Synthesis of a compound having the essential structural unit for the hetisine-type of aconite alkaloids

Article abstract of DOI:10.1016/S0040-4039(02)00453-7

A hexacyclic compound 1, which carries an almost full structure of the hetisine skeleton lacking only the six-membered ring with an exo methylene group, was synthesized by applying an acetal-ene reaction on 5 for the bond formation of C-14 and C-20 as wel

Full text of DOI:10.1016/S0040-4039(02)00453-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 2913–2917
Synthesis of a compound having the essential structural unit for
the hetisine-type of aconite alkaloids
Hideaki Muratake* and Mitsutaka Natsume*
Research Foundation Itsuu Laboratory, 2-28-10 Tamagawa, Setagaya-ku, Tokyo 158-0094, Japan
Received 6 February 2002; revised 27 February 2002; accepted 1 March 2002
Abstract—A hexacyclic compound 1, which carries an almost full structure of the hetisine skeleton lacking only the six-membered
ring with an exo methylene group, was synthesized by applying an acetal-ene reaction on 5 for the bond formation of C-14 and
C-20 as well as stereoselective hydrocyanation reaction on 7 for construction of the azabicyclo ring system. © 2002 Elsevier
Science Ltd. All rights reserved.
More than four hundred aconite alkaloids have so far
been isolated from genera Aconitum, Delphinium, Con-
solida and Spiraea. Their chemical structures are gener-
ally classified into five types, which possess the
frameworks of atidane, veatchane, cycloveatchane,
aconitane, and hetisan (Fig. 1).1,2 Hetisine3 is a repre-
sentative of nearly a hundred hetisan alkaloids, and its
heptacyclic structure is characteristic of the chemical
bonds between N and C-6, and also C-14 and C-20 in
addition to the atidane structure.
been accomplished for 40 years since the first clarifica-
tion of the hetisine structure. Nominine, kobusine, and
pseudokobusine are particularly our synthetic targets.
On the way to these objectives, we report here a
synthesis of the hexacyclic compound 1, which carries
the above essential structural unit for the hetisine struc-
ture. Okamoto and co-worker once synthesized a pen-
tacyclic compound 2.4,5
The synthetic plan of 1 is based on (i) our previous
finding of a palladium-catalyzed intramolecular aryla-
tion reaction of aldehyde compounds,6 i.e. 3“4 (Fig.
2), (ii) application of the Lewis acid-catalyzed ene
reaction7 to an acetal 5 for the formation of the bond
We commenced synthetic work aiming at the hetisine-
type of aconite alkaloids, whose total synthesis has not
13
HO
17
11
HO
20
MeO
OHC
OMe
14
1
OHC
H
15
HO
N
H
6
H
N
H
Br
N
H
(i)
H
H
H
O
O
O
O
19 18
3
4
Hetisan
Atidane
Hetisine
R
O
H
O
H
O
HO
(ii)
R1
OBz
OH
R1
H
N
H
X
N
H
N
H
H
O
O
H
H
O
6
8
5
Nominine: R=X=H
R2O
R2O
1
2
Kobusine: R=OH, X=H
Pseudokobusine: R=X=OH
R1
R1
(iv)
(iii)
1
H
H
Figure 1.
