Biomimetic total synthesis of (-)-neroplofurol and (+)-ekeberin D4triggered by hydrolysis of terminal epoxides

Article abstract of DOI:10.1246/cl.140618

To accumulate the chemi cal basis of epoxide-opening cascade biogenesis, chemical syntheses of sesqui- and triterpenoids were performed. The biomi metic total syntheses of (-)-neroplofurol (1) and (+)-ekeberin D₄(2) were accomplished by protic acid-catalyzed hydrolysis of the terminal epoxide from nerolidol diepoxide 3 and squalene tetraepoxide 4 through single and double 5-exo cyclizations in intermediates 5 and 6, respectively. This chemical reaction mimics the direct hydrolysis mechanism of epoxide hydrol ases, enzymes that catalyze an epoxide-opening reaction to finally produce vicinal diols.

Full text of DOI:10.1246/cl.140618

CL-140618
Received: June 24, 2014 | Accepted: July 22, 2014 | Web Released: October 5, 2014
Biomimetic Total Synthesis of (¹)-Neroplofurol and (+)-Ekeberin D4
Triggered by Hydrolysis of Terminal Epoxides
Takeshi Kodama, Shingo Aoki, Tomoki Matsuo, Yoshimitsu Tachi, Keisuke Nishikawa, and Yoshiki Morimoto*
Department of Chemistry, Graduate School of Science, Osaka City University, Sumiyoshi-ku, Osaka 558-8585
(E-mail: morimoto@sci.osaka-cu.ac.jp)
H+
O
To accumulate the chemical basis of epoxide-opening cascade
biogenesis, chemical syntheses of sesqui- and triterpenoids were
performed. The biomimetic total syntheses of (¹)-neroplofurol (1)
and (+)-ekeberin D4 (2) were accomplished by protic acid-
catalyzed hydrolysis of the terminal epoxide from nerolidol
diepoxide 3 and squalene tetraepoxide 4 through single and
double 5-exo cyclizations in intermediates 5 and 6, respectively.
This chemical reaction mimics the direct hydrolysis mechanism of
epoxide hydrolases, enzymes that catalyze an epoxide-opening
reaction to finally produce vicinal diols.
O
R1
R2
R3
R4
O
3
OR
H2O
H2O
epoxide hydrolase?
epoxide
hydrolase
H+
HO
O
OH
R3
R4
HO
5
R1
R2
OR
5-exo
OH
direct hydrolysis
mechanism
OH
Recently, the epoxide-opening cascade biogenesis of natural
polyethers,1 known as the Cane-Celmer-Westley hypothesis,2
has progressively been evidenced experimentally. An epoxide
hydrolase Lsd19 found by Oikawa and co-workers has realized
transformation of the prelasalocid diepoxide to lasalocid A in
the final stage of the biosynthesis.3 The amino acid residues
constituting the active site of Lsd19 resemble those of epoxide
hydrolases catalyzing direct hydrolysis (Figure 1).4 The epoxide-
opening cascades have also been utilized by synthetic chemists as
a method to rapidly construct polyether frameworks.5 In these
examples of epoxide-opening cascades, however, it is almost
always the case that the first epoxide-opening is initiated by an
intramolecular nucleophilic attack to the neighboring epoxide.6
Recently, Qu’s7 and our8 groups reported epoxide-opening
cascades triggered by an intermolecular nucleophilic attack of
water to the epoxide under basic and acidic conditions, respec-
tively, that mimics the intrinsic role of epoxide hydrolases
catalyzing direct hydrolysis. To understand the biogenetic mecha-
nism of epoxide-opening cascades in the absence of intramolecular
nucleophiles, we think that it might be important to accumulate
the chemical basis. In this contribution, we show a further two
examples of the chemical epoxide-opening cascade mimicking the
direct hydrolysis mechanism of epoxide hydrolases.
