Enantioselective syntheses of heliannuols G and H

Article abstract of DOI:10.3987/COM-05-S(K)21

Enantioselective total syntheses of the proposed structures for heliannuols G and H have been accomplished. However, their spectral properties were not identical with those of the natural products.

Full text of DOI:10.3987/COM-05-S(K)21

HETEROCYCLES, Vol. 66, 2005
69
HETEROCYCLES, Vol. 66, 2005, pp. 69 – 73. © The Japan Institute of Heterocyclic Chemistry
Received, 23rd August, 2005, Accepted, 7th October, 2005, Published online, 11th October, 2005. COM-05-S(K)21
ENANTIOSELECTIVE SYNTHESES OF HELIANNUOLS G AND H#
Sachie Morimoto,† Mitsuru Shindo,‡ and Kozo Shishido*†
†Graduate School of Pharmaceutical Sciences, The University of Tokushima,
1-78-1
Sho-machi,
Tokushima
770-8505,
Japan.
E-mail:
shishido@ph.tokushima-u.ac.jp
‡Institute for Materials Chemistry and Engineering, Kyushu University, 6-1,
Kasugako-en, Kasuga 816-8580, Japan
Abstract – Enantioselective total syntheses of the proposed structures for
heliannuols G and H have been accomplished. However, their spectral properties
were not identical with those of the natural products.
The heliannane type sesquiterpenes heliannuols G and H were isolated by Macías from a fresh leaf
aqueous extract of Helianthus annuus L. SH-222 and YPP and exhibited phytotoxic allelopathic activity.1
Because of their interesting structural features, an eight-membered oxygen-containing heterocycle fused
to an aryl ring2 and two contiguous tertiary stereogenic centers, we decided to try to synthesize these
diastereomeric natural products in optically active form. We report here the first enantioselective total
syntheses of the proposed structures for heliannuols G (1) and H (2), which are not identical with the
natural products.
OH
OH
HO
Me
HO
Me
O
O
Heliannuol G (1)
Heliannuol H (2)
Figure 1
#Dedicated to the memory of the late Professor Kenji Koga.

70
HETEROCYCLES, Vol. 66, 2005
Our strategy is illustrated in Scheme 1. We thought that the heliannuols G and H might be derived from a
ring closing metathesis (RCM)2, 3 of the diene (3), which could be constructed by a palladium catalyzed
dimethylallylation4 of the phenol (4), a mixture of diastereoisomers at the future C8 position. This in turn
would be synthesized from the optically active alcohol (5), which has already been prepared by
lipase-mediated desymmetrization of the prochiral σ-symmetrical diol (6).2 (Scheme 1)
OH
OH
OH
RO
Me
RCM
HO
Me
RO
Me
8
OH
O
O
1, 2
3
4
OH
OH
OH
MeO
Me
OAc
MeO
Me
OMe
OMe
5
6
Scheme 1. Retrosynthetic Analysis
PPL-Mediated desymmetrization of the prochiral diol (6) provided the optically active acetate [(R)-5]2
([α]D + 12.6˚ (c 1.01, CHCl3)) in 34% yield. The major product of this reaction was the starting diol and
the yield of 5 based on the consumed diol was quantitative. The enantiomeric excess (ee) was determined
to be 84% by HPLC analysis on a Chiralcel AD column. Sequential tosylation, NaBH4 reduction in hot
DMSO and alkaline hydrolysis of 5 produced the alcohol (7)2 in 55% overall yield. Fortunately, it was
obtained as a crystalline compound and only one recrystallization was necessary to obtain it optically pure.
Dess-Martin oxidation of the optically pure alcohol (7) provided the aldehyde which was immediately
reacted with vinylmagnesium chloride to give an inseparable 4:1 diastereoisomeric mixture of the alcohol
(8) in 77% yield for the 2 steps. At this point, a decrease in optical purity (ca. 80% ee by HPLC analysis
on a Chiralcel OD column) was observed. Since separation of the diastereoisomers proved to be difficult
at this stage, the mixture was treated with TBSOTf. The resulting silyl ether was sequentially oxidized
with CAN and reduced with Na2S2O4 to provide the hydroquinone (9) in good overall yield. The sterically
less hindered phenolic hydroxyl group was protected successfully as the TBDPS ether to give a mixture
of 10 and 11. The two diastereomers were separated cleanly by preparative HPLC to provide the anti-
(10) and syn-alkenylphenol (11) in 69% and 20% yields, respectively. Although the stereostructures of
the products could not be determined at this stage, they were deduced in terms of the Felkin-Anh model
for the addition of the Grignard reagent to the aldehyde. Confirmation was made by the conversion of the
major anti-isomer into the final compound, whose structure was established by X-Ray crystallographic

