Synthesis of (4R,15R,16R,21S)- and (4R,15S,16S,21S)-rollicosin

Article abstract of DOI:10.1016/j.tetlet.2005.04.144

A convergent synthesis of (4R,15R,16R,21S)-rollicosin (1) and (4R,15S,16S,21S)-rollicosin (2) was accomplished. Hydroxy lactone 6a and/or 6b were synthesized from 4-pentyn-1-ol, and α,β-unsaturated lactone 7 was synthesized from γ-lactone 8 and 5-hexen-1-

Full text of DOI:10.1016/j.tetlet.2005.04.144

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 4671–4675
Synthesis of (4R,15R,16R,21S)- and (4R,15S,16S,21S)-rollicosin
Hidefumi Makabe,a,* Masaharu Higuchi,a Hiroyuki Konno,b
Masatoshi Muraic and Hideto Miyoshic
aSciences of Functional Foods, Graduate School of Agriculture, Shinshu University, 8304 Minami-minowa, Kamiina,
Nagano 399-4598, Japan
bDepartment of Chemistry, Graduate School of Medical Sciences, Kyoto Prefectural University of Medicine,
Kita-ku, Kyoto 603-8334 Japan
cDivision of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kita-shirakawa,
Sakyo-ku, Kyoto 606-8502, Japan
Received 3 March 2005; revised 13 April 2005; accepted 22 April 2005
Available online 23 May 2005
Abstract—Aconvergent synthesis of (4 R,15R,16R,21S)-rollicosin (1) and (4R,15S,16S,21S)-rollicosin (2) was accomplished.
Hydroxy lactone 6a and/or 6b were synthesized from 4-pentyn-1-ol, and a,b-unsaturated lactone 7 was synthesized from c-lactone
8 and 5-hexen-1-ol. Inhibitory activity of these compounds was examined with bovine heart mitochondrial complex I.
Ó 2005 Elsevier Ltd. All rights reserved.
Annonaceous acetogenins, that have been isolated from a
number of tropical or subtropical plants of the Annon-
aceae, have attracted much attention due to a wide vari-
ety of biological activities, for example, cytotoxic,
antitumoral, antimalarial, antibiotic, antiparasitic, and
antifeedant. So far, more than 400 compounds have
been isolated.1 Most of them possess one or more tetra-
hydrofuran (THF) rings, together with an a,b-unsatu-
rated c-lactone part on a C-35 or C-37 carbon chain.
Rollicosin (1) was isolated from Rollinia mucosa by
Wu and co-workers.2 This compound possesses a partial
skeleton of A. acetogenins containing two c-lactone moi-
eties on both sides of an aliphatic chain. Rollicosin may
be generated from oxidative degradation of classical
acetogenins such as murisolin (3)3 and cis-murisolin
(4).4 Moreover, this compound may help investigate
the role of the terminal hydroxylated lactone moiety
for its bioactivity in stead of the hydroxylated THF moi-
ety with long aliphatic chain that can be seen in the clas-
sical acetogenins. The absolute stereochemistry of 1 was
reported to be (4R,15R,16R,21S). The absolute chemis-
try at C-4 and C-21 position was assigned in the R- and
-S by the CD spectrum and the configurations of the C-
15 and C-16 position of 1 were determined by compari-
son of optical rotation between 1 and (ꢀ)-muricatacin
(5).5 Recently, rollicosin analogue, (15S,16S,21R)-4-
deoxyrollicosin was synthesized by Wu and co-workers6
(Fig. 1).
In this letter, we wish to report the synthesis of
(4R,15R,16R,21S)-rollicosin (1) and (4R, 15S,16S,21S)-
rollicosin (2) and their inhibitory activity against bovine
mitochondrial complex I.
Scheme 1 outlines our synthetic strategy. The target
compound 1 would be derived from hydroxy lactone
6a and the a,b-unsaturated lactone 7. Hydroxy lactone
6a could be synthesized from 4-pentyn-1-ol by applica-
tion of Wu and co-workers.6 Synthesis of 7 could be
accomplished by 5-hexen-1-ol and c-lactone 8 which
could be prepared by White et al.Õs method.7
As shown in Scheme 2, the hydroxy lactone 6a was
constructed via a six-step process using Wu and
co-workers method6 with slight modification. The syn-
thesis was started from 4-pentyn-1-ol, which was treated
with trimethylsilyl chloride to afford alcohol 9. Com-
pound 9 was oxidized with SO3 pyridine to give alde-
hyde 10, which was reacted with vinylmagnesium
chloride to afford allylic alcohol 11. Johnson–Claisen
rearrangement of 11 with triethyl orthoacetate and a
catalytic amount of propionic acid gave 12, followed
Keywords: Annonaceous acetogenin; Antitumor; Mitochondrial com-
plex I.
