Design, synthesis and in vitro antimalarial activity of spiroperoxides

Article abstract of DOI:10.1016/j.tet.2006.05.065

Several spiroperoxy antimalarial compounds were designed and synthesized using the hydrogen peroxide in UHP (urea-H₂O₂ complex) as the source of the peroxy bond. Incorporation of the H₂O₂ into the organic molecule framework through ketal exchange reaction in the present cases was greatly facilitated by the potential to form a five- or six-membered cyclic hemiketal due to the presence of a hydroxyl group γ or δ to the ketone carbonyl group. When the electron-withdrawing group in the Michael acceptor was a nitro group, the closure of the peroxy ring occurred readily under the hydroxidation conditions. Presence of a benzene ring fused to the peroxy ring effectively reduced the degrees of freedom in the transition state for the ring-closure step and made the otherwise very difficult seven-membered 1,2-dioxepane rather easy to form through the intramolecular Michael addition.

Full text of DOI:10.1016/j.tet.2006.05.065

Tetrahedron 62 (2006) 7699–7711
Design, synthesis and in vitro antimalarial activity
of spiroperoxides
Hong-Xia Jin,a Qi Zhang,a Hye-Sook Kim,b Yusuke Wataya,b He-Hua Liua and Yikang Wua,
*
aState Key Laboratory of Bio-organic & Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy
of Sciences, 354 Fenglin Road, Shanghai 200032, PR China
bFaculty of Pharmaceutical Sciences, Okayama University, Tsuhima, Okayama 700-8530, Japan
Received 20 April 2006; revised 25 May 2006; accepted 26 May 2006
Available online 15 June 2006
Abstract—Several spiroperoxy antimalarial compounds were designed and synthesized using the hydrogen peroxide in UHP (urea–H2O2
complex) as the source of the peroxy bond. Incorporation of the H2O2 into the organic molecule framework through ketal exchange reaction
in the present cases was greatly facilitated by the potential to form a five- or six-membered cyclic hemiketal due to the presence of a hydroxyl
group g or d to the ketone carbonyl group. When the electron-withdrawing group in the Michael acceptor was a nitro group, the closure of the
peroxy ring occurred readily under the hydroxidation conditions. Presence of a benzene ring fused to the peroxy ring effectively reduced the
degrees of freedom in the transition state for the ring-closure step and made the otherwise very difficult seven-membered 1,2-dioxepane rather
easy to form through the intramolecular Michael addition.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
most common ways of creating organic peroxy species. In
2001, Kobayashi5e,f and co-workers reported a new variant
of this methodology, which utilized solid UHP (urea–H2O2
complex) instead of liquid H2O2 as the source of the peroxy
bond. In MeOH with Sc(OTf)3 as catalyst, they obtained
MeO/OOH mixed ketals in high yields and successfully
closed the peroxy ring by an intramolecular Michael addi-
tion catalyzed by HNEt2 in F3CCH2OH. Impressed by the
convenience of using UHP, we also tried to adopt Kobaya-
shi’s methodology in our own work. In the preliminary
exploration6 on the scope and limitation of the UHP proto-
col, we noticed that the original conditions appeared to be
applicable only to synthesis of five- or six-membered peroxy
rings in the monocyclic cases. And, the presence of, e.g., an
ester group on the ketal carbonyl carbon also led to complete
failure in the hydroperoxidation step. These observations,
together with the relatively high cost of the essential catalyst
Sc(OTf)3, prompted us to turn to new structures with extra
rings. Here in this article we wish to detail7 some of our
recent work along this line.
Prompted by the excellent activity observed with qinghaosu1
(artemisinin, 1) and closely related peroxy compounds in the
treatment of malaria including the multi drug-resistant vari-
ants, more and more organic chemists around the world are
actively engaged in design and synthesis of novel organic
peroxides since the late 1980’s. Due to lack of generally
feasible2 synthetic methods for forming the O–O bond
directly from two separate one oxygen-containing groups,
up to now the peroxy bonds in organic peroxides are almost
all derived from some species that already contain the O–O
bond such as dioxygen3 (O2), ozone4 (O3), or hydrogen per-
oxide5 (H2O2). Thus, what is special of synthesis of organic
peroxides is that the way and the timing of introducing the
fragile O–O bond must be taken into consideration in addi-
tion to all other factors encountered in synthesis in general.
The limited source of the O–O bond-containing species de-
cides that the span of the reactions that one can choose from
is narrow. And the fragile nature of O–O bonds demands cir-
cumventing many transformations routinely employed in
organic synthesis. Hence, development of new approaches
to make facile use of the readily available O–O containing
species is critical for synthesis of organic peroxides.
2. Results and discussions
As formation of the hydroperoxy hemiketal is a key step in
incorporating H2O2 into an organic framework, we chose
this as an entry point for this work. The first thing we looked
into was whether the hydroperoxidation could be made
easier by switching from the open-chain hemiketals to cyclic
ones, i.e., to replace the methoxyl group in the monocyclic
Formation of a carbon–oxygen bond to an inorganic O–O
bond-containing reagent through ketalization is one of the
Keywords: Peroxides; Spiroketals; Cyclizations; Hemiketals; Antimalarials.
* Corresponding author. E-mail: yikangwu@mail.sioc.ac.cn
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.05.065

7700
H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
a
CO2R
c
MeO
R
MOMO
S
MOMO
S
O
O
OMe
OMe
CO2R'
CO2R'
O
O
OH
c
O
O
OH
a
b
RO
b
S
S
S
S
4
5 R = Et
6 R = Bn
2 R = H
3 R = MOM
CO2R'
NO2
OH
O
OH
CO2R
d
HO
CO2R
CO2R
Figure 1. General illustration of the plans for this work: (a) the fundamental
idea—switching from the open-chain hemiketal to cyclic hemiketals to
facilitate hydroperoxyl group attachment. (b) Introducing UV chromo-
phores to improve molecule-tracing property and facilitate synthesis. (c)
Examining other electron-withdrawing groups in place of the ester group
in the Michael acceptor.
CO2R
e
HO
O
H
f
O
O
O
O
S
S
O
O
7 R = Et
8 R = Bn
9 R = Et
13 R = Et
14 R = Bn
11 R = Et
12 R = Bn
10 R = Bn
structures with an alkoxyl one that was connected to the
ketal carbonyl carbon through a covalent chain (Fig. 1).
Scheme 1. (a) BuLi/HMPA/I(CH2)2CH(OMe)2/0 ꢀC/48 h, 88%; (b) i.
PPTS/acetone/reflux/12 h; ii. Ph3P]CHCO2Et, 86% for 5 or Ph3P]
CHCO2Bn/CH2Cl2/rt/12 h, 94% for 6 (yields from 4); (c) p-TsOH/EtOH/
reflux, 90% for 7 or 2 N HCl/THF/40–50 ꢀC, 64% for 8; (d) I2/acetone/
NaHCO3/0 ꢀC/30 min, 83% for 9 or 77% for 10; (e) UHP/p-TsOH/
MeOH/rt, 82% for 11 or 75% for 12 and (f) HNEt2/CF3CH2OH/rt/24 h,
52% for either 13 or 14.
To facilitate future mechanistic investigations, it would be
also desirable to incorporate some UV chromophore into
the molecule. Such chromophore was better directly bonded
to the carbon framework so that it would not be lost due to
hydrolysis. From synthetic viewpoint, presence of such
a chromophore as a benzene ring may also facilitate rapid
construction of the molecule because the building blocks
could be more readily accessible. Finally, whether the ester
group in the Michael acceptor could be replaced by other
electron-withdrawing groups was also among what we
wished to explore.
of 4 was done as mentioned above. Then, the condensation
was realized by using MeNO2 instead of a Wittig reagent.
The hydroxyl group was converted to the corresponding
mesylate and eliminated immediately, yielding 16. Up to
this stage, the remaining tasks were removing the MOM
and dithiane protecting groups. Attempts to hydrolyze the
MOM group under similar conditions to those employed in
the synthesis of 7/8 completely failed, presumably due to the
too high reactivity of the Michael acceptor –CH]CHNO2.
Although this property turned out to be an advantage in
a later stage of the synthesis when an intramolecular Michael
addition was needed to construct the 1,2-dioxane ring, at this
point we had to switch to an alternative approach—to free
the carbonyl group first.
Five- or six-membered cyclic hemiketals are generally ex-
pected to form easier than the non-cyclic ones. Although
according to our experience that the population of the cyclic
hemiketal can be rather small in solution (i.e., most of the
substrate is present in the open chain form) in some cases,
we believe that even very small amount of such cyclic hemi-
ketals would also facilitate the critical exchange of the hemi-
ketal hydroxyl group with the hydroperoxyl group. Once the
hydroperoxyl group is introduced into the substrate, it is not
so easy to leave as hydroxyl group. Therefore, the hydroper-
oxidation precursors of all target molecules in this work
were designed to carry a hydroxyl group either g or d to
the ketone carbonyl group.
NO2
c
MOMO
NO2
OH
MOMO
MOMO
OMe
OMe
a
b
S
S
S
S
S
S
15
4
16
The simplest targets (13 and 14) were synthesized using the
route shown in Scheme 1. The MOM protected dithiane 3,
which could be readily prepared from the corresponding
NO2
NO2
alcohol (28), was alkylated with I(CH2)2CH(OMe)2
9
NO2
e
NO2 HO
d
MOMO
O
H
using the standard umpolung technique to give acetal 4.
Selective hydrolysis of the dimethyl acetal with PPTS in
aqueous acetone afforded an intermediate aldehyde, which
was immediately treated with either Ph3P]CHCO2Et or
Ph3P]CHCO2Bn to yield 5 or 6, respectively. The MOM
protecting group was then removed and the resulting hy-
droxy ketone 9 or 10 was converted into the corresponding
hydroperoxyl hemiketal 11 or 12. The presence of an addi-
tional THF ring (cyclic hemiketal) did lead to facile incorpo-
ration of the hydroperoxyl group as expected. The expensive
catalyst Sc(OTf)3 that was essential in the non-cyclic cases
were thus no longer necessary.
O
O
O
O
O
O
O
18
17
19
20
Scheme 2. (a) i. PPTS/acetone/reflux/12 h; ii. CH3NO2/DBU/rt/15 h, 94%
from 4; (b) MsCl/NEt3/rt/24 h, 97%; (c) I2/NaHCO3/0 ꢀC/30 min, 87%;
(d) concd HCl/acetone–H2O/rt/17 h, 21% and (e) UHP/p-TsOH/DME/rt/
10.5 h, 35%.
Under the conditions for the similar deprotections of 7 and 8,
the ketone 17 was readily obtained in 87% yield. The subse-
quent hydrolysis of the MOM group, however, was still very
complicated. We managed to isolate the desired 18 in only
21%. Further treatment of 18 with the UHP/p-TsOH/DME
Next we examined to replace the ester functionality with a
nitro group (Scheme 2). The hydrolysis of the acetal end

