Total synthesis and absolute configuration of (-)-gummiferol

Article abstract of DOI:10.1021/ol201301b

Stereoselective synthesis of two possible diastereomers of (-)-gummiferol was accomplished by the stepwise epoxidation and Cadiot-Chodkiewicz reaction as the key transformations. Detailed comparison of their ¹H and ¹³C NMR data and specific rotation with those of the natural product led to the absolute structural elucidation of (-)-gummiferol.

Full text of DOI:10.1021/ol201301b

ORGANIC
LETTERS
2011
Vol. 13, No. 14
3644–3647
Total Synthesis and Absolute
Configuration of (ꢀ)-Gummiferol
Hiroyoshi Takamura,* Hiroko Wada, Nan Lu, and Isao Kadota*
Division of Chemistry and Biochemistry, Graduate School of Natural Science
and Technology, Okayama University, 3-1-1 Tsushimanaka, Kita-ku,
Okayama 700-8530, Japan
takamura@cc.okayama-u.ac.jp; kadota-i@cc.okayama-u.ac.jp
Received May 13, 2011
ABSTRACT
Stereoselective synthesis of two possible diastereomers of (ꢀ)-gummiferol was accomplished by the stepwise epoxidation and
1
CadiotꢀChodkiewicz reaction as the key transformations. Detailed comparison of their H and 13C NMR data and specific rotation with those
of the natural product led to the absolute structural elucidation of (ꢀ)-gummiferol.
(ꢀ)-Gummiferol was isolated from the 50% MeOH/
CHCl3 extract of the leaves of Adenia gummifera in 1995by
Wall et al.1 This molecule exhibits a cytotoxicity against
various cell lines including strong activity against P-388
(ED50: 0.03 μg/mL) and U-373 (ED50: 0.05 μg/mL). The
planarstructure of(ꢀ)-gummiferol, which hasfeatured the
triacetylene and diepoxide moieties, was elucidated on the
CadiotꢀChodkiewicz reaction3 through a combination
of the bromoacetylenes 2 and 3 and diacetylene 4 to yield
the triacetylenes 5 and 6, respectively.
3
1
1
basis of JH,H analysis and the Hꢀ H COSY, HMQC,
and HMBC spectra (Figure 1). However, the stereoche-
mistries of the contiguous epoxide moiety at the C8 to C11
positions were not determined. Herein, we report the
stereoselective synthesis of two possible diastereomers of
(ꢀ)-gummiferol and comparison of their spectroscopic data
with those of the natural product, which has resulted in the
absolute structural determination of (ꢀ)-gummiferol.
Figure 1. Planar structure of (ꢀ)-gummiferol.
First, we examined the stereocontrolled synthesis of
the syn-diepoxide 5. Sharpless asymmetric epoxida-
tion2 of dienol 74 with (þ)-diisopropyl tartrate (DIPT)
provided epoxy alcohol 8 as a single stereoisomer
(Scheme 2).5 Although Sharpless asymmetric epoxidation
Our synthetic stategy toward two possible diastereomers
of (ꢀ)-gummiferol, diepoxides 5 and 6, is outlined in
Scheme 1. The key synthetic intermediate 1 would be
stereoselectively converted to syn-diepoxide
anti-diepoxide 3 via Sharpless epoxidation, respectively.2
The triacetylene unit could be constructed by
2 and
(3) (a) Brandsma, L. Preparative Acetylenic Chemistry, 2nd ed.; Else-
vier: Amsterdam, 1988. (b) Siemsen, P.; Livingston, R. C.; Diederich, F.
Angew. Chem., Int. Ed. 2000, 39, 2632.
a
(4) Zhang, P.; Morken, J. P. J. Am. Chem. Soc. 2009, 131, 12550.
(5) The optical purity of 8 was judged by comparison of the 1H NMR
spectrum of its (R)-MTPA ester with that of the (R)-MTPA ester
prepared from the racemic epoxy alcohol, which was available from 7
using m-CPBA.
(1) Fullas, F.; Brown, D. M.; Wani, M. C.; Wall, M. E.; Chagwedera,
T. E.; Farnsworth, N. R.; Pezzuto, J. M.; Kinghorn, A. D. J. Nat. Prod.
1995, 58, 1625.
(2) Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5974.
r
10.1021/ol201301b
Published on Web 06/21/2011
2011 American Chemical Society