H
O
O
NC
7
* Corresponding authors. E-mail: itsumura@wa.catv.ne.jp; itsnatsu@
ca.catv.ne.jp
Figure 2.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00453-7

2914
H. Muratake, M. Natsume / Tetrahedron Letters 43 (2002) 2913–2917
MeO
R
MeO
OHC
OHC
MeO
OMe
O
H
(c)
(d)
(a)
Br
Br
Br
H
H
OMe
ref. 6
O
O
O
O
O
O
I
O
12a: trans
12b: cis
10: R=H
11: R=CH2NO2
4
9
3
(b)
12
O
H
O
O
O
1
O
O
H
O
O
H
O
O
O
O
O
H
O
O
H
O
OMe
OMe
11
O
(f)
(e)
10
8
H
H
4
OH
3
O
5
O
O
O
O
O
O
O
O
O
O
O
18
15
16
17
13
14
(e)
(i)
(g)
(h)
Cl
Cl
H3
CeB
H
O
OH
O
H
O
O
O
O
H
O
H
O
O
O
(j)
(k)
H
H
H
OR
NMe2
H
H
O
O
O
O
O
O
O
O
20
21: R=H
5: R=Piv
19
16
O
O
S*
O
O
MOMO
MOMO
RO
MOMO
(q)
OR
OAc
OPiv
OAc
H
H
(n)
(p)
(l)
H
H
H
X
O
X
Y
O
O
NC
23: R=Piv, X, Y=O
24: R=H, X=Me, Y=OH
25: R=H, X=OH, Y=Me
6a: X=β-H, R=H
6b: X=α-H, R=H
22: X=Br (α, β mixture),
8
7
(o)
(m)
R=TMS
RO
HO
HO
OR
-
N
OAc
(s)
OR
(v)
I
OBz
OBz
+
H
H
H
H
H
(r)
(u)
(w)
N
H
Me
H
H
H
H
O
NC
N
N
Boc
Boc
26: R=CH2CH2OH
27: R=CH2CH2Br
28: R=H
31: R=H
32: R=Bz
33
29: R=H
30: R=Boc
1
(t)
Scheme 1. Reagents and yields: (a) N-cyclohexylidenecyclohexylamine, LDA, THF, −18°C, then oxalic acid, THF/H2O (4:1), rt,
10 95%. (b) TMSCl, NaI, (TMS)2NH, CH3CN, rt, then NBS, THF, −18°C to rt, 80%; Li2CO3, LiBr, DMF, 120°C, 73%;
CH3NO2, KF, 18-crown-6, CH3CN, reflux, 11 96%. (c) (CH2OH)2, p-TsOH, benzene, reflux 97%; KOH, MeOH, 0°C, then
KMnO4, MgSO4, MeOH/H2O (7:1), 0°C–rt, 3 75%. (d) PdCl2(Ph3P)2, Cs2CO3, THF, reflux, 4 65% (cis/trans=4.2), 12a 11%, 12b
6%. (e) (CH2OH)2, p-TsOH, benzene, reflux, 13 61%, 14 34% from 4; 13 61%, 14 35% from 13. (f) Li, EtOH, liq. NH3/THF, −78
to −60°C, 92%; 0.5% HCl in THF/H2O (4:1), 0°C, 15 88%, 16 7%. (g) NaOMe, MeOH, rt, 16 55%, 17 10%, recovery of 15 10%.
(h) NaOMe, Me2S, MeOH, rt, 16 58%, 18 14%, recovery of 15 13%. (i) NaBH4, CeCl3·7H2O, MeOH, 0°C, 19 94%. (j)
N,N-Dimethylacetamide dimethyl acetal, toluene, 165–170°C (sealed tube), 20 68%. (k) BH3·NH3, BuLi, THF, 0°C–rt, 21 94%;
Piv2O, Et3N, 4-DMAP, CH2Cl2, −18°C to rt, 5 98%. (l) BF3·OEt2, toluene, −18°C; p-TsOH, acetone, rt, 6a 66%, 6b 3%. (m)
TMSCl, NaI, (TMS)2NH, CH3CN, reflux; NBS, THF, 0°C–rt. (n) 0.2% HCl, THF/H2O (12:1), 0°C; MOMCl, i-Pr2NEt, CH2Cl2,
−20°C to rt; DBU, benzene, reflux, 23 52% overall from 6a, 45% overall from 6b. (o) MeLi, THF, −78°C, 24 59%, 25 11%. (p)
Ac2O, pyridine, CH2Cl2, rt; 20 wt.% PCC/Al2O3, benzene, 5°C–rt, 7 93% and 63% for each step from 24, 7 90% and 65% for each
step from 25. (q) Et2AlCN, toluene, 0°C–rt, 8 94%. (r) TMSCl, NaI, CH3CN, 0°C, 26 92%; CBr4, Ph3P, CH2Cl2, rt, 27 93%; Zn,
NH4Cl, 2-propanol/H2O (14:1), reflux, 28 95%. (s) TMSCl, LDA, THF, −78 to −67°C; LiAlH4, THF, reflux; Boc2O, Et3N,
CH2Cl2, rt, 29 31%, 30 25%. (t) K2CO3, MeOH, reflux, 29 89%. (u) NaBH3CN, 1% HCl, MeOH/H2O (8:1), 0°C–rt, 31 85%;
BzCl, pyridine, CH2Cl2, rt, 32 95%. (v) CF3COOH, CH2Cl2, 0°C; SOCl2, pyridine, CH2Cl2, rt, 1 78%. (w) MeI, MeOH, rt,
33 73%.

