Structure-activity relationships of the antimalarial agent artemisinin. 4. Effect of substitution at C-3

Article abstract of DOI:10.1021/jm960200e

Novel antimalarial artemisinin analogs, 3-alkylartemisinins as well as 3-(arylalkyl)- and 3-(carboxyalkyl)artemisinins, were prepared via the synthetic intermediate 2. Formation of the N,N-dimethylhydrazones 5 and 24 and then regio- and chemoselective dep

Full text of DOI:10.1021/jm960200e

2900
J . Med. Chem. 1996, 39, 2900-2906
Str u ctu r e-Activity Rela tion sh ip s of th e An tim a la r ia l Agen t Ar tem isin in . 4.
Effect of Su bstitu tion a t C-3
Mitchell A. Avery,* Sanjiv Mehrotra, J ason D. Bonk, J effrey A. Vroman, D. Keith Goins, and Robert Miller†
Department of Medicinal Chemistry and Research Institute of Pharmaceutical Sciences, School of Pharmacy, University of
Mississippi, University, Mississippi 38677, and Division of Experimental Therapeutics, Walter Reed Army Institute of
Research, Washington, D.C. 20307
Received March 11, 1996X
Novel antimalarial artemisinin analogs, 3-alkylartemisinins as well as 3-(arylalkyl)- and
3-(carboxyalkyl)artemisinins, were prepared via the synthetic intermediate 2. Formation of
the N,N-dimethylhydrazones 5 and 24 and then regio- and chemoselective deprotonation
followed by alkylation provided initially alkylated hydrazones that upon chromatography gave
ketones 6-13 and 25-30. Direct ozonolysis of the ketones followed by in situ acidification
lead directly to the formation of title compounds 14-21 and 31-36. The analogs were tested
in vitro against W-2 and D-6 strains of Plasmodium falciparum and found to be in some cases
much more active than the natural product (+)-artemisinin. The results were included in
structure-activity relationship (CoMFA) studies for further analog design.
(+)-Artemisinin (1), a naturally occurring peroxidic
cadinane sesquiterpene possessing good antimalarial
activity,1,2 has been the subject of many total syntheses3
and structure-activity relationship (SAR) studies.4-6
Much of this intense interest stems from the need for
new antimalarial drugs with unconventional structures
and novel modes of action to be used for the treatment
of pervasive strains of drug-resistant Plasmodium fal-
ciparum.7
larial potency. These studies have proven to be useful
in the design of new antimalarials.13
In addition to mechanism-based design which focuses
upon a single determinant in drug activity, empirical
methods, such as quantitative structure-activity rela-
tionship (QSAR) studies, are useful in addressing more
elusive structure-activity components. Very few re-
ports of a QSAR model for the artemisinin class of
antimalarials exist; a classical QSAR study based upon
derivatives of dihydroartemisinin was reported a num-
ber of years ago16 and was only recently followed by 3D-
QSAR studies from our laboratory.17,6 In an effort to
expand upon our pharmacophore model, we felt that it
would be necessary to obtain analogs substituted in as
many available positions as possible about the tetracy-
clic framework of the parent structure 1. Thus far we
have reported modifications to C-917 as well as the
heteroatom-modified N-11 analogs.18 In this paper we
focus upon the effect of modification to the C-3 position
of artemisinin on antimalarial potency.19 The results
are intended for inclusion into a comparative molecular
field analysis (3D-QSAR) which will be reported sepa-
rately.