*
Corresponding author. Tel.: +81 265 77 1630; fax: +81 265 77
1700; e-mail: makabeh@gipmc.shinshu-u.ac.jp
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.04.144

4672
H. Makabe et al. / Tetrahedron Letters 46 (2005) 4671–4675
O
O
OH
1
16
15
2
3
O
O
4
O
O
OH
(4R, 15R, 16R, 21S)-rollicosin (1)
O
O
OH
OH
(4R, 15S, 16S, 21S)-rollicosin (2)
oxidative cleavage
O
O
OH
C12H25
O
OH
OH
murisolin (3)
oxidative cleavage
rollicosin (1)
O
O
OH
C
12H25
O
OH
OH
cis-murisolin (4)
C12H25
OH
O
O
(-)-muricatacin (5)
Figure 1. The structure of rollicosin.
The a,b-unsaturated lactone 7 was prepared as shown in
Scheme 3. 5-Hexen-1-ol was treated with TBSCl and
imidazole in DMF afforded the silyl ether 14. Sharpless
asymmetric dihydroxylation of 14 using (DHQD)2AQN
as a ligand9 gave diol 15, which showed 93% ee based on
O
OH
O
O
O
O
O
1
OH
O
O
OTBS
1
a H NMR analysis of the corresponding Mosher ester
I
derivative. The 1,2-diol of 15 was protected as an aceto-
nide with dimethoxypropane in the presence of p-TsOH
to furnish 16. Deprotection of the TBS group with
TBAF gave 17. Compound 17 was transformed into
iodide 18 via mesylation followed by iodination. Alkyl-
ation of 18 with lithium acetylide, ethylenediamine com-
plex gave terminal acetylene 19. Compound 19 was
treated with n-Bu3SnH and subsequently iodine to af-
ford an EZ mixture (E/Z = 8/1) of vinyl iodide 20.
Deprotection of the acetonide group of 20 with methan-
olic HCl gave diol 21. Selective sulfonylation of the
primary hydroxyl group in 21 with triisopropyl-
benzenesulfonyl chloride in pyridine afforded the sulfo-
nate 22, which was then treated with NaH in THF to
give epoxide 23. Iodination of 23 with LiI gave hydroxyl
+
OH
6a
7
SPh
4-pentyn-1-ol
5-hexen-1-ol
+
O
O
8
Scheme 1. Synthetic strategy of rollicosin.
by Sharpless asymmetric dihydroxylation using AD mix
b8 to furnish lactone 13. Removal of TMS group of 13
with TBAF gave hydroxy lactone 6a.
CHO
b
a
OH
OH
TMS
TMS
9
10
4-pentyn-1-ol
OH
d
c
CO2Et
TMS
O
11
TMS
f
12
TMS
e
O
O
O
OH
13
OH
6a
Scheme 2. Synthesis of hydroxy lactone 6a of 1. Reagents and conditions: (a) (i) n-BuLi, THF, 0 °C; (ii) TMSCl, 0 °C to rt (71%); (b) SO3 pyridine
(96%); (c) vinylmagnesium chloride, THF, 0 °C (99%); (d) CH3C(OEt)3, propionic acid, 150 °C (64%); (e) AD mix b, CH3SO2NH2, t-BuOH/H2O,
0 °C (94%); (f) TBAF, THF, 0 °C (95%).