H. Muratake, M. Natsume / Tetrahedron Letters 43 (2002) 2913–2917
2915
between C-14 and C-20 to give 6, (iii) stereoselective
Michael addition of a cyano group into the enone
function of 7, and (iv) construction of the azabicyclo
ring system by utilizing primary amine derived from the
cyano group in 8. Preparation of 3 was started by
condensation of 2-bromo-5-methoxyphenethyl iodide8
(9) with an anion derived from N-cyclohexylidenecyclo-
hexylamine and LDA (Scheme 1).9 The resulting 10 was
converted to a thermodynamically stable enol silyl
ether,10 which was brominated11 to afford 2-bromo-2-
phenethylcyclohexanone. Regioselective dehydrobromi-
nation of the latter gave a cyclohexenone derivative,
and conjugate addition of nitromethane12 afforded 11.
The ketone group was protected and the nitromethyl
group was transformed into the aldehyde function13
providing 3 as a mixture of two stereoisomers.
ene reaction occurred by interaction of BF3 with the
acetal oxygen as shown in A (Scheme 2), where the
angular acetal group possessed the conformation of
restricted rotation so as to avoid the steric environment
of the C-4 acetal group to result in the 20S* configura-
tion in 6a and 6b via B. Realization of the bond
between C-14 and C-20 implied that the C-14 of 5 was
situated in the same face as the angular acetal at the a
side, and further, the origin of this stereochemical
situation could be traced back to the configuration of
the hydroxy group of 19, which regulated stereochem-
istry of the dimethylacetamide function of 20. Eventu-
ally, reduction of 16 took place coveniently to give
stereoselectively 19 with the b-hydroxy group, probably
due to complex formation of the borohydride with the
angular acetal oxygen in aid of the ceric species to
deliver hydride from the a-side.18 The ene reaction
products 6a and 6b were separately converted to enone
23 in 52 and 45% yields, respectively, by successive
operations of the enol silyl ether formation between C-4
and C-5, bromination with NBS to afford 22, hydroly-
sis of TMSꢀO bond with diluted acid, protection of the
primary OH with MOM group, and dehydrobromina-
tion with DBU.
As reported previously,6 the palladium-catalyzed reac-
tion of aldehyde 3 afforded an inseparable mixture of
cis and trans isomers 4 (ca. 4:1) in 65% yield together
with by-products 12a and 12b. When this mixture 4 was
treated under the usual conditions of acetal formation,
we observed that an acid-catalyzed equilibration took
place at the C-5 position adjacent to the original acetal
group, and the ratio of the resulting products 13 and 14
was changed to ca. 2:1. This knowledge was useful for
enrichment of the trans isomer, and the cis isomer 13
once isolated in 61% yield in addition to 14 in 34%
yield was submitted again to the same acetalization
conditions to give 13 and 14 in 61 and 35% yields. Thus
this recycling process of 13 made it possible to accumu-
late the important trans compound 14 for further use.
The next task was to construct the two substituents at
the C-4 position, i.e. a-cyano and b-methyl groups as
shown by 8. The methyl group was first introduced by
reaction of 23 with MeLi to afford 24 and 25 in 59 and
1
11% yields, respectively. In the H NMR spectrum of
the major product 24, NOESY data exhibiting a corre-
lation between CH3 and H-20 revealed that the methyl
group was situated at the a-side, and the attack of
MeLi mostly occurred from the side of C-14 and C-20
bridge. Both products 24 and 25 were separately con-
verted to their acetates in 93 and 90% yields and further
Birch reduction of 14 with Li metal afforded b,g-enone
15 in 88% yield after quenching with dilute aq. HCl,
accompanied by a,b-enone 16 in 7% yield. As conver-
sion of 15 into 16 had to be carried out under the basic
conditions in order to avoid hydrolysis of the acetal
group, 15 was treated with NaOMe in MeOH. The
required 16 was obtained in 55% yield in addition to
the recovery of 15 in 10% yield, accompanied by the
formation of oxetane 17 in 10% yield. Probably autoxi-
dation took place partially at the C-9 position of 16,
and the resulting hydroperoxide group would react to
the enolate anion at C-12 to afford 17. This assumption
was supported by the fact that g-hydroxy-a,b-enone 18
was obtained in 14% yield in addition to 16 and 15 in
respective yields of 58 and 13%, when Me2S was added
to the treatment of 15 with NaOMe in MeOH.14
19
oxidized with PCC/Al2O3 to provide 7 in 63 and 65%
yields, respectively. The cyano group was introduced to
7 by reaction with Et2AlCN20 to obtain stereoselectively
in 94% yield the desired 8, whose stereochemical proof
was secured by NOE data (6.5% from H-5 to H-9 as
1
well as 1.2% from CH3 to H-5 and vice versa) in the H
NMR spectrum of 26 produced by removal of the
MOM group from 8 with TMSI21 in 92% yield. Again,
approach of the cyano group to the C-4 position of
enone 7 was observed exclusively to take place from the
side of C-14 and C-20 bridge, giving the required
stereochemistry in 8. Here the ethanol function, which
served as a protecting group of the C-20 hydroxy
group, was removed by treatment of 26 with CBr4/Ph3P
to afford 27 in 93% yield, followed by reductive cleav-
age with Zn in the presence of NH4Cl to produce 28 in
95% yield.
