A diversified approach to the synthesis of highly functionalized novel azines and azoles from 2H-pyran-2-ones

Article abstract of DOI:10.3987/COM-12-S(N)1

One-pot regioselective synthesis of 2,6-diarylpyrimidines (4), 2,6-disubstituted pyridines (5 and 6), 5-aryl-3-cyanomethyl-1H-pyrazoles (7) through base catalyzed ring transformation of suitably functionalized 2H-pyran-2-ones by arylamidines, cyanamide, ammonium carbonate, hydrazines respectively has been developed. Our synthetic approach opened a new avenue for the regioselective synthesis of 2,6-diarylpyrimidines and 2,6-disubstituted pyridines and cyanomethylpyrazoles. The procedure is very simple, efficient and economically viable. The protocol provides an efficient methodology for the preparation of new class of medicinally useful and highly functionalized azines and azoles from single precursor 2H-pyran-2-ones.

Full text of DOI:10.3987/COM-12-S(N)1

HETEROCYCLES, Vol. 75, No. 2, 2008
285
HETEROCYCLES, Vol. 75, No. 2, 2008, pp. 285 - 290. © The Japan Institute of Heterocyclic Chemistry
Received, 3rd September, 2007, Accepted, 29th October, 2007, Published online, 30th October, 2007. COM-07-11211
ASYMMETRIC SYNTHESIS OF STELLETTAMIDES A AND C
Naoki Yamazaki,* Takeshi Suzuki, Yuta Yoshimura, Chihiro Kibayashi, and
Sakae Aoyagi*
School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, 1432-1
Horinouchi, Hachioji, Tokyo 192-0392, Japan
Abstract – The first synthesis of the marine indolizidine alkaloids,
(+)-stellettamide A and (–)-stellettamide C, has been achieved through a coupling
reaction with the aminomethylindolizidine fragment and the chiral or achiral
trienecarboxylic acid fragment, both of which could be derived from farnesol as a
common precursor.
Stellettamide A (1) was isolated from the marine sponge genus Stelletta, which was collected near
Shikinejima island of Japan in 1990 by Fusetani and coworkers,1 as the first example of an indolizidine
class of marine alkaloids. Stellettamide A exhibits calmodulin inhibitory activity2 as well as antifungal
and cytotoxic activities against tumor cell lines.1 The absolute stereochemistry of stellettamide A was
established on the basis of the enantioselective synthesis of the antipodal (–)-1 by Carreira3 in 1997. In
1997, Shin reported a closely related alkaloid stellettamide B,4 which was obtained from the same genus
Stelletta collected near Keomun island of Korea, and disclosed structure 2. The configuration at the C6”
position in stellettamide B was revised to be R by our total synthesis5 as shown in (–)-3. Soon after, a
new compound of this class, stellettamide C (4) was isolated from the Japanese Stelletta sponge.6 The
H
+
N
Me
H
H
N
H
N
+
N
Me
Cl–
H
H
O
O
H2PO4–
stellettamide A [(+)-1]
proposed structure 2 for stellettamide B
H
+
N
+
H
H
N
H
N
Me
Cl–
Me
N
6"
H
H
O
O
X–
revised structure (–)-3 for stellettamide B
proposed structure ! for stellettamide C

286
HETEROCYCLES, Vol. 75, No. 2, 2008
chemical features of this family are recognized to consist of trienecarboxylic acid fragments as structural
variations and the aminomethylindolizidine fragment as a common component. Though there are, so far,
no additional examples of the synthesis of stellettamides, two synthetic approaches focusing on the
synthesis of the aminomethylindolizidine fragment in an enantiomerically pure form7 and a racemic form8
have been reported. In this paper we present the first synthesis of the enantiomerically correct form of
natural stellettamide A and (1S,4S,8aR)-stellettamide C (4).
Our synthetic plan was based on a straightforward application employing the connection of
trienecarboxylic acid fragment 6 with the formerly synthesized aminomethylindolizidine fragment 7 via
amide linkage formation (Scheme 1). For the synthesis of aminomethylindolizidine fragment 7, we
H
H
+
N
Me
+
N
Me
H
N
R
OH
H2N
R
X–
X–
+
H
H
O
O
6
7
5
Scheme 1
intend to improve some points, and significant improvements (11→12 and 13→14) were made as
illustrated in Scheme 2. Firstly, in the formation of N,O-acetal 12, the formation required two steps
from glutaric anhydride (8) and the combined chemical yield was somewhat poor (41%). The yield was
improved by the condensation of 2-(1-aminoethyl)phenol 9 with formyl ester 11, giving the cyclic
NH2
Me
OH
O
N
O
9
Me
ref. 5
O
O
O
8
OH
10
Vitride®, then HCl
(2 steps, 41%)
H
O
O
O
ref. 5
9
O
N
O
N
OMe
MeO
H
benzene reflux
(91%)
Me
Boc
11
13
12
Br
CO2Me
H
(5 equiv)
H
H
ref. 5
LHMDS (2 equiv)
H2N
N
–78 °C ! rt
(86% de, 73% combined yield)
N
Boc
15
14
Scheme 2