The mechanism of action of these antimalarials has
also been the subject of scrutiny as this knowledge is
likely to provide the basis for rational drug design
efforts.8-10 While the mode of action of artemisinin and
its analogs is not known with absolute certainty, it is
generally accepted from in vitro studies that scissioning
of the peroxide bond by Fe(II) leads to an intermediate
oxy radical that rapidly abstracts a neighboring hydro-
gen atom from the C-4 position resulting in a more
stable carbon radical.11 The fate of this radical inter-
mediate is not understood but is actively under inves-
tigation in vivo.12 Activity is improved if groups which
stabilize radicals are placed at C-4â but not C-4R.13 This
behavior is also observed in simple trioxanes which
fragment to carbon radicals.14 Furthermore, total syn-
thesis of 13-carbaartemisinin, in which the nonperoxidic
trioxane ring oxygen atom at O-13 is replaced by a
methylene, leads to an abrupt loss of activity, signaling
the mechanistic importance of an intact trioxane ring.15
While the intermediate carbon radical at C-4 is capable
of fragmentation by a deketalization-like process, it
would appear that stabilization of the radical at C-4 by
the adjacent oxygen atom at O-13 is crucial for antima-
Ch em istr y
Both the C-9 and O-11 positions of the artemisinin
nucleus have been shown to be available from the ketal
acid 2 (Scheme 1), a synthetic intermediate in the total
synthesis of (+)-artemisinin (1).20 Because this inter-
mediate could be readily produced in large quantities,
investigation into the chemistry of the ketal-bearing side
chain seemed a worthwhile goal as it could in principle
provide analogs substituted at the C-3, C-4, and C-5
positions. Previously, we had demonstrated that a
similar 5,5-dimethyl-1,3-dioxane ketal, differing only in
the remote carboxylate side chain, could be selectively
deketalized with aqueous oxalic acid-treated silica gel.
Similarly, ketal 2 could be readily converted into the
keto acid 4. Yields in this reaction were found to be
sensitive to conditions; at higher acid concentrations,
protodesilylation led to formation of unwanted exo-
methylene byproduct 3. When conducted carefully,
yields were typically around 80%. In these circum-
† Walter Reed Army Institute of Research.
X Abstract published in Advance ACS Abstracts, J uly 1, 1996.
S0022-2623(96)00200-2 CCC: $12.00 © 1996 American Chemical Society

3-Substituted Antimalarial Artemisinin Analogs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 2901
Sch em e 1a
Sch em e 2a
a
Key: (a) aq oxalic acid, silica gel, CH2Cl2, 82%; (b) Me2NNH2,
∆, 92%; (c) 2 LDA, THF, HMPA, -78-20 °C, then R1-X; (d) O3,
CH2Cl2, -78 °C, then aq H2SO4, silica gel, CH2Cl2.
a
Key: (a) aq oxalic acid, silica gel, CH2Cl2, 85%; (b) Me2NNH2,
∆, 93%; (c) LDA, THF, HMPA, -78-20 °C, then R1-X; (d) O3,
CH2Cl2, -78 °C, then aq H2SO4, silica gel, CH2Cl2.
stances there appeared to be very little protodesilyla-
tion, the yields being less than theoretical due to facile
adsorption of the keto acid 4 onto silica gel.
6-13, could then be stored indefinitely or used directly
for the final stage of the synthetic sequence.
Initial attempts to alkylate the kinetic enolate of keto
acid 4 (generated upon treatment with 2 equiv of LDA)
with methyl iodide were unsuccessful due to formation
of mono-, di-, and trimethylated ketones. This problem
would presumably have worsened with less reactive
alkylating agents, and therefore an alternate methodol-
ogy was sought. Hydrazone chemistry appeared to offer
the desired regioselectivity as it has been reported that
the less substituted side of N,N-dimethylhydrazones is
preferentially deprotonated, regardless of the hydrazone
isomer (E or Z).21,22
The N,N-dimethylhydrazone 5, furnished from keto
acid 4 upon treatment with N,N-dimethylhydrazine,
was found to be extremely water sensitive. Attempts
to form the hydrazone were thwarted by low yields
under a number of conditions in which solvents were
present. Azeotropic removal of water, with or without
molecular sieves, was also unsatisfactory. Eventually,
it was found most convenient to simply dissolve the keto
acid in neat dimethylhydrazine without dessicant. After
heating for a number of hours followed by cooling and
removal of excess dimethylhydrazine, formation of the
desired hydrazone was apparent by NMR due to loss of
the methyl ketone resonance at δ 2.14. This initially
formed hydrazone existed as a dimethylhydrazonium
carboxylate, but it was found that reversion to the free
carboxylic acid 5 occurred in vacuo, as evidenced by the
proton NMR run in dry CDCl3.
