Controlled syntheses of 12-oxo-PDA and its 13-epimer

Article abstract of DOI:10.1016/S0040-4039(02)00786-4

Stereoselective synthesis of 12-oxo-PDA starting with (1R,3S)-cyclopenten-1,3-diol monoacetate is accomplished. Key transformations are copper-catalyzed installation of the C(1)-C(8) chain onto the cyclopentene ring and construction of the C(14)-C(18) cha

Full text of DOI:10.1016/S0040-4039(02)00786-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 4361–4364
Controlled syntheses of 12-oxo-PDA and its 13-epimer
Yuichi Kobayashi* and Michitaka Matsuumi
Department of Biomolecular Engineering, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501,
Japan
Received 29 March 2002; revised 18 April 2002; accepted 19 April 2002
Abstract—Stereoselective synthesis of 12-oxo-PDA starting with (1R,3S)-cyclopenten-1,3-diol monoacetate is accomplished. Key
transformations are copper-catalyzed installation of the C(1)–C(8) chain onto the cyclopentene ring and construction of the
C(14)–C(18) chain. Similarly, 13-epi-12-oxo-PDA was synthesized, and its [h]D was used to confirm the purity of 12-oxo-PDA
(>95%) obtained by 1H NMR spectroscopy. In addition, stability of 12-oxo-PDA was investigated to eliminate possibility of
epimerization to the 13-epimer. © 2002 Elsevier Science Ltd. All rights reserved.
The plant metabolism of linolenic acid produces epi-
jasmonic acid through 12-oxo-phytodienoic acid (12-
oxo-PDA),1 cyclopentanoic acid (OPC-8:0), and two
b-oxidation products (OPC-6:0, OPC-4:0) (Fig. 1).2 A
significant role of this cascade in the regulation of
process in plant physiology has been unveiled partially
and is summarized in the reviews.3 To support the
biological study by supplying the natural acids as well
as their analogues, synthesis of these metabolites has
received attention. Among the metabolites, epi-jas-
monic acid has been a synthetic target of active investi-
gation,4 while only few syntheses of the other acids are
reported.5 Although there is a structural similarity
between epi-jasmonic acid and the other metabolites,
difficulty would be encountered when the methods
developed for epi-jasmonic acid synthesis are applied to
synthesis of the other metabolites because of the specifi-
city of the methods.
1
3
RMgCl, CuCN
THF or Et2O
HO
OAc
3a
1
HO
4
1
HO
R
+
2
R
4
5
1,4-isomer
1,2-isomer
(1)
a (1R)-enantiomer is shown.
1
9
(CH2)nCO2H
Recently, we reported a reaction of acetate 3 and the
reagent derived from RMgCl and CuCN to afford
either 1,4-isomer 4 or 1,2-isomer 5 with high regioselec-
tivity (Eq. (1)).6 This selectivity is controlled by the
ratio of RMgCl/CuCN and the solvent used (THF or
Et2O). Thus, a reasonable approach to 12-oxo-PDA (1)
is installation of one of the alkyl chains on the
cyclopentene ring by using this reaction and the other
chain by another method taking advantage of the allylic
moiety on the ring. This idea was realized by a
sequence of reactions illustrated in Scheme 1. In a
similar manner, the 13-epimer of 1 (i.e. 2) was synthe-
sized stereoselectively and used to secure the purity of
1. In addition, stability of 1 was studied briefly.
CO2H
12
O
O
O
OPC-8:0, n = 7
OPC-6:0, n = 5
OPC-4:0, n = 3
epi-jasmonic acid, n = 1
12-oxo-PDA (1)
CO2H
13
13-epi-12-oxo-PDA (2)
Figure 1. Metabolites of linolenic acid via the lipoxygenase
pathway.
In order to prepare the key intermediate 7 in our
synthesis (Scheme 1) by using the reaction shown in Eq.
(1), a Grignard reagent derived from 8-chlorooctan-1-ol
through protection of the hydroxyl group is required.
Keywords: asymmetric synthesis; 12-oxo-PDA; cyclopenten-1,3-diol
monoacetate; copper catalyst; coupling reaction.