HETEROCYCLES, Vol. 75, No. 2, 2008
287
N,O-acetal 12 in a 91% yield.9 Secondly, in the allylation process of the piperidinylmethyl ester
(13→14), we sometimes experienced substantial loss of the optical purities (~90–60% ee) of the resulting
allylated products. This serious drawback was avoided by adding an excess amount of allylbromide (5
equiv) to the base (LHMDS) (2 equiv).10 No racemization with 14 occurred in the allylation reaction
(>99% ee, Chiralpak AD, hexane–IPA–Et2NH, 1900:100:5, v/v). With the aminomethylindolizidine 15
having high enantiomeric purity in hand, we next investigated the synthesis of trienecarboxylic acid
fragments for stellettamides A and C.
Farnesol (16) was treated under Sharpless asymmetric epoxidation conditions using L-DET and t-BuOOH
to give the epoxy alcohol 17.11 Reductive cleavage of the epoxide in the presence of BF3·OEt2 gave
glycol 18,12 which was consecutively applied to oxidative cleavage using NaIO4 to yield aldehyde 19.
Wittig olefination followed by basic hydrolysis afforded trienecarboxylic acid 21 for the fragment of
stellettamide A.
The acid 21 and the indolizidine fragment 15 were connected by the DCC–DMAP coupling to yield
amide 22. Finally iodomethylation of 22 with iodomethane, followed by a counter-ion exchange3 from
the iodide to the phosphate with sodium biphosphate furnished (+)-stellettamide A. The synthetic
material coincided in all aspects with those of the reported stellettamide A as well as in terms of optical
NaBH3CN, BF3·OEt2
AE, L-DET
(87%)
O
OH
OH
(72%)
farnesol (16)
17
NaIO4
Ph3P=CHCO2Me
(97%)
H
OH
dioxane–H2O
(76%)
O
OH
18
19
H
H
N
15, DCC, DMAP
N
CO2R
(41%)
H
O
LiOH
H2O–THF
(97%)
20: R = Me
21: R = H
22
1) MeI, MeOH
2) NaH2PO4
H
Me
+
N
H
N
H2PO4–
(58%, 2 steps)
H
O
(+)-stellettamide A [(+)-1]
Scheme 3
rotation [[α]22D +23.8 (c 0.79, EtOH), lit.,1 [α]D +23.1 (c 0.3, EtOH)]. This sequence established the first
total synthesis of (+)-stellettamide A [(+)-1] as the enantiomerically correct form of a natural product.

288
HETEROCYCLES, Vol. 75, No. 2, 2008
NaOH, EtOH
(82%)
NaCN, DMSO
(83%)
CN
Cl
23
24
H
H
N
15, DCC, DMAP
N
CO2H
(72%)
H
O
25
26
1) MeI, MeOH
2) AgCl
H
+
N
H
N
Me
X–
H
(76%)
O
(–)-stellettamide C (4, X = Cl)
Scheme 4
Trienecarboxylic acid fragment 25, for the synthesis of stellettamide C, was obtained by homologation of
the farnesyl chloride (23) followed by hydrolysis under usual conditions (Scheme 4).
Dehydrocondensation of the resulting carboxylic acid 25 with the indolizidine 15 gave nor-stellettamide
C (26). By treatment of 26 with iodomethane, followed by silver chloride,13 yielded stellettamide C (4,
X = Cl). The complete exchange of the counter ion from the iodide ion to the chloride ion was
1
confirmed by the measurement of an energy dispersive X-ray spectrum as shown in Figure 1. Both H
and 13C NMR spectra of the synthetic material were found to be identical to those of natural stellettamide
C, though the careful measurement of the optical rotation14 [[α]20 –1.5 (c 0.34, MeOH)] showed the
D
opposite sign to the natural stellettamide C (4, X = undefined) [lit.,6 [α]23D +1.1 (c 0.32, MeOH)]. These
results suggest that the absolute stereochemistry of natural stellettamide C was the antipodal structure of 4,
having (1R,4R,8aS)-indolizidine moiety, although the absolute stereochemistry for the indolizidine part of
stellettamides A and B was already established to be 1S,4S,8aR by asymmetric syntheses.3,5 This
Figure 1. Energy dispersive X-ray spectrum of 4 was obtained by electron
bombardments at 20 kV on a Horiba EMAX scanning electron microscope.