(¹)-Neroplofurol (1), a nerolidol sesquiterpene bearing one
THF ring, was isolated from anti-TB active fractions of the inner
stem bark of Oplopanax horridus, an abundant deciduous shrub
found along the Northern Pacific coast of North America, by Pauli
and co-workers (Figure 1).9 The molecular structure and relative
configuration of 1 were elucidated on the basis of spectroscopic
studies. The absolute configuration was determined by Huo and
co-workers through the total synthesis of (+)-neroplofurol,
enantiomeric to natural neroplofurol.10 (+)-Ekeberin D4 (2) was
isolated from the stem bark of Ekebergia capensis, a tree widely
distributed in Kenya, by Miyase and co-workers (Figure 1).11
Ekeberin D4 (2) exhibits antiplasmodial activity against FRC-3
with IC50 = 40 ¯M, and the triterpenoid structure possesses C2
symmetry and two THF ring moieties with the same relative
configuration as that of 1. Based on the epoxide-opening cascade
mimicking the direct hydrolysis mechanism of epoxide hydrolases,
O
HO
H
(–)-neroplofurol (1)
OH
same relative configuration
C2
OH
H
OH
O
O
HO
H
OH
H+
(+)-ekeberin D4 (2)
5-exo
HO
O
OH
HO
O
6
OH
H+
H2O
H+
O
epoxide hydrolase?
O
O
O
H+
4
H2O
Figure 1. Hypothetical biogenesis of (¹)-neroplofurol (1)
and (+)-ekeberin D4 (2) based on epoxide-opening cascades
triggered by direct hydrolysis of the terminal epoxide.
we envisioned that 1 and 2 could biogenetically be derived by
hydrolysis of the terminal epoxide from nerolidol diepoxide 3 and
squalene tetraepoxide 4 via single and double 5-exo cyclizations in
intermediates 5 and 6, respectively. We attempted the biomimetic
synthesis of 1 and 2 along this line.
Iodination of the known optically active epoxy alcohol 7
(er = 95.5:4.5)12 followed by zinc reduction of the resulting iodide
furnished (3R,6E)-nerolidol (8) in 88% yield over two steps
(Figure 2). After triethylsilyl (TES) protection of the tertiary
alcohol, triene 9 was subjected to Shi asymmetric epoxidation with
a chiral L-ketone catalyst to afford the required diepoxide 3
(R = TES) in an approximately 5:1 dr.13 Treatment of 3 (R = TES)
under our previous epoxide-opening cascade conditions8 (0.3 equiv
of TfOH, THF/H2O (9:1), 0 °C, 42 h) gave the desired (¹)-neroplo-
furol (1) in 29% yield along with many other complex mixtures.14
The spectral data (1H and 13C NMR) and optical rotation of
25
synthetic 1, ½ꢀꢀD ¹23.4 (c 0.026, MeOH), were consistent with
25
those reported for the natural product, lit.,9 ½ꢀꢀD ¹23.7 (c 0.013,
1662 | Chem. Lett. 2014, 43, 1662–1664 | doi:10.1246/cl.140618
© 2014 The Chemical Society of Japan

by another method,18 we assumed that synthetic 2 could be isolated
from the complex mixtures, albeit in only 6% yield.14,19 The
spectral data (1H and 13C NMR) and optical rotation of synthetic 2,
O
1) I2, PPh3, imidazole
CH2Cl2, 0 °C, 3 h
OH
OH
8
7
27
2) Zn, EtOH, 65 °C
18 h, 88% (2 steps)
½ꢀꢀD +5.94 (c 0.245, CHCl3), were identical to those reported
for the natural and another synthetic product, lit.,11 [α]D +8.00
27
(c 0.420, CHCl3); another synthetic:18 ½ꢀꢀD +7.80 (c 0.19,
O
O
O
CHCl3). However, when we performed derivatization of synthetic
2 to diMTPA esters, we noticed that 2 was not a single isomer but
a 2:1 mixture. We guessed that meso-compound 11 was present as
an impurity in 2, by comparing the 1H NMR data of a 2:1 mixture
of synthetic (R)-MTPA-2 and (R)-MTPA-11, respectively, with
those of (R)- and (S)-MTPA-2 derived from the natural product by
Miyase et al.11 We confirmed that the impurity was 11 through
independently synthesizing (R)-MTPA-2 and (R)-MTPA-11 as a
single diastereomer by another method.18,21
Although it was found from these results that the production
of (+)-ekeberin D4 (2) could be reproduced by an intermolecular
nucleophilic attack of water to both terminal epoxides of 4, it was
envisaged that regioselectivity would be considerably low com-
pared to that of other squalene polyepoxides.8 Therefore, if the
hypothetical intermediate 6, an immediate product obtained after
hydrolysis of both terminal epoxides, could only be generated, the
yield of 2 would be improved. To demonstrate this, we attempted
the synthesis of the hypothetical intermediate tetraol 6 and its
conversion to (+)-ekeberin D4 (2).