* Corresponding author. Tel./fax: +81-45-924-5789; e-mail: ykobayas
@bio.titech.ac.jp
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00786-4

4362
Y. Kobayashi, M. Matsuumi / Tetrahedron Letters 43 (2002) 4361–4364
1
RO
TBDPSO(CH2)8MgCl (6)
HO
(CH2)8OTBDPS
HO
OAc
(CH2)8OTBDPS
b
d
a
93%
85%
8: R = Ac
9: R = H
7
c
84%
(1R)-3
91%
(CH2)8OTBDPS
(CH2)8OTBDPS
COOR
(CH2)8OTBDPS
R
f, g
h
j
HO
i
O
65%
from 10
10: R = CHO
11: R = CO2H
13: R = H
14: R = Me
O
e
12
(CH2)8OTBDPS
(CH2)8OTBDPS
m,e
53%
l
12-oxo-PDA (1)
COR
TESO
k
TESO
81%
17
15: R = OMe, 69%
16: R = H, 85%
Scheme 1. Synthesis of 12-oxo-PDA (1). Reagents and conditions: (a) 6, CuCN (15 mol%), THF, −18°C; (b) AcOH, DEAD, PPh3,
toluene, −78 to −60°C; (c) MeLi (3 equiv.), Et2O, 0°C; (d) CH2ꢀCHOEt, Hg(OAc)2 (0.5 equiv.), benzene, 170°C, 60 h; (e) CrO3,
H2SO4, acetone; (f) I2 (2 equiv.), KI (6 equiv.), 0.5N NaHCO3 (3 equiv.), rt, 13 h; (g) DBU (2.5 equiv.), THF; (h) LiOH,
MeOH/H2O/THF (1:1:3); (i) CH2N2; (j) TESCl, imidazole; (k) DIBAL, CH2Cl2, −78°C; (l) [Ph3PPr]+Br−, NaN(TMS)2,
,
THF/DMF (9:1), 0°C to rt; (m) TBAF, 4 A MS, 55°C.
Among common protecting groups we tested,7 t-
BuPh2Si group (TBDPS) showed excellent stability
throughout the synthesis. The Grignard reagent,
ClMg(CH2)8OTBDPS (6), was prepared from
Cl(CH2)8OTBDPS and Mg in THF, and subjected to
the CuCN-catalyzed reaction with (1R)-38 of >99% ee
(checked by MTPA ester) to furnish 7 as a major
product in 84% yield. The ratio of 7 and the corre-
Previously, conversion of a closely related lactone 20
into olefin 21 was reported by Crombie,5b who used a
sequence through reduction to g-lactol followed by
Wittig olefination. Though this protocol is surely appli-
cable to our lactone 12, the observed stereoselectivity in
the Wittig reaction is only 80:20 for the cis and trans
olefins,14 which is probably insufficient for a biological
study. Consequently, the free aldehyde 16, which is not
involved in the equilibrium with lactol, was chosen as a
substrate for the Wittig reaction.
1
sponding 1,2-isomer 189 was 97:3 by H NMR spec-
troscopy (300 MHz, CDCl3): 7, l 5.79–5.84 (m, 1H)
and 5.93–5.98 (m, 1H); 18, l 5.64–5.77 (m, 2H). After
separation of the regioisomer 18, the hydroxyl group of
7 was inverted by the Mitsunobu reaction10 to afford
acetate 8 with a 98:2 ratio of 8 and the trans isomer
CO2H
CO2H
1
1911 by H NMR spectroscopy: 8, l 2.42–2.66 (m, 2H);
O
HO
19, l 2.78–2.92 (m, 1H). In contrast to the case of a
O
related cyclopentenol,12 formation of the regioisomeric
21
1
20
acetate was not observed by TLC and H NMR spec-
troscopy. Acetate 8 was hydrolyzed, and the resulting
alcohol 9 was submitted to the Claisen rearrangement
using CH2ꢀCHOEt and Hg(OAc)2 as a catalyst.
Although reaction temperature of 140°C is sufficient for
the rearrangement of a related compound,13 higher
temperature of 170°C for 60 h was necessary for alco-
hol 9 to furnish the corresponding aldehyde 10 in 85%
yield. Jones reagent was used for oxidation of 10 to
acid 11, and iodolactonization of 11 followed by elimi-
nation of HI from the resulting iodolactone under the
given conditions afforded lactone 12 in 65% yield from
aldehyde 10.
Lactone 12 was hydrolyzed, and the resulting acid 13,
after acidification, was extracted with Et2O. Without
concentration,15 the solution was treated with CH2N2
to produce the hydroxyl methyl ester 14, and protection
of the hydroxyl group with TESCl furnished TES ether
15 in 69% yield from lactone 12. Little contamination
of lactone 12 was detected by TLC. Reduction of ester
15 with DIBAL at −78°C afforded the key aldehyde 16
in 85% yield. Wittig reaction of 16 with [Ph3PPr]+Br−
was best carried out under the modified conditions of
Santelli16 using NaN(TMS)2 in THF/DMF (9:1) to
produce 17 in good yield with a 99:1 ratio of 17 (l 4.50,
HO
AcO
(CH2)8OTBDPS
1
dd, J=5.7, 2.4 Hz, 1H in H NMR (300 MHz, CDCl3)
spectrum) and the corresponding trans isomer (ca. l
4.48). Note that use of n-BuLi as a base gave an 89:11
ratio of 17 and the trans isomer.