H. Makabe et al. / Tetrahedron Letters 46 (2005) 4671–4675
4673
OH
b
a
c
HO
OH
OTBS
OTBS
5-hexen-1-ol
14
15
O
O
O
O
d
e
O
O
OTBS
19
OH
I
16
17
18
O
O
g
f
h
j
O
O
I
20
OH
OH
OH
i
HO
TrisO
I
I
I
I
21
22
24
O
l
k
I
I
23
25
O
O
OTBS
TBSO SPh
m
n
I
I
SPh
O
26
O
8
O
O
OTBS
I
7
Scheme 3. Synthesis of the c-lactone part 7 of 1. Reagents and conditions: (a) TBSCl, DMF, imidazole (89%); (b) (DHQD)2AQN, K2CO3,
K3Fe(CN)6, t-BuOH/H2O, 0 °C (95%); (c) dimethoxypropane, p-TsOH (93%); (d) TBAF, THF, 0 °C (94%); (e) (i) MsCl, Et3N, CH2Cl2; (ii) NaI,
NaHCO3, acetone (85%); (f) lithium acetylide, ethylenediamine complex, DMSO (73%); (g) (i) n-Bu3SnH, AIBN; (ii) I2, THF, 0 °C (92%); (h) concd
HCl, MeOH (86%); (i) TrisCl, pyridine (98%); (j) NaH, THF (85%); (k) LiI, THF–water–AcOH (93%); (l) TBSCl, imidazole, DMF (83%); (m) LDA,
THF–HMPA(16%); (n) (i) m-CPBA; (ii) toluene, reflux (67%).
iodide 24, protection of the alcohol with TBSCl gave
25.10 The lactone 26 was obtained in 16% yield by alkyl-
ation of the enolate prepared by mixing 84 and LDA
with 25. Unreacted 8 and 25 were recovered in 70%
and 71% yield, respectively. These compounds could
be used for the same reaction again. The a,b-unsatu-
rated lactone 7 was obtained after oxidation of 26 with
m-CPBAfollowed by thermal elimination of sulfoxide
under reflux in toluene.
O
O
OTBS
O
O
+
a
I
OH
6a
7
O
O
OTBS
OTBS
OH
O
O
b
c
OH
27
O
O
Both segments were coupled by the Sonogashira
cross-coupling reaction11 to furnish cross-coupled
product 27 in 64% yield. Diimide reduction with
p-TsNHNH2 and NaOAc in ethylene glycol diethyl
ether under reflux afforded saturated product 28.12
Finally, deprotection of TBS ethers with HF afforded
1 (Scheme 4).13
O
O
OH
28
O
O
O
O
1
OH
1
The H NMR and 13C NMR spectra of the synthetic 1
were in good agreement with those of natural 1 kindly
provided by Professor Wu.14 The specific rotation of
synthetic 1 showed different value compared to that of
Scheme 4. Synthesis of 1. Reagents and conditions: (a) 5 mol%
Cl2Pd(PPh3)2, 10 mol% CuI, Et3N (64%); (b) p-TsNHNH2, NaOAc,
ethylene glycol diethyl ether (96%); (c) HF, CH3CN (85%).
24
24
reported (½aꢁ +2.5 (c 0.29, CHCl3)). The ½aꢁ value of
D
D
natural 1 was reported to be ꢀ26.0 (c 0.05, CHCl3).15
Therefore we also prepared (4R,15S,16S,21S)-rollicosin
(2), which has same relative stereochemistry (15,16-
threo) of 1,16 starting from 6b using the same procedure
as that employed for 1 (Scheme 5).17
The specific rotation of 2 showed sharp contrast
20
D
(½aꢁ +24 (c 0.43, CHCl3)) compared to natural 1.
Taking into account that the optical rotation of natu-
ral product was measured at low concentration, the

4674
H. Makabe et al. / Tetrahedron Letters 46 (2005) 4671–4675
OTBS
O
O
I
OH
4-pentyn-1-ol
O
O
+
OH
6b
7
O
O
OH
O
O
OH
2
Scheme 5. Synthesis of 2.
difference may be due to experimental error. To clarify
this, a direct comparison of our synthetic sample with
the authentic natural product would be necessary.
Therefore, we prepared the corresponding MTPAesters
from synthetic 1 and 2 (Fig. 2).
bition.19 We therefore conclude that the hydroxylated
lactone cannot be substituted for the hydroxylated
mono- or bis-THF ring moiety of ordinary acetogenins
for the activity.
In summary, we have achieved total synthesis of
(4R,15R,16R,21S)-rollicosin (1) and (4R,15S,16S,21S)-
rollicosin (2). Inhibitory action of these compounds
was examined with bovine heart mitochondrial complex
I. Both compounds showed almost the same activity.
1
The H NMR chemical shifts of H-15 position of (R)-
MTPA-1 and (R)-MTPA-2 showed clear difference
(Table 1).
