Synthetic studies of the HIV-1 protease inhibitive didemnaketals: Stereocontrolled synthesis of an ester side chain

Article abstract of DOI:10.1016/j.tetlet.2004.03.103

The stereocontrolled synthesis of the C₁-C₈ portion, the ester side chain of the HIV-1 protease inhibitive didemnaketals from the ascidian Didemnum sp., has been carried out through 15 steps starting from (S)-carvone as the chiral template. This approach involved the diastereoselective construction of three conjoint chiral centers by intramolecular chiral inducement, and generation of allylic alcohol intermediate through a key Grob fragmentation reaction.

Full text of DOI:10.1016/j.tetlet.2004.03.103

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 3713–3716
Synthetic studies of the HIV-1 protease inhibitive didemnaketals:
stereocontrolled synthesis of an ester side chainq
Xue Zhi Zhao, Yong Qiang Tu,* Lei Peng, Xue Qiang Li and Yan Xing Jia
Department of Chemistry & State Key Laboratory of Applied Organic Chemistry, Lanzhou University, Lanzhou 730000, PR China
Received 1 March 2004; revised 12 March 2004; accepted 17 March 2004
Abstract—The stereocontrolled synthesis of the C1–C8 portion, the ester side chain of the HIV-1 protease inhibitive didemnaketals
from the ascidian Didemnum sp., has been carried out through 15 steps starting from (S)-carvone as the chiral template. This
approach involved the diastereoselective construction of three conjoint chiral centers by intramolecular chiral inducement, and
generation of allylic alcohol intermediate through a key Grob fragmentation reaction.
Ó 2004 Elsevier Ltd. All rights reserved.
In 1991, Faulkner and co-workers reported the novel
didemnaketals A (1) and B (2) (Scheme 1) isolated from
the ascidian Didemnum sp. at Auluptagel Island, Palau,
and demonstrated A and B exhibited to be highly
inhibitive to HIV-1 protease with the IC50 being 2 and
10 lM, respectively.1 In-depth investigation proved the
ester side chain might be the main activity portion of the
HIV-1 protease didemnaketals.2 As their important
biological activity and particular structure, we have
already performed the synthesis of the spiroketal moi-
ety.3 In 2002, Faulkner and co-workers finished the full
assignment of relative and absolute stereochemistry of
the didemnaketals,1c which benefits our further studies
on the asymmetric total synthesis of this kind of com-
pounds. Here, we present our efficient and stereocon-
trolled synthetic of C1–C8 side-chain
4 of the
didemnaketals, which have three conjoint stereocenters
(C5–C7) and the conjugated unsaturated ester.
O
1(A) X= O
O
O
O
O
O
OMe
1
16
MeO2C
2(B) X=
O
X
10
4
8
18
O
O
12
O
O
SO3Na
3(C) X=
O
21
O
OH
23
2
8
O
R1O
R1O
OR2
OH
OR1 OR2
OR1 OR2
1
1
MeO2C
OH
6
5
8
5
7
HO
OR3
7
O
9
6 C6 - βOH
7 C6 - αOH
9
5
4
8
Scheme 1.
Keywords: Ester side chain; HIV-1 protease; Intramolecular chiral inducement; Grob fragmentation reaction.
q Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.tetlet.2004.03.103
* Corresponding author. Tel.: +86-931-8912410; fax: +86-931-8912582; e-mail: tuyq@lzu.edu.cn
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.03.103