TBDPSO(CH2)8
19
18

Y. Kobayashi, M. Matsuumi / Tetrahedron Letters 43 (2002) 4361–4364
4363
The remaining steps for the synthesis of 1 were depro-
tection of the two silyl groups and oxidation of the
resulting alcohol. This transformation was accom-
plished easily under the conditions presented in Scheme
1.
and 26, in turn, was prepared from racemic 1 and
BuMgCl according to Eq. (1).
Bu
HO
Bu
O
1
26
The H NMR spectrum of 1 thus synthesized was in
27
good agreement with data reported,1a and showed
>95% purity of 1 with <5% contamination of the thermo-
dynamically more stable epimer 2.17 However, values of
the specific rotation of 1 measured twice ([h]2D8 +127 (c
0.496, CHCl3) and [h]2D9 +127 (c 0.198, CHCl3)) were
inconsistent with that reported5a ([h]D25 +104.0 (c 9.5,
CHCl3)). It seemed likely that a large [h]D of the epimer
2, though less than 5% contamination in 1, was respon-
sible for the larger [h]D of synthetic 1 we observed.
Two solutions of 26 in CDCl3 in NMR tubes were
prepared. Commercial CDCl3 was used without any
purification in one tube (slightly acidic conditions due
to the presence of DCl20), while DCl-free21 CDCl3 was
added into the other tube (neutral conditions). The
solutions were left on a bench at room temperature,
and the diagnostic olefin protons in the 1H NMR
spectra17 were monitored. During 1 month, no change
was detected. A possibility of the DCl-assisted epimer-
ization was thus eliminated. In a similar manner, a
solution of synthetic 1 in commercial CDCl3 was left at
room temperature for 1 month, and the stable nature of
1 under these conditions was thus established.22
Since the specific rotation of 2 was not reported, a
method to obtain 2 was investigated. Epimerization of
1 would be a more convenient method than a total
synthesis.18 Although, the methyl ester of 1 has been
subjected to the epimerization,1b,19 that of 1 to 2 is not
reported. In addition, the yield of the 13-epimer (methyl
ester of 2), the product ratio after the epimerization,
and the details of the reaction conditions are not pre-
sented. Consequently, it seemed a better way for us to
follow the sequence of reactions shown in Scheme 1
without conducting the Mitsunobu reaction in this case.
Although the yields were not optimized, 2 was synthe-
sized stereoselectively as depicted in Scheme 2. In con-
trast to our assumption, observed [h]D of 2 ([h]2D5 +93 (c
0.176, CHCl3)) was smaller than that of 1. Taking
together this result and the >95% purity of 1 measured
by 1H NMR spectroscopy, the [h]D of 1 is revised
herewith to the value of +127 mentioned above.
In conclusion, synthesis of 12-oxo-PDA (1) and the
13-epimer of 1 (i.e. 2) is achieved. One of the key
reactions in the present synthesis is the installation of
the C(1)ꢁC(8) chain, which was accomplished by the
reaction shown in Eq. (1). This reaction is applicable to
a number of alkylmagnesium chloride, and thence,
other metabolites and analogues thereof as well as
other cyclopentanoids possessing two side chains with
the cis relationship would be synthesized by the present
method.
Acknowledgements
Examined next was the stability of 1. Preliminarily, a
model compound 27 was synthesized in racemic form
from cyclopentene 26 in a similar manner to Scheme 1,
This research work was supported by Grant-in-Aid for
Scientific Research from the Ministry of Education,
Culture, Sports, Science and Technology, Japan.
R
a
R
b,c,d
36%
References
7
O
CHO
79%
1. (a) Baertschi, S. W.; Ingram, C. D.; Harris, T. M.; Brash,
A. R. Biochemistry 1988, 27, 18–24; (b) Crombie, L.;
Morgan, D. O. J. Chem. Soc., Perkin Trans. 1 1991,
581–587; (c) Laudert, D.; Henning, P.; Stelmach, B. A.;
Mu¨ller, A.; Andert, L.; Weiler, E. W. Anal. Biochem.
1997, 246, 211–217; (d) Hamberg, M.; Hughes, M. A.
Lipids 1988, 23, 469–475; (e) Hamberg, M.; Miersch, O.;
Sembdner, G. Lipids 1988, 23, 521–524.
O
22
23
R
R
e,f,g,h
i
CHO
51%
82%
TESO
TESO
24
25
j,b
2. Vick, B. A.; Zimmerman, D. C. Biochem. Biophys. Res.
R = (CH2)8OTBDPS
2
Commun. 1983, 111, 470–477.
68%
3. Reviews: (a) Creelman, R. A.; Mullet, J. E. Annu. Rev.
Plant Physiol. Plant Mol. Biol. 1997, 48, 355–381; (b)
Sembdner, G.; Parthier, B. Annu. Rev. Plant Physiol.