a,b-Enone 16 was exclusively reduced to 19 in 94%
yield with NaBH4 in the presence of CeCl3.15 Introduc-
tion of a two-carbon unit at the C-8 position was
achieved by Claisen rearrangement on 19 with
16
MeC(OMe)2NMe2 to afford 20 in 68% yield. Subse-
BF3
17
O
+
quent reduction of 20 with LiH2NBH3 produced 21 in
–
BF3
O
O
14
20
H
:
94% yield, and the hydroxy function of 21 was pro-
tected by the pivaloyl group to give 5 in 98% yield.
When 5 was treated with BF3·OEt2 in toluene, acetal-
ene reaction7 took place to afford 6a and 6b in 66 and
3% yields, respectively. In these compounds 6a and 6b,
stereochemical relationship between 14-H and 20-H
was shown to be cis (J14,20=6 Hz). This meant that the
O
5-Exo-Trig
6a
6b
H
H
10
H
O
O
PivO
8
O
4
O
PivO
A
B
Scheme 2.

2916
H. Muratake, M. Natsume / Tetrahedron Letters 43 (2002) 2913–2917
Having arrived at the final stage of the synthesis, the
formation of a primary amine from the cyano group
was to be attempted. For that purpose, prior protection
of the ketone group in 28 was carried out by converting
it to enol silyl ether by treatment with LDA and
TMSCl. This product was reduced with LiAlH4 in
refluxing THF, and after quenching of the reaction with
H2O, the crude reaction product was treated with
Boc2O and Et3N to isolate 29 (31% yield) and 30 (25%
yield), and the latter was easily converted to 29 in 89%
yield with K2CO3 in refluxing MeOH. Structure of 29
was supported by the 1H NMR signal of the vinyl
proton involved in the ene carbamate system at 5.39
ppm (br s). Compound 29 was reduced with NaBH3CN
in 85% yield, and the product 31 was benzoylated to 32
in 95% yield.22 The final compound 1 was readily
obtained in 78% yield by two reactions, i.e. removal of
the Boc group from 32 and dehydration of the resulting
amine with SOCl2 in the presence of pyridine. The
tertiary amine nature of 1 was supported by the forma-
Rep. 1986, 3, 452–475; (c) Yunuzov, M. S. Nat. Prod.
Rep. 1991, 8, 499–526; (d) Yunuzov, M. S. Nat. Prod.
Rep. 1993, 10, 471–486; (e) Atta-ur-Rahman; Choud-
hary, M. I. Nat. Prod. Rep. 1995, 12, 361–379; (f) Atta-
ur-Rahman; Choudhary, M. I. Nat. Prod. Rep. 1997,
14, 191–203; (g) Atta-ur-Rahman; Choudhary, M. I.
Nat. Prod. Rep. 1999, 16, 619–635.
2. (a) Stern, E. S. In The Alkaloids; Manske, R. H. F.;
Holmes, H. L., Eds.; Academic Press: New York, 1954;
Vol. 4, pp. 275–333; (b) Stern, E. S. In The Alkaloids;
Manske, R. H. F., Ed.; Academic Press: New York,
1960; Vol. 7, pp. 473–521; (c) Pelletier, S. W.; Keith, L.
H. In The Alkaloids; Manske, R. H. F., Ed.; Academic
Press: New York, 1970; Vol. 12, pp. 1–206; (d) Pel-
letier, S. W.; Mody, N. V. In The Alkaloids; Manske,
R. H. F.; Rodrigo, R. G. A., Eds.; Academic Press:
New York, 1979; Vol. 17, pp. 1–103; (e) Pelletier, S.
W.; Mody, N. V. In The Alkaloids; Manske, R. H. F.;
Rodrigo, R. G. A., Eds.; Academic Press: New York,
1981; Vol. 18, pp. 99–216; (f) Amiya, T.; Bando, H. In
The Alkaloids; Brossi, A., Ed.; Academic Press: New
York, 1988; Vol. 34, pp. 95–179; (g) Wang, F.-P.;
Liang, X.-T. In The Alkaloids; Cordell, G. A., Ed.;
Academic Press: New York, 1992; Vol. 42, pp. 151–
247.
