Structure-activity relationships of synthetic tricyclic trioxanes related to artemisinin: The unexpected alkylative property of a 3- (methoxymethyl) analog

Article abstract of DOI:10.1002/(sici)1099-0690(199908)1999:8<1935::aid-ejoc1935>3.0.co;2-y

A clear-cut correlation between antimalarial potency and the alkylative property of synthetic tricyclic trioxanes 5-10 is reported. Thus, trioxanes 5 and 7, substituted at the C-5a angular position by a methyl or a cyano group, proved to be completely devoid of antimalarial activity, and did not alkylate the heme model Mn(II)TPP. In contrast, both the anti-Plasmodium activity and the alkylative property were restored in the C-5a-unsubstituted analog 8, bearing a methoxymethyl group at C-3. Reaction of 8 with Mn(II)TPP furnished the covalent adduct 18, resulting from trapping of the methoxymethyl radical by the heine model. All these results reinforce the hypothesis that the metalloporphyrin closely interacts with the peroxide bond of the drug to bring about activation of these trioxane antimalarial agents.

Full text of DOI:10.1002/(sici)1099-0690(199908)1999:8<1935::aid-ejoc1935>3.0.co;2-y

FULL PAPER
Structure–Activity Relationships of Synthetic Tricyclic Trioxanes Related
to Artemisinin: The Unexpected Alkylative Property of a
3
-(Methoxymethyl) Analog
[
a]
[a]
[a]
[a]
Olivier Provot,* Boris Camuzat-Dedenis, Mohamed Hamzaoui, Henri Moskowitz,
[
a]
[b]
[b]
[b]
[c]
Jo e¨ lle Mayrargue, Anne Robert, J e´ r oˆ me Cazelles, Bernard Meunier, Fatima Zouhiri,
[
c]
[c]
[c]
[d]
Didier Desma e¨ le, Jean d’Angelo,* Jacqueline Mahuteau, Fr e´ d e´ rick Gay,
and Liliane Cic e´ ron[
d]
Keywords: Antiprotozoals / Artemisinin / Trioxanes / Heme / Structure–activity relationships / Drug research
A clear-cut correlation between antimalarial potency and the
alkylative property of synthetic tricyclic trioxanes 5–10 is
reported. Thus, trioxanes 5 and 7, substituted at the C-5a
angular position by a methyl or a cyano group, proved to be
completely devoid of antimalarial activity, and did not
restored in the C-5a-unsubstituted analog 8, bearing a
methoxymethyl group at C-3. Reaction of 8 with Mn TPP
II
furnished the covalent adduct 18, resulting from trapping of
the methoxymethyl radical by the heme model. All these
results reinforce the hypothesis that the metalloporphyrin
closely interacts with the peroxide bond of the drug to bring
about activation of these trioxane antimalarial agents.
II
alkylate the heme model Mn TPP. In contrast, both the anti-
Plasmodium activity and the alkylative property were
Introduction
oxide moiety, which initially gives rise to oxyl radicals.
These radicals then rearrange to produce the C-centered
radicals 3 and/or 4. The latter species would seem very
likely to be the entities responsible for antimalarial activity,
which function by ultimately alkylating nearby macromo-
The protozoan Plasmodium falciparum is the pathogen
responsible for the most life-threatening form of malaria,
causing the deaths of in excess of 2 million people annually.