H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
7701
(MeO(CH2)2OMe) led to direct formation of the end product
20. Compound 19 could not be isolated at all. Here the non-
alcohol solvent was essential for the hydroperoxidation. Us-
ing MeOH as in Kobayashi’s original protocol, for instance,
led to completely failure.
aldehyde 2711 with lithium species 28 (prepared in situ
from the corresponding bromide) gave diol 29. A Swern
oxidation yielded the intermediate dicarbonyl species 30,
which was immediately treated with Ph3P]CHCO2Et to
afford 31. The TBS protecting group was then removed
and the resulting hydroxy ketone 25 was subjected to the
hydroperoxidation.
After completing the syntheses of 13 and 14, we began to
consider the possibility to introduce a benzene ring into
the spiro[4,5]decane framework. The first designed structure
was 21, partially because the retrosynthetic analysis
(Scheme 3) showed a possible rapid access from three read-
ily available building blocks. Based on this design, we
started the synthesis as shown in Scheme 4.
OH
b
CHO
OLi
a
+
OH
OTBS
27
Li
TBSO
29
28
CO2Et
CO2Et
CHO
O
CO2Et
CO2Et
CO2Et
c
d
O
O
O
O
O
30
TBSO
OH
O
O
25
HO
31
TBSO
OH
21
CO2Et
CO2Et
e
f
Ph3P=CHCO2Et
OBn
O
O
O
O
O
21
O
HOO
26
O
Scheme 5. (a) ꢁ78 ꢀC/40min, 61%; (b) (COCl)2/DMSO/CH2Cl2/NEt3;
(c) Ph3P]CHCO2Et/rt/3 h, 74% from 29; (d) p-TsOH/rt/5 h, 96%; (e)
UHP/CSA/DME/rt/18 h, 86% and (f) HNEt2/CF3CH2OH/rt/2 h, 56%
from 26.
Scheme 3.
O
O
O
Because as mentioned above we already failed to obtain 26
in MeOH, this time tried to use DME as the solvent. Under
such conditions, the desired 26 was formed in 86% yield.
Further treatment with HNEt2 in F3CCH2OH at the ambient
temperature gave the end product 21 as a pair of separable
diastereomers. It is interesting to note that this is the only
case so far we found, where the diastereomers could be sep-
arated on silica gel. Interestingly, the two diastereomers
showed different antimalarial activity in the in vitro testing.
O
b
a
O
O
MeO
MeO
O
OBn
HO
24
22
23
CO2Et
CO2Et
d
c
O
OH
26
O
HOO
25
Previously, we observed that using the intramolecular
Michael addition of the hydroperoxyl group to close
a seven-membered ring was impossible if all the bonds in-
volved in the process were s-bonds without any restrictions
on their rotation. In this work, in combination with our inten-
tion to incorporate a UV chromophore into the carbon
framework, we attempted to fuse a benzene ring to a 1,2-di-
oxepane ring. Such an additional ring would greatly reduce
the degrees of freedom in the transition state for the forma-
tion of the seven-membered ring and therefore should
increase the possibility to reach the 1,2-dioxepane.
Scheme 4. (a) i. n-BuLi/HC^CCH2OBn/THF/ꢁ25 ꢀC/1.5 h, then 22/
ꢁ70 ꢀC/2 h; ii. CH2N2/0 ꢀC, 43% from 22; (b) H2/Pd–C/MeOH/rt, 38%;
(c) i. DIBAL-H/CH2Cl2/ꢁ75 ꢀC/3 h; ii. MnO2/CHCl3/rt/4 d; iii. Ph3P]
CHCO2Et/CHCl3/rt/10 h, 47% from 24 and (d) UHP/p-TsOH or CSA/
MeOH/rt.
Introduction of the three-carbon chain was performed as
reported10 in the literature. The resulting acid was converted
to lactone-ketal 23 by treatment with diazomethane. The
benzyl protecting group was then removed along with the
hydrogenation of the triple bond, leading to 24 in low yield.
Conversion of the lactone 24 into the intermediate aldehyde
was achieved by reduction with DIBAL-H to the correspond-
ing triol, followed by a MnO2 oxidation. A subsequent Wittig
reaction with Ph3P]CHCO2Et on the aldehyde carbonyl
group gave the desired 25 in 47% yield (from 24). The hydro-
peroxidation under the conditions for syntheses of 7 and 8,
however, did not lead to any discernible amounts of 26.
As shown in Scheme 6, from the commercially available in-
expensive triol 32 after a five-step sequence, an allyl group
protected terminal epoxide 33 was conveniently obtained
in 64% overall yield. Ring-opening of the epoxidewith a lith-
ium species prepared in situ from known bromide 34 led to
alcohol 35, which on hydrolytic removal of the MOM pro-
tecting group and a Swern oxidation followed by treatment
with Ph3P]CHCO2Et gave the 37. Cleavage of the allyl
protecting group was then achieved using PdCl2/MeOH.
The product was a mixture of several interchangeable
species, which was therefore all utilized in the next step.
Because the yields along the sequence shown in Scheme 4
were too low, before proceeding we decided to adopt another
more practical approach (Scheme 5). Thus, the reaction of

7702
H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
O
OH
O
b
O
a
c
OMOM
O
a
b
c
O
Br
OH
32
OH
OH
OH
33
OMOM
45
35
OMOM
43
34
MOMO
O
O
d
e
OH
O
O
36
37
O
OH
OH
O
d
CO2Et
e
Br
O
CO2Et
OH
HO
HO
MeO
O
44
O
46
47
+
+
O
40
CO2Et
CO2Et
39
38
CO2Et
OH
HO
O
HO
O
CO2Et
O
CO2Et
O
CO2Et
O
OMOM
+
+
O
g
f
Br
O
49
O
48
50
34
41
CO2Et
42
CO2Et
Scheme 6. (a) i. Me2C(OMe)2/acetone/p-TsOH/rt/22 h; ii. NaH/
CH2]CHCH2Br/rt/10 h; iii. 2 N HCl/MeOH/rt/2 h; iv. p-TsCl/NEt3/rt/
12 h; v. KOH/Et2O/rt/5.5 h, 64% from 32; (b) n-BuLi/34/ꢁ78 ꢀC/1 h,
then 33/BF3$OEt2/ꢁ78 ꢀC/1 h, 80%; (c) concd HCl/MeOH/rt/4 d, 93%;
(d) i. (COCl)2/DMSO/NEt3; ii. Ph3P]CHCO2Et/rt/12 h, 85% from 36;
(e) PdCl2/MeOH/rt/24 h, then 60 ꢀC/4 h; (f) UHP/p-TsOH/DME/rt/22 h,
71% from 37 and (g) HNEt2/CF3CH2OH/rt/19 h, 59%.
O
O
f
g
O
O
OH
O
52
51
CO2Et
CO2Et
Scheme 7. (a) i. n-BuLi/BF3$Et2O/epichlorohydrin/ꢁ78 ꢀC/1 h; ii. NaH/rt/
2.5 h, 67% from 34; (b) n-BuLi/44/ꢁ78 ꢀC/1 h, then 43/BF3$OEt2/ꢁ78 ꢀC/
1 h, 76%; (c) concd HCl/MeOH/rt/3 d, 99%; (d) i. (COCl)2/DMSO/NEt3;
ii. Ph3P]CHCO2Et/rt/12 h, 77% from 46; (e) PdCl2/CuCl/DMF–H2O/air/
rt/12 h; (f) UHP/CSA/DME–EtOH/rt/7 d, 33% from 47 and (g) HNEt2/
CF3CH2OH/rt/21 h, 24%.
The hydroperoxidation turned out to be a smooth reaction as
we had expected. The starting mixture of 38, 39, and 40 led
to the hydroperoxy hemiketal 41 in 71% yield. Due to the
presence of the additional benzene ring, the intramolecular
Michael also proceeded very well, rendering the desired
1,2-dioxepane in 59% yield.
shown in Scheme 8, alkylation of furfuryl alcohol with 53
under the conditions similar to those reported14 in the litera-
ture, followed by manipulation of the furan ring by MCPBA
oxidation15 and hydrogenation16 led to compound 55.
Using a similar strategy we also synthesized 52, which con-
tained two fused benzene rings. The synthetic sequence is
depicted in Scheme 7. The initial steps were rather conve-
nient due to the possibility to use 34,12 44,13 and epichloro-
hydrin. Like in the case of deprotection of 37, removal of the
allyl protecting group in 47 also resulted in a mixture of
several closely related species (48, 49, and 50). Treatment
of this mixture with UHP/DME/CSA led to partial conver-
sion of 48 into the anticipated hydroperoxy hemiketal 51,
along with recovered 49 and 50 (which reacted much more
slowly than 48). In order to speed up the reaction, higher
doses of UHP (13 equiv) and CSA (5 equiv) were required.
Increased amounts of UHP and CSA, however, created sol-
ubility problem. Fortunately we found that addition of a care-
fully controlled amount of EtOH could facilitate dissolution
of the reagents without interfering the desired transforma-
tion. The presence of an additional benzene ring in 51 also
created solubility problem in the final Michael addition
step. Unlike 41, the 51 was not so soluble in F3CCH2OH.
Some CH2Cl2 must be added before the ring closure could
proceed to a synthetically useful extent.
MeO
CO2Et
O
O
61
The hydroxyl group was then masked as an acetate by treat-
ment with Ac2O before selectively converting the carbonyl
group closer to the acetate to an a,b-unsaturated ester. The
isomer with ester group cis to the hydroxyl group formed
the corresponding lactone 57 easily, which allowed for facile
separation of the trans isomer 56 for further transformations.
Again, the hydroperoxidation was apparently facilitated by
the possibility to form a cyclic hemiketal. The hydroperoxy
hemiketal 58 could be obtained in 44% yield. However, no
peroxide 60 could be detected at all. Instead, an epoxide
(59) was formed as the major product and isolated in 58%
yield.
Apart from the spirosystems mentioned above, we also
attempted to synthesize such bridged peroxides as 60. As
The results of the preliminary in vitro tests are shown in
Table 1. The EC50 values for Plasmodium falciparum were

H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
7703
OH
to formation of a cyclic hemiketal. When the electron-with-
drawing group in the Michael acceptor was a nitro group, the
closure of the peroxy ring occurred readily under the hydrox-
idation conditions. Presence of a benzene ring fused to the
peroxy ring effectively reduced the degrees of freedom in
the transition state for the ring-closure step and made the
otherwise very difficult seven-membered 1,2-dioxepane
rather easy to form through the intramolecular Michael
addition.
Ph
O
O
Ph
I
a
b
+
O
Ph
HO
CO2Et
OH
O
53
56
54
55
O
OH
O
Ph
d
c
Ph
HOO
CO2Et
CO2Et
58
O
O
59
4. Experimental
4.1. General
O
Ph
O
O
O
O
CO2Et
Ph
The 1H NMR and 13C NMR spectrawere recorded in deutero-
chloroform at ambient temperature using a Varian Mercury
300 or a Bruke Avance 300 instrument (operating at
300 MHz for proton). The FTIR spectra were scanned with
a Nicolet Avatar 360 FTIR. EIMS and EIHRMS were
recorded with an HP 5989A and a Finnigan MAT 8430
mass spectrometer, respectively. The ESIMS and ESIHRMS
were recorded with a PE Mariner API-TOF and a APEX III
(7.0 tesla) FTMS mass spectrometer, respectively. Elemental
analyses were performed on an Elementar VarioEL III instru-
ment. The melting point was uncorrected. Dry THF was
distilled from Na/Ph2CO under N2. Dry CH2Cl2 was distilled
60
57
Scheme 8. (a) i. n-BuLi/ꢁ40 ꢀC/4 h, then 53/0 ꢀC/43 h, 42%; ii. m-CPBA/
rt/2 h, 95%; iii. H2/Pd–C/rt/7 h, 40%; (b) i. Ac2O/NEt3/rt/5 h; ii. NaH/
(EtO)2(O)P]CHCO2Et/rt/7 h; iii. 2 N HCl/EtOH/rt/2 d, 18% from 55;
(c) UHP/p-TsOH/DME/rt/24 h, 44% and (d) HNEt2/CF3CH2OH/rt/10 h,
58% for 59.
Table 1. The results of the preliminary in vitro testsa
Entry
Compd
EC50(M)
Ratiob
P. falciparum
FM3A
˚
over CaH2 and kept over 4 A molecular sieves. UHP was
1
13
8ꢂ10ꢁ6
>3.5ꢂ10ꢁ5
(100% growth)
1ꢂ10ꢁ5
>5
purchased from Acros. All other solvents and reagents
were commercially available and used as received without
any further purification.
2
3
14
20
7ꢂ10ꢁ6
2
>4.8ꢂ10ꢁ5
(70% growth)
4ꢂ10ꢁ7
>4.8ꢂ10ꢁ5
(100% growth)
>2.7ꢂ10ꢁ5
(100% growth)
1ꢂ10ꢁ5
4
21ac
>63
4.2. MOM protection of 2 (3)
5
6
7
8
21bd
42
6ꢂ10ꢁ6
7ꢂ10ꢁ6
5ꢂ10ꢁ6
6ꢂ10ꢁ7
2
1
1
12
8ꢂ10ꢁ6
A solution of 2 (14.24 g, 80 mmol) and anhydrous LiBr
(1.392 g, 16 mmol) in CH(OMe)2 (160 mL) was stirred at
the ambient temperature for 2 d. The mixture was partitioned
between brine and Et2O. The aqueous layer was back-
extracted with Et2O thrice. The combined ethereal phases
were concentrated to dryness on a rotary evaporator. The
residue was chromatographed on silica gel (20:1 n-hexane/
EtOAc) to give 3 as a colorless oil (14.448 g, 81% yield).
FTIR (film) 2932, 1422, 1276, 1147, 1111, 1041, 919,
771 cmꢁ1; 1H NMR d 4.61 (s, 2H), 4.05 (t, J¼6.6 Hz, 1H),
3.55 (t, J¼6.1 Hz, 2H), 3.36 (s, 3H), 2.89–2.84 (m, 4H),
1.89–1.80 (m, 6H); EIMS m/z (%) 222 (M+, 1), 177
(M+ꢁCH2OCH3, 26), 103 (49), 45 (100). EIHRMS calcd
for C9H18O2S2 222.0748, found 222.0750.
52
7ꢂ10ꢁ6
61e
7ꢂ10ꢁ6
a
For detailed information about the biological testing, see the experimental
part.
b
c
d
e
The EC50 (P. falciparum)/EC50 (FM3A).
The less polar diastereomer of 21.
The more polar diastereomer of 21.
Published in Ref. 6a (listed here for comparison).
within the range between 10ꢁ7 to 10ꢁ6 M, which were more
or less the same as those reported5d,6a for the monocyclic
simple analogues of peroxyplakoric acid. The nitro analogue
20, however, was apparently less potent than other com-
pounds tested, perhaps because the nitro group somehow
interfered the radical reactions after the cleavage of the
peroxy bond in the living cell. More work needs to be done
before it is possible to give a more plausible explanation. It
is also interesting to note that the two diastereomers of 21
showed significantly different activities.
4.3. Alkylation of 3 (4)
With cooling (ꢁ78 ꢀC bath) and stirring n-BuLi (30.0 mL,
1.6 M) was added dropwise (via a syringe) to a solution of
the dithiane 3 (9.033 g, 40.5 mmol) and Ph3CH (15 mg, as
an indicator) in dry THF (120 mL) under argon, followed
by HMPA (7 mL, 40.7 mmol) (via a syringe). The bath was
allowed to warm to ꢁ20 ꢀC and the magenta solution was
stirred at that temperature for 4 h, before the iodide
(10.295 g, 44.8 mmol) was introduced via a syringe. After
further stirring at 0 ꢀC for 48 h, the mixture was diluted
with Et2O and washed with water. The aqueous layer was
back-extracted with Et2O. The combined organic layers
werewashed withbrine, driedoverNa2SO4, andconcentrated
3. Conclusions
Several spiroperoxy antimalarial compounds were designed
and synthesized using the hydrogen peroxide in UHP (urea–
H2O2 complex) as the source of the peroxy bond. Incorpora-
tion of the H2O2 into the organic molecule framework
through ketal exchange reaction was greatly facilitated due

