Synthesis of (±)-mispyric acid, a triterpene inhibitor of DNA polymerase β isolated from Mischocarpus pyriformis

Article abstract of DOI:10.1016/S0040-4039(02)01193-0

The first synthesis of (±)-mispyric acid, an inhibitor of DNA polymerase β with a novel triterpene skeleton, was achieved by starting from isoprene, geraniol and 1,5-dimethoxy-1,4-cyclohexadiene.

Full text of DOI:10.1016/S0040-4039(02)01193-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 5743–5746
Synthesis of ( )-mispyric acid, a triterpene inhibitor of DNA
polymerase b isolated from Mischocarpus pyriformis
Yusuke Imamura,a Hirosato Takikawab and Kenji Moric,*
aDepartment of Chemistry, Science University of Tokyo, Kagurazaka 1-3, Shinjuku-ku, Tokyo 162-8601, Japan
bDepartment of Biofunctional Chemistry, Kobe University, Rokkodai 1-1, Nada-ku, Kobe 657-8501, Japan
cInsect Pheromone and Traps Division, Fuji Flavor Co. Ltd., Midorigaoka 3-5-8, Hamura, Tokyo 205-8503, Japan
Received 24 May 2002; accepted 20 June 2002
Abstract—The first synthesis of ( )-mispyric acid, an inhibitor of DNA polymerase b with a novel triterpene skeleton, was
achieved by starting from isoprene, geraniol and 1,5-dimethoxy-1,4-cyclohexadiene. © 2002 Elsevier Science Ltd. All rights
reserved.
In 1999, Hecht and co-workers isolated mispyric acid
(1), a structurally unique triterpene dicarboxylic acid
with a novel skeleton, from stem bark of Australian
plant Mischocarpus pyriformis as a DNA polymerase b
inhibitor.1 This structurally unique monocyclic triter-
pene is presumably derived via a new biogenetic path-
way including direct coupling of two farnesyl units.1 In
continuation of our works on synthesis of DNA poly-
merase inhibitors2 and biosynthetically unique triter-
penoids,3 we initiated the synthesis of mispyric acid (1).
Herein we report the first synthesis of ( )-1.
target compound 1 is readily obtainable from A, which
may be prepared from B by Michael addition of a
methyl group to the enone and subsequent methylena-
tion of the carbonyl group. Our early studies, however,
have shown that this Michael addition is quite difficult
to achieve due to steric hindrance.4 Cyclopropanation–
reduction methodology is therefore to be adopted to
overcome this difficulty. For the construction of the key
intermediate B, two alkylations (C with D, and E with
F) are chosen as key steps. Preparation of the two
alkylating agents (D and F), corresponding to side
chains, is considered to be possible by conventional
methods.
Our synthetic plan for ( )-1 is shown in Scheme 1. The
30
24
28
26 27
29
CO2H
10
6
HO2C
R1
R2
20
11
8
16
22
3
5
12
14
18
1
13
4
15
17
19
21
23
2
9
7
25
Mispyric acid (1)
A
O
OMe
R2
R1
R1
O
+ R2X
+ R1X
MeO
MeO
B
E
C
D
F
OR
R1 =
R2 =
RO
Scheme 1. Structure of mispyric acid (1) and synthetic plan for ( )-1.
Keywords: terpenes and terpenoids; DNA polymerase b; enzyme inhibitors; Mischocarpus pyriformis.
* Corresponding author. Fax: +81-42-555-7920; e-mail: kjk-mori@arion.ocn.ne.jp
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01193-0