This indicates that if the bis-(R)-MTPAester of natural
1 would be available, the absolute configuration of rol-
licosin (1) would be determined very clearly.
Acknowledgements
Inhibitory effect of 1 and 2 on bovine heart mitochon-
drial complex I (NADH-ubiquinone oxidoreductase)
was examined according to the previous method.18 Both
compounds exhibited almost the same inhibitory
potency (1: IC50 = 0.66 0.03 lM, 2: IC50 = 0.68 0.03 lM),
indicating that the stereochemistry around the hydroxyl-
ated lactone moiety does not affect the inhibitory action.
It is noteworthy that compared to potent natural aceto-
genins like bullatacin (IC50 = 0.8 nM), both compounds
are much weaker inhibitors of the enzyme. Rollicosin
does not have a long hydrophobic alkyl tail which is
one of the common structural features of a large number
of natural acetogenins. However this may not be a rea-
son for the weak inhibitory activity since the long alkyl
tail is not a crucial structural factor for the potent inhi-
We thank Professor Yang-Chang Wu (Kaohsiung Medi-
cal University, Taiwan) for providing 1H and 13C NMR
spectra of natural rollicosin. This work was supported in
part by a Grant-in-aid from the Japan Society for the
Promotion of Science (15780084). We thank Ms. Keiko
Hashimoto and Ms. Megumi Miyazawa of the Faculty
of Agriculture, Shinshu University, for the 500 MHz
NMR measurements.
References and notes
1. For recent reviews for Annonaceous acetogenins, see: (a)
Alali, F. Q.; Liu, X.-X.; McLaughlin, J. L. J. Nat. Prod.
`
1999, 62, 504–540; (b) Zafra-Polo, M. C.; Figadere, B.;
Gallardo, T.; Tormo, J. R.; Cortes, D. Phytochemistry
1998, 48, 1087–1117.
2. Liaw, C.-C.; Chang, F.-R.; Wu, M.-J.; Wu, Y.-C. J. Nat.
Prod. 2003, 66, 279–281.
O
O
(R)-MTPA-O
1
16
15
2
3
´
3. Myint, S. H.; Laurens, A.; Hocquemiller, R.; Cave, A .;
O
4
O
Davoust, D.; Cortes, D. Heterocycles 1990, 31, 861–867.
4. Woo, M. H.; Ye, Q.; Gu, Z.-M.; Zhao, G.-X.; McLaugh-
lin, J. L. Bioorg. Med. Chem. Lett. 1995, 5, 1135–1140.
5. For isolation: (a) Rieser, M. J.; Kozlowski, J. F.; Wood,
K. V.; McLaughlin, J. L. Tetrahedron Lett. 1991, 32,
1137–1140; For synthesis: (b) Keinan, E.; Sinha, S.;
Anjana, S. B.; Wang, Z. M.; Zhang, X. L.; Sharpless, K.
B. Tetrahedron Lett. 1992, 33, 6307–6410; (c) Makabe, H.;
Tanaka, A.; Oritani, T. Biosci. Biotechnol. Biochem. 1993,
57, 1028–1029.
O-(R)-MTPA
(R)-MTPA-1
O
O
(R)-MTPA-O
O
O
O-(R)-MTPA
(R)-MTPA-2
Figure 2. Bis-(R)-MTPAesters of 1 and 2.
6. Lee, J.-L.; Lin, C.-F.; Hsieh, L.-Y.; Chiu, H.-F.; Wu,
Y.-C.; Wang, K.-S.; Wu, M.-J. Tetrahedron Lett. 2003, 44,
7833–7835.
7. White, J. D.; Somers, T. C.; Reddy, G. N. J. Org. Chem.
1992, 57, 4991–4998.
Table 1. 1H NMR chemical shifts of the bis-(R)-MTPAesters of 1 and 2
8. Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino,
G. A.; Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Morik-
awa, K.; Wang, Z.-M.; Xu, D.; Zhang, X.-L. J. Org.
Chem. 1992, 57, 2768–2771.
MTPAester
15-H
16-H
(R)-MTPA-1
(R)-MTPA-2
5.17
5.08
4.60
4.60

H. Makabe et al. / Tetrahedron Letters 46 (2005) 4671–4675
4675
1
9. Becker, H.; Sharpless, K. B. Angew. Chem., Int. Ed. 1996,
35, 448–451.