7704
H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
on a rotary evaporator. The residue was chromatographed
on silica gel (10:1 n-hexane/EtOAc) to give compound 4
(11.654 g, 88%). FTIR (film) 2933, 1451, 1383, 1151,
4.6. Hydrolysis of the dithiane protecting group in 7 (9)
NaHCO3 (6.183 g, 73.6 mmol) and I2 (9.209 g, 36.3 mmol)
were added to a solution of 7 (3.34 g, 11.0 mmol) in aqueous
acetone (94 mL of acetone plus 17 mL water). The resulting
mixture was stirred at 0 ꢀC for 30 min until TLC showed
complete disappearance of 7. The reaction was quenched
with saturated aqueous Na2S2O3 and the mixture was diluted
with Et2O. The phases were separated and the aqueous layer
was extracted with Et2O. The combined organic layers were
washed with saturated aqueous Na2S2O3, water, and brine,
dried over Na2SO4, and concentrated on a rotary evaporator.
The residue was purified by column chromatography on
silica gel (2:1 n-hexane/EtOAc) to give 9 as a yellowish oil
1128, 1041, 919 cmꢁ1 1H NMR d 4.61 (s, 2H), 4.38
;
(t, J¼5.4 Hz, 1H), 3.54 (t, J¼6.0 Hz, 2H), 3.36 (s, 3H),
3.32 (s, 6H), 2.84–2.79 (m, 4H), 1.98–1.92 (m, 6H), 1.81–
1.72 (m, 4H); 13C NMR (75 MHz, CDCl3) d 104.2, 96.3,
67.4, 55.1, 52.7, 52.5, 35.1, 32.8, 27.2, 25.9, 25.3, 24.4;
EIMS m/z (%) 324 (M+, 73), 293 (M+ꢁOCH3, 22), 75
(100); ESIHRMS calcd for C14H28O4S2Na ([M+Na]+)
347.1321, found 347.1326.
4.4. Synthesis of 5
(3.892 g, 83%). FTIR (film) 1716, 1655, 1043, 979 cmꢁ1
;
A mixture of 4 (3.202 g, 9.9 mmol) and PPTS (588 mg,
2.4 mmol) in aqueous acetone (100 mL of acetone plus
10 mL water) was heated to reflux with stirring for 12 h,
when TLC showed the hydrolysis to be complete. Acetone
was removed on a rotary evaporator, and the residue was di-
luted with Et2O and washed with water. The aqueous layer
was back-extracted with Et2O. The combined organic layers
were washed with brine, dried over Na2SO4, and concen-
trated on a rotary evaporator to give aldehyde as a yellowish
oil (2.86 g), which was dissolved in CH2Cl2 (68 mL) and
treated with Ph3P]CHCO2Et (6.0 g, 18.9 mmol). The mix-
ture was stirred at ambient temperature until TLC showed
complete disappearance of the aldehyde. The solvent was re-
moved on a rotary evaporator, and the residue was diluted
with Et2O. The solid–liquid mixture was filtered and the fil-
trate was concentrated on a rotary evaporator. The residue
was chromatographed on silica gel (6:1 n-hexane/EtOAc)
to give compound 5 as a colorless oil (3.033 g, 86%).
1H NMR d 6.93 (dt, J¼16, 6.9 Hz, 1H), 5.83 (d, J¼16 Hz,
1H), 4.18 (q, J¼7.4 Hz, 2H), 3.66 (dt, J¼10, 5.7 Hz, 2H),
2.65–2.56 (m, 4H), 2.53–2.45 (m, 2H), 1.91–1.82 (m, 2H),
1.28 (t, J¼7.2 Hz, 3H); ESIMS m/z 232 ([M+NH4]+);
EIHRMS calcd for C11H18O4Na ([MNa]+) 237.1097, found
237.1089.
4.7. Synthesis of 13
p-TsOH (3.27 g, 17.2 mmol) and UHP (9.927 g,
105.5 mmol) were added to a solution of 9 (2.995 g,
14.0 mmol) in MeOH (280 mL). The resulting mixture
was stirred at ambient temperature for 20 h. When TLC
showed complete disappearance of 9, the mixture was
diluted with Et2O and washed with water. The aqueous layer
was back-extracted with Et2O. The combined organic layers
were washed with brine, dried over Na2SO4, and concen-
trated on a rotary evaporator. The residue was chromato-
graphed on silica gel (4:1 n-hexane/EtOAc) to give
compound 11 (2.633 g, 82%), which gave the following
1
FTIR (film) 1717, 1269, 1042, 919 cmꢁ1; H NMR d 6.99
(dt, J¼16, 6.8 Hz, 1H), 5.87(d, J¼16 Hz, 1H), 4.63 (s,
2H), 4.20 (q, J¼6.9 Hz, 2H), 3.56 (t, J¼5.7 Hz, 2H), 3.37
(s, 3H), 2.84–2.81 (m, 4H), 2.43–2.35 (m, 2H), 2.07–1.95
(m, 6H), 1.80–1.73 (m, 2H), 1.30 (t, J¼6.9 Hz, 3H); EIMS
m/z (%) 348 (M+, 17), 303 (M+ꢁOCH2CH3, 4), 45 (100);
ESIHRMS calcd for C16H28O4S2Na ([M+Na]+) 371.1321,
found 371.1312. Anal. calcd for C16H28O4S2: C, 55.14; H,
8.10. Found: C, 55.26; H, 8.23.
1
data: H NMR d 8.06 (s, 1H), 6.99 (dt, J¼16, 6.7 Hz, 1H),
5.85 (dt, J¼16, 1.3 Hz, 1H), 4.19 (q, J¼7.1 Hz, 2H),
4.02–3.97 (m, 2H), 2.27–2.05 (m, 4H), 2.05–1.79 (m, 4H),
1.29 (t, J¼7.2 Hz, 3H). The 11 (2.633 g, 11.5 mmol) was
then dissolved in CF3CH2OH (64 mL) containing NHEt2
(100 mL). The mixture was stirred at the ambient tempera-
ture until TLC showed complete disappearance of 11. The
solvent was removed on a rotary evaporator. The residue
was chromatographed on silica gel (5:1 n-hexane/EtOAc)
to give compound 13 as a colorless oil (1.369 g, 52%).
4.5. Removal of the MOM protecting group in 5 (7)
A mixture of 5 (1.946 g, 5.6 mmol) and p-TsOH (106 mg,
0.56 mmol) in EtOH (40 mL) was heated to reflux with stir-
ring until TLC showed the hydrolysis to be complete. The
mixture was diluted with Et2O and washed with water. The
aqueous layer was back-extracted with Et2O. The combined
organic layers were washed with brine, dried over Na2SO4,
and concentrated on a rotary evaporator. The residue was
chromatographed on silica gel (3:1 n-hexane/EtOAc) to
give compound 7 as a colorless oil (1.52 g, 90%). FTIR
1
FTIR (film) 1737, 1445, 1370, 1289, 1184, 1032 cmꢁ1; H
NMR d 4.56–4.54 (m, 1H), 4.15 (q, J¼7.1 Hz, 2H), 4.08–
4.02 (m, 2H), 2.56 (dd, J¼7.4, 16 Hz, 1H), 2.39 (dd,
J¼6.0, 16 Hz, 1H), 2.05–1.74 (m, 8H), 1.25 (t, J¼7.1 Hz,
3H); ESIMS m/z 248 ([M+NH4]+). Anal. calcd for
C11H18O5: C, 57.38; H, 7.88. Found: C, 57.56; H, 7.87.
4.8. Synthesis of compound 6
1
(film) 3438, 1713, 1273, 1046, 908 cmꢁ1; H NMR d 6.99
(dt, J¼15, 6.8 Hz, 1H), 5.88 (d, J¼15 Hz, 1H), 4.20 (q,
J¼7.2 Hz, 2H), 3.71–3.68 (m, 2H), 2.85–2.81 (m, 4H),
2.43–2.35 (m, 2H), 2.07–1.94 (m, 6H), 1.78–1.69 (m, 2H),
1.30 (t, J¼7.1 Hz, 3H); EIMS m/z (%) 304 (M+, 9), 259
(M+ꢁOEt, 3), 245 (28), 197 (100); EIHRMS calcd for
C14H24O3S2Na ([M+Na]+) 327.1059, found 327.1063.
Anal. calcd for C14H24O3S2: C, 55.23; H, 7.95. Found: C,
55.39; H, 7.82.
Prepared from 4 in 94% yield using the same procedure
for the synthesis of 5 described above except that
Ph3P]CHCO2Bn was utilized instead of Ph3P]CHCO2Et.
Data for 6 (a colorless oil): FTIR (film) 1720, 1653, 1040,
1
919 cmꢁ1; H NMR d 7.39–7.32 (m, 5H), 7.04 (dt, J¼16,
6.7 Hz, 1H), 5.92 (d, J¼16 Hz, 1H), 5.18 (s, 2H), 3.54 (t,
J¼6.2 Hz, 2H), 3.34 (s, 3H), 2.81 (t, J¼5.7 Hz, 4H), 2.43–
2.35 (m, 2H), 2.05–1.93 (m, 6H), 1.78–1.89 (m, 2H); ESIMS