3714
X. Z. Zhao et al. / Tetrahedron Letters 45 (2004) 3713–3716
Our synthesis, shown in the retrosynthetic analysis
(Scheme 1), was based on the selection of the menthane
kind of monoterpene (S)-carvone 9 as starting material,
for its C1–carbonyl, C2–Me, and the isopropene (C7–C9)
could generate the corresponding C7–OH, C6–Me, and
the terminal isopropyl ester of 4, respectively. So a C6–
hydroxy epoxide intermediate 6 (or 7) was designed and
a key Grob fragmentation would be used to generate the
open-chained allylic alcohol 5.4
trans-OH and epoxyisopropyl located at the six-mem-
bered ring (prepared from 11 via direct m-CPBA epox-
idation) was promoted with Al(iPrO)3, only (E)-8-
hydroxycitronellol 13 was produced.6 However, if the
substrate 14 possessed the cis-OH and epoxyisopropyl
(prepared from 12 via Mitsunobu reaction), a (Z)-8-
hydroxycitronellol 15 was obtained as a single product.
This important fact provided alternative procedure for
synthesis of the hard-to-available Z-double bond, and
here suggested that the intermediates 7 could be a
favorable precursor for constructing the side-chain 4
with E-double bond by use of Grob fragmentation.
In the initial investigation on the possibility of above
synthetic analysis, we tested the key Grob fragmentation
using the model substrate 12 prepared from (ꢀ)-citro-
nellal 10.5 It was very interesting that we observed the
configuration of the double bond formed in Grob
fragmentation of the epoxide 12 was highly dependent
upon the inducement of its OH group. As shown in
Scheme 2, if fragmentation of the epoxide 12 bearing the
On the basis of above model testing, the key interme-
diate 7 was designed firstly. The synthesis began with the
stereoselective epoxidation of commercially available
(S)-(þ)-carvone (9) using basic hydrogen peroxide7
(Scheme 3). Successive organoselenium-mediated cata-
lytic reductive ring opening of the resulting oxirane 16
led to the a-hydroxyl ketone 17.8 Therefore the corre-
sponding C5–OH of 4 was stereoselectively introduced.
As the direct protection of C3–OH in 17 with TBDPSCl
led to substantial elimination to give the starting carv-
one 9, protection of the C3–OH had to be performed
and the product 18 was obtained in three steps: pro-
tecting the carbonyl group in 17 with ethylene glycol,
then protecting the hydroxyl group with TBDPSCl and
again releasing carbonyl group with TsOH. Subse-
quently elimination of the carbonyl of 18 by hydrazon-
ization with TsNHNH2 followed by treatment the
formed hydrazone with n-BuLi afforded diene 8. With 8
in hand we turned our attention to gain the key inter-
mediate 23. Thus deprotection of TBDPS of 8 with
TBAF followed by stereoselective epoxidation of the
formed homoallylic alcohol 19 with TBHP through the
inducement of the hydroxyl group and then protection
of the hydroxy in 20 afforded the epoxide 21. Here, the
*
b
a
c
CHO
OH
OH
OH
OH
O
12
13
11
10
d
c
OH
OH
O
HO
15
14
Scheme 2. Reagents and conditions: (a) ZnBr2, CH2Cl2, 90%; (b)
m-CPBA, CH2Cl2; 0 °C, 92%; (c) Al(iPrO)3, iPrOH, reflux, 90%; (d) (i)
p-NO2C6H4CO2H, PPh3, DEAD, benzene; (ii) K2CO3, MeOH, 72% (2
steps).
O
R1O
O
R1O
HO
O
O
O
d
c
b
a
8
18
17
16
9
BnO
OH
OAc
BnO
HO
HO
g
h
f
e
O
O
22
BnO
OAc
OH
21
20
19
7
R1 = TBDPS
23
Scheme 3. Reagnets and conditions: (a) H2O2, NaOH, MeOH, 95%; (b) (PhSe)2, N-acetylcysteine, NaOH, MeOH, 70%; (c) (i) (CH2OH)2, PPTS,
benzene, reflux; (ii) TBDPSCI, NaH, THF, 0 °C; (iii) TsOH, acetone, 72% (three steps); (d) (i) TsNHNH2, MeOH; (ii) n-BuLi, THF, ꢁ78 °C, rt, 52%
(2 steps); (e) TBAF, THF, reflux, 84%; (f) t-BuO2H, V(acac)2, 89%; (g) BnBr, NaH, TBAI, THF, 0 °C, 68%; (h) HOAc, NaOAc, reflux, 72%.

