Synthesis of the PMB ether of 5,6-epoxyisoprostane E2 through aldol reaction of the α-bromocyclopentanone

Article abstract of DOI:10.1021/ol500654g

5,6-Epoxyisoprostane E2 was synthesized via bromohydrination of the cyclopentene and aldol reaction of the α-bromocyclopentanone with the epoxyaldehyde. High regioselectivity in the bromohydrination was attained with recrystallized NBS and pyridine in aqu

Full text of DOI:10.1021/ol500654g

Letter
pubs.acs.org/OrgLett
Synthesis of the PMB Ether of 5,6-Epoxyisoprostane E2 through
Aldol Reaction of the α‑Bromocyclopentanone
Hidehisa Kawashima and Yuichi Kobayashi*
Department of Bioengineering, Tokyo Institute of Technology, Box B52, Nagatsuta-cho 4259, Midori-ku, Yokohama 226-8501, Japan
S
* Supporting Information
ABSTRACT: 5,6-Epoxyisoprostane E2 was synthesized via
bromohydrination of the cyclopentene and aldol reaction of
the α-bromocyclopentanone with the epoxyaldehyde. High
regioselectivity in the bromohydrination was attained with
recrystallized NBS and pyridine in aqueous DMSO. The
enolate for the aldol reaction was generated by adding t-BuLi
to the mixture of the α-bromocyclopentanone and ZnI2. This
aldol protocol was applied successfully to several cyclo-
pentanones and aldehydes.
xidation products of 1-palmitoyl-2-arachidonoyl-sn-glyc-
ero-3-phosphocholine (PAPC) have been isolated from
exhibit anti-inflammatory activity by suppressing IL-6 and IL-
12 production.6 Taken together, these results suggest that
further studies of these products as well as their stereo- and
regioisomers are needed to better understand their biological
roles.7
O
rabbit atherosclerotic lesions, mildly oxidized low-density
lipoproteins (ox-LDL), and autoxidized PAPC.1,2 The
oxidation products were separated by LC/MS, and the two
of them were identified as those that possess 5,6-epoxyiso-
prostanes E2 and A2 at the sn-2 position of the phospholipid.
Based on UV analysis, the other compounds were suggested
to be the stereo- and regioisomers of 5,6-epoxyisoprostanes
E2 and A2. The relative stereochemistry of 5,6-epoxyiso-
prostane A2 was assigned as 5 by total synthesis3 (Figure 1),
To support such biological investigations, the development
of an efficient synthesis of these compounds is the subject of
organic synthesis. After synthesizing 5,6-PECPC (4) and its
regioisomer,3,4 we have shifted our focus toward synthesis of
the PMB derivative of 2, i.e., 3, which is esterified to 5,6-
PEIPC (1) by Jung.5b Herein, a new protocol for aldol
reaction of α-bromocyclopentanones with aldehydes was
developed and successfully utilized for the synthesis of 3.
Previously, Stork8 developed a scheme for the conversion of
olefin A to aldol F (Scheme 1), in which the key reactions
were the bromohydrination of olefin A (eq 1) and aldol
reaction of α-bromocyclopentanone D (eq 2). This strategy
was used in the recent synthesis of ophiobolin A, in which the
two substituents are cis oriented (diastereomer of A).9
Regarding the regioselectivity of the bromohydrination, pure
B was obtained by Stork,8 while a ca. 5:1 ratio of B and C
Scheme 1. Highlights of the Present Study
Figure 1. 5,6-PEIPC, -PECPC, and related compounds.