H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
7705
m/z 428 ([M+NH4]+). ESIHRMS calcd for C21H30O4S2Na
([M+Na]+) 433.1478, found 433.1479. Anal. calcd for
C21H30O4S2: C, 61.43; H, 7.36. Found: C, 61.45; H, 7.39.
18 mL of water) was heated to reflux with stirring until
TLC showed completion of the hydrolysis (12 h). Acetone
was removed on a rotary evaporator. The residue was diluted
with Et2O and washed with water. The aqueous layer was
back-extracted with Et2O. The combined organic layers
were washed with brine, dried over anhydrous Na2SO4,
and concentrated on a rotary evaporator to give the interme-
diate aldehyde as a yellowish oil (4.92 g), which was dis-
solved in MeCN (8 mL) and added to a mixture of DBU
(0.39 mL, 2.64 mmol), MeNO2 (3.22 g, 26.4 mmol, a com-
mercially available 50% solution in MeOH), and MeCN
(40 mL). The mixture was then stirred at the ambient tem-
perature until TLC showed complete disappearance of the
aldehyde (ca. 15 h). The reaction was quenched with satu-
rated aqueous NH4Cl. The phases were separated. The
aqueous layer was back-extracted with Et2O. The combined
organic layers were washed with water and brine, dried over
anhydrous Na2SO4, and concentrated on a rotary evaporator.
The residue was purified by column chromatography on
silica gel (3:1 n-hexane/EtOAc) to give 15 as a colorless
oil (5.625 g, 94% overall). FTIR (film) 3425, 2932, 1553,
4.9. Removal of the MOM group in 6 (8)
A solution of 6 (3.941 g, 9.6 mmol) in THF (30 mL) and 2 N
HCl (15 mL) was stirred at 40–50 ꢀC until TLC showed
complete disappearance of 6. The reaction was quenched
with saturated aqueous NaHCO3. The layers were separated
and the aqueous layer was extracted with Et2O. The com-
bined organic layers were washed with water and brine,
dried over anhydrous Na2SO4. The solvent was removed
on a rotary evaporator. The residue was chromatographed
on silica gel (3:1 n-hexane/EtOAc) to give 8 (2.246 g,
64%). FTIR (film) 3421, 1717, 1652, 1272, 1163, 1017,
1
978 cmꢁ1; H NMR d 7.39–7.27 (m, 5H), 7.04 (dt, J¼16,
6.7 Hz, 1H), 5.92 (d, J¼16 Hz, 1H), 5.18 (s, 2H), 3.67 (t,
J¼6.1 Hz, 2H), 2.83–2.79 (m, 4H), 2.43–2.35 (m, 2H),
2.06–1.92 (m, 6H), 1.76–1.66 (m, 2H); 13C NMR
(75 MHz, CDCl3) d 166.2, 148.8, 135.9, 128.4, 128.1,
121.2, 66.0, 62.4, 52.4, 36.2, 36.8, 27.2, 25.8, 25.1; ESIMS
m/z 384 ([M+NH4]+); ESIHRMS calcd for C19H27O3S2
([M+H]+) 367.1396, found 367.1394.
1
1383, 1109, 1036 cmꢁ1; H NMR d 4.63 (s, 2H), 4.46–
4.42 (m, 2H), 3.56 (t, J¼6.0 Hz, 2H), 3.37 (s, 3H), 2.83 (t,
J¼5.7 Hz, 4H), 2.73 (d, J¼5.0 Hz, 1H), 2.25–2.15 (m,
1H), 1.99–1.94 (m, 6H), 1.74–1.70 (m, 4H); EIMS m/z (%)
339 (M+, 4), 294 (M+ꢁCH2OCH3), 45 (100); ESIHRMS
calcd for C13H25NO5S2Na ([M+Na]+) 362.1066, found
362.1075.
4.10. Hydrolysis of the dithiane protecting group
in 8 (10)
Compound 10 was obtained in 77% yield from 8 using the
same procedure described above for the synthesis of 9.
Data for 10 (a colorless oil): FTIR (film) 3420, 1716,
4.13. Elimination of the hydroxyl group in 15 (16)
1
1654, 1267, 1171, 1026, 699 cmꢁ1; H NMR d 7.38–7.27
With cooling (0 ꢀC bath) and stirring, MsCl (2.3 mL,
30 mmol) and NEt3 (5 mL, 36 mmol) were added to a solu-
tion of 15 (6.056 g, 17.86 mmol) in CH2Cl2 (90 mL). The
stirring was continued at the ambient temperature for 24 h.
The reaction was quenched with saturated aqueous NH4Cl.
The phases were separated. The aqueous layer was back-
extracted with Et2O. The combined organic layers were
washed with water and brine, dried over anhydrous
Na2SO4, and concentrated on a rotary evaporator. The resi-
due was purified by column chromatography on silica gel
(6:1 n-hexane/EtOAc) to give 16 as a yellowish colorless
oil (5.588 g, 97%). FTIR (film) 2933, 1648, 1524, 1350,
(m, 5H), 6.98 (dt, J¼15, 7.0 Hz, 1H), 5.88 (d, J¼15 Hz,
1H), 5.17 (s, 2H), 3.64 (t, J¼5.8 Hz, 2H), 2.64–2.45 (m,
6H), 1.89–1.80 (m, 2H); ESIMS m/z 294 ([M+NH4]+);
ESIHRMS calcd for C16H20O4Na ([M+Na]+) 299.1254,
found 299.1240.
4.11. Synthesis of compound 14
Using the same procedure described above for converting 9
into 11, the intermediate hydroperoxide 12 was obtained in
75% from 10, which showed the following: 1H NMR
d 8.14 (s, 1H), 7.39–7.32 (m, 5H), 7.05 (dt, J¼15.3,
6.6 Hz, 1H), 5.91 (d, J¼15.6 Hz, 1H), 5.18 (s, 2H), 4.01–
3.96 (m, 2H), 2.41–2.15 (m, 3H), 2.08–1.77 (m, 5H). The
12 was then converted into 14 (1:4 mixture of two diastereo-
mers) in 52% yield using the same procedure as described
above for the conversion of 11 into 13. Data for 14 (a color-
less oil): FTIR (film) 1724, 1455, 1291,1033 cmꢁ1; 1H NMR
d 7.37–7.31 (m, 5H), 5.14 (s, 2H), 4.63–4.56 (m, 1H), 4.11–
4.00 (m, 2H), 2.88 and 2.64 (two doublets in 1:4 ratio, J¼16,
7.4 Hz, 1H altogether, part of an AB system), 2.60 and 2.47
(two doublets in 1:4 ratio, J¼16, 7.4 Hz, 1H altogether, part
of an AB system), 2.06–1.75 (m, 8H); ESIMS m/z 310
([M+NH4]+); ESIHRMS calcd for C16H20O5Na ([M+Na]+)
315.1203, found 315.1210. Anal. calcd for C16H20O5: C,
65.74; H, 6.90. Found: C, 65.86; H, 6.97.
1
1039 cmꢁ1; H NMR d 7.34–7.27 (m, 1H), 7.04 (d, J¼
13 Hz, 1H), 4.63 (s, 2H), 3.55 (t, J¼3.1 Hz, 2H), 3.37 (s,
3H), 2.83 (t, J¼5.7 Hz, 4H), 2.53–2.45 (m, 2H), 2.10–2.04
(m, 2H), 2.01–1.96 (m, 4H), 1.79–1.72 (m, 2H); 13C NMR
(75 MHz, CDCl3) d 141.8, 139.7, 96.3, 67.3, 55.1, 52.2,
36.1, 35.4, 25.9, 24.9, 24.5, 23.8; EIMS m/z (%) 321 (M),
275 (M+ꢁNO2, 32), 218 (M+ꢁC2H6OMOM, 31), 45
(100); ESIHRMS calcd for C13H23NO4S2Na ([M+Na]+)
344.0961, found 344.0966.
4.14. Hydrolysis of the dithiane protecting group
in 16 (17)
Compound 17 was obtained in 87% yield from 16 using the
same procedure described above for the synthesis of 9. Data
for 17 (a colorless oil): FTIR (film): 1715, 1649, 1525, 1352,
4.12. Conversion of 4 into 15
1040 cmꢁ1 1H NMR d 7.29–7.20 (m, 1H), 7.01 (d,
;
J¼14 Hz, 1H), 4.59 (s, 2H), 3.54 (t, J¼6.3 Hz, 2H), 3.35
(s, 3H), 2.68 (t, J¼6.7 Hz, 2H), 2.58–2.51 (m, 4H), 1.90
(quint, J¼6.6 Hz, 2H); ESIMS m/z 254 ([M+Na]+). Anal.
A mixture of 4 (6.3582 g, 19.6 mmol) and PPTS (523 mg,
2.0 mmol) in aqueous acetone (178 mL of acetone plus