With simple methodology in hand for formation of the
requisite hydrazone 5, we returned to our alkylation
studies. Treatment of the hydrazone 5 in tetrahydro-
furan (THF) with 2 equiv of LDA at low temperature
led to formation of hydrazone enolate as was evidenced
by the slow appearance of alkylated products by TLC.
However, these reactions were reluctant to go to comple-
tion, and it was found that if hexamethylphosphoric
triamide (HMPA) were added to the intermediate met-
alloenamine, alkylation would reach completion in a
matter of hours. While the preliminary alkylation
product(s) were now relatively stable regioisomeric
hydrazones, alkylated exclusively on the methyl group,
and could be used crude in the ensuing ozonolysis
reaction, products were easier to characterize after silica
gel chromatography in which the hydrazone group was
cleaved. The products of chromatography, keto acids
It was anticipated that ozonolysis of the keto acids
6-13 would be uncomplicated leading to an intermedi-
ate dioxetane or hydroperoxy lactol and that in situ
acidification would then result in multiple cyclizations
to afford desired tetracyclic artemisinin analogs modi-
fied at the C-3 position.17,23-25 In fact, exposure of the
keto acids to ozone at low temperature followed by
purging of excess ozone and addition of silica gel and
aqueous sulfuric acid led, over several days at room
temperature, to clean formation of desired targets 14-
21. Yields were generally quite good for production of
the target analogs, being in the range of 55-66% with
one notable exception, the ethyl ester 18 for which the
yield was substantially lower at 26%.
We were also interested in the possibility of synthe-
sizing derivatives modified in both the carboxylic acid
side chain and ketone-bearing side chain; the resulting
tetracyclic analogs of artemisinin would then be modi-
fied at both C-3 and C-9. As the requisite ketal(s) were
available,17 their deketalization, hydrazone formation,
alkylation, and subsequent ozonolysis and cyclization
were explored as shown in Scheme 2.
Starting with known “butylated” ketal acid 22, readily
synthesized in high yield directly from ketal acid 2,
deketalization as before with oxalic acid-treated silica
gel gave comparable results to before, with keto acid
23 being produced in 85% yield. Derivatization of this
ketone as the hydrazone provided, in high yield, the
expected hydrazone 24, once again as an apparently
pure regioisomer as inferred by the appearance in the
NMR spectra of 24 of a singlet for the vinylsilane proton
at δ 5.30. Alkylation of 24 by treatment first with 2
equiv of LDA followed by addition of HMPA and the
alkylating agent gave intermediate hydrazones which
upon chromatography underwent hydrolysis and af-
forded the penultimate intermediate keto acids 25-30
in reasonable yields ranging from 42% to 66%. Finally,
low-temperature ozonolysis followed by admission of
silica gel and aqueous acid furnished the target tetra-
cyclic analogs 31-36 substituted at both C-9 and C-3.
Unfortunately, yields for targets 31-36 were roughly
one-half of those obtained for tetracycles not substituted
at C-9 (14-21).
It is noteworthy (Scheme 3) that cyclization of inter-
mediate hydroperoxy aldehyde equivalents such as 40,

2902 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Avery et al.
