, J. Am. Chem. Soc., 1978, 100, 2226–2227.
1133–1148.
23 V. A. Vu, L. Bérillon and P. Knochel, Tetrahedron Lett., 2001,
42, 6847–6850.
24 J. R. Hudlicky, L. Werner, V. Semak, R. Simionescu and
T. Hudlicky, Can. J. Chem., 2011, 89, 535–543.
13 R. Obata, T. Sunazuka, Z. M. Tian, H. Tomoda, Y. Harigaya
and S. Omura, J. Antibiot., 1997, 50, 229–236.
14 M. Ohtawa, H. Yamazaki, D. Matsuda, T. Ohshiro,
L. L. Rudel, S. Omura, H. Tomoda and T. Nagamitsu, 25 S. Horii, M. Torihata, T. Nagasawa and S. Kuwahara, J. Org.
Bioorg. Med. Chem. Lett., 2013, 23, 2659–2662. Chem., 2013, 78, 2798–2801.
15 M. Ohtawa, H. Yamazaki, S. Ohte, D. Matsuda, T. Ohshiro, 26 E. J. Corey, A. GuzmanPerez and S. E. Lazerwith, J. Am.
L. L. Rudel, S. Omura, H. Tomoda and T. Nagamitsu,
Bioorg. Med. Chem. Lett., 2013, 23, 3798–3801.
16 M. Ohtawa, H. Yamazaki, S. Ohte, D. Matsuda, T. Ohshiro,
Chem. Soc., 1997, 119, 11769–11776.
27 M. Ohtawa, S. Omura, H. Tomoda and T. Nagamitsu, J. Syn.
Org. Chem. Jpn., 2013, 71, 830–843.
L. L. Rudel, S. Omura, H. Tomoda and T. Nagamitsu, 28 M. Ohtawa, S. Omura, H. Tomoda and T. Nagamitsu,
Bioorg. Med. Chem. Lett., 2013, 23, 1285–1287. Japanese Patents, 2014, JP2014-144922A.
17 T. Ohshiro, D. Matsuda, K. Sakai, C. Degirolamo, H. Yagyu, 29 T. L. Aaron, D. Matthew, K. Carol, C. James, T. Hiroshi,
L. L. Rudel, S. Omura, S. Ishibashi and H. Tomoda,
O. Satoshi and L. R. Lawrence, J. Lipid. Res., 2004, 45, 378–
Arterioscler., Thromb., Vasc. Biol., 2011, 31, 1108–1115.
386.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem., 2016, 14, 747–751 | 751