7706
H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
calcd for C10H17NO5: C, 51.94; H, 7.41; N, 6.06. Found: C,
52.13; H, 7.13; N, 6.06.
4.17. Synthesis of compound 24
A mixture of 23 (1.208 g, 3.9 mmol) and 10% palladium on
charcoal (500 mg) in MeOH (20 mL) was stirred at the
ambient temperature under H2 (1 atm) until TLC showed
complete disappearance of 23 (5.5 h). The catalyst was fil-
tered off and the filtrate was concentrated on a rotary evap-
orator. The residue was chromatographed on silica gel (4:1
n-hexane/EtOAc) to give compound 24 as a colorless oil
(332 mg, 38%). FTIR (film) 3402, 2932, 1723, 1434, 1264,
4.15. Synthesis of 20
Removal of the MOM group in 17, utilizing the same proce-
dure described above for transforming 5 into 7, led to 18 in
21% yield (chromatography on silica gel eluting with 1:1
n-hexane/EtOAc). FTIR (film) 3411, 1712, 1650, 1525,
1
1352, 1058 cmꢁ1; H NMR d 7.30–7.20 (m, 1H), 7.01 (d,
1
J¼14 Hz, 1H), 3.85–3.91 (m, 1H), 3.67 (t, J¼5.9 Hz, 2H),
2.72–2.67 (m, 2H), 2.61–2.54 (m, 4H), 1.93–1.85 (m, 2H).
The 18 thus obtained (59 mg, 0.32 mmol) was immediately
dissolved in DME (3.2 mL). p-TsOH (61 mg, 0.32 mmol)
and UHP (208 mg, 2.2 mmol) were then added. The mixture
was stirred at the ambient temperature until TLC showed
complete disappearance of 18 (ca. 10.5 h). The mixture
was diluted with Et2O and washed with water. The phases
were separated. The aqueous layer was back-extracted
with Et2O. The combined organic layers were washed with
brine, dried over anhydrous Na2SO4, and concentrated on
a rotary evaporator. The residue was chromatographed on
silica gel (5:1 n-hexane/EtOAc) to give compound 20
1091 cmꢁ1; H NMR d 7.90 (d, J¼7.1 Hz, 1H), 7.46–7.40
(m, 1H), 7.28–7.23 (m, 2H), 3.90 (s, 3H), 3.72 (t,
J¼5.6 Hz, 2H), 2.97 (t, J¼7.7 Hz, 2H), 1.72–1.68 (m, 2H);
ESIMS m/z 245 ([M+Na]+); ESIHRMS calcd for
C12H14O4Na ([M+Na]+) 245.0784, found 245.0783.
4.18. Synthesis of compound 25 (from 24)
DIBAL-H (1.0 M solution in cyclohexane, 1.8 mL) was
added (via a syringe) to a solution of 24 (132 mg,
0.59 mmol) in dry CH2Cl2 (3 mL) stirred at ꢁ75 ꢀC under
N2. The stirring was continued at that temperature for 3 h.
Then the bath was allowed to warm to the ambient tem-
perature before the reaction was quenched with MeOH.
The reaction mixture was shaken with aqueous saturated
potassium sodium tartrate. The phases were separated. The
aqueous layer was back-extracted with EtOAc. The com-
bined organic layers were washed with brine, dried over
anhydrous Na2SO4, and concentrated on a rotary evapora-
tor to give the intermediate triol (160 mg), which was dis-
solved in CHCl3 (3 mL) and treated with activated MnO2
(262 mg, 3 mmol). The mixture was stirred at the ambient
temperature for 4 d. The solids were filtered off (rinsed
with CHCl3) and the combined filtrate/washings were con-
centrated on a rotary evaporator and added to a solution of
Ph3P]CHCO2Et (150 mg, 0.43 mmol) in CHCl3 (3 mL).
The mixture was stirred at the ambient temperature for
10 h. The solvent was removed on a rotary evaporator.
The solid residue was triturated with Et2O. The solids
were filtered off and the filtrate was concentrated on a ro-
tary evaporator. The residue was chromatographed on silica
gel (4:1 n-hexane/EtOAc) to give compound 25 as a color-
less oil (73 mg, 47%). FTIR (film) 3417, 2928, 1683, 1634,
1478, 1175, 1036 cmꢁ1; 1H NMR d 8.05 (d, J¼16 Hz, 1H),
7.72 (d, J¼7.7 Hz, 1H), 7.62–7.09 (m, 3H), 6.31 (d,
J¼16 Hz, 1H), 4.30 (q, J¼7.2 Hz, 2H), 3.76 (t, J¼5.8 Hz,
2H), 3.08 (t, J¼6.8 Hz, 2H), 2.05–1.98 (m, 2H), 1.69
(br s, 1H), 1.34 (t, J¼7.2 Hz, 3H); EIMS m/z (%) 245
(M+ꢁOH, 0.6), 189 (M+ꢁCO2Et, 100), 217 (M+ꢁOEt,
2). Anal. calcd for C15H18O4: C, 68.68; H, 6.92. Found:
C, 68.32; H, 6.79.
(23 mg, 35%). FTIR (film) 2960, 1556, 1371, 1041 cmꢁ1
;
1H NMR d 4.95–4.86 (m, 1H), 4.54 (dd, J¼9.1, 14 Hz,
1H), 4.37 (dd, J¼3.5, 13.8 Hz, 1H), 4.09–4.03 (m, 2H),
2.08–1.88 (m, 6H), 1.82–1.76 (m, 2H); ESIMS m/z 221
([M+NH4]+); EIHRMS calcd for C8H13NO5 (M+)
203.0806, found 203.0818.
4.16. Synthesis of 23
With cooling (ꢁ70 ꢀC bath), n-BuLi (13.7 mL, 1.6 M) was
added dropwise (via a syringe) to a solution of HC^
CCH2OBn (3.193 g, 21.9 mmol) in dry THF (45 mL) stirred
under argon. The bath was allowed to warm to ꢁ25 ꢀC and
kept at that temperature for 1.5 h before being added via
a cannula to a solution of 22 (4.855 g, 32.8 mmol) in dry
THF (55 mL) stirred at ꢁ70 ꢀC under argon. The mixture
was stirred at that temperature for 2 h. The reaction was
quenched with water and acidified to pH 4 with 2 N HCl.
The phases were separated. The aqueous layer was back-
extracted with EtOAc. The combined organic layers were
washed with brine, dried over anhydrous Na2SO4, and con-
centrated on a rotary evaporator to give the intermediate
ketone-acid as a yellowish oil (3.906 g), which was dis-
solved in Et2O (48 mL) and treated with an excess of
CH2N2 (an ethereal solution prepared using standard proce-
dure prior to use) at 0 ꢀC. When TLC showed complete dis-
appearance of the ketone-acid, the cooling bath was allowed
to warm to the ambient temperature. The stirring was contin-
ued for another hour before removal of the solvent by rotary
evaporation. The residue was chromatographed on silica gel
(8:1 n-hexane/EtOAc) to give compound 23 as a yellowish
oil (2.103 g, 43%). FTIR (film) 3031, 2951, 2221, 1732,
4.19. Synthesis of compound 29
n-BuLi (50 mL, 1.2 M) was added dropwise (via a syringe)
to a solution of o-BrC6H4CH2OH (5.578 g, 29.8 mmol) in
dry THF (150 mL) stirred at ꢁ78 ꢀC under N2. After the
completion of the addition, the stirring was continued at
that temperature for 1 h. Then, a solution of 27 (prepared
by Swern oxidation from the corresponding alcohol
(6.78 g, 33.2 mmol)) in dry THF (20 mL) was added drop-
wise via a syringe. The mixture was stirred at the same tem-
perature for another 40 min before saturated aqueous NH4Cl
1650, 1260, 1074, 698 cmꢁ1; H NMR d 7.94–7.91 (m,
1
1H), 7.69–7.59 (m, 3H), 7.38–7.27 (m, 5H), 4.65 (s, 2H),
4.41 (s, 2H), 3.91 (s, 3H); 13C NMR (75 MHz, CDCl3)
d 180.6, 140.2, 139.4, 135.2, 135.1, 133.7, 132.8, 131.7,
131.3, 130.88, 130.86 (2C), 93.4, 87.4, 75.0, 59.8, 55.5;
EIMS m/z (%) 293 (M+ꢁMe, 1), 263 (M+ꢁOEt, 54), 91
(M+ꢁBn, 64), 170 (100); MALDIHRMS calcd for
C19H16O4Na ([M+Na]+) 331.0941, found 331.0942.

H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
7707
was introduced to quench the reaction. The mixture was di-
4.22. Synthesis of compound 21
luted with Et2O. The phases were separated. The aqueous
layer was back-extracted with Et2O. The combined organic
layers were washed with water and brine, dried over anhy-
drous Na2SO4, and concentrated on a rotary evaporator.
The residue was chromatographed on silica gel (4:1 n-
hexane/EtOAc) to give compound 29 as a yellowish oil
(5.577 g, 61% from 28). FTIR (film) 3351, 2927, 1463,
A solution of 25 (1.905 g, 7.3 mmol), CSA (5.0 g,
22.0 mmol), and UHP (4.7 g, 51 mmol) in DME (116 mL)
was stirred at the ambient temperature until TLC showed
complete disappearance of 25 (ca.18 h). The mixture was
partitioned between Et2O and water. The phases were sepa-
rated. The aqueous layer was back-extracted with Et2O. The
combined organic layers were washed with brine, dried over
anhydrous Na2SO4, and concentrated on a rotary evaporator.
The residue was chromatographed on silica gel (3:1 n-hex-
ane/EtOAc) to give compound 26 (1.744 g, 86%), which
gave the following data: FTIR (film) 3375, 1712, 1633,
1
1255, 1100 cmꢁ1; H NMR d 7.43–7.40 (m, 1H), 7.35–
7.26 (m, 3H), 4.95 (dd, J¼7.9, 5.0 Hz, 1H), 4.72 (s, 2H),
3.79–3.71 (m, 2H), 2.00–1.97 (m, 2H), 2.79–1.74 (m, 2H),
0.93 (s, 9H), 0.10 (s, 6H); EIMS m/z (%): 292 (M+ꢁH2O,
0.9), 195 (M+ꢁTBS, 1.3), 143 (100); ESIHRMS calcd for
C17H30O3SiNa ([M+Na]+) 333.1856, found 333.1868.
Anal. calcd for C17H30O3Si: C, 65.76; H, 9.74. Found: C,
65.98; H, 9.64.
1
1316, 1183 cmꢁ1; H NMR d 8.32 (d, J¼15.8 Hz, 1H),
8.17 (s, 1H), 7.79–7.76 (m, 1H), 7.59–7.56 (m, 1H), 7.39–
7.35 (m, 1H), 6.28 (d, J¼15.8 Hz, 1H), 4.31–4.19 (m, 4H al-
together, including a quartet at d 4.26, J¼7.1 Hz), 2.52–2.45
(m, 1H), 2.18–2.00 (m, 3H), 1.34 (t, J¼7.1 Hz, 3H). The 26
was then transformed into 21a (the less polar diastereomer,
44.4% yield) and 21b (the more polar diastereomer, 12.3%
yield) using the same procedure described above for convert-
ing 9 into 11. Data for 21a (the less polar diastereomer):
4.20. Synthesis of compound 31
With cooling (ꢁ70 ꢀC bath) and stirring a solution of DMSO
(7 mL, 97.4 mmol) was added dropwise (via a syringe) to
a solution of (COCl)2 (3.9 mL, 44.3 mmol) in CH2Cl2
(100 mL). The mixture was stirred at that temperature for
30 min before the alcohol 29 (3.269 g, 10.58 mmol) was
introduced via a syringe. The bath was allowed to warm to
ꢁ60 ꢀC and the stirring was continued at that temperature
for 30 min. Then, Et3N (27 mL, 199.2 mmol) was added to
the mixture and the reaction was continued at that tempera-
ture for another 30 min. After the reaction was quenched
with 2 N HCl (40 mL), the cold bath was removed and the
mixture was stirred at ambient temperature for 5 min. The
phases were separated and the aqueous layer was extracted
with CH2Cl2. The combined organic layers were washed
with saturated NaHCO3, water, and brine, dried over
Na2SO4, and concentrated on a rotary evaporator to give
crude ketone–aldehyde 30 as a yellowish oil (3.578 g),
which was then converted into a,b-unsaturated ester 31 as
a red-orange oil (3.064 g, 74% from 29) by a similar Wittig
reaction as described above for the synthesis of 25. Data for
31: FTIR (film) 1716, 1256, 1177 cmꢁ1; 1H NMR d 8.02 (d,
J¼15.9 Hz, 1H), 7.70 (d, J¼7.6 Hz, 1H), 7.55 (d, J¼7.3 Hz,
1H), 7.49–7.41 (m, 2H), 6.25 (d, J¼15.8 Hz, 1H), 4.24
(q, J¼7.1 Hz, 2H), 3.66 (t, J¼6.1 Hz, 2H), 2.98 (t,
J¼7.5 Hz, 2H), 1.95–1.85 (m, 2H), 1.29 (t, J¼7.1 Hz, 3H),
0.85 (s, 9H), 0.00 (s, 6H); ESIMS m/z: 377 ([M+H]+).
Anal. calcd for C21H32O4Si: C, 66.98; H, 8.57. Found: C,
67.10; H, 8.23.
1
FTIR (film) 1736, 1453, 1374, 1281, 1053 cmꢁ1; H NMR
d 7.37–7.30 (m, 3H), 7.13–7.10 (m, 1H), 5.70 (t, J¼6.4 Hz,
1H), 4.26–4.19 (m, 4H), 2.91 (d, J¼6.7 Hz, 2H), 2.43–2.36
(m, 4H), 1.29 (t, J¼7.1 Hz, 3H); ESIMS m/z 296
([M+NH4]+); ESIHRMS calcd for C15H18O5Na ([M+Na]+)
301.1046, found 301.1042. Anal. calcd for C15H18O5: C,
64.74; H, 6.52. Found: C, 64.77; H, 6.52. Data for 21b (the
more polar diastereomer): FTIR (film) 1735, 1454, 1372,
1
1279, 1050 cmꢁ1; H NMR d 7.36–7.21 (m, 3H), 7.16–
7.13 (m, 1H), 5.50 (dd, J¼4.4, 9.0 Hz, 1H), 4.28–4.16 (m,
4H altogether, including a quartet at d 4.20, J¼7.2 Hz),
3.24 (dd, J¼8.8, 16 Hz, 1H), 2.80 (dd, J¼4.4, 16 Hz, 1H),
2.38–2.35 (m, 1H), 2.23–2.17 (m, 3H), 1.31 (t, J¼7.2 Hz,
3H); ESIMS m/z 296 ([M+NH4]+); ESIHRMS calcd for
C15H18O5Na ([M+Na]+) 301.1046, found 301.1042.
4.23. Synthesis of compound 33
A mixture of 1,2,6-hexatriol (14.5 g, 108 mmol), p-TsOH
(2.05 g, 10.8 mmol), and DMOP (20 mL, 162 mmol) in ace-
tone (300 mL) was stirred at the ambient temperature for
22 h. Aqueous NaHCO3 was added, followed by Et2O.
The phases were separated. The aqueous layer was back-
extracted with Et2O. The combined organic layers were
washed with water and brine, dried over anhydrous
Na2SO4, and concentrated on a rotary evaporator to give
a yellowish oil. This oil was added to a mixture of NaH
(8.4 g, 210 mmol) in DMF (300 mL) stirred at 0 ꢀC. CH2]
CHCH2Br (25.41 g, 210 mmol) was then introduced. The
mixture was stirred at the ambient temperature for 10 h be-
fore being partitioned between water and Et2O. The phases
were separated. The aqueous layer was back-extracted
with Et2O. The combined organic layers were washed with
water and concentrated on a rotary evaporator. The residue
was dissolved in MeOH (160 mL). HCl 2 N (80 mL) was
then added. The mixture was stirred at the ambient temper-
ature for 2 h before the solvent was removed on a rotary
evaporator. The residue was extracted with EtOAc. The
combined organic layers were washed with saturated aque-
ous NaHCO3, water, and brine, and dried over Na2SO4.
Removal of the solvent on a rotary evaporator left a yellowish
4.21. Removal of the TBS group in 31 (25)
A mixture of 31 (2.916 g, 7.46 mmol) and p-TsOH
(380 mg, 2 mmol) in aqueous THF (40 mL of THF plus
10 mL water) was stirred at ambient temperature for 5 h,
when TLC showed the hydrolysis to be complete. The
reaction was quenched with saturated aqueous NaHCO3.
The mixture was diluted with Et2O, and the phases were
separated and the aqueous layer was extracted with Et2O.
The combined organic layers were washed with water
and brine, dried over Na2SO4, and concentrated on a rotary
evaporator. The residue was chromatographed on silica
gel (2:1 n-hexane/EtOAc) to give 25 (1.827 g, 96%).
Data for 25 are given above under ‘synthesis of 25 (from
24)’.