1
tion of its quaternary salt 33. H NMR spectral study
confirmed the structure of 1, and the existence of both
long-range coupling (0.8 Hz) between H-6 and H-20
(Fig. 3), and NOE enhancement of the Hb-19 signal on
irradiation of the H-20 signal ascertained the N–C-20
1
bond connection during the SOCl2 treatment. In the H
3. (a) Isolation: Jacobs, W. A.; Craig, L. C. J. Biol.
Chem. 1942, 143, 605–609; (b) X-ray structure determi-
nation: Przybylska, M. Can. J. Chem. 1962, 40, 566–
568.
NMR spectra of nominine and kobusine, spin–spin
coupling between H-6 and H-20 was also observed as
shown in Fig. 3.23
4. Shibanuma, Y.; Okamoto, T. Chem. Pharm. Bull. 1985,
33, 3187–3194.
J6,20 = 0.8 Hz
2%
NOE
HO
H
H
H
5. Two papers entitled a synthesis of hetisine-type alka-
loids have been reported. (a) Van der Bean, J. L.; Bick-
elhaupt, F. Recl. Trav. Chim. Pays-Bas 1975, 94,
109–112; (b) Kwak, Y.-S.; Winkler, J. D. J. Am. Chem.
Soc. 2001, 123, 7429–7430.
OBz
OH
OH
N
H
H
N
H
H
N
H
H
H
H
H
Hb
Ha
J6,20 = 1.1 Hz
3% NOE
J6,20 = 1.0 Hz
6. Muratake, H.; Nakai, H. Tetrahedron Lett. 1999, 40,
2355–2358.
1
Nominine
Kobusine
7. Reviews: (a) Snider, B. B. Acc. Chem. Res. 1980, 13,
426–432; (b) Mikami, K.; Shimizu, M. Chem. Rev.
Figure 3.
1992, 92, 1021–1050.
Johnson, W. S.; Kinnel, R. B. J. Am. Chem. Soc. 1966,
88, 3861–3862.
A similar acetal-ene reaction:
Here we have succeeded for the first time in the synthe-
sis of a hexacyclic compound 1, which has an almost
full structure of hetisan lacking only the six-membered
ring with the exo methylene group. Applying this work,
synthetic studies of nominine, kobusine, and pseu-
dokobusine are now in progress.
8. Snyder, L.; Meyers, A. I. J. Org. Chem. 1993, 58,
7507–7515.
9. Wittig, G.; Reiff, H. Angew. Chem. 1968, 80, 8–15.
10. Miller, R. D.; McKean, D. R. Synthesis 1979, 730–
732.
11. Reuss, R. H.; Hassner, A. J. Org. Chem. 1974, 39,
1785–1787.
Acknowledgements
12. Belsky, I. J. C. S. Chem. Commun. 1977, 237.
13. Stelieu, K.; Poupart, M. J. Org. Chem. 1985, 50, 4971–
4973.
14. NOESY data of 16 showed a correlation between 5-H
and 9-H, which confirmed the structure of 16.
15. Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226–
2227.
The authors thank Professor Emeritus Shin-ichiro
Sakai and Professor Hiromitsu Takayama of Chiba
University for providing the precious samples of natu-
ral nominine and kobusine.
16. Wick, A. E.; Felix, D.; Steen, K.; Eschenmoser, A.
Helv. Chim. Acta 1964, 47, 2425–2429.
17. Myers, A. G.; Yang, B. H.; Kopecky, D. J. Tetra-
hedron Lett. 1996, 37, 3623–3626.
References
1. (a) Pelletier, S. W.; Page, S. W. Nat. Prod. Rep. 1984,
1, 375–386; (b) Pelletier, S. W.; Page, S. W. Nat. Prod.

H. Muratake, M. Natsume / Tetrahedron Letters 43 (2002) 2913–2917
2917
Lett. 1984, 25, 2515–2518.
18. Jung, R. E.; Usai, Y.; Vu, C. T. Tetrahedron Lett. 1987,
28, 5977–5980.
19. Cheng, Y.; Liu, W.; Chen, S. Synthesis 1980, 223–224.
20. Nagata, W.; Yoshioka, M. Organic Reactions; Wiley:
New York, 1977; Vol. 25, pp. 255–476.
22. The C-20 hydroxy group was sterically hindered as far as
the nitrogen atom was protected, and was inert to any
acylation or sulfonation reagent.
23. NOE experiments were impossible due to close vicinity of
the chemical shifts of Hb-19 and H-20.
21. Hanessian, S.; Delorme, D.; Dufresne, Y. Tetrahedron