HETEROCYCLES, Vol. 66, 2005
71
analysis. These compounds were treated independently with i-butyl-2-methyl-3-buten-2-yl carbonate in
the presence of catalytic Pd(Ph3P)4 to give the desired dienes (12) and (13) in 84% and 72% yields,
respectively. (Scheme 2)
OH
OH
OH
OAc
PPL
vinyl acetate
MeO
Me
OH
MeO
Me
MeO
Me
ref. 2
Et2O, rt, 34 h
OMe
OMe
OMe
7
6
5
84% ee
(HPLC, Chiralcel AD)
>99% ee
(HPLC, Chiralcel OD)
OH
OTBS
MeO
Me
HO
c - e
f
a, b
OMe
Me
OH
8
9
OTBS
OTBS
OTBS
OTBS
TBDPSO
Me
TBDPSO
g
OH
Me
O
O
10
12
13
TBDPSO
Me
TBDPSO
Me
g
OH
11
Scheme 2. Reagents and Conditions: (a) Dess-Martin periodinane, CH2Cl2, rt, 30 min; (b)
vinylmagnesium chloride, THF, 0 °C, 10 min, 77% (2 steps); (c) TBSOTf, 2,6-lutidine, CH2Cl2, rt, 15
min, 83%; (d) Ce(NH4)2(NO3)6, MeCN, H2O, rt, 20 min, 89%; (e) Na2S2O4, THF, rt, 50 min, 98%; (f)
TBDPSCl, imidazole, 4-DMAP, CH2Cl2, rt, 1.5 h, 69% for 10, 20% for 11; (g)
i-BuOCOOC(CH3)2CH=CH2, Pd(Ph3P)4, THF, rt, 84% for 12, 72% for 13.
The stage was now set to construct the oxygen-containing eight-membered heterocycle by the key RCM
reaction2 of the dienes (12) and (13). Treatment of the major diastereomer (12) with 60 mol% of
(tricyclohexyl)phosphine[1,3-bis(2,4,6-trimethylphenyl)-4,5-dihydroimidazol-2-ylidene][benzylidene]rut-
henium(IV) dichloride (15) in refluxing CH2Cl2 provided the cyclized product (14) in 77% yield.
Deprotection of 14 with HF•pyridine complex in pyridine at room temperature gave 25 with the structure
proposed for heliannuol H in 90% yield. The structure was confirmed by the X-Ray crystallographic
analysis6 as shown in Figure 2. We also examined the opposite sequence, namely deprotection followed
by RCM. Interestingly, in the case of 16, the RCM proceeded smoothly even at room temperature in the