As the present deterioration in malaria control is largely
due to the fact that this parasite is acquiring multidrug re-
sistance, there is a great impetus for the development of new
effective antimalarials. Currently used in China and South-
East Asia, the sesquiterpene lactone 1,2,4-trioxane artemisi-
nin (1) (Qinghaosu), isolated from Artemisia annua L., con-
stitutes the prototype of endoperoxides, a new class of pro-
mising antimalarial agents. Artemisinin and semi-synthetic
derivatives clear the peripheral blood of the parasite at nan-
omolar concentrations, and have proved to be highly selec-
tive, safe antimalarials.[ There is strong evidence to sug-
gest that once inside the Plasmodium, artemisinin reacts
first with intraparasite heme, giving rise to free radicals.[
Computational studies have shown that artemisinin and
heme first dock in configuration 2, which allows the heme
iron center to cleave the endoperoxide bridge.[ This acti-
vation phase is followed by homolytic cleavage of the per-
[4]
lecules within the Plasmodium. Specific alkylations of
proteins have thus been identified at pharmacologically rel-
[5]
evant concentrations of artemisinin derivatives. Support-
ing evidence that these alkylation reactions involve the gen-
eration of drug radicals in the vicinity of heme was recently
provided by the unambiguous characterization of a coval-
1]
2]
3]
[
a]
Laboratoire de Chimie Organique Associ e´ au CNRS, Centre
dЈEtudes Pharmaceutiques, Universit e´ Paris Sud,
5
rue J.-B. Cl e´ ment, F-92296 Ch aˆ tenay-Malabry, France
Fax: (internat.) ϩ 33-1/46835223
E-mail: provot@pop.u-psud.fr
[
[
b]
c]
Laboratoire de Chimie de Coordination du CNRS,
2
05 route de Narbonne, F-31077 Toulouse cedex 4, France
Unit e´ de Synth e` se Organique Associ e´ e au CNRS, Centre dЈEtu-
des Pharmaceutiques, Universit e´ Paris Sud,
Figure 1. Proposed hemeϪartemisinin interaction
5
rue J.-B. Cl e´ ment, F-92296 Ch aˆ tenay-Malabry, France
Fax: (internat.) ϩ 33-1/46835652
II
[6Ϫ8]
E-mail: jean.dangelo@cep.u-psud.fr
ent adduct between radical 4 and an Mn heme model.
[
d]
´
Laboratoire dЈEpid e´ miologie et de Chimior e´ sistance du Paludi-
As part of a systematic program devoted to the develop-
ment of new simplified artemisinin analogs, we recently syn-
thesized tricyclic trioxanes 5 and 6 bearing a methyl group
sme, Service de Parasitologie, Groupe Hospitalier Piti e´ -Salp e´ -
tri eˆ re,
4
7Ϫ83 Boulevard de lЈH oˆ pital, F-75651 Paris, France
Eur. J. Org. Chem. 1999, 1935Ϫ1938

WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1434Ϫ193X/99/0808Ϫ1935 $ 17.50ϩ.50/0
1935

O. Provot, J. d’Angelo et al.
FULL PAPER
at the C-5a angular position and reported that these com-
When tested against the “H” clone of P. falciparum,[14]
pounds were devoid of antimalarial activity.[
9][10]
In this enantiomerically pure trioxanes 5, 6 and 7 proved to be
paper, we now report that compound 7, bearing a cyano completely devoid of biological activity in the concentration
[
11][12]
group at C-5a,
is also inactive, whereas analogs 8, 9, range 20Ϫ500 n. In comparison, artemisinin (1) exhibited
and 10, lacking the angular C-5a substituent, show substan- an IC50 value of 19 n on the same strain of Plasmodium.
tial anti-Plasmodium potencies. Accordingly, the 5a-unsub- In sharp contrast, racemic trioxanes 8 and 9 display sub-
stituted analog 8, after reductive activation, was found to stantial antimalarial activities, having IC50 values of 0.7 µ
II
be capable of alkylating a heme model, namely Mn tetra- and 1.3 µ, respectively, on the chloroquine-sensitive Plas-
phenylporphyrin, to give the covalent adduct 18. On the modium falciparum FCR3 (reference compound chloro-
contrary, when 5a-substituted trioxanes 5 and 7 were used quine showed an IC50 value of 0.19 µ on this strain). A
under similar conditions, no porphyrinϪdrug adducts high antiprotozoal potency has also been reported for ra-
could be identified.
cemic trioxane 10, structurally analogous to 5, but lacking
the methyl group at C-5a [IC50 value of 8 n on the (W-2)
Indochina clone of P. falciparum].[
13]
Results and Discussion
Trioxanes 7, 8, and 9 were prepared by the photooxi-
dation of enol ethers 13, according to the method of Jefford
Alkylative Properties
[
13]
and Posner,
involving acidic rearrangement of the inter-
The heme model used in the alkylation experiments was
the hydrophobic complex Mn TPP, generated in situ by
mediate dioxetanes 14. The requisite starting enol ethers 13
were in turn prepared by addition of Wittig reagent 12 to
II
III
nBu NBH reduction of Mn tetraphenylporphyrin chlo-
4
4
[
11][12]
oxo nitriles 11.