Ta ble 1. Relative in Vitro Antimalarial Activity of
3-Substituted Analogs of Artemisinin against P. falciparum
Sch em e 3
relative
activitya
structure
R1
R
D-6 W-2 anal. (C,H)
1
14
15
16
17
18
19
20
21
31
32
33
34
35
36
H
CH3
CH3CH2
CH3
H
H
H
H
H
H
H
H
100 100 C15H22O5
88 112 C15H22O5
2102 673 C16H24O5
CH3(CH2)2
(CH3)2CH
EtO2CCH2
C6H5CH2
p-ClC6H4(CH2)2
C6H5(CH2)3
CH3
CH3(CH2)2
C6H5CH2
p-ClC6H4(CH2)2 CH3(CH2)3
20
53
18 C17H26O5
45 C17H26O5
232 232 C18H26O7
C21H26O5
3
1
114 127 C22H27ClO5
220 281 C23H32O5
CH3(CH2)3 184 257 C19H30O5
CH3(CH2)3
CH3(CH2)3
28
1
33 C21H34O5
C25H34O5
1
whether from total synthetic studies,3 analog work,6,17
or this work, bearing a substituent in the carboxylate
side chain at C-1′ (e.g., 40, R ) alkyl) typically occur in
a range of 15-35%. On the other hand, those lacking
a substituent (e.g., 39 or 41, R ) H) are more efficiently
converted to tetracycles with yields roughly from 50%
to 65%. We have noted that this effect is unrelated to
the efficiency with which initial oxidative addition
occurs providing oxetanes 38. Molecular modeling
studies26 indicate that increased torsional strain from
the C-9 R substituent to the C-8 methylene in product
43 is reflected in less rapid cyclization of intermediate
40, resulting in an increase in competing decomposition
of 40 and thus lower yields of desired products such as
43. When the R substituent is H, as in 39, the
energetically unfavorable C-9/C-8 interaction in the
product 42 is relieved, and thus the reaction is more
efficient.
43
39
53 C26H35ClO5
48 C27H38O5
C6H5(CH2)3
EtO2CCH2
CH3(CH2)3
CH3(CH2)3 1382 2285 C22H34O7
a
Relative activity ) 100 × [IC50(artemisinin, control value)/
IC50(analog)]MW(analog)/MW(artemisinin).
influence on antimalarial potency in vitro. Addition of
another carbon at C-3 giving the ethyl analog 14 (R1 )
Et, but lacking the C-9 methyl group, R ) H) results in
little effect on activity, while the next homolog 15 (R1
) propyl) is quite potent compared to artemisinin.
Interestingly, 3-butyl homolog 16 and its branched
counterpart 17 have diminished antimalarial activity
relative to control. The overall trend for this series is
qualitatively similar to that observed for C-9 substitu-
tion: A peak and trough in activity is seen upon
substitution of n-alkyl groups at C-9 of two to six
carbons in chain length. For C-9, maximal activity was
achieved at ethyl and propyl (12 × activity of artemisi-
nin) but had dropped by hexyl to around 5 times the
potency of artemisinin. In this C-3-substituted series,
the activity profile is attenuated with a peak occurring
at propyl 15 (7-21 times activity of artemisinin) but
with lower activity on either side such as ethyl 14 and
butyl 16. Interestingly, placement of an ester group at
C-3 as in the propionate ethyl ester 18 results in an
analog with better activity than artemisinin (2 ×
artemisinin), while the closest homolog to 18 is the
relatively inactive isobutyl derivative 17.
Next, the effect of aryl substitution was considered.
The series substituted at C-3: 2-phenylethyl 19, 3-(p-
chlorophenyl)propyl 20, and 4-phenylbutyl 21, displayed
roughly 1%, 100%, and 300% the activity of artemisinin,
respectively. Compared to the analogous series at C-9
(2-phenylethyl, 100%; 3-phenylpropyl, 500%; 4-phenyl-
butyl, 300%), these results are only substantially at
variance for the phenylethyl analog 19. Furthermore,
as was the case for C-9-substituted arylalkyls, the steric
bulk typified by a aryl ring did not adversely effect
activity, but branched hydrocarbons did lower potency
appreciably.