and the Yamaguchi esterification with lyso-PC afforded 5,6-
PECPC (4).3,4 Subsequently, 5,6-PEIPC (1), with the
absolute structure (as depicted in Figure 1), was synthesized
and the occurrence of 1 in the oxidation products was
identified by HPLC and MS analysis,5 thus establishing the
trans stereochemistry between the epoxide part and the 2-
octenyl chain as is seen in 5. Compounds 1 and 4 have been
reported to increase the production of pro-inflammatory
mediatorsinterleukin-8 (IL-8) and monocyte chemotactic
protein-1 (MCP-1).2 In contrast, recent biological studies with
synthesized 1, 2, 4, and 5 showed that these compounds
Received: March 3, 2014
Published: April 24, 2014
© 2014 American Chemical Society
2598
dx.doi.org/10.1021/ol500654g | Org. Lett. 2014, 16, 2598−2601

Organic Letters
Letter
Scheme 2. Synthesis of the 5,6-Epoxyisoprostanes E2 Derivative
was described by Walker,10 who converted B to the
prostaglandin-like heterocycles via D. In the present study,
the motif A occurs in the form of the PMB ether 13 (Scheme
2).
Table 1. Bromohydrination of Olefin 13
We envisaged a Claisen rearrangement on 8 as a means of
installing the requisite side chain shown in olefin 13 (Scheme
2). Toward this, Mitsunobu inversion of the cyclopentene
monoacetate 6 (99% ee) gave alcohol 7, which was converted
to alcohol 8. Eschenmoser−Claisen rearrangement11 on 8
with MeC(OMe)2NMe2 afforded amide 9 in 95% yield.
Reduction of 9 with DIBAL at −105 °C (liquid nitrogen and
MeOH) produced aldehyde 10 in 80% yield, whereas the
yield was only 40% at −78 °C. The Corey−Fuchs reaction of
10 followed by debromination with n-BuLi afforded the Li
anion 12, which was quenched to produce 11. We avoided
the direct use of the anion 12 for alkylation with C5H11I
because of the possible competition with n-BuBr generated in
situ (from n-BuLi). Thus, alkylation of 11 with C5H11I under
the given conditions afforded 13 in 86% yield. Previously,
similar intermediates possessing the alkoxy group and the side
chain at C11 and C12, respectively, have been synthesized
with the 2−4:15b,7 and 9:16 diastereomeric ratios, whereas our
synthesis of 13 is completely stereoselective.
Various aqueous conditions were evaluated for the
bromohydrination of 13 with NBS (Table 1). The product
distributions of the desired bromohydrin 14, the regioisomer
15, 4-MeOC6H4CHO (22), and the substrate 13 were
determined by 1H NMR integration and used to assess
regioselectivity (14 over 15), the extent of decomposition of
the products and/or 13, and conversion. Initial attempts
(entries 1−3) indicated the unstable nature of 13 under these
conditions since bromohydrination of a simple olefin 2312
proceeded smoothly (with ca. 2:1 rations of 24 and 25 in the
all entries) (eq 3). Gratifyingly, further investigation in
a
b
entry purity grade of NBS
additive
solvent
14:15:13:22
c
1
2
3
4
5
6
7
8
purchased
−
−
acetone
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
−
d
purchased
52:<10:14:24
d
purchased
NaHCO3
pyridine
pyridine
−
15:<10:7:68
purchased
74:15:9:2
e
recrystallized
recrystallized
recrystallized
recrystallized
91(81) :5:2:2
67:26:7:0
d
NaHCO3
21:<10:20:49
f
Et3N
0:0:11:89
a
b
c
d
6 equiv. Solvent/H2O (4:1). Unidentified compounds. Unidenti-
e
f
fied compounds were also produced. Isolated yield. A similar result
with i-Pr2NEt.