7708
H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
oil, which was dissolved in a mixture of CH2Cl2 (422 mL)
and NEt3 (45 mL). The mixture was cooled in a 0 ꢀC bath.
p-TsCl (23.7 g, 123.5 mmol) was then added. The mixture
was stirred at the ambient temperature for 12 h before dilut-
ing with CH2Cl2, washed with water and brine, and dried
over anhydrous Na2SO4. The solvent was removed on a ro-
tary evaporator. The residue was added to Et2O (440 mL),
followed by KOH (7.056 g, 126 mmol). The mixture was
stirred at the ambient temperature for 5.5 h. Water was added
and the phases were separated. The aqueous layer was back-
extracted with Et2O. The combined organic layers were
washed with brine, dried over anhydrous Na2SO4, and con-
centrated on a rotary evaporator. The residue was chromato-
graphed on silica gel (10:1 n-hexane/EtOAc) to give 33 as
a colorless oil (10.761 g, 69% from 1,2,6-hexatriol). FTIR
1H), 4.48 (d, J¼12 Hz, 1H), 3.97 (dt, J¼5.6, 1.4 Hz, 2H),
3.85–3.77 (m, 1H), 3.46 (t, J¼6.4 Hz, 2H), 2.84 (s, 1H),
2.82 (d, J¼3.8 Hz, 1H), 1.71–1.46 (m, 8H); EIMS m/z
(%): 246 (M+ꢁH2O, 0.86), 143 (20), 104 (100); ESIHRMS
calcd for C16H24O3Na ([M+Na]+) 287.1618, found
287.1624. Anal. calcd for C16H24O3: C, 72.69; H, 9.15.
Found: C, 72.44; H, 9.36.
4.26. Synthesis of 37
Compound 37 was prepared in 85% yield (chromatography
on silica gel eluting with 8:1 n-hexane/EtOAc) from com-
pound 36 using the same procedure described above for con-
verting 29 into 31. Data for 37 (a yellowish oil): FTIR (film)
1713, 1634, 1314, 1178 cmꢁ1; 1H NMR d 7.81 (d, J¼16 Hz,
1H), 7.61 (d, J¼7.2 Hz, 1H), 7.37–7.26 (m, 2H), 7.23 (d,
J¼7.1 Hz, 1H), 6.36 (d, J¼16 Hz, 1H), 5.89 (ddt, J¼17,
11, 5.3 Hz, 1H), 5.25 (br d, J¼17 Hz, 1H), 5.16 (br d,
J¼11 Hz, 1H), 4.26 (q, J¼7.2 Hz, 2H), 3.93 (d, J¼5.6 Hz,
2H), 3.86 (s, 2H), 3.40 (t, J¼6.2 Hz, 2H), 2.51 (t,
J¼7.0 Hz, 2H), 1.71–1.51 (m, 4H), 1.33 (t, J¼7.2 Hz, 3H);
EIMS m/z (%): 330 (M+, 0.97), 273 (M+ꢁOallyl), 141
(100). Anal. calcd for C20H26O4: C, 72.73; H, 7.88. Found:
C, 72.71; H, 8.02.
(film) 2932, 1458, 1360, 1179, 1105 cmꢁ1 1H NMR
;
d 5.92 (ddt, J¼17, 10.6, 5.5 Hz, 1H), 5.72 (br d, J¼17 Hz,
1H), 5.18 (br d, J¼10 Hz, 1H), 3.97 (d, J¼5.6 Hz, 2H),
3.45 (t, J¼6.2 Hz, 2H), 2.92–2.89 (m, 1H), 2.75 (t,
J¼4.5 Hz, 1H), 2.48 (dd, J¼2.8, 4.9 Hz, 1H), 1.70–1.47
(m, 6H); 13C NMR (75 MHz, CDCl3) d 134.9, 116.7, 71.7,
70.0, 52.1, 47.0, 32.2, 29.4, 22.6; EIMS m/z: (%) 155
(M+ꢁH, 6), 99 (M+ꢁOallyl, 1), 41 (100); EIHRMS calcd
for C9H16O2 (M+) 156.1103, found 156.1154.
4.24. Synthesis of compound 35
4.27. Synthesis of compound 42
n-BuLi (8.8 mL, 1.6 M) was added dropwise (via a syringe)
to a solution of 34 (3.253 g, 14.1 mmol) in dry THF (50 mL)
stirred at ꢁ70 ꢀC under N2. After the completion of the ad-
dition, the mixture was stirred at ꢁ70 ꢀC for 1 h. A solution
of epoxide 33 (2.197 g, 14.1 mmol) in dry THF (10 mL) was
added, followed by BF3$EtO2 (2.0 mL, 14.1 mmol). The
stirring was continued at the same temperature for another
hour. The reaction was quenched with saturated aqueous
NH4Cl. The mixture was diluted with Et2O. The phases were
separated. The aqueous layer was back-extracted with Et2O.
The combined organic layers were washed with water and
brine, dried over anhydrous Na2SO4, and concentrated on
a rotary evaporator. The residue was chromatographed on
silica gel (7:1 n-hexane/EtOAc) to give 35 (3.477 g, 80%).
A mixture of 37 (3.551 g, 10.76 mmol) and PdCl2 (383 mg,
2.15 mmol) in MeOH (54 mL) was stirred at 60 ꢀC until
TLC showed the complete disappearance of 37. The reaction
mixturewas filtered. The filtrate was concentrated on a rotary
evaporator to give an orange oil (a mixture containing 38, 39,
and 40, 2.458 g). This crude oil was added to the mixture of
UHP (6.9 g, 73.4 mmol) and p-TsOH (1.85 g, 9.74 mmol) in
DME (110 mL). The mixture was stirred at the ambient tem-
perature until TLC showed disappearance of the starting
material. The mixture was partitioned between Et2O and
water. The phases were separated. The aqueous layer was
back-extracted with Et2O. The combined organic layers
were washed with brine, dried over anhydrous Na2SO4,
and concentrated on a rotary evaporator. The residue was
chromatographed on silica gel (5:1 n-hexane/EtOAc) to
give compound 41 as a colorless oil (2.314 g, 71% from
37), which gave the following data: FTIR (film) 3361,
1
FTIR (film) 3470, 2936, 1453, 1100, 1040 cmꢁ1; H NMR
d 7.36–7.23 (m, 4H), 5.92 (ddt, J¼17, 10.2, 5.8 Hz, 1H),
5.27 (br d, J¼18 Hz, 1H), 5.18 (br d, J¼11 Hz, 1H), 4.71
(s, 4H), 4.67 (d, J¼11 Hz, 2H, part of an AB system), 4.60
(d, J¼11 Hz, 2H, part of an AB system), 3.97 (dt, J¼4.3,
1.4 Hz, 2H), 3.83–3.80 (m, 1H), 3.45 (t, J¼6.4 Hz, 2H),
3.42 (s, 3H), 2.91 (dd, J¼15, 3.6 Hz, 1H), 2.75 (dd, J¼14,
9.0 Hz, 1H), 2.42 (d, J¼3.3 Hz, 1H), 1.68–1.58 (m, 6H);
EIMS m/z: (%) 309 (M++H, 0.03), 309 (M+, 0.02), 104
(100); ESIHRMS calcd for C18H28O4Na ([M+Na]+)
331.1880, found 331.1881. Anal. calcd for C18H28O4: C,
70.10; H, 9.15. Found: C, 69.72; H, 8.78.
2943, 1712, 1691, 1632, 1316, 1186 cmꢁ1 1H NMR
;
d 8.23 (s, 1H), 8.20 (d, J¼16 Hz, 1H), 7.58 (d, J¼7.7 Hz,
1H), 7.40–7.24 (m, 3H), 6.34 (d, J¼16 Hz, 1H), 4.28 (q,
J¼7.1 Hz, 2H), 3.88–3.77 (m, 2H), 3.41 (d, J¼14 Hz, 1H,
part of AB system), 3.13 (d, J¼14 Hz, 1H, part of AB sys-
tem), 1.51–1.03 (m, 6H), 0.88 (t, J¼7.1 Hz, 3H). The 41
was then transformed into 42 (a 1:1 mixture of diastereo-
mers) in 51% total yield (chromatography on silica gel elut-
ing with 15:1 n-hexane/EtOAc) using the same procedure
described above for converting 11 into 13. Data for 42
(a colorless oil): FTIR (film) 1736, 1445, 1372, 1216,
1180, 1045 cmꢁ1; 1H NMR d 7.28–7.07 (m, 4H), 5.93 (dd,
J¼4.9, 8.4 Hz, 0.5H), 5.88 (dd, J¼4.2, 9.1 Hz, 0.5H),
4.29–4.20 (m, 2H, including a quartet at d 4.17, J¼
7.1 Hz), 4.15–3.94 (m, 1H), 3.84–3.66 (m, 2H altogether, in-
cluding a doublet at d 3.86, J¼14 Hz), 3.22 (dd, J¼9.1,
16 Hz, 0.5H), 3.01 (dd, J¼3.9, 16 Hz, 0.5H), 2.84 (dd,
J¼8.0, 16 Hz, 0.5H), 2.71 (dd, J¼4.9, 16 Hz, 0.5H), 2.59
(d, J¼14 Hz, 1H), 1.80–1.40 (m, 6H), 1.31 and 1.25 (two
4.25. Removal of the MOM group in 35 (36)
Compound 36 was prepared in 93% yield (chromatography
on silica gel eluting with 2:1 n-hexane/EtOAc) from 35
using the same procedure described above for converting 6
into 8. Data for 36 (a yellowish oil): FTIR (film) 3331,
1
2936, 1453, 1100, 1010 cmꢁ1; H NMR d 7.34–7.20 (m,
4H), 5.92 (ddt, J¼17, 10, 5.8 Hz, 1H), 5.27 (br d, J¼
18 Hz, 1H), 5.18 (br d, J¼10 Hz, 1H), 4.76 (d, J¼12 Hz,