O
O
TESOTf, Et3N
DMAP, CH2Cl2
O
OTES
L-ketone, Oxone, Bu4HSO4
9
0 °C, 15 min
94%
CH2(OMe)2, MeCN, H2O
pH 10.5, 0 °C, 20 min, 77% (dr = 5:1)
OH
TfOH
THF/H2O (9:1)
OTES
O
O
O
HO
H
OH
(–)-neroplofurol (1)
0 °C, 42 h
29%
3 (R = TES)
TfOH
O
O
THF/H2O (9:1)
TMSO
rt, 20 min
61%
ent-3 (R = TMS)
H+
O
O
O
O
HO
H
H
HO
ent-3 (R = H)
10
The known diol 1222 with a high optical purity (er = 97.5:2.5)
was protected as diTBS ether 13, which was subjected to Shi
epoxidation using D-ketone to afford monoepoxide 14 in a regio-
and diastereoselective manner (Figure 4).16,18 Deacetylation of 14
and bromination of the resulting allylic alcohol produced unstable
allylic bromide 15, which was substituted with sodium benzene-
sulfinate to yield allylic sulfone 16. Alkylation of a lithio derivative
of 16 with 15 followed by reductive desulfonylation23 of 17
resulted in dimerization in a good yield. Desilylation of tetraTBS
ether 18 with TBAF in refluxing THF in the presence of AcOH
generated the hypothetical biogenetic intermediate 6, which was
exposed to the same cyclization conditions as those in Figure 3 to
furnish the desired (+)-ekeberin D4 (2) in 63% yield over two
steps. The spectral characteristics (1H and 13C NMR) of synthetic
Figure 2. Biomimetic total synthesis of (¹)-neroplofurol (1).
MeOH). To successfully obtain (¹)-1 in the last reaction, the
hydrolysis of the terminal epoxide to generate 5 (R = TES) should
occur prior to desilylation; this is because when diepoxide ent-3
(R = TMS),15 enantiomeric to 3 (R = TMS), was exposed to the
cyclization conditions, compound 10 reported by Marshall and
Hann17 was produced in 61% yield, indicating that deprotection of
the TMS group has occurred prior to the epoxide-opening.
Next, we turned our attention to the biomimetic total synthesis
of (+)-ekeberin D4 (2). Squalene tetraepoxide 415,16 reported by
McDonald et al. was treated under the cyclization conditions and
a large number of spots were observed on the TLC (Figure 3).
However, since we have already synthesized (+)-ekeberin D4 (2)
30
(+)-ekeberin D4 (2), ½ꢀꢀD +8.22 (c 0.19, CHCl3), from 6 were
consistent with those of the natural and another synthetic product.
O
O
O
O
4 (dr = ~2:1)
TfOH, THF/H2O (9:1), rt, 2 h, 6%
C2
meso
OH
OH
H
OH
H
OH
O
O
+
O
O
HO
HO
H
HO
H
OH
OH
an inseparable mixture of C2 2 and meso 11 indiscernible by NMR
11
2
MTPACl = α-methoxy-α-(trifluoromethyl)phenylacetyl chloride (S)-MTPACl, py, rt, 6 h, then 3-(dimethylamino)-1-propanamine, 54%
3.53 ppm
3.55 ppm
Ph R OMe
C2
(R)-MTPAO
CF3
H
OH
O
R
O
O
R
H
OH
O
+
O
H
R
O
F3C
O
S
O
HO
H
(R)-MTPA-2
O(R)-MTPA
3.53 ppm
MeO
Ph
3.53 ppm
(R)-MTPA-11
R
an inseparable mixture of (R)-MTPA-2:(R)-MTPA-11 = ~2:1 discernible by the MeO peaks in the NMR
Figure 3. Biomimetic total synthesis of (+)-ekeberin D4 (2).