aqueous DMSO disclosed acceleration of the reaction by
pyridine (entry 4). Furthermore, the use of recrystallized NBS
with pyridine produced 14 with high regioselectivity and
product selectivity in 81% isolated yield (entry 5, 14/15 =
18:1). In contrast, the use of recrystallized NBS alone
(without pyridine) reduced the regioselectivity (entry 6),
while its use in the presence of other bases, such as NaHCO3,
Et3N, and i-Pr2NEt, retarded or prevented the reaction
(entries 7 and 8, footnote f). These results suggest the slightly
acidic conditions realized by HBr (generated in situ) and
pyridine probably assist the nucleophilic attack by DMSO
and/or decomposition of the sulfoxonium ion to bromohydrin
14. The 1H NMR analysis of 14 showed it to be a 4:1
diastereomeric ratio (dr) though this ratio had no meaning for
further transformation. Indeed, this diastereomeric mixture
was subjected to oxidation with PCC to afford ketone 16
2599
dx.doi.org/10.1021/ol500654g | Org. Lett. 2014, 16, 2598−2601

Organic Letters
Letter
1
(93%), which was a 3:2 diastereomeric mixture by H NMR
analysis (δ 4.31 (d) and 4.61 (d) ppm for the major and
minor isomers).
selectivity (28a/29a) (entry 9) and Zn(OTs)2 gave
unidentified products (data not shown). Comparable out-
comes were also obtained in Et2O with t-BuLi alone and t-
BuLi/ZnX2 (X = I, Br) (entries not shown). In contrast, the t-
BuLi−Br exchange of cyclohexanone 26b (n = 6) proceeded
cleanly without ZnI2, and the subsequent aldol reaction
afforded 28b efficiently (entry 12; cf. entry 13).
Among the several successful combinations of t-BuLi and
the zinc salts, t-BuLi/ZnI2 was applied to the real α-
bromoketone 16 because of the comparatively nonhygroscopic
nature and the ready-to-use convenience of commercial ZnI2.
The exchange proceeded well, and the subsequent aldol
reaction with epoxyaldehyde 18 (99% ee) (see Supporting
Information for the preparation) afforded 19 as a diastereo-
meric mixture at C7 (and C8). Dehydration of 19 via
mesylate according to the procedure developed by us3 for the
synthesis of PECPC afforded exoenone 20 as the sole product
in 48% yield over three steps. The acetylene in 20 was
reduced to the cis olefin, and the TBS group was removed to
produce alcohol 21 in 82% yield.17 Finally, a two-step
Among the possible methods for conversion of α-
bromoketones to enolates, organometal−bromine exchange14
was examined for the present investigation, whereas the
radical−bromine exchange15 was not considered because of
the presence of a triple bond which is generally radical-
sensitive. The other possible two-step procedures8 (formation
of an enol derivative followed by regeneration of enolate)
were also excluded from our examination. First, EtMgBr−Br
exchange of a model α-bromocyclopentanone 26a (n = 5)
was attempted according to the previous generation of enolate
from α-iodocyclohexanones with EtMgBr.14,16 However, an
attempted aldol reaction with aldehyde 27 gave a mixture of
unidentified products (Table 2, entry 1). In comparison,
cyclohexanone 26b (n = 6) afforded 28b in good yield (entry
2). The use of n- and t-BuLi in place of EtMgBr was also
unsuccessful (entries 4 and 6). In contrast, the presence of
ZnI2 in the exchange with n- and t-BuLi followed by the aldol
reaction with 27 produced aldol 28a (entries 5 and 7). As the
t-BuLi/ZnI2 system gave slightly higher product selectivity
(28a over 29a) than n-BuLi/ZnI2, combinations of t-BuLi and
other zinc salts were also investigated. The t-BuLi−Br
exchange in the presence of ZnI2·TMEDA and ZnX2 (X =
Br, OAc) proceeded with an efficiency similar to that of entry
7 (entries 8, 10, and 11), whereas ZnCl2 lowered the
1
oxidation of 21 furnished acid 3. The H NMR spectrum of 3
was consistent with that reported.5b
The ZnI2-assisted protocol for t-BuLi−Br exchange (Table
2, entry 7) was applied to various α-bromoketones. Cyclo-
penta-nones 26c−e possessing i-Pr, t-Bu, and Ph substituents
underwent the exchange, and the subsequent aldol reaction
with Ph(CH2)2CHO (27) afforded aldols 28c−e in good
yields (Scheme 3, eq 4). Similarly, the enolate was derived
from 26a and the aldol reaction with c-C6H11 and PhCHO
furnished aldols 28f and 28g, respectively. In contrast, the t-
BuLi−Br exchange of acyclic ketone 26h was accomplished
without ZnI2 and the aldol reaction with 27 afforded aldol
28h in good yield (eq 5).