Configurational assignments for triox-
III
[6Ϫ8]
ride [Mn (TPP)Cl].
This synthetic analog of heme was
1
anes 8 and 9 were established by H-NMR spectroscopy,
including long-range coupling between 5a-H and 12-H in
isomer 8 (J ϭ 1.5 Hz).
expected to give a limited number of alkylated products ow-
ing to its four-fold symmetry. No covalent adducts could be
characterized when trioxanes 5 and 7 were exposed to this
heme model. In contrast, trioxane 8 furnished, after demet-
allation under mild conditions, adduct 18 in 20% yield. For-
mation of 18 can be rationalized by invoking initial one-
electron reduction and homolytic cleavage of the peroxide
bridge of 8, giving oxyl radical 15. However, subsequent
rearrangement of 15 did not involve the breaking of C-
Antimalarial Activities
3
4
ϪC-4 bond to give a primary C-centered radical of type
, as observed for artemisinin (1),[ but rather cleavage of
7]
the C-3ϪC-14 bond, which furnished an unidentified neu-
tral fragment (presumably lactone 16 or a derivative there-
of) and methoxymethyl radical 17. Trapping of this radical
II
with Mn TPP, followed by demetallation of the porphyrin
ring, finally afforded adduct 18. This result indicates that
trioxane 8 is able to generate C-centered radicals, acting as
alkylating species toward the heme model, but in a different
way from that reported for artemisinin and other trioxane
Figure 2. Structures of trioxanes 5؊10
[8]
analogs. Justification for the fragmentation [15 Ǟ 16 ϩ
1
7] was provided by the difference in stability between the
free ethyl radical (as a model for the primary C-centered
radical of type 4) and methoxymethyl radical 17, namely 5
Ϫ1
±
1 kcal mol in favor of the latter, estimated on the basis
of the homolytic dissociation energies of the correspond-
ing alkanes.[
15]
The above results clearly established that the replacement
of the hydrogen atom at C-5a by a methyl or a cyano group
in our synthetic trioxanes was detrimental to their biologi-
cal activity and also, under the present conditions, to their
alkylative property. Thus, the inactive 5a-substituted triox-
anes 5 and 7 did not alkylate the heme model. In contrast,
both anti-Plasmodium activity and alkylative property were
restored in the 5a-unsubstituted analog 8. This lack of reac-
tivity of 5a-substituted trioxanes can be rationalized by ex-
amining the activation phase of these endoperoxides. In-
Scheme 1. Synthesis of trioxanes 7؊9
1936
Eur. J. Org. Chem. 1999, 1935Ϫ1938

StructureϪActivity Relationships of Synthetic Tricyclic Trioxanes Related to Artemisinin
FULL PAPER
nuclei in 13C-NMR spectra were recognized on the basis of the J-
modulated spin-echo sequence. Optical rotations were measured at
2
0°C with a PerkinϪElmer 241 MC polarimeter in a 1-dm cell.