is a well-established method, the unambiguous stereoche-
mical elucidation of 8 was performed by the derivatization
and the modified Mosher method.6 When the epoxy
alcohol 8 was treated with Red-Al,7 epoxide ring opening
occurred regioselectively at the C11 position due to the
CꢀO bond activation by the adjacent π-orbital to afford
1,2-diol 9. The primary hydroxy group of 9 was selectively
protected to give the corresponding TBDPS ether. The
resulting secondary alcohol was converted to MTPA esters
(S)- and (R)-10 by the standard conditions (MTPACl/
Et3N/DMAP). Figure 2 describes the ΔδSꢀR values of the
(S)- and (R)-10.6 The signs at the left side of the C10 positon
were negative, and those at the right side were positive.
Therefore, the absolute configuration of 10 was assigned to
be 10R, which resulted in the structural elucidation of 8.
WadsworthꢀEmmons reaction,9 and (3) DIBAL-H
reduction.10 Second Sharpless asymmetric epoxidation2
with (þ)-DIPT proceeded smoothly to provide syn-diep-
oxide 12 as a sole product in 80% yield. The diepoxide 12
was derivatized for the structural determination of the
resulting 8,9-epoxide moiety. Thus, diimide reduction of
the olefin unit and subsequent regioselective reduction of
the 8,9-epoxide moiety with Red-Al7 gave 1,3-diol 13.
Protection of the primary alcohol of 13 as the TBDPS
ether followed by reaction with MTPACl afforded esters
(S)- and (R)-14. In the observed ΔδSꢀR values of the (S)-
and (R)-14, the signs at the left side of the C9 positon were
positive, and those at the right side were negative (Figure 3).
Therefore, the absolute stereochemistry of 14 was eluci-
dated to be 9R, leading to the structural determination of
the 8,9-epoxide moiety of 12.
Scheme 1. Synthetic Plan of 5 and 6
Figure 2. Chemical shift differences (ΔδSꢀR) of (S)- and (R)-10.
R = MTPA. MTPA = R-methoxy-
R-(trifluoromethyl)phenylacetyl.
Scheme 3. Synthesis and Derivatization of 12
Scheme 2. Synthesis and Derivatization of 8
Further transformation of 12 to the syn-diepoxide 5 is
described in Scheme 4. ParikhꢀDoering oxidation8 of 12
As shown in Scheme 3, the epoxy alcohol 8 was con-
verted to allylic alcohol 11 by the following three step
sequence: (1) ParikhꢀDoering oxidation,8 (2) Hornerꢀ
(9) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25,
2183.
(6) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092.
(7) Finan, J. M.; Kishi, Y. Tetrahedron Lett. 1982, 23, 2719.
(8) Parikh, J. R.; Doering, W. v. E. J. Am. Chem. Soc. 1967, 89, 5505.
(10) When we tried DIBAL-H reduction (2.2 equiv), we observed the
epoxide opening as a side reaction. Therefore, we chose the stepwise
DIBAL-H reduction. The first reaction (1.1 equiv) gave the correspond-
ing aldehyde, and the second reduction (1.1 equiv) provided 11.
Org. Lett., Vol. 13, No. 14, 2011
3645

Next, we investigated the stereoselective synthesis of the
anti-diepoxide 6 (Scheme 5). The transformation from the
key synthetic intermediate 11 to 6 is similar to that toward
5. The allylic alcohol 11 was subjected to Sharpless asym-
metric epoxidation2 using (ꢀ)-DIPT, giving anti-diepoxide
19 as a single stereoisomer in 81% yield.15 Parikhꢀ
Doering oxidation8 of 19 followed by dibromoole-
fination11 and desilylation/dehydrobromination12b pro-
vided bromoacetylene 20 in 57% yield in three steps. The
bromoacetylene 20 (1.0 equiv) was coupled with the dia-
cetylene 17 (1.1 equiv) by CadiotꢀChodkiewicz reaction3
under the optimized conditions affording triacetylene 21.14
The final transformation, acetylation and desilylation, was
performed to provide the anti-diepoxide 6.
Figure 3. Chemical shift differences (ΔδSꢀR) of (S)- and (R)-14.
R = MTPA. MTPA = R-methoxy-
R-(trifluoromethyl)phenylacetyl.
and subsequent one-carbon homologation with CBr4/
PPh3/Et3N11,12 gave dibromoolefin 15. Treatment of 15
with TBAF (4.0 equiv) induced desilylation and dehydro-
bromination simultaneously to afford bromoacetylenic
alcohol 16.12b After the detailed investigation on the
introduction of the triacetylenic moiety, it was found that
CadiotꢀChodkiewicz reaction3 between 16 (1.0 equiv) and
diacetylene 17 (1.1 equiv)13 with CuCl/NH2OH•HCl/
EtNH2 at ꢀ78 °C proceeded smoothly to provide triace-
tylene 18.14 Finally, acetylation of the alcohol 18 and
The synthetic diepoxides 5 and 6 were submitted to
extensive NMR analysis. The selected Δδ values in ppm
between natural (ꢀ)-gummiferol and the synthetic pro-
ducts inthe1Hand13C NMR spectra are depicted in Table 1.
The 1H and 13C NMR data of the synthetic syn-diepoxide 5
were in good agreement with those of natural (ꢀ)-
gummiferol.1,16 On the other hand, the 1H and 13C NMR
data of the synthetic anti-diepoxide 6 were different from
those of the natural product.1,16 It was observed that the
chemical shift differences between the natural product and
synthetic 6 at the C9 and C10 positions were significant:
þ0.08 (H-9), þ0.12 (H-10), ꢀ1.23 (C-9), and ꢀ0.88 (C-10).
subsequent deprotection of the TBS group with HF pyr
afforded the syn-diepoxide 5.
3
The measured specific rotation of the synthetic 5, [R]28
D
Scheme 4. Synthesis of 5
ꢀ62.5 (c 0.07, CH3OH), was consistent with that of the
natural product.17,18 Therefore, we concluded that the
absolute configuration of (ꢀ)-gummiferol was that de-
scribed in 5.
Scheme 5. Synthesis of 6
(11) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769.
(12) In the absence of Et3N, only a trace amount of 15 was obtained
because the epoxide opening occurred as a side reaction. For the related
examples, see: (a) Grandjean, D.; Pale, P.; Chuche, J. Tetrahedron Lett.
ꢀ
1994, 35, 3529. (b) Gonzalez, I. C.; Forsyth, C. J. J. Am. Chem. Soc.
2000, 122, 9099. (c) Dıaz, D.; Martın, T.; Martın, V. S. J. Org. Chem.
2001, 66, 7231.
(14) The homocoupling byproduct was not observed at all.
(15) The diastereomeric purity of 19 was determined by the 1H and
13C NMR spectra, which were clearly different from those of 12.
(16) See Supporting Information for details.
€
(13) Reber, S.; Knopfel, T. F.; Carreira, E. M. Tetrahedron 2003, 59,
6813.
3646
Org. Lett., Vol. 13, No. 14, 2011