10. Yang, W.-Q.; Kitahara, T. Tetrahedron Lett. 1999, 40,
7827–7830.
11. Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron
Lett. 1975, 16, 4467–4470.
from ours. The H and 13C NMR of 1 synthesized by us
matched very well with those by K. J. Quinn and co-workers.
15. The discrepancy in the optical rotation value of the
synthetic product with that reported for the natural
product has been reported.: (a) Sinha, S. C.; Sinha, S.
C.; Keinan, E. J. Org. Chem. 1999, 64, 7067–7073; (b)
Takahashi, S.; Kubota, A.; Nakata, T. Org. Lett. 2003, 5,
1353–1356; (c) Ref. 12.
12. Makabe, H.; Miyawaki, A.; Takahashi, R.; Hattori, Y.;
Konno, H.; Abe, M.; Miyoshi, H. Tetrahedron Lett. 2004,
45, 973–977.
16. The relative stereochemistry of hydroxy lactone moiety
(15,16 position of 1) can be determined by 1H and 13C
NMR. We also prepared 15-epi-1 (15,16-erythro) from 6a.
The 1H and 13C NMR data for 15-epi-1: 1H NMR
(CDCl3, Me4Si) d: 1.20–1.65 (19H, m), 1.43 (3H, d,
J = 6.8 Hz), 1.96 (1H, br –OH), 2.10–2.45 (4H, m), 2.24
(1H, br –OH), 2.45–2.65 (3H, m), 3.85 (1H, m), 3.92 (1H,
m), 4.42 (1H, td, J = 7.4, 4.5 Hz), 5.05 (1H, qd, J = 6.8,
1.4 Hz), 7.16 (1H, d, J = 1.4 Hz). 13C NMR (CDCl3,
Me4Si) d: 19.12, 21.17, 25.41, 25.55, 28.70, 29.41, 29.47,
31.98, 33.41, 33.46, 37.43, 69.99, 71.49, 77.98, 82.82,
131.23, 151.81, 174.59, 177.38.
24
13. The data for compound 1a: mp 104–106 °C, ½aꢁD +2.5 (c
0.29, CHCl3). IR (KBr) mmax cmꢀ1: 3420, 2923, 2851, 1748,
1458, 1321, 1192, 1084, 1026. 1H NMR (CDCl3, Me4Si) d:
1.20–1.60 (19H, m), 1.43 (3H, d, J = 6.8 Hz), 1.86 (1H, d,
J = 5.8 Hz, –OH), 2.10–2.45 (4H, m), 2.23 (1H, br –OH),
2.50–2.65 (3H, m), 3.57 (1H, m), 3.85 (1H, m), 4.41 (1H,
td, J = 7.4, 4.5 Hz), 5.05 (1H, qd, J = 6.8, 1.4 Hz), 7.16
(1H, d, J = 1.4 Hz). 13C NMR (CDCl3, Me4Si) d: 19.13,
24.12, 25.41, 25.55, 28.70, 29.41, 29.47, 33.02, 33.40, 33.47,
37.44, 70.03, 73.69, 77.97, 82.90, 131.24, 151.79, 174.59,
177.05. HRFABMS (M+H): calcd for C22H36O6:
397.2592. Found 397.2601.
20
17. The data for compound 2: mp 91–92 °C, ½aꢁD +24 (c
14. The proton chemical shifts of C-4 and C-15 showed 3.85 and
3.57 ppm in the 1H NMR spectrum kindly provided by
Professor Y.-C. Wu. Those of reported values were 3.75 and
3.48 ppm, respectively. There might be some miswriting in
the 1H NMR of 1 in Ref. 2. After submission of this
manuscript, the paper by K. J. Quinn and co-workers
appeared in Org. Lett. 2005, 7, 1243–1245, in which they
reported the synthesis of 1 through a procedure different
0.43, CHCl3). The IR, 1H NMR, 13C NMR were identical
with those of synthetic 1. HR-FABMS (M+H): calcd for
C22H36O6: 397.2592. Found 397.2586.
18. Miyoshi, H.; Ohshima, M.; Shimada, H.; Akagi, T.;
Iwamura, H.; McLaughlin, J. L. Biochim. Biophys. Acta
1998, 1365, 443–452.
19. Motoyama, T.; Yabunaka, H.; Miyoshi, H. Bioorg. Med.
Chem. Lett. 2002, 12, 2089–2092.