Analytical thin-layer chromatography was performed on Merck sil-
ica gel 60F254 precoated glass plates (0.25 mm layer). All liquid
chromatography separations were performed using Merck silica gel
6
0 (230Ϫ400 mesh ASTM). Tetrahydrofuran (THF) was distilled
from sodium/benzophenone ketyl. CH Cl was distilled from cal-
2
2
cium hydride under nitrogen. All reactions involving air- or moist-
ure-sensitive compounds were routinely conducted in glassware
that had been flame-dried under a positive pressure of argon. Or-
ganic layers were dried with anhydrous MgSO . Chemicals ob-
4
tained from commercial suppliers were used without further purifi-
cation. Mass-spectral analyses were recorded by electron impact at
ϩ
7
0 eV or DCI/NH
3
with a Nermag R10Ϫ10H spectrometer. All
elemental analyses were performed by the Service de Microanalyse,
Centre dЈEtudes Pharmaceutiques, Ch aˆ tenay-Malabry, France,
with a PerkinϪElmer 2400 analyzer.
II
Scheme 2. Reaction of trioxane 8 with heme model Mn TPP
(
E)- and (Z)-1-Methoxy-4-(2-methoxymethylenecyclohexyl)butan-2-
one (13a): To an ice-cooled solution of (methoxymethyl)triphen-
ylphosphonium chloride (3.4 g, 10 mmol) in dry THF (50 mL), a
solution of phenyllithium (2  in benzene/diethyl ether, 3:1, 5 mL,
spection of complex 19 indeed reveals that close docking of
the drug onto the metalloporphyrin should be impeded,
owing to the destabilizing steric interaction between the ax-
ial substituent at C-5a (syn to the endoperoxide bridge) and
the porphyrin nucleus.
1
0 mmol) was added dropwise. The red solution was stirred at 0°C
[12]
for 2 h and then nitrile 11a
(2.5 mmol) in THF (10 mL) was
added dropwise. After stirring at 20°C for 24 h, the solution was
poured into water and extracted with diethyl ether. The combined
organic phases were washed with brine, dried, and concentrated
under reduced pressure. Chromatographic purification on silica gel
(
1
1
cyclohexane/ethyl acetate, 50:50) afforded 330 mg of enol ethers
Ϫ1
3a as a colorless oil (58%). Ϫ IR (neat): ν˜ ϭ 1710 cm (CϭO),
1
3
680 (CϭC). Ϫ H NMR (CDCl , 200 MHz) [the presence of an
(
E)/(Z) mixture of isomers in a 30:70 ratio complicates the spec-
trum]: Less polar (Z) isomer: δ ϭ 1.40Ϫ2.00 (m, 10 H), 2.25Ϫ2.55
m, 2 H), 2.80 (m, 1 H), 3.41 (s, 3 H, CH OCH CO), 3.50 (s, 3
H, ϭCHOCH ), 4.00 (s, 2 H, CH OCH CO), 5.80 (d, J ϭ 1.0 Hz,
H, ϭCHOMe). Ϫ C NMR (CDCl , 50 MHz): Less polar (Z)
isomer: δ ϭ 21.6 (CH ), 24.8 (CH ), 26.4 (CH ), 28.3 (CH ), 31.6
(CH2), 32.5 (CH), 37.0 (CH2), 59.1 (2 CH3), 77.5 (CH2,
CH OCH CO), 128.0 (C, MeOCHϭC), 140.4 (CH, MeOCHϭC),
01.8 (C, CO). Ϫ C13 (226.3): calcd. C 68.99, H 9.82; found
C 69.01, H 9.63.
(
3
2
3
3
2
1
3
1
3
Figure 3. Presumed interaction of 5a-substituted trioxanes with the
porphyrin ring
2
2
2
2
3
2
2
22 3
H O
Conclusion
(
1S,2E)- and (1S,2Z)-[2-Methoxymethylene-1-(4-methoxy-3-oxobu-
[
12]
tyl)]cyclohexyl-1-carbonitrile (13b): Treatment of nitrile 11b
as
Characterization of the covalent adduct 18 between tri-
II
described above afforded enol ether 13b (60%) as a 65:35 mixture
of (Z)/(E) isomers. Ϫ Less polar (E) isomer, R ϭ 0.59 (petroleum
oxane 8 and the heme model Mn TPP confirms that the
f
alkylative property of artemisinin is not limited to this natu-
ral compound, but is a common feature probably required
for the antimalarial activity of most endoperoxide-contain-
ing drugs. The foregoing results provide an understanding
of the mode of action of 1,2,4-trioxanes. Moreover, the in-
2
2
ether/ethyl acetate, 1:1), colorless oil. Ϫ [α]
D
ϭ ϩ15 (c ϭ 1.4,
(CN), 1715 (CϭO), 1675
, 200 MHz): δ ϭ 1.10Ϫ1.90 (m, 8 H),
.00Ϫ2.60 (m, 4 H), 3.30 (s, 3 H, CH OCH CO), 3.50 (s, 3 H, ϭ
), 3.90 (s, 2 H, CH OCH CO), 6.10 (s, 1 H, ϭCHOMe).