Chem. Lett. 2014, 43, 1662–1664 | doi:10.1246/cl.140618
© 2014 The Chemical Society of Japan | 1663

O
O
O
O
O
O
OH
OTBS
TBSOTf, 2,6-lutidine
CH2Cl2
D-ketone, Oxone, Bu4HSO4
HO
TBSO
OAc
OAc
12
13
0 °C, 2 h, 83%
CH2(OMe)2, MeCN, H2O
pH 10.5, 0 °C, 2 h, 74% (dr > 95:5)
NaSO2Ph, Bu4NI, DMF
OTBS
OTBS
1) K2CO3, MeOH, rt, 13 h, 96%
O
O
TBSO
TBSO
TBSO
2) MsCl, Et3N, THF, –40 °C, 1 h
then LiBr, 0 °C, 30 min
rt, 12 h, 87% (2 steps)
Br
15
14
OAc
OTBS
OTBS
O
OTBS
BuLi, HMPA, THF
–78 °C, 30 min
TBSO
then 15, –78 °C
30 min, 88%
TBAF
AcOH
O
O
R
OTBS
THF
reflux, 30 h
SO2Ph
17 (R = SO2Ph)
16
[PdCl2(dppp)], LiBHEt3
THF, 0 °C, 1 h, 87%
dppp: 1,3-bis(diphenylphosphino)propane
18 (R = H)
OH
TfOH
THF/H2O (9:1)
H+
H
OH
HO
O
O
OH
O
HO
H
rt, 2 h
63% (2 steps)
OH
HO
O
6
OH
(+)-ekeberin D4 (2)
H+
Figure 4. Synthesis of hypothetical intermediate tetraol 6 and its conversion to (+)-ekeberin D4 (2).
122, 4831. d) Y. Morimoto, T. Okita, H. Kambara, Angew. Chem., Int.
Ed. 2009, 48, 2538.
P. Yang, P.-F. Li, J. Qu, L.-F. Tang, Org. Lett. 2012, 14, 3932.
Y. Morimoto, E. Takeuchi, H. Kambara, T. Kodama, Y. Tachi, K.
Nishikawa, Org. Lett. 2013, 15, 2966.
In conclusion, we have achieved biomimetic epoxide-opening
cascades of (¹)-neroplofurol (1) and (+)-ekeberin D4 (2) through
single and double 5-exo cyclizations, respectively, triggered by
triflic acid-catalyzed hydrolysis of terminal epoxides. In the case of
squalene polyepoxide precursors, 4 and other substrates,8 it was
found that the yields of the cascade reaction were of a wide range
depending on the number, position, and configuration of the
epoxides. It would be significant to understand the biogenetic
mechanism so that the epoxide-opening cascade mimicking the
direct hydrolysis mechanism of epoxide hydrolases could chemi-
cally be reproduced. Improvement of the reaction efficiency is
under investigation.
7
8
9
T. Inui, Y. Wang, D. Nikolic, D. C. Smith, S. G. Franzblau, G. F. Pauli,
J. Nat. Prod. 2010, 73, 563.
10 X. Huo, X. Pan, G. Huang, X. She, Synlett 2011, 1149.
11 T. Murata, T. Miyase, F. W. Muregi, Y. Naoshima-Ishibashi, K.
Umehara, T. Warashina, S. Kanou, G. M. Mkoji, M. Terada, A. Ishih,
J. Nat. Prod. 2008, 71, 167.
12 H. Kigoshi, M. Ojika, Y. Shizuri, H. Niwa, K. Yamada, Tetrahedron
1986, 42, 3789.