X. Z. Zhao et al. / Tetrahedron Letters 45 (2004) 3713–3716
3715
R1O
OH
OH
R1O
OR2
OH
R1O
R1O
b
c
a
O
O
O
25
24
OR2
6
8
OR1
OR1
OR2
OH
OH
OHC
f
e
4 (R3 = Ac)
d
HO
26
27
R1 = TBDPS
R2 = TBDMS
t
Scheme 4. Reagents and conditions: (a) m-CPBA, CH2Cl2, 0 °C, 65%; (b) K2OsO4, NMO, BuOH–acetone–H2O (1:2:1) 75%; (c) TBSCl, imid.,
i
DMF, 75%; (d) Al(iPrO)3, PrOH, 72%; (e) SeO2, EtOH (95%), reflux, 72%; (f) (i) Ac2O, DMAP, Py; (ii) NaClO2, 2-methyl-2-butene, NaH2PO4,
iBuOH/H2O; (iii) CH2N2, Et2O, 70% (three steps).
deprotection of TBDPS group in 8 was necessary for the
regioselective monoepoxidation of 19 because a direct
epoxidation of 8 without inducement of C3–OH only
gave a terminal epoxy product. Unfortunately, the fol-
lowing ring opening of the epoxide 21 with NaOAc
could not afford the desired homoallylic alcohol 23
(which could be easily converted to the key intermediate
7) but the alcohol 22, which had two opposite stereo-
centers C1 and C6 to 23 and was of no use to our syn-
thesis. The possible reason for this result would be
that the diaxial transition state for yielding 22 is more
stable than the diequatorial one, which lead to the
product 23.9
synthetic studies on the didemnaketals are ongoing in
our group.
Acknowledgements
We are grateful for the financial support of the Natural
Science Foundation of China (NSFC Nos 29925205,
30271488, 20021001, 203900501).
References and notes
1. (a) Potts, B. C. M.; Faulker, D. J. J. Am. Chem. Soc. 1991,
113, 6321; (b) Pika, J.; Faulkner, D. J. Nat. Prod. Lett.
1995, 7, 291; (c) Salomon, C. E.; Williams, D. H.;
Lobkovsky, E.; Clardy, J. C.; Faulker, D. J. Org. Lett.
2002, 4, 1699–1702.
Subsequently we had to modify the above route for first
demanding the desired C7 stereochemistry of
4
(R3 ¼ Ac). A viable route as shown in Scheme 4 was
performed by regioselective epoxidation of the diene 8
with m-CPBA and the terminal epoxide 24 in two iso-
mers (1:1) was obtained. Then the stereoselective
osmylation10 of the epoxide 24 followed by selective
protection of the equatorial hydroxyl group in the
formed diol 25 afforded the intermediate 6, in which
three desired stereocenters corresponding C5–C7 have
been constructed successfully. Then fragmentation of
epoxide alcohol 6 with Al(iPrO)3 afforded the Z-allylic
alcohol 26, whose stereochemistry for the double bond
and three stereocenters was confirmed through spec-
troscopy inspection of its derivatives and the configu-
ration of the double bond was just what we anticipated
before.11 The formation of only one isomer of 26 from
two isomers of 6 indicated that the configuration of
epoxide has no effect on the Grob fragmentation reac-
tion. Fortunately, the following selective oxidation of
the Z-primary allylic alcohol 26 with selenium (IV)
oxide afforded the E-unsaturated aldehyde 27 in 70%
yield.12 Protection of the hydroxyl group in the aldehyde
27 with acetic anhydride followed by oxidation of the
formed aldehyde with sodium chlorite and then esteri-
fication of the formed acid with CH2N2 afforded the
methyl ester 4 (R3 ¼ Ac) successively.13 Further total
2. Fan, X.; Flentke, G. R.; Rich, D. H. J. Am. Chem. Soc.
1998, 120, 6321.
3. (a) Jia, Y. X.; Wu, B.; Li, X.; Ren, S. K.; Tu, Y. Q.; Chan,
A. S. C.; Kitching, W. Org. Lett. 2001, 3, 847–849; (b) Jia,
Y. X.; Li, X.; Wu, B.; Zhao, X. Z.; Tu, Y. Q. Tetrahedron
2002, 58, 1697–1708.
4. Waddell, T. G.; Ross, P. A. J. Org. Chem. 1987, 52, 4802–
4804.
5. Misono, M.; Nojiri, N. Appl. Catal. 1990, 64, 1–30.
6. The configuration of the double bond in the compound 4,
13, 15, 26, 27 was confirmed through 1H NMR NOE
experiment.
7. Nicolaou, K. C.; Xu, J. Y.; Kim, S.; Pfefferkorn, J.;
Ohshima, T.; Vourloumis, D.; Hosokawa, S. J. Am. Chem.
Soc. 1998, 120, 8661–8673.
8. Engman, L.; Stern, D. J. Org. Chem. 1994, 59, 5179–5183.
9. Parker, R. E.; Isaacs, N. S. Chem. Rev. 1959, 59, 737–799.
10. Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. Rev. 1994, 94, 2483–2547.
11. The compound 26 was first made into derivative 28
through three steps, and the stereochemistry of 28 was
confirmed through 1H NMR NOE experiment and NO-
ESY experiment as shown below. For example, irradiation
of C9–H (d: 1.76 ppm) lead to 4% enhancement of C3–H
(d: 5.48 ppm), irradiation of C5–H (d: 3.28 ppm) lead to 5%
enhancement of C10–H (d: 0.86 ppm), and irradiation of