Ϫ1
CH
(
2
2
Cl
2
). Ϫ IR (neat): ν˜ ϭ 2250 cm
1
CϭC). Ϫ H NMR (CDCl
3
3
2
CHOCH
3
3
2
sights so gained are likely to aid the design and develop- Ϫ C H NO (251.3): calcd. C 66.90, H 8.42, N 5.57; found C
1
4
21
3
ment of new artemisinin-like antimalarials.
66.75, H 8.47, N 5.54. Ϫ More polar (Z) isomer, R
f
ϭ 0.53 (petro-
ϭ ϩ42.5 (c ϭ
2
2
leum ether/ethyl acetate, 1:1), colorless oil. Ϫ [α]
D
Ϫ1
1
1
8
.7, CH
675 (CϭC). Ϫ H NMR (CDCl
H), 2.20Ϫ2.70 (m, 4 H), 3.30 (s, 3 H, CH
2 2
Cl
). Ϫ IR (neat): ν˜ ϭ 2250 cm (CN), 1715 (CϭO),
1
3
, 200 MHz): δ ϭ 1.20Ϫ1.90 (m,
OCH CO), 3.45 (s, 3
3 3 2
), 4.00 (s, 2 H, CH OCH CO), 5.85 (s, 1 H, ϭ
Experimental Section
General: Infrared (IR) spectra were obtained with a PerkinϪElmer
3
2
H, ϭCHOCH
CHOMe).
8
41 spectrometer using neat films between NaCl plates or KBr pel-
1
13
lets. Only significant absorptions are listed. The H- and C-NMR
spectra were recorded with Bruker AC 200 P (200 MHz and
[3R*-(3␣,5aβ,9a␣,10R*)]- and [3R*-(3␣,5aβ,9a␣,10S*)]-3,9a-(Epoxy-
methano)-5a,6,7,8,9,9a-hexahydro-10-methoxy-3-methoxymethyl-
9aH-1,2-benzodioxepin (8 and 9): A solution of enol ether 13a
1
13
50 MHz, for H and C, respectively), Bruker AM 250, or Bruker
1
13
ARX 400 (400 MHz and 100 MHz, for H and C, respectively)
spectrometers. Methyl, methylene, methine, and quaternary carbon
(321 mg, 1.42 mmol) in anhydrous CH
2
Cl
2
(30 mL) was placed in a
Eur. J. Org. Chem. 1999, 1935Ϫ1938
1937