In summary, the PMB ether of 5,6-epoxyisoprostane E2 3
was synthesized in 11.6% overall yield through bromohydri-
nation of 13 and the subsequent aldol reaction of α-
bromocyclopentanone 16. Furthermore, the t-BuLi−Br
Table 2. Aldol Reaction of Model Bromoketones
a
entry
ketone
n
RM (equiv)
additive (equiv)
solvent
temp1 (°C)
temp2 (°C)
28:29
bc
yield of 28 (%)
,
1
26a
26b
26a
26a
26a
26a
26a
26a
26a
26a
26a
26b
26b
5
6
5
5
5
5
5
5
5
5
5
6
6
EtMgBr (1.1)
EtMgBr (1.1)
EtMgBr (1.1)
n-BuLi (2.2)
n-BuLi (2.2)
t-BuLi (2.2)
t-BuLi (2.2)
t-BuLi (2.2)
t-BuLi (2.2)
t-BuLi (2.2)
t-BuLi (2.2)
t-BuLi (2.2)
t-BuLi (2.2)
−
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
0
0
−
−
71
−
−
2
−
0
0
98:2
d
b
3
ZnX2 (2)
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−78
−
b
4
−
−
e
f
5
ZnI2 (2)
95:5
nd
b
6
−
−
−
70
g
7
ZnI2 (2)
ZnBr2 (2)
ZnCl2 (2)
ZnI2·TMEDA (2)
Zn(OAc)2 (2)
−
−78
99:1
98:2
87:13
99:1
97:3
99:1
95:5
8
−78
−78
−78
−78
−78
−78
67
f
9
nd
10
11
12
13
69
61
65
62
h
ZnI2 (2)
a
b
c
d
e
Mixture of ca. 1:1 diastereomers. Complex mixture was obtained. Reaction at −78 °C also gave a complex mixture. X = Cl, Br, I. ZnBr2 and
f
g
ZnCl2 produced a 91:9 ratio of 28a:29a and a complex mixture, respectively. Not determined. Aldol reaction at 0 °C gave a complex mixture.
h
ZnBr2 produced a 94:6 ratio of 28b:29b in 60% yield, whereas ZnCl2 gave a complex mixture.
2600
dx.doi.org/10.1021/ol500654g | Org. Lett. 2014, 16, 2598−2601

Organic Letters
Letter
a
,b
Berliner, J. A.; Koroniak, L.; Gugiu, B. G.; Watson, A. D. Org. Lett.
2008, 10, 4207−4209.
Scheme 3. t-BuLi-Br Exchange and Aldol Reaction
(6) Egger, J.; Bretscher, P.; Freigang, S.; Kopf, M.; Carreira, E. M.
Angew. Chem., Int. Ed. 2013, 52, 5382−5385.
(7) Zhong, W.; Springstead, J. R.; Al-Mubarak, R.; Lee, S.; Li, R.;
Emert, B.; Berliner, J. A.; Jung, M. E. J. Med. Chem. 2013, 56, 8521−
8532.
(8) Stork, G.; Isobe, M. J. Am. Chem. Soc. 1975, 97, 4745−4746.
(9) (a) Tsuna, K.; Noguchi, N.; Nakada, M. Angew. Chem., Int. Ed.
2011, 50, 9452−9455. (b) Tsuna, K.; Noguchi, N.; Nakada, M.
Chem.Eur. J. 2013, 19, 5476−5486.
(10) (a) Bradbury, R. H.; Walker, K. A. M. J. Org. Chem. 1983, 48,
1741−1750. (b) Bradbury, R. H.; Walker, K. A. M. Tetrahedron Lett.
1982, 23, 1335−1338.
(11) Castro, A. M. M. Chem. Rev. 2004, 104, 2939−3002.