(ꢀ)-gummiferol, wherein the stepwise stereoselective
epoxidation and CadiotꢀChodkiewicz reaction were
utilized for the efficient introduction of the contiguous
epoxide unit and the triacetylenic moiety, respective-
Table 1. Selected Chemical Shift Differences in ppm between
Natural (ꢀ)-Gummiferol and the Synthetic Diepoxides 5 and 6
in the 1H and 13C NMR (CDCl3)a
1
ly. The H and 13C NMR data of the synthetic syn-
1H NMR (ΔδNꢀS
)
13C NMR (ΔδNꢀS
)
diepoxide 5 matched with those of natural (ꢀ)-gum-
miferol. On the other hand, the 1H and 13C NMR data
of the synthetic anti-diepoxide 6 were significantly
different, wherein the chemical shift deviations were
clearly distinguishable at the C9 and C10 positions.
Finally, comparison of the specific rotation between
the synthetic 5 and the natural product established the
absolute stereochemistry of (ꢀ)-gummiferol as shown
in 5.
position
5
6
5
6
8
þ0.02
þ0.01
þ0.02
þ0.01
0
þ0.03
þ0.08
þ0.12
þ0.04
þ0.01
þ0.01
0
ꢀ0.01
ꢀ0.03
ꢀ0.02
þ0.04
þ0.06
þ0.09
þ0.05
ꢀ0.42
ꢀ1.23
ꢀ0.88
ꢀ0.71
þ0.32
þ0.03
þ0.10
9
10
11
12
13
14
þ0.01
ꢀ0.01
a NMR spectra of the natural product and the synthetic products
were recorded at 500 MHz (125 MHz) and 400 MHz (100 MHz),
respectively. Chemical shifts are reported in ppm with reference to
tetramethylsilane. δN and δS are chemical shifts of the natural product
and the synthetic product, respectively.
Acknowledgment. We are grateful to Professor Hideki
Okamoto (Okayama University) for the NMR measure-
ment. This research was financially supported in part by a
Grant-in-Aid for Scientific Research from Japan Society
for the Promotion of Science (JSPS) and The Ministry of
Education, Culture, Sports, Science and Technology
(MEXT).
In conclusion, we have accomplished the stereo-
controlled synthesis of two possible diastereomers of
(17) The natural product: [R]25D ꢀ170 (c 0.2, CH3OH). Our synthetic
ent-5: [R]25 þ76.0 (c 0.03, CH3OH). Our synthetic 6: [R]23 þ32.0
Supporting Information Available. Experimental pro-
cedures, spectroscopic data, and copies of NMR spectra
for all new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
D
(c 0.12, CH3OH).
D
(18) The optical purity of the synthetic product 5 was confirmed at
the stage of 8 as described in ref 5. Furthermore, the purity of 5 was
verified by NMR spectroscopy.
Org. Lett., Vol. 13, No. 14, 2011
3647