13 a) Z.-X. Wang, Y. Tu, M. Frohn, J.-R. Zhang, Y. Shi, J. Am. Chem.
Soc. 1997, 119, 11224. b) M.-X. Zhao, Y. Shi, J. Org. Chem. 2006, 71,
5377.
14 In both reactions starting materials 3 (R = TES) and 4 were completely
consumed, and many other products than 1 and 2 were also observed;
however, we could not isolate identifiable ones.
15 Diepoxide ent-3 (R = TMS) was prepared according to the same
sequence of reactions as those of 3 (R = TES). McDonald’s tetraepoxide
4 was prepared by almost the same method as their one (ref 16). See the
Supporting Information.
16 R. Tong, M. A. Boone, F. E. McDonald, J. Org. Chem. 2009, 74, 8407.
17 J. A. Marshall, R. K. Hann, J. Org. Chem. 2008, 73, 6753.
18 T. Kodama, S. Aoki, S. Kikuchi, T. Matsuo, Y. Tachi, K. Nishikawa, Y.
Morimoto, Tetrahedron Lett. 2013, 54, 5647.
19 We investigated the double cyclization of tetraepoxide 4 using Brønsted
acids such as perchloric acid, («)-10-camphorsulfonic acid, and bulky
1-[bis(triflyl)methyl]-2,3,4,5,6-pentafluorobenzene (ref 20) in place of
triflic acid in aq solvents such as 1,4-dioxane, DME, DMF, and
dimethoxymethane instead of THF; however, these reactions resulted in
only complex mixtures or traces of 2.
We are grateful to Prof. T. Miyase for generously providing
1
the H NMR spectrum of natural (+)-ekeberin D4. This work was
financially supported by a Grant-in-Aid for Scientific Research
from the Japan Society for the Promotion of Science and the Asahi
Glass Foundation.
Supporting Information is available electronically on J-STAGE.
References and Notes
1
2
3
4
a) A. R. Gallimore, Nat. Prod. Rep. 2009, 26, 266. b) J. J. Fernández,
M. L. Souto, M. Norte, Nat. Prod. Rep. 2000, 17, 235.
D. E. Cane, W. D. Celmer, J. W. Westley, J. Am. Chem. Soc. 1983, 105,
3594.
Y. Shichijo, A. Migita, H. Oguri, M. Watanabe, T. Tokiwano, K.
Watanabe, H. Oikawa, J. Am. Chem. Soc. 2008, 130, 12230.
a) A. Minami, A. Migita, D. Inada, K. Hotta, K. Watanabe, H. Oguri, H.
Oikawa, Org. Lett. 2011, 13, 1638. b) K. Hotta, X. Chen, R. S. Paton, A.
Minami, H. Li, K. Swaminathan, I. I. Mathews, K. Watanabe, H.
Oikawa, K. N. Houk, C.-Y. Kim, Nature 2012, 483, 355.
20 K. Ishihara, A. Hasegawa, H. Yamamoto, Angew. Chem., Int. Ed. 2001,
40, 4077.
5
6
I. Vilotijevic, T. F. Jamison, Angew. Chem., Int. Ed. 2009, 48, 5250.
a) W. C. Still, A. G. Romero, J. Am. Chem. Soc. 1986, 108, 2105.
b) S. L. Schreiber, T. Sammakia, B. Hulin, G. Schulte, J. Am. Chem.
Soc. 1986, 108, 2106. c) Z. Xiong, E. J. Corey, J. Am. Chem. Soc. 2000,
21 See the Supporting Information.
22 G. Vidari, A. Dapiaggi, G. Zanoni, L. Garlaschelli, Tetrahedron Lett.
1993, 34, 6485.
23 M. Mohri, H. Kinoshita, K. Inomata, H. Kotake, Chem. Lett. 1985, 451.
1664 | Chem. Lett. 2014, 43, 1662–1664 | doi:10.1246/cl.140618
© 2014 The Chemical Society of Japan