H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
7709
triplets in 1:1 ratio, J¼7.3 Hz, 3H altogether); ESIMS m/z
324 ([M+NH4]+). Anal. calcd for C17H22O5: C, 66.65; H,
7.24. Found: C, 66.89; H, 7.37.
(61), 104 (100). Anal. calcd for C20H24O3: C, 76.89; H,
7.74. Found: C, 76.77; H, 8.10.
4.31. Synthesis of 47
4.28. Synthesis of 43
Compound 47 was prepared from 46 in 77% yield (chroma-
tography on silica gel eluting with 10:1 n-hexane/EtOAc)
using the same procedure described above for converting
29 into 31). Data for 47 (a red-orange oil): FTIR (film)
Using the same procedure described above for the synthesis
of 35 (except that epichlorohydrin (3.2 g, 41.1 mmol) was
utilized as the epoxide instead of 33) an intermediate alcohol
(5.058 g) was obtained from 34 (6.329 g, 27.4 mmol). The
crude alcohol–chloride was added to a mixture of NaH
(914 mg, 22.8 mmol) in THF (104 mL) stirred at 0 ꢀC. The
bath was allowed to warm to ambient temperature and the
reaction was continued until TLC showed complete disap-
pearance of the chloride (2.5 h). The reaction was quenched
with water and the reaction mixture was diluted with Et2O.
The phases were separated. The aqueous layer was back-
extracted with Et2O. The combined organic layers were
washed with water and brine, dried over anhydrous
Na2SO4, and concentrated on a rotary evaporator. The resi-
due was chromatographed on silica gel (15:1 n-hexane/
EtOAc) to give 43 (3.827 g, 67% from 34). FTIR (film)
2929, 1452, 1212, 1150, 1100 cmꢁ1; 1H NMR d 6.99–6.95
(m, 1H), 6.90–6.85 (m, 3H), 4.29 (s, 2H), 4.24 (s, 2H),
3.01 (s, 2H), 2.81–2.76 (m, 1H), 2.61 (dd, J¼5.5, 15 Hz,
1H), 2.53 (dd, J¼5.4, 15 Hz, 1H), 2.39 (t, J¼7.4 Hz, 1H),
2.15–2.13 (m, 1H); EIMS m/z (%) 163 (M+ꢁCH2OCH3,
14), 133 (M+ꢁCH2OCH2OCH3, 6.7), 91 (Bn, 31), 45
(100); MALDI-HRMS calcd for C12H16O3Na ([M+Na]+)
231.0992, found 231.0999. Anal. calcd for C12H16O3: C,
69.21; H, 7.74. Found: C, 69.17; H, 7.59.
1711, 1634, 1314, 1178 cmꢁ1
;
1H NMR d 7.79 (d, J¼
16 Hz, 1H), 7.61–7.58 (m, 1H), 7.34–7.25 (m, 5H), 7.18–
7.11 (m, 2H), 6.35 (d, J¼16 Hz, 1H), 5.89 (ddt, J¼17, 11,
6.0 Hz, 1H), 5.26 (br d, J¼18 Hz, 1H), 5.18 (br d,
J¼12 Hz, 1H), 4.41 (s, 2H), 4.26 (q, J¼7.2 Hz, 2H), 3.93–
3.86 (m, 6H), 1.34 (t, J¼7.1 Hz, 3H); ESIMS m/z 378
([M+Na]+); ESIHRMS calcd for C24H26O4Na ([M+Na]+)
401.1723, found 401.1719. Anal. calcd for C24H26O4: C,
76.17; H, 6.92. Found: C, 75.97; H, 7.04.
4.32. Synthesis of 52
A mixture of 47 (2.51 g, 6.64 mmol), PdCl2 (1.175 g,
6.64 mmol), and CuCl (657 mg, 6.64 mmol) in DMF–H2O
(50 mL/5 mL) was stirred at the ambient temperature in an
open flask (because air was needed in the reaction) until
TLC showed completion of the deallylation (ca.12 h). The
solids were filtered off. The filtrate was concentrated on a
rotary evaporator to give an orange liquid–solid mixture
(2.3 g), which was directly added to a mixture of UHP
(7.94 g, 84.4 mmol) and CSA (7.83 g, 33.8 mmol) in
DME–EtOH (96 mL/22 mL). The resulting mixture was
stirred at the ambient temperature until the starting material
disappeared on TLC. The mixture was partitioned between
Et2O and water. The phases were separated. The aqueous
layer was back-extracted with Et2O. The combined organic
layers were washed with brine, dried over anhydrous
Na2SO4, and concentrated on a rotary evaporator. The resi-
due was chromatographed on silica gel (8:1 n-hexane/
EtOAc) to give compound 51 (775 mg, 33% from 47), which
4.29. Synthesis of 45
Compound 45 was prepared from epoxide 43 and 44 in 76%
yield (chromatography on silica gel eluting with 6:1 n-
hexane/EtOAc) using the same procedure described above
for converting 33 into 35 (but with 43 and 44 replacing 33
and 34, respectively). Data for 45 (a colorless oil): FTIR
1
gave the following data: H NMR d 8.70 (s, 1H), 8.20 (d,
(film) 3447, 2929, 1452, 1149, 1041 cmꢁ1 1H NMR
;
J¼16 Hz, 1H), 7.58 (d, J¼7.6 Hz, 1H), 7.41–7.27 (m, 3H),
7.14–7.03 (m, 3H), 6.89 (d, J¼6.6 Hz, 1H), 6.34 (d,
J¼16 Hz, 1H), 5.00 (d, J¼14 Hz, 1H, part of an AB system),
4.77 (d, J¼15 Hz, 1H, part of an AB system), 4.27 (q,
J¼7.1 Hz, 2H), 3.64 (d, J¼14 Hz, 1H, part of AB system),
3.24 (d, J¼15 Hz, 1H, part of AB system), 2.88 (d,
J¼17 Hz, 1H, part of an AB system), 2.41 (d, J¼17 Hz,
1H, part of an AB system), 1.33 (t, J¼7.1 Hz, 3H). A solu-
tion of 51 (105 mg, 0.28 mmol) and NHEt2 (35 mL) in
CF3CH2OH–CH2Cl2 (1.2 mL/0.6 mL) was stirred at ambi-
ent temperature for 21 h. The solvent was removed on a
rotary evaporator. The residue was chromatographed on
silica gel (16:1 n-hexane/EtOAc) to give compound 52 as
a colorless oil (46 mg, 24% from 51). FTIR (film) 1735,
1449, 1372, 1249, 1160, 1027 cmꢁ1; 1H NMR d 7.31–7.04
(m, 8H), 5.94 (dd, J¼4.0, 8.5 Hz, 1H), 5.01 (dd, J¼9.1,
15 Hz, 1H), 4.77 (dd, J¼12.3, 15 Hz, 1H), 4.25–4.13 (m,
2H, including a quartet at d 4.21, J¼6.9 Hz), 4.03
(d, J¼14 Hz, 0.5H), 3.94 (br d, J¼14 Hz, 0.5H), 3.22 (dd,
J¼9.2, 16 Hz, 0.5H), 3.02 (dd, J¼4.2, 16 Hz, 1H), 2.95
(d, J¼17 Hz, 0.5H), 2.87 (dd, J¼8.3, 16 Hz, 0.5H), 2.75
(dd, J¼4.2, 16 Hz, 1H), 2.64 (d, J¼14 Hz, 1H), 2.48 (d,
J¼17 Hz, 0.5H), 1.28 and 1.25 (two triplets in 1:1 ratio,
d 7.38–7.21 (m, 8H), 5.93 (ddt, J¼17, 11, 5.5 Hz, 1H), 5.31
(br d, J¼17 Hz, 1H), 5.22 (br d, J¼9.5 Hz, 1H), 4.67 (s,
2H), 4.63 (d, J¼2.7 Hz, 1H), 4.55 (d, J¼11 Hz, 1H, part of
an AB system), 4.60 (d, J¼11 Hz, 1H, part of an AB system),
4.06–4.00 (m, 3H altogether, including a doublet at d 4.01,
J¼5.4 Hz), 3.40 (s, 3H), 3.05 (d, J¼3.8 Hz, 1H), 2.95–2.86
(m, 4H); EIMS m/z (%) 293 (M+ꢁOCH2OMe, 1), 119
(ꢁCOBn, 7), 104 (100), 91 (Bn, 23). Anal. calcd for
C22H28O4: C, 74.13; H, 7.92. Found: C, 74.15; H, 7.85.
4.30. Removal of the MOM group in 45 (46)
Compound 46 was prepared from 45 in 99% yield (chroma-
tography on silica gel eluting with 4:1 n-hexane/EtOAc)
using the same procedure described above for converting 6
into 8. Data for 46 (a colorless oil): FTIR (film) 3325,
1
2923, 1452, 1053 cmꢁ1; H NMR d 7.34–7.19 (m, 8H),
5.85 (ddt, J¼17, 10, 6.0 Hz, 1H), 5.27 (br d, J¼17 Hz,
1H), 5.22 (br d, J¼11 Hz, 1H), 4.72 (d, J¼12 Hz, 1H, part
of an AB system), 4.55–4.46 (m, 2H), 4.40 (d, J¼11 Hz,
1H, part of an AB system), 4.05–3.97 (m, 5H), 3.04–2.85
(m, 4H); EIMS m/z (%) 313 (M++H, 0.3), 191 (6), 105

7710
H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
J¼7.2 Hz, 3H altogether); ESIMS m/z 372 ([M+NH4]+);
ESIHRMS calcd for C21H22O5Na ([M+Na]+) 377.1359,
found 377.1346. Anal. calcd for C21H22O5: C, 71.17; H,
6.26. Found: C, 71.52; H, 6.54.
on silica gel (3:1 n-hexane/EtOAc) to give 56 (34 mg, 18%
yield) and 57 (86 mg, 54% yield). Data for compound 56 (a
colorless oil): FTIR (film) 3469, 2929, 1709, 1178 cmꢁ1
;
1H NMR d 7.31–7.15 (m, 5H), 5.95 (s, 1H), 4.15 (q,
J¼7.1 Hz, 2H), 4.17 (d, J¼14.0 Hz, 2H), 2.75–2.65
(m, 4H), 2.61 (t, J¼7.5 Hz, 2H), 2.43 (t, J¼7.2 Hz, 2H),
2.34 (t, J¼6.1 Hz, 1H), 1.95–1.85 (m, 2H), 1.27 (t,
J¼7.0 Hz, 3H); ESIMS m/z 305 ([M+H]+); ESIHRMS calcd
for C18H24O4Na ([M+Na]+) 327.1567, found 327.1568. Data
for compound 57 (a white solid): mp 69–71 ꢀC. FTIR 1789,
4.33. Synthesis of 55
n-BuLi (1.6 M, in hexanes, 6.67 mL, 10 mmol) was added
dropwise to a solution of 54 (0.43 mL, 5.0 mmol) in dry
THF (22 mL) stirred at ꢁ40 ꢀC under N2. The mixture was
stirred at the temperature for 4 h before a soluiton of 53
(2.139 g, 8.15 mmol) in dry THF (2 mL) was introduced.
The stirring was continued at 0 ꢀC for another 43 h. The
reaction was quenched by addition of water. The mixture was
extracted with Et2O, washed with water, and brine, and dried
over anhydrous Na2SO4. The residue after removal of the
solvent was chromatographed on silica gel (4:1 n-hexane/
EtOAc) to give the alkylation product as a yellow-greenish
liquid (453 mg, 42% yield). A solution of this compound
(300 mg, 1.38 mmol) and m-CPBA (261 mg, 1.53 mmol)
in CH2Cl2 (7 mL) was stirred at the ambient temperature for
3 h. The mixture was partitioned between aqueous Na2CO3
and Et2O. The ethereal layer was washed with brine and
dried over anhydrous Na2SO4. Removal of the solvent and
chromatography of the residue on silica gel (5:1 n-hexane/
EtOAc) gave the furan ring opened product as a yellowish
liquid (306 mg, 95% yield). A solution of this liquid
(200 mg, 0.86 mmol) and 10% Pd–C (23 mg) in MeOH
(5 mL) was stirred under H2 (1 atm) for 7 h. The catalyst and
the solvent were removed by filtration and rotary evapora-
tion, respectively. The residue was chromatographed on
silica gel (3:1 n-hexane/EtOAc) to give 55 as a yellowish
oil (81 mg. 40% yield). FTIR (film) 3426, 2924, 1711,
1
1748, 1638, 1496, 1176 cmꢁ1; H NMR d 7.33–7.16 (m,
5H), 5.78 (s, 1H), 4.75 (s, 2H), 2.75–2.62 (m, 6H), 2.47 (t,
J¼7.4 Hz, 2H), 2.00–1.90 (m, 2H); ESIMS m/z 259
([M+H]+); ESIHRMS calcd for C16H19O3Na ([M+H]+)
259.1329, found 259.1327.
4.35. Hydroperoxidation of 56 and attempted cyclization
(59)
A solution of 56 (26 mg, 0.085 mmol), p-TsOH (10 mg,
0.52 mmol), and UHP (30 mg, 0.32 mmol) in DME (1 mL)
was stirred at the ambient temperature for 24 h. The mixture
was diluted with Et2O, washed with water and brine, and
dried over anhydrous Na2SO4. The solvent was removed
and the residue was chromatographed on silica gel (3:1
n-hexane/EtOAc) to give 58 as a colorless oil (12 mg, 44%
yield), on which the following data were recorded: FTIR
1
(film) 3480, 1747, 1714, 1200 cmꢁ1; H NMR d 7.65 (s,
1H), 7.30–7.18 (m, 5H), 5.70 (s, 1H), 4.46 (d, J¼14 Hz,
1H, part of an AB system), 4.15 (q, J¼7.2 Hz, 2H), 3.99 (d,
J¼14 Hz, 1H, part of an AB system), 3.12–3.05 (m, 1H),
2.84–2.76 (m, 1H), 2.64 (t, J¼7.1 Hz, 2H), 2.05–1.63 (m,
6H), 1.25 (t, J¼7.2 Hz, 3H). The 58 (12 mg, 0.037 mmol)
was dissolved in CF3CH2OH (1 mL) and HNEt2 (10 mL).
The mixture was stirred at the ambient temperature for
10 h. The solvent was removed by rotary evaporation and
the residue was chromatographed on silica gel (3:1 n-hex-
ane/EtOAc) to give 59 as a colorless oil (7 mg, 58% yield).
FTIR (film) 3485, 1750, 1713, 1498, 1456, 1282, 1012,
1406 cmꢁ1 1H NMR d 7.32–7.16 (m, 5H), 4.34 (d,
;
J¼5.0 Hz, 2H), 3.00 (t, J¼4.8 Hz, 1H), 2.80–2.76 (m, 2H),
2.64–2.59 (m, 4H), 2.47 (t, J¼7.3 Hz, 2H), 1.97–1.87 (m,
2H); ESIMS m/z 235 ([M+H]+); ESIHRMS calcd for
C14H18O3Na ([M+Na]+) 257.1148, found 257.1148.
1
4.34. Wittig reaction of 55 (56 and 57)
896, 753, 699 cmꢁ1; H NMR d 7.32–7.15 (m, 5H), 4.28
(q, J¼7.1 Hz, 2H), 3.74 (dd, J¼13, 5.2 Hz, 1H), 3.63 (s,
1H), 3.61 (dd, J¼13, 8.3 Hz, 1H), 2.65–2.55 (m, 4H), 2.44
(t, J¼7.3 Hz, 2H), 2.21–2.16 (m, 1H), 2.04–1.86 (m, 4H),
1.30 (t, J¼7.2 Hz, 3H); EIMS m/z (%) 292 (M+ꢁCH2]CH2,
5), 277 (M+ꢁC2H4CH3, 11), 57 (100); MALDI-HRMS calcd
for C18H24O5Na ([M+Na]+) 343.1516, found 343.1523.
A solution of 55 (144 mg, 0.61 mmol) in Ac2O (2 mL) and
NEt3 (0.3 mL) was stirred at the ambient temperature for
5 h. Aqueous saturated Na2CO3 was carefully added. The
mixture was extracted with Et2O thrice. The combined ethe-
real phases were washed with aqueous NaHCO3 and brine,
dried over anhydrous Na2SO4. Removal of the solvent
left a yellowish oil (170 mg, ca. 0.6 mmol), which was dis-
solved in dry THF (0.5 mL) and added to a mixture of
(EtO)2(O)PCHCO2Et (167 mg, 0.72 mmol) and NaH
(33 mg, 0.72 mmol, washed with dry THF three times) in
THF (0.5 mL) stirred at 0 ꢀC. After completion of the addi-
tion, the mixture was stirred at the ambient temperature for
7 h. The reaction was quenched by addition of water and
the mixture was extracted with Et2O thrice. The combined
ethereal phases were washed with aqueous NaHCO3 and
brine, dried over anhydrous Na2SO4. Removal of the solvent
left a yellowish oil (245 mg), which was dissolved in EtOH
(4 mL) and 2 N HCl (0.5 mL). After stirring at the ambient
temperature for 2 d, the mixture was extracted with Et2O
thrice. Thecombinedetherealphaseswerewashed withaque-
ous NaHCO3 and brine, dried over anhydrous Na2SO4. The
solvent was removed and the residue was chromatographed
4.36. Antimalarial activities of peroxides in vitro
4.36.1. Malaria parasites. P. falciparum (ATCC 30932,
FCR-3 strain) was used. P. falciparum was cultivated by
a modification of the method of Trager and Jensen17 using
a 5% hematocrit of type A human red blood cells suspended
in RPMI 1640 medium (Gibco, NY) supplemented with
heat-inactivated 10% type A human serum. The plates
were placed in a CO2–O2–N2 incubator (5% CO2, 5% O2,
and 90% N2 atmosphere) at 37 ꢀC, and the medium was
changed daily until 5% parasitemia (which means that 5
parasite-infected erythrocytes in every 100 erythrocytes
were existing).
4.36.2. Mammalian cells. Mouse mammary tumor FM3A
cells (wild-type, subclone F28-7) 16 were supplied by the