(12) Olefin 23 was prepared from 2-cyclopenten-1-one in 67%
yield: (i) NaBH4, CeCl3·7H2O; (ii) PyCO2H, DCC, DMAP; (iii)
Ph(CH2)3MgBr/CuI (2:1).
1
(13) Determination of the stereochemistry of 14 by the H NMR
a
t-BuLi (2.2 equiv), ZnI2 (2 equiv), and RCHO (1.3 equiv) were used.
Diastereomeric ratios of 26a,c−e were 1:1, 4:1, 6:1, and 1:1.
decoupling study was unsuccessful because of the overlap in signals.
(14) Aoki, Y.; Oshima, K.; Utimoto, K. Chem. Lett. 1995, 463−464.
(15) Nozaki, K.; Oshima, K.; Utimoto, K. Bull. Chem. Soc. Jpn.
1991, 64, 403−409.
b
exchange in the presence of ZnI2 was successfully applied to
several α-bromocyclopentanones.
(16) William, A. D.; Kobayashi, Y. J. Org. Chem. 2002, 67, 8771−
8782.
(17) The 1H and 13C NMR spectra of 21 were consistent with
those scanned,5b and the 13C NMR data were revised.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures and spectral data of compounds
described herein. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: ykobayas@bio.titech.ac.jp.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by a Grant-in-Aid for Scientific
Research from MEXT, Japan. Cyclopentenone (converted to
23) was kindly gifted by Zeon Corporation, Jpn.
REFERENCES
■
(1) (a) Jahn, U.; Galano, J.-M.; Durand, T. Angew. Chem., Int. Ed.
2008, 47, 5894−5955. (b) Leitinger, N. Mol. Nutr. Food Res. 2005,
49, 1063−1071. (c) Navab, M.; Ananthramaiah, G. M.; Reddy, S. T.;
Van Lenten, B. J.; Ansell, B. J.; Fonarow, G. C.; Vahabzadeh, K.;
Hama, S.; Hough, G.; Kamranpour, N.; Berliner, J. A.; Lusis, A. J.;
Fogelman, A. M. J. Lipid Res. 2004, 45, 993−1007.
(2) (a) Watson, A. D.; Leitinger, N.; Navab, M.; Faull, K. F.;
Horkko, S.; Witztum, J. L.; Palinski, W.; Schwenke, D.; Salomon, R.
̈
̈
G.; Sha, W.; Subbanagounder, G.; Fogelman, A. M.; Berliner, J. A. J.
Biol. Chem. 1997, 272, 13597−13607. (b) Watson, A. D.;
Subbanagounder, G.; Welsbie, D. S.; Faull, K. F.; Navab, M.; Jung,
M. E.; Fogelman, A. M.; Berliner, J. A. J. Biol. Chem. 1999, 274,
24787−24798. (c) Subbanagounder, G.; Wong, J. W.; Lee, H.; Faull,
K. F.; Miller, E.; Witztum, J. L.; Berliner, J. A. J. Biol. Chem. 2002,
277, 7271−7281.
(3) Acharya, H. P.; Kobayashi, Y. Angew. Chem., Int. Ed. 2005, 44,
3481−3484.
(4) (a) Acharya, H. P.; Kobayashi, Y. Tetrahedron Lett. 2005, 46,
8435−8438. (b) Acharya, H. P.; Miyoshi, K.; Kobayashi, Y. Org. Lett.
2007, 9, 3535−3538. (c) Acharya, H. P.; Miyoshi, K.; Takashima, Y.;
Ogawa, N.; Kobayashi, Y. Heterocycl. 2008, 76, 1181−1189.
(5) (a) Jung, M. E.; Berliner, J. A.; Angst, D.; Yue, D.; Koroniak, L.;
Watson, A. D.; Li, R. Org. Lett. 2005, 7, 3933−3935. (b) Jung, M. E.;
2601
dx.doi.org/10.1021/ol500654g | Org. Lett. 2014, 16, 2598−2601