O. Provot, J. d’Angelo et al.
Pyrex photochemical apparatus. Methylene Blue (2 mg, 0.03 mmol) cadmium(II) porphyrin. Air was then admitted into the flask and
FULL PAPER
was added and the reaction mixture was cooled to Ϫ60°C. Oxygen
was then bubbled into the solution under irradiation by a 400-W
high-pressure sodium lamp set at a distance of 10 cm. After 2 h,
TMSOTf (115 mg, 0.52 mmol) was added and the reaction mixture
was allowed to warm to room temperature. After stirring for
10% aqueous acetic acid (5 mL) was added. After stirring for 5 min
to ensure complete demetallation of the cadmium(II) porphyrin,
the mixture was extracted with CH
washed with water, dried with MgSO
duced pressure. The covalent adduct was purified by chromatogra-
phy on silica gel (CH Cl , R ϭ 0.60) to give ca. 3 mg of adduct 18
(ca. 20%). Ϫ UV/Vis (CH Cl ): λmax (rel. ε) ϭ 418 nm (100), 514
, 250 MHz): δ ϭ 3.37 (s, 3 H, OCH ), 4.58
OCH ), 7.79 (m, 12 H, phenyl), 8.09 (m,
2 H, phenyl), 8.21 (m, 6 H, phenyl), 8.64 (d, J ϭ 5.0 Hz, 1 H, 18Ј-
]benzene, 400 MHz): δ ϭ 1.05 H), 8.79 (d, J ϭ 5 Hz, 2 H, 8Ј-H and 17Ј-H), 8.84 (d, J ϭ 5.0 Hz,
m, 1 H, 6-Hax), 1.28 (td, J ϭ 13.8, 5.0 Hz, 1 H, 9-Hax), 1.42Ϫ1.75 1 H, 7Ј-H), 8.88 (br. s, 3 H, 3Ј-H, 12Ј-H and 13Ј-H). Ϫ Upon
m, 6 H, 7-H, 8-Hax, 5-Heq, 4a-H, 6-Heq), 1.74 (m, 1 H, 8-Heq), irradiation of the broad singlet at δ ϭ 8.88, containing the 3Ј-H
2 2
Cl , the organic layer was
4
, and concentrated under re-
1
0 min, triethylamine (150 mg, 1.5 mmol) was added and the reac-
tion mixture was filtered through Celite. The filtrate was washed
with water, dried with MgSO , and concentrated in vacuo. The resi-
due was purified by chromatography on silica gel (cyclohexane then (d, J ϭ 1.5 Hz, 2 H, CH
cyclohexane/ethyl acetate, 95:5) to give 142 mg of trioxane 8 (40%)
2
2
f
2
2
1
4
(6). Ϫ H NMR (CD
2
Cl
2
3
2
3
1
as a colorless oil. Ϫ H NMR ([D
6
(
(
1
1
2
3
.81 (br. d, J ϭ 13.8 Hz, 1 H, 9-Heq), 1.95 (dddd, J ϭ 14.5, 13.4,
3.4, 4.4 Hz, 1 H, 5-Hax), 2.22 (td, J ϭ 14.5, 4.4 Hz, 1 H, 4-Heq), ance of the CH
.41 (td, J ϭ 14.5, 3.7 Hz, 1 H, 4-Hax), 3.20 (s, 3 H, CH OCH ), (DCI/NH ); m/z (%): 662 (5), 661 (19), 660 (56), 659 (100) [MH ],
.30 (s, 3 H, 10-COCH ), 3.40 (s, 2 H, CH OCH ), 5.04 (d, 1 H, 627 (3) [M Ϫ (OCH
, 50 MHz): δ ϭ 24.1
, C-5), 31.1 (CH , C-7),
, C-9), 47.8 (CH, C-5a), 56.5 (CH
, 10-COCH ), 76.3 (CH , CH OCH ), 83.7
(258.3):
resonance, the doublet at δ ϭ 4.58 assigned to the methylene reson-
3
OCH fragment collapsed to a sharp singlet. Ϫ MS
2
ϩ
3
2
3
ϩ
3
3
2
3
) ].
13
J ϭ 1.5 Hz, 10α-H). Ϫ C NMR (CDCl
CH , C-8), 25.4 (CH , C-6), 26.6 (CH
4.1 (CH , C-4), 36.0 (CH
), 59.5 (CH
3
(
3
2
2
2
2
[1]
[1a]
For recent reviews, see:
man Prog. Chem. Org. Nat. Prod. 1997, 72, 121Ϫ214. Ϫ
R. Meshnick, T. E. Taylor, S. Kamchonwongpaisan, Microbiol.
H. Ziffer, R. J. Highet, D. L. Klay-
2
2
3
,
[1b]
S.