72
HETEROCYCLES, Vol. 66, 2005
presence of 10 mol% of 15 to give 2 in 84% yield for the 2 steps. For the minor isomer (13), however,
sequential deprotection and RCM did not give the desired product (1) but instead a mixture of
unidentified products. On the contrary, RCM followed by deprotection of the resulting 17 furnished 1 in
61% overall yield. However, the 1H-NMR spectra7 of the compounds (1) and (2) thus prepared were not
identical with those of the natural heliannuols G and H. (Scheme 3)
Figure. 2. The molecular structure of 2.
OTBS
OTBS
OH
TBDPSO
Me
TBDPSO
Me
HO
Me
15 (60 mol%)
HF•pyridine
CH2Cl2
reflux, 2 d
pyridine
rt, 13 h
O
O
O
12
14
2
77%
90%
OH
Mes N
N Mes
HO
Me
HF•pyridine
15 (10 mol%)
Cl
Cl
12
2
Ru
PCy3
pyridine
rt, 11 h
CH2Cl2
rt, 6 h
Ph
O
15
16
91%
92%
OTBS
OTBS
OH
TBDPSO
Me
TBDPSO
Me
HO
Me
15 (50 mol%)
TBAF
CH2Cl2
reflux, 2 d
THF, rt, 4 d
64%
O
O
O
17
1
13
95%
Scheme 3
In summary, we have completed the first enantiocontrolled total syntheses of the compounds with the
structures proposed for the natural heliannuols G and H. However, it was revealed by the present

HETEROCYCLES, Vol. 66, 2005
73
synthesis that the initial structural assignments were incorrect. Determination of the real structures for the
natural heliannuols G and H is now underway in our laboratories.
ACKNOWLEDGEMENTS
We thank Professor Francisco A. Macías (University of Cádiz, Spain) for providing us with the spectral
data of the natural heliannuols G and H. This work was supported financially by a Grant-in-Aid for
Scientific Research (A) (No. 16209001) from the Japan Society for the Promotion of Science.
REFERENCES AND NOTES
1. F. A. Macías, R. M. Varela, A. Torres, and J. M. Molinillo, J. Nat. Prod., 1999, 62, 1636.
2. H. Kishuku, M. Shindo, and K. Shishido, Chem. Commun., 2003, 350.
3. T. Kamei, M. Shindo, and K. Shishido, Tetrahedron Lett., 2003, 44, 8505.
4. T. Kaiho, T. Yokoyama, H. Mori, J. Fujiwara, T. Nobori, H. Odaka, J. Kamiya, M. Maruyama, and
T. Sugawara, JP 06128238, 1994 [Chem. Abstr., 1995, 123, 55900].
1
5. The optical purity was determined to be 79% ee by the H-NMR spectrum of the corresponding
di-MTPA ester of 2.
6. Crystal data for 2: C15H20O3, M = 248.32, orthorhombic, a=8.397(1)Å, b=10.458(2)Å, c=16.462(2)Å,
V=1445.6(4)Å3, Dcalc=1.141 g/cm3, P212121 (#19), Z = 4, µ(MoKa)=0.78 cm-1. No. of observed
reflections, 13037, No. of variable parameters, 183. The final R=0.051 (all data), R1=0.044
(I>2.00s(I)), wR(F2)=0.134 (all data).
1
7. 1: H-NMR (400 MHz, CDCl3, -40 ˚C) δ 1.35 (d, 3H, J=7.6 Hz), 1.35 (s, 3H), 1.69 (s, 3H), 2.20 (s,
3H), 3.04-3.06 (m, 1H), 4.30-4.36 (m, 1H), 5.40 (d, 1H, J=12.4 Hz), 5.77-5.82 (m, 1H), 6.62 (s, 1H),
6.77 (s, 1H). 2: 1H-NMR (400 MHz, CDCl3) δ 1.32 (d, 3H, J=6.8 Hz), 1.36 (s, 3H), 1.58 (d, 3H,
J=7.6 Hz), 1.86 (s, 1H, D2O exchangeable), 2.16 (s, 3H), 2.62 (dq, 1H, J=6.8, 9.2 Hz), 4.55 (s, 1H,
D2O exchangeable), 5.19 (d, 1H, J=11.2 Hz), 5.35 (dd, 1H, J=7.6, 8.0 Hz), 5.54 (dd, 1H, J=7.6, 11.2
Hz), 6.48 (s, 1H), 6.68 (s, 1H).