3716
X. Z. Zhao et al. / Tetrahedron Letters 45 (2004) 3713–3716
1
C10–H (d: 0.86 ppm) lead to 4% enhancement of C5–H (d:
3.28 ppm) and 4% enhancement of C8–H (d: 4.05 ppm)
13. Analytical data of the compound 4: H NMR (300 MHz,
CDCl3): d 7.66–7.62 (m, 4H), 7.44–7.32 (m, 6H), 6.71
(t, 1H, J ¼ 6:7 Hz), 3.90–3.75 (m, 4H), 3.67 (s, 3H), 2.37–
2.27 (m, 2H), 1.93 (s, 3H), 1.87–1.83 (m, 1H), 1.65 (s, 3H),
1.04 (s, 9H), 0.95 (d, 3H, J ¼ 7:2 Hz), 0.82 (s, 9H), ꢁ0.01
(s, 3H), ꢁ0.08 (s, 3H); 13C NMR (75 MHz): d 170.6, 168.3,
139.9, 136.0 (4C), 133.9, 133.8, 129.7 (2C), 128.4, 127.6
(2C), 127.5 (2C), 74.6, 71.7, 67.0, 51.5, 42.5, 32.9, 27.0
(3C), 25.8 (3C), 20.8, 19.3, 18.1, 12.5, 10.0, ꢁ4.3,
ꢁ4.6; FAB-MS (MþH)þ, m=z 627; HR-SIMS:
m=z calcd for C35H55Si2O6 (MþH) 627.3532; found
627.3523.
OAc
(i) Ac2O, Py
(ii) TBAF
(iii) DMP, TsOH
H
H
H
Me
9
H
Me
3
8
1
26
5
O
Me
Me
10
AcO
O
7
6
28
12. (a) Bhalerao, U. T.; Rapoport, H. J. Am. Chem. Soc. 1971,
93, 5311–5313; (b) Kalsi, P. S.; Chhabra, B. R.; Singh, J.;
Vig, R. Synlett 1992, 425–426.