T. T. Hien, N. J. White, Lancet 1993,
CH
3
OCH
2
3
3
2
3
2
[
1c]
(C, C-9a), 105.0 (C, C-10), 105.8 (C, C-3). Ϫ C13
H
22
O
5
Rev. 1996, 301Ϫ315. Ϫ
[1d]
3
41, 603Ϫ608. Ϫ
S. Kamchonwongpaisan, S. R. Meshnick,
calcd. C 60.44, H 8.58; found C 60.58, H 8.42. Ϫ Further elution
gave 40 mg of trioxane 9 as a colorless oil (12%). Ϫ H NMR
[1e]
Gen. Pharmac. 1996, 27, 587Ϫ592. Ϫ
Opin. Ther. Pat. 1998, 8, 1487Ϫ1493.
G. H. Posner, Expert
1
[
2]
(
[D
listed): δ ϭ 2.20Ϫ2.50 (m, 2 H), 3.00 (s, 3 H, CH
H, 10-COCH ), 3.20 (s, 2 H, CH OCH ), 4.70 (s, 1 H, 10α-H).
6
]benzene, 200 MHz) (only the most significant resonances are
S. R. Meshnick, Y.-Z. Yang, V. Lima, F. Kuypers, S. Kamchon-
wongpaisan, Y. Yuthavong, Antimicrob. Agents Chemother.
3
OCH ), 3.15 (s,
2
1
993, 37, 1108Ϫ1114.
3
3
3
2
[
[
3]
M. A. Avery, S. Mehrota, J. D. Bonk, J. A. Vroman, D. K.
[
3S-(3␣,5aβ,9a␣,10S*)]-3,9a-(Epoxymethano)-5a,6,7,8,9,9a-
Goins, R. Miller, J. Med. Chem. 1996, 39, 2900Ϫ2906.
4] [4a]
G. H. Posner, C. H. Oh, J. Am. Chem. Soc. 1992, 114,
hexahydro-10-methoxy-3-methoxymethyl-9aH-1,2-benzodioxepin-
[4b]
8
328Ϫ8329. Ϫ
G. H. Posner, J. N. Cumming, P. Ploypradith,
5a(3H)-carbonitrile (7): Treatment of enol ether 13b as described
[4c]
C. H. Oh, J. Am. Chem. Soc. 1995, 117, 5885Ϫ5886. Ϫ
C. W.
2
2
above afforded trioxane 7 (50%) as a colorless oil. Ϫ [α]
D
ϭ ϩ62
Jefford, M. G. H. Vicente, Y. Jacquier, F. Favarger, J. Mareda, P.
Ϫ1
1
(CH
2
Cl
2
, c ϭ 0.6). Ϫ IR (neat): ν˜ ϭ 2250 cm (CN). Ϫ H NMR
Millasson-Schmidt, G. Brunner, U. Burger, Helv. Chim. Acta
1996, 79, 1475Ϫ1487.
(
1
4
CDCl
3
, 400 MHz): δ ϭ 1.60Ϫ1.75 (m, 5 H, 7-H, 8-H, 6-Hax),
.82Ϫ1.96 (m, 3 H, 4-Heq, 9-Heq, 6-Heq), 2.10 (ddd, J ϭ 13.7, 11.7,
.8 Hz, 1 H, 9-Hax), 2.22 (td, J ϭ 13.7, 3.8 Hz, 1 H, 4-Hax), 2.26
[
[
5] [5a]
Y.-Z. Yang, W. Asawamahasakda, S. R. Meshnick, Biochem.
[5b]
Pharmacol. 1993, 46, 336Ϫ339. Ϫ
W. Asawamahasakda, I.
Ittarat, Y.-M. Pu, H. Ziffer, S. R. Meshnick, Antimicrob. Agents
Chemother. 1994, 38, 1854Ϫ1858.