5744
Y. Imamura et al. / Tetrahedron Letters 43 (2002) 5743–5746
910 was treated with 4 to give the alkylation product
(75%), which was then converted to 10 in conventional
three steps (41%); treatment with acid, enol ether for-
mation and TBS protection. Alkylation of 10 with 8
was followed by treatment with MeMgBr and subse-
quent acidic work-up to furnish the key intermediate 11
(83% in two steps).
Our synthetic route to the key intermediate B (=11) is
illustrated in Scheme 2. First, we synthesized the two
alkylating agents as follows. According to the reported
procedure, allylic chloride 3 was prepared from iso-
prene 2 in three steps.5 The chloride 3 was converted to
the corresponding bromide, which was exposed to the
homologation reaction developed by Knochel et al.,6
furnishing homoallylic iodide 4 (43% in two steps). It
should be noted that this homologation was not suc-
cessful with the allylic chloride 3. Next, geraniol 5 was
converted to aldehyde 6 in four steps (41%); TBS
protection, regioselective epoxidation, oxidative cleav-
age of the epoxide and re-protection with TBSCl. The
aldehyde 6 was then subjected to Corey–Yamamoto’s
modified Wittig reaction with concomitant hydroxy-
methylation.7 In spite of all our efforts, the reaction of
6 with 5-methyl-4-hexenylidenetriphenylphosphorane8
and paraformaldehyde gave adduct 7 in rather low
yield. The major product (30% yield) of this reaction
was 7% without the hydroxymethyl group. However, we
could obtain geometrically pure 7 without contamina-
tion of the undesired (E)-isomer. Although other
methodologies9 might be applicable for the preparation
of 7, this route was adopted due to its brevity. After
protection of the hydroxy group of 7 as TBDPS ether
(14% based on 6), the resulting product was converted
to the corresponding allylic bromide 8 in two steps
(80%).
As mentioned in the synthetic plan, cyclopropanation–
reduction methodology11 was adopted for the addition
of a methyl group to 11, because our initial attempts
with Michael addition met with failure. The key inter-
mediate 11 was reduced under Luche’s conditions12 to
give a mixture of two diastereomeric alcohols 12a/b
(98%, ratio=ca. 2:1). The structures of major and
minor isomers were elucidated to be 12a and 12b,
1
respectively, based on the similarity of H NMR data
between 12a/b and the structurally related com-
pounds.13 Although these two diastereomers were sepa-
rable, we used them as a mixture, because there was no
need of separation for the synthesis of ( )-1. This
diastereomeric mixture was then subjected to
cyclopropanation14 to furnish a mixture of 13a and 13b
(94%),15 which was immediately oxidized with PDC to
afford a mixture of 14a and 14b (86%). At this stage,
two different protecting groups (TBS and TBDPS) were
converged to TBS in two steps (quant.), because the
TBDPS group with two phenyl groups was not appro-
priate for the next dissolving metal reduction. Reduc-
tive cleavage of cyclopropane-ring16 was performed
under the classical Birch conditions (Li, NH3, THF,
t-BuOH) to yield the desired 16 (56%) as a single
With two alkylating agents (4 and 8) in hand, we
turned to construct the basic framework of mispyric
acid. The lithiated 1,5-dimethoxy-1,4-cyclohexadiene
ref. 5
a, b
Cl
TBSO
OTBS
I
TBSO
3 steps
2
3
4 (= F)
c, d, e, f
g
OH
OHC
5
6
OTBDPS
R
h, i, j
OTBS
Br
R = CH2OH
R = H
7
7'
8 (= D)
OMe
O
o, p
k, l, m, n
TBSO
MeO
MeO
10 (= C)
9 (= E)
OTBDPS
TBSO
(±)-11 (= B)
O
Scheme 2. Synthesis of the key intermediate 11. Reagents and conditions: (a) NaBr, DMF, rt; (b) ICH2ZnI, CuI, LiI, THF, −25°C
to rt (43% in two steps); (c) TBSCl, imidazole, DMF (98%); (d) mCPBA, CHCl3 (81%); (e) HIO4·2H2O, THF/H2O; (f) TBSCl,
imidazole, DMF (52% in two steps); (g) 5-methyl-4-hexenylidenetriphenylphosphorane, THF, −78°C; s-BuLi, −78°C; then
(CH2O)n, 0°C to rt; (h) TBDPSCl, imidazole, DMF (14%, from 6); (i) AcOH, H2O, THF (92%); (j) PBr3, pyridine, Et2O (87%);
(k) t-BuLi, THF, −78°C; then HMPA, 4, (75%); (l) 1
M HCl, THF; (m) CH2N2, Et2O/MeOH, 0°C; (n) TBSCl, imidazole, DMF
(41% in three steps); (o) LDA, THF, −78°C; then 8, −90°C to rt (93%); (p) MeMgBr, THF, 5°C to rt; then NH4Cl aq. (89%).