H.-X. Jin et al. / Tetrahedron 62 (2006) 7699–7711
7711
Japanese Cancer Research Resources Bank (JCRB). FM3A
References and notes
cells were maintained in suspension culture at 37 ꢀC in a
5% CO2 atmosphere in plastic bottles containing ES medium
(Nissui Pharmaceuticals, Tokyo, Japan) supplemented with
2% heat-inactivated fetal bovine serum (Gibco, NY).
1. See: e.g., (a) Li, Y.; Wu, Y.-L. Curr. Med. Chem. 2003, 10,
2197–2230 and the references therein; (b) Klayman, D. L.
Science 1985, 228, 1049–1055; (c) Haynes, R. K.; Vonwiller,
S. C. Acc. Chem. Res. 1997, 30, 73–79.
4.36.3. In vitro antimalarial activity of peroxides. The
following procedures were used for assay of antimalarial
activity.18,19 Asynchronously cultivated P. falciparum were
used. Various concentrations of compounds in dimethylsulf-
oxide were prepared. Five microliters of each solution was
added to individual wells of a multidish 24 wells. Erythro-
cytes with 0.3% parasitemia were added to each well
containing 995 mL of culture medium to give a final hemato-
crit level of 3%. The plates were incubated at 37 ꢀC for 72 h
in a CO2–O2–N2 incubator (5% CO2, 5% O2, and 90% N2
atmosphere). To evaluate the antimalarial activity of test
compound, we prepared thin blood films from each culture
and stained them with Giemsa (E. Merck, Germany).
Total 1ꢂ104 erythrocytes per one thin blood film were
examined under microscopy. All of the test compounds
were assayed in duplicate at each concentration. Drug-free
control cultures were run simultaneously. All data points
represent the mean of three experiments. Parasitemia in
control was reaching between 4 and 5% at 72 h. The EC50
value refers to the concentration of the compound necessary
to inhibit the increase in parasite density at 72 h by 50% of
control.
2. For an example of forming peroxy bond from two oxygen
radicals, see: Porter, N. A.; Caldwell, S. E.; Lowe, J. R.
J. Org. Chem. 1998, 63, 5547–5554.
3. (a) O’Nill, P. M.; Pugh, M.; Davies, J.; Ward, S. A.; Park, B. K.
Terahedron Lett. 2001, 42, 4569–4571; (b) O’Nill, P. M.;
Hindley, S.; Pugh, M.; Davies, J.; Bray, P. G.; Park, B. K.;
Kapu, D. S.; Ward, S. A.; Stocks, P. A. Terahedron Lett.
2003, 44, 8135–8138; (c) Tokuyasu, T.; Kunikawa, S.; Abe,
M.; Masuyama, A.; Nojima, M.; Kim, H.-S.; Begum, K.;
Wataya, Y. J. Org. Chem. 2003, 68, 7361–7367; (d) Tokuyasu,
T.; Kunikawa, S.; McCullough, K. J.; Masuyama, A.; Nojima,
M. J. Org. Chem. 2005, 70, 251–260.
4. See: e.g., (a) Ushigoe, Y.; Kano, Y.; Nojima, M. J. Chem. Soc.,
Perkin Trans. 1 1997, 5–10; (b) Avery, M. A.; Jennings-White,
C.; Chong, W. K. M. J. Org. Chem. 1989, 54, 1792–1795.
5. See: e.g., (a) Kerr, B.; McCullough, K. J. J. Chem. Soc., Chem.
Commun. 1985, 590–591; (b) Subramanyam, V.; Brizulela,
C. L.; Soloway, A. H. J. Chem. Soc., Chem. Commun. 1976,
508–509; (c) Kovac, F. Tetrahedron Lett. 2001, 42, 5313–
5314; (d) Terent’ev, A. O.; Kutkin, A. V.; Platonov, M. M.;
Ogibin, Y. N.; Nikishin, G. I. Tetrahedron Lett. 2003, 44,
7359–7363; (e) Murakami, N.; Kawanishi, M.; Itagaki, S.;
Horii, T.; Kobayashi, M. Tetrahedron Lett. 2001, 42, 7281–
7285; (f) Murakami, N.; Kawanishi, M.; Itagaki, S.; Horii, T.;
Kobayashi, M. Bioorg. Med. Chem. Lett. 2002, 12, 69–72.
6. (a) Liu, H.-H.; Jin, H.-X.; Zhang, Q.; Wu, Y.-K.; Kim, H.-S.;
Wataya, Y. Chin. J. Chem. 2005, 23, 1469–1473; (b) Liu, H.-H.;
Jin, H.-X.; Wu, Y.-K. Chin. J. Chem. 2004, 22, 1029–1033.
7. (a) Jin, H.-X.; Liu, H.-H.; Zhang, Q.; Wu, Y.-K. Tetrahedron
Lett. 2005, 46, 5767–5769; (b) Jin, H.-X.; Wu, Y.-K.
Tetrahedron Lett. 2005, 46, 6801–6803.
4.36.4. Toxicity against mammalian cell line. FM3A cells
grew with a doubling time of about 12 h. Prior to exposure to
drugs, cell density was adjusted to 5ꢂ104 cells/mL. A cell
suspension of 995 mL was dispensed to the test plate, and
compound at various concentrations suspended in dimethyl-
sulfoxide (5 mL), were added to individual wells of a multi-
dish 24 wells. The plates were incubated at 37 ꢀC in a 5%
CO2 atmosphere for 48 h. All of the test compounds were
assayed in duplicate at each concentration. Cell numbers
were measured using a microcell counter CC-130 (Toa
Medical Electric Co., Japan). All data points represent the
mean of three experiments. The EC50 value refers to the con-
centration of the compound necessary to inhibit the increase
in cell density at 48 h by 50% of control. Selectivity refers
to the mean of EC50 value for FM3A cells per the mean of
EC50 value for P. falciparum.
8. Wu, Y.-K.; Ohlsson, E. Acta Chem. Scand. 1993, 47, 422–424.
9. Clive, D. L. J.; Paul, C. C.; Wang, Z. R. J. Org. Chem. 1997, 62,
7028–7032.
10. Nemoto, H.; Kikuishi, J.; Shibuya, M. Chem. Pharm. Bull.
1996, 44, 2186–2188.
11. Taillier, C.; Gille, B.; Bellosta, V.; Cossy, J. J. Org. Chem.
2005, 70, 2097–2108.
12. Meyers, A. I.; Himmelsbach, Richard J. J. Am. Chem. Soc.
1985, 107, 682–685.
13. Padwa, Albert; Hornbuckle, Susan F.; Fryxell, Glen E.; Stull,
Paul D. J. Org. Chem. 1989, 54, 817–824.
Acknowledgements
14. Masahiro, T.; Satoko, A.; Michiko, S.; Takeshi, H. Agric. Biol.
Chem. 1982, 46, 291–292.
15. Chen, I.-C.; Wu, H.-J. J. Chin. Chem. Soc. (Taipei) 2000, 47,
263–270.
16. Bach, J.; Berenguer, R.; Garcia, J.; Lopez, M.; Manzanal, J.
Tetrahedron 1998, 54, 14947–14962.
17. Trager, W.; Jensen, J. B. Science 1976, 193, 673–675.
18. Kim, H.-S.; Miyake, H.; Arai, M.; Wataya, Y. Parasitol. Int.
1998, 47, 59–67.
This work was supported in China by the National Natural
Science Foundation of China (20025207, 20272071,
20372075, 20321202), the Chinese Academy of Sciences
(‘Knowledge Innovation’ project, KGCX2-SW-209), and
the Major State Basic Research Development Program
(G2000077502). In Japan, this work was supported in part
by a Grant-in-Aid for Scientific Research (B) (16390031)
from Japan Society for the Promotion of Sciences (JSPS)
and by a Grant-in-Aid for Scientific Researches on Priority
Areas from the Ministry of Education, Science, Culture,
and Sports of Japan (14021072, 16017266).
19. Takaya, Y.; Kurumada, K.; Takeuji, Y.; Kim, H.-S.; Shibata, Y.;
Ikemoto, N.; Wataya, Y.; Oshima, Y. Tetrahedron Lett. 1998,
39, 1361–1364.