(
dt, 15.3, 3.8 Hz, 1 H, 5-Heq), 2.51 (ddd, J ϭ 15.3, 13.5, 3.8, 1 H,
-Hax), 3.39 (d, J ϭ 11.7 Hz, 1 H, CH OCH C), 3.40 (s, 3 H,
CH OCH C), 3.45 (d, J ϭ 11.7 Hz, 1 H, CH OCH C), 3.52 (s, 3
H, 10-COCH ), 4.99 (s, 1 H, H-10). Ϫ C NMR (CDCl
00 MHz): δ ϭ 21.4 (CH , C-7 or C-8), 22.6 (CH , C-8 or C-7),
0.8 (CH , C-4), 32.2 (CH , C-5), 33.3 (CH , C-6), 34.3 (CH , C-
), 50.3 (C, C-5a), 57.1 (CH 10-COCH ), 60.0 (CH
OCH C), 75.4 (CH , CH OCH C), 81.5 (C, C-9a), 103.0 (CH,
C-10), 105.1 (C, C-3), 121.7 (CN). Ϫ C14 (283.3): calcd. C
9.35, H 7.47, N 4.94; found C 59.21, H 7.32, N 4.97.
6]
A. Robert, B. Meunier, J. Am. Chem. Soc. 1997, 119,
5
3
2
5
968Ϫ5969.
3
2
3
2
[7]
[8]
A. Robert, B. Meunier, Chem. Soc. Rev. 1998, 27, 273Ϫ279.
A. Robert, B. Meunier, Chem. Eur. J. 1998, 4, 1287Ϫ1296.
F. Zouhiri, D. Desma e¨ le, J. dЈAngelo, C. Riche, F. Gay, L.
Cic e´ ron, Tetrahedron Lett. 1998, 39, 2969Ϫ2972.
13
3
3
,
[9]
1
3
9
2
2
2
2
2
2
[10]
F. Zouhiri, D. Desma e¨ le, J. dЈAngelo, J. Mahuteau, C. Riche,
F. Gay, L. Cic e´ ron, Eur. J. Org. Chem. 1998, 2897Ϫ2906.
3
,
3
3
,
CH
3
2
2
3
2
[11]
´
M. Hamzaoui, O. Provot, F. Gregoire, C. Riche, A. Chiaroni,
H
21NO
5
F. Gay, H. Moskowitz, J. Mayrargue, Tetrahedron: Asymmetry
1997, 8, 2085Ϫ2088.
5
[
[
12]
M. Hamzaoui, O. Provot, B. Camuzat-Dedenis, H. Moskowitz,
J. Mayrargue, L. Cic e´ ron, F. Gay, Tetrahedron Lett. 1998, 39,
Reaction of Trioxane 8 with Mn (TPP): MnIIItetraphenylporphyrin
chloride (15.5 mg, 22 µmol) and trioxane 8 (17.8 mg, 0.07 mmol)
II
4
029Ϫ4030.
13] [13a]
C. W. Jefford, J. A. Velarde, G. Bernardinelli, D. H. Bray,
D. C. Warhurst, W. K. Milhous, Helv. Chim. Acta 1993, 76,
were dissolved in CH
gassed, and then solid tetra-n-butylammonium borohydride
60.3 mg, 235 µmol) was added. The resulting solution was stirred
at 25°C under nitrogen for 1 h. A solution of Cd(NO )2·4H2O
138 mg, 447 µmol) in deoxygenated DMF (1 mL) was then added
2 2
Cl (4 mL). This solution was carefully de-
[13b]
2
775Ϫ2788. Ϫ
G. H. Posner, C. H. Oh, L. Gerena, W. K.
(
Milhous, Heteroatom Chem. 1995, 6, 105Ϫ116.
P. Mirovsky, F. Gay, D. Bustos, D. Mazier, M. Gentilini, Trans.
R. Soc. Trop. Med. Hyg. 1990, 84, 511Ϫ515.
D. M. Golden, S. W. Benson, Chem. Rev. 1969, 69, 125Ϫ134.
Received February 1, 1999
[
14]
3
(
[15]
and the resulting mixture was stirred under nitrogen for 15 min in
order to achieve transmetallation from the manganese(II) to the
[O99056]
1938
Eur. J. Org. Chem. 1999, 1935Ϫ1938