HETEROCYCLES, Vol. 75, No. 2, 2008
289
contradiction might be the consequence of the weak degrees of the optical rotation value of stellettamide
C. In addition, the counter ion of the natural material (4, X = undefined) has not been established and
thereby it is possible that the difference of the counter ions might influence the sign of the optical
rotation.
In summary, we have synthesized the natural form of (+)-stellettamide A and the proposed structure of
(1S,4S,8aR)-(–)-stellettamide C by applying a convergent strategy to aminomethylindolizidine fragment
with chiral or achiral trienecarboxylic acid fragments.
ACKNOWLEDGEMENTS
This work was partly supported by a Grant-in-Aid for Science Research from Japan Society for the
Promotion of Sciences.
REFERENCES AND NOTES
1. H. Hirota, S. Matsunaga, and N. Fusetani, Tetrahedron Lett., 1990, 31, 4163.
2. Y. Abe, S. Saito, M. Hori, H. Ozaki, N. Fusetani, and H. Karaki, Brit. J. Pharmacol., 1997, 121,
1309.
3. G. A. Whitlock and E. M. Carreira, J. Org. Chem., 1997, 62, 7916.
4. J. Shin, Y. Seo, K. W. Cho, J.-R. Rho, and C. J. Sim, J. Nat. Prod., 1997, 60, 611.
5. N. Yamazaki, W. Dokoshi, and C. Kibayashi, Org. Lett., 2001, 3, 193.
6. S. Matsunaga, T. Yamashita, S. Tsukamoto, and N. Fusetani, J. Nat. Prod., 1999, 62, 1202.
7. R. A. Pilli, P. R. Zanotto, and M. A. Böckelmann, Tetrahedron Lett., 2001, 42, 7003.
8. K. Ishii, T. Sone, T. Shigeyama, M. Noji, and S. Sugiyama, Tetrahedron, 2006, 62, 10865.
9. A solution of 11 (98.0 mg, 0.753 mmol) and 9 (69.0 mg, 0.503 mmol) in benzene (5 mL) was heated
to reflux using a Dean-Stark apparatus for water removal. After 24 h, the mixture was concentrated
in vacuo. The residue was purified by column chromatography on silica gel (hexane–EtOAc, 2:1)
to give 12 (99.5 mg, 91%) as a colorless oil.
10. To a stirred solution of 13 (100 mg, 0.389 mmol) in THF (4 mL) was added LHMDS (1.0 M in THF,
0.78 mL, 0.780 mmol) at –78 °C. After stirring at –78 °C for 30 min, allylbromide (238 mg, 1.96
mmol) was added and stirring was continued for additional 30 min. The reaction mixture was
allowed to warm to room temperature and diluted with ether (20 mL). Water (10 mL) was added,
and the mixture was extracted with ether (3 × 30 mL). The combined ethereal extracts were
washed with brine, dried (MgSO4), and evaporated in vacuo. Purification of the residue by column
chromatography (hexane–EtOAc, 10:1) gave 14 (77.7 mg, 67%) as a colorless oil.
11. E. J. Corey and D.-C. Ha, Tetrahedron Lett., 1988, 29, 3171.

290
HETEROCYCLES, Vol. 75, No. 2, 2008
12. Data for selected new compounds: Compound 18: [α]28D –13.2 (c 1.01, CHCl3); IR (neat) 3376, 2966,
1
2924, 1668 cm–1; H NMR (400 MHz, CDCl3) δ 0.93 (3H, d, J = 6.8 Hz), 1.14–1.27 (1H, m),
1.41–1.75 (2H, m), 1.60 (6H, s), 1.68 (3H, d, J = 0.5 Hz), 1.91–2.13 (6H, m), 3.51–3.68 (3H, m),
5.03–5.15 (2H, m). Compound 19: [α]23D +23.6 (c 1.37, CHCl3); IR (neat) 2968, 2922, 2856, 1727
cm–1; 1H NMR (500 MHz, CDCl3) δ 1.08 (3H, d, J = 7.1 Hz), 1.35–1.43 (1H, m), 1.58 (3H, s), 1.59
(3H, s), 1.66 (3H, s), 1.72–1.80 (1H, m), 1.95–2.01 (2H, m), 2.01–2.09 (4H, m), 2.34 (1H, tdd, J =
13.8, 6.9, 1.8 Hz), 5.04–5.11 (2H, m), 9.61 (1H, d, J = 1.8 Hz). Compound 21: [α]30D +47.7 (c 1.21,
CHCl3); IR (neat) 2965, 2918, 1696, 1651, 1419 cm–1; 1H NMR (400 MHz, CDCl3) δ 1.07 (3H, d, J
= 6.8 Hz), 1.35–1.53 (2H, m), 1.58 (3H, s), 1.60 (3H, s), 1.68 (3H, d, J = 0.5 Hz), 1.93–2.03 (4H, m),
2.03–2.11 (2H, m), 2.29–2.42 (1H, m), 5.04–5.13 (2H, m), 5.79 (1H, dd, J = 15.7, 0.8 Hz), 6.99 (1H,
dd, J = 15.6, 7.9 Hz).
13. R. D. Hawarth, W. H. Perkin, Jr., and H. S. Pink, J. Chem. Soc., 1926, 1709.
14. The optical rotation was measured using a 100 mm length quartz cell with a methanolic solution,
which was prepared 1 hour in advance by using a 2-mL volumetric flask.