Plant Mol. Biol. 1993, 44, 569–589; (c) Hamberg, M.;
Gardner, H. W. Biochim. Biophys. Acta 1992, 1165, 1–18.
4. Recent syntheses: (a) Roth, G. J.; Kirschbaum, S.; Best-
mann, H. J. Synlett 1997, 618–620; (b) Stadtmu¨ller, H.;
Knochel, P. Synlett 1995, 463–464; (c) Kitahara, T.;
Nishi, T.; Mori, K. Tetrahedron 1991, 47, 6999–7006.
Scheme 2. Synthesis of 13-epi-12-oxo-PDA (2). Reagents and
conditions: (a) CH2ꢀCHOEt, Hg(OAc)2 (0.5 equiv.), benzene,
170°C, 65 h; (b) CrO3, H2SO4, acetone; (c) I2 (2 equiv.), KI (6
equiv.), 0.5N NaHCO3 (3 equiv.), rt, 17 h; (d) DBU (2.5
equiv.), THF; (e) LiOH, MeOH/H2O/THF (1:1:3); (f) CH2N2;
(g) TESCl, imidazole; (h) DIBAL, CH2Cl2, −78°C; (i)
[Ph3PPr]+Br−, NaN(TMS)2, THF/DMF (9:1), 0°C to rt; (j)
,
TBAF, 4 A MS.

4364
Y. Kobayashi, M. Matsuumi / Tetrahedron Letters 43 (2002) 4361–4364
Bu
5. (a) Grieco, P. A.; Abood, N. J. Org. Chem. 1989, 54,
Bu
CO2H
6008–6010; (b) Crombie, L.; Mistry, K. M. J. Chem. Soc.,
Perkin Trans. 1 1991, 1981–1991.
6. Ito, M.; Matsuumi, M.; Murugesh, M. G.; Kobayashi, Y.
O
HO
O
i
ii
J. Org. Chem. 2001, 66, 5881–5889.
7. The THP protection was somewhat unstable during the
preparation of the corresponding Grignard reagent in
THF, while the t-BuMe2Si group was partially depro-
tected with Jones reagent.
16. Viala, J.; Santelli, M. Synthesis 1988, 395–397.
17. Diagnostic signals in the 1H NMR (300 MHz, CDCl3)
spectra of 1 and 2 are protons at C(11) and C(10): 1, l
6.19 (dd, J=6, 1.5 Hz, 1H), 7.74 (dd, J=6, 3 Hz, 1H); 2,
l 6.13 (dd, J=6, 2 Hz, 1H), 7.61 (dd, J=6, 2.5 Hz, 1H).
18. Synthesis of racemic 2: Kru¨ger, G.; Harde, C.;
Bohlmann, F. Tetrahedron Lett. 1985, 26, 6027–6030.
19. Vick, B. A.; Zimmerman, D. C.; Weisleder, D. Lipids
1979, 14, 735–740.
20. Cf. We reported DCl-catalyzed isomerization of a cis
double bond in the conjugated triene of leukotriene B4:
Kobayashi, Y.; Shimazaki, T.; Taguchi, H.; Sato, F. J.
Org. Chem. 1990, 55, 5324–5335.
8. Sugai, T.; Mori, K. Synthesis 1988, 19–22.
9. The regioisomer 18 was synthesized from 3 regioselec-
tively by using the reagent derived from 6 (3 equiv.) and
CuCN (3.5 equiv.) in THF at 0°C for 5 h.
10. Mitsunobu, O. Synthesis 1981, 1–28.
11. Authentic 19 was prepared from 7 by the standard acetyl-
ation (Ac2O, pyridine).
12. Hodgson, D. M.; Gibbs, A. R. Synlett 1997, 657–658.
13. (a) Nara, M.; Terashima, S.; Yamada, S. Tetrahedron
1980, 36, 3161–3170; (b) Laumen, K.; Schneider, M. P. J.
Chem. Soc., Chem. Commun. 1986, 1298–1299.
14. A similar insufficient stereoselectivity has been reported:
(a) Ref. 4c; (b) Takahashi, K.; Sato, H.; Mikami, K.;
Nakai, T. Chem. Lett. 1989, 247–250.
15. Concentration of an extract containing a model car-
boxylic acid i furnished lactonization to ii. The present
operation described in the text reproducibly furnished 14
and, after the silylation, 15 in good yield.
21. Conveniently prepared by passing commercial CDCl3
through a pad of basic Al2O3 before use.
22. Since the stability of 1 under the almost neutral condi-
tions is established, it might be likely that the strongly
acidic conditions of Jones oxidation to produce 1 and
subsequent purification by silica gel chromatography
brought about epimerization of 1, but to a small extent
(<5%).