Y. Imamura et al. / Tetrahedron Letters 43 (2002) 5743–5746
5745
Scheme 3. Synthesis of ( )-1. Reagents and conditions: (a) NaBH4, CeCl3·7H2O, MeOH (98%); (b) CH2I2, Et2Zn, Et2O (94%); (c)
PDC, MS 4A, CH2Cl2 (86%); (d) TBAF, THF (e) TBSCl, imidazole, DMF (quant. in two steps); (f) Li, NH3, THF, t-BuOH,
−78°C (56%); (g) Ph3PꢀCH2, THF, 0°C to rt (72%); (h) TBAF, THF (97%); (i) PDC, MS 4A, CH2Cl2; (j) NaClO2, NaH2PO4,
2-methyl-2-butene, t-BuOH, H2O (31% in two steps).
isomer.17 The ketone 16 was then exposed to Wittig
methylenation to give 17 (72%). After removal of two
TBS groups (97%), the resulting diol was oxidized
stepwise to give the target compound ( )-1 (31%, two
steps) (Scheme 3). The various spectral data of syn-
thetic ( )-mispyric acid (1) are in good accord with
those of the natural product.18
5. (a) Ohsugi, M.; Takahashi, S.; Ichimoto, I.; Ueda, H.
Nippon Noˆgeikagaku Kaishi 1973, 47, 807–811; (b)
Kurosawa, S.; Mori, K. Eur. J. Org. Chem. 2000, 955–
962.
6. Knochel, P.; Chou, T.-S.; Chen, H. G.; Yeh, M. C. P.;
Rozema, M. J. J. Org. Chem. 1989, 54, 5202–5204.
7. Corey, E. J.; Yamamoto, H. J. Am. Chem. Soc. 1970, 92,
6637–6638.
In conclusion, the first synthesis of ( )-mispyric acid (1)
was accomplished by starting from isoprene (2), geran-
iol (5) and 1,5-dimethoxy-1,4-cyclohexadiene (9). Fur-
ther optimization of each step and the enantioselective
version for the synthesis of 1 are now in progress.
8. For the preparation of 5-methyl-4-hexenyltriphenylphos-
phonium iodide, see: Cane, D. E.; Tandon, M. Tetra-
hedron Lett. 1994, 35, 5355–5358.
9. For example: (a) Tago, K.; Arai, M.; Kogen, H. J. Chem.
Soc., Perkin Trans. 1 2000, 2073–2078; (b) Takikawa, H.;
Koizumi, J.; Kato, Y.; Mori, K. J. Chem. Soc., Perkin
Trans. 1 1999, 2271–2276.
10. Piers, E.; Grierson, J. R. J. Org. Chem. 1977, 42, 3755–
3757.
References
11. Hara, R.; Furukawa, T.; Kashima, H.; Kusama, H.;
Horiguchi, Y.; Kuwajima, I. J. Am. Chem. Soc. 1999,
121, 3072–3082.
12. Gemal, A. L.; Luche, J.-L. J. Am. Chem. Soc. 1981, 103,
5454–5459.
13. Gottschalk, F.-J.; Marschall-Weyerstahl, H.; Weyerstahl,
P. Liebigs Ann. Chem. 1985, 462–467.
14. Nishimura, J.; Kawabata, N.; Furukawa, J. Tetrahedron
1. Sun, D.-A.; Deng, J.-Z.; Starck, S. R.; Hecht, S. M. J.
Am. Chem. Soc. 1999, 121, 6120–6124.
2. Takikawa, H.; Nozawa, D.; Kayo, A.; Muto, S.-E.; Mori,
K. J. Chem. Soc., Perkin Trans. 1 1999, 2467–2478.
3. (a) Mori, K.; Tamura, H. Liebigs Ann. Chem. 1990,
361–368; (b) Watanabe, H.; Mori, K. J. Chem. Soc.,
Perkin Trans. 1 1991, 2919–2934; (c) Mori, K.; Takikawa,
H.; Kido, M. J. Chem. Soc., Perkin Trans. 1 1993,
169–179; (d) Yoshida, M.; Takikawa, H.; Mori, K. J.
Chem. Soc., Perkin Trans. 1 2001, 1007–1017.
4. Michael addition was attempted using the model com-
pound 18 as a substrate under various conditions as
follows: Me2CuLi-TMSCl-TMEDA, Me2CuLi-Me2S-
Et2O, Me2Zn-Ni(acac)2, Me3Al-Ni(acac)2, etc. The
observed best yield was less than 20%.
1969, 25, 2647–2659.
15. It was observed that the chromatographically purified
12a gave 13a as a single product by the cyclopropanation
reaction.
16. Dauben, W. G.; Deviny, E. J. J. Org. Chem. 1966, 31,
3794–3798.
17. Unexpectedly and fortunately, the undesired trans-isomer
could not be obtained.
18. Properties of synthetic ( )-1: colorless oil; IR wmax (film)
3500–2500 (s, OꢁH), 1690 (s, CꢀO), 1640 (s, CꢀC) cm−1
;
EIMS (m/z) 470, 452, 434, 419, 383, 365, 303, 285, 243,
O
(±)-18
235, 217, 203, 189, 175, 161, 147, 135, 121, 107, 93, 81,

5746
Y. Imamura et al. / Tetrahedron Letters 43 (2002) 5743–5746
69; HREIMS obsd. 470.3394 calcd. for C30H46O4
470.3396; 1H NMR (400 MHz, CDCl3) l=0.58 (3H, s,
26-H3), 1.03 (3H, s, 27-H3), 1.12 (1H, m, 9-H), 1.20 (1H,
m, 10-H), 1.58 (3H, s, 28-H3), 1.59 (3H, s, 30-H3),
1.59–1.67 (3H, m, 5-,6-,12-H), 1.68 (3H, s, 23-H3), 1.69–
1.78 (2H, m, 5-,9-H), 1.89 (1H, m, 8-H), 1.98-2.08 (2H, m,
15-H2), 2.08–2.16 (3H, m, 4-H2, 12-H), 2.17 (3H, br s,
24-H3), 2.22–2.39 (5H, m, 8-H, 19-,20-H2), 2.61 (2H, q,
J=7.3Hz, 16-H2), 4.51 (1H, s, 25-H), 4.86 (1H, s, 25-H),
5.06–5.16 (2H, m, 13-,21-H), 5.68 (1H, s, 2-H), 6.00 (1H,
t, J=7.3 Hz, 17-H); 13C NMR (100 MHz, CDCl3) l=
15.1, 16.1, 17.7, 19.2, 22.8, 25.7, 26.6, 27.8, 28.1, 29.1, 30.2,
34.6, 37.4, 39.2, 39.6, 40.0, 48.6, 53.0, 106.5, 114.8, 123.4,
125.3, 130.5, 132.3, 134.5, 145.6, 148.5, 164.0, 171.7, 173.0.