Biologica l Activity
The analogs 14-21 and 31-36 were tested in vitro
in parasitized whole blood (human) against drug-
resistant strains of P. falciparum at the Walter Reed
Army Institute of Research by a modification of the
procedure of Desjardins27,28 involving uptake of tritiated
hypoxanthine. Two P. falciparum malaria parasite
clones, designated as Indochina (W-2) and Sierra Leone
(D-6), were utilized in susceptibility testing. The W-2
clone is chloroquine-resistant and mefloquine-sensitive,
while the D-6 clone is chloroquine-sensitive but meflo-
quine-resistant. The relative potency values for these
analogs were derived from the IC50 value for artemisinin
(1) divided by their IC50 values (Table 1) and then
adjusted for molecular weight differences by multiplica-
tion of the ratio of the molecular weight of the analog
divided by the molecular weight of artemisinin. This
approach to reporting activity was based in part on the
fact that the analogs were tested on different occasions
in which the IC50 for the control, artemisinin, had varied
anywhere from 0.4 to 2 ng/mL based on parasitemia
levels.
On the whole, analogs of artemisinin substituted at
C-3 were found to be less active than those substituted
at C-9. Because it was a simple matter to combine
methodologies for synthesis of analogs at both C-3 and
Str u ctu r e-Activity Rela tion sh ip s
Substitution solely at C-3 and the effect of dual
substitution at C-3 and C-9 were examined for their

3-Substituted Antimalarial Artemisinin Analogs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 2903
acceptors with appropriate tethers at the C-3 position
of artemisinin might be expected to have enhanced
activity.
The results of these studies have been incorporated
into an evolving QSAR model and will be reported when
additional data has been collected for the remaining
unexplored positions about the artemisinin framework.
Exp er im en ta l Section
All solvents were purchased as reagent grade and where
appropriate were distilled from CaH2 and stored over dry 4 Å
sieves for at least 1 day prior to use. Solvent and reagent
transfers were accomplished via dried syringe, and all reac-
tions were routinely conducted under an inert atmosphere
unless otherwise indicated. Flash chromatography was ac-
complished using silica gel (Whatman 60, 230-400 mesh).
Preparative thin-layer chromatography utilized 1-, 1.5-, or
2-mm thick Analtech Uniplates with F-256, and 250-µm silica
gel thin-layer chromatography plates were also purchased from
Analtech. Unless otherwise noted, all NMR analyses were
conducted in CDCl3, on a Varian VXR-300 spectrometer, and
referenced to chloroform at δ 7.27. IR spectra were recorded
on a Digilab FTS-40 FT-IR instrument and run neat unless
otherwise indicated. Mass spectral data was obtained on a
VG 7070E-HF instrument. Elemental analyses were within
0.4% as determined by Desert Analytics, Tucson, AZ.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxobu tyl)-2′(E)-[(tr i-
m eth ylsilyl)m eth ylen e]cycloh exyl]eth a n oic Acid (4). To
a vigorously stirring suspension of silica gel 60 (230-400 mesh,
2.33 g) in CH2Cl2 (40 mL) was added 10% aqueous oxalic acid
(820 µL). After 15 min, a solution of ketal acid 2 (0.65 g, 1.64
mmol) in CH2Cl2 (10 mL) was added. The reaction mixture
was stirred for 24 h, filtered, and washed with CH2Cl2 (10 mL)
and then EtOAc (10 mL). The organic layer was washed with
water, dried (MgSO4), and evaporated to give 0.414 g (82%) of
pure keto acid 4. 1H NMR: δ 5.42 (s, 1H), 2.74 (br s, 1H),
2.52-2.64 (m, 1H), 2.38-2.48 (m, 1H), 2.14 (s, 3H), 0.91 (d,
3H, J ) 6.8 Hz), 0.078 (s, 9H). IR: 2949, 1703, 1601, 1408,
1246, 1163, 834, 747, 689 cm-1. DCIMS (NH3): m/z 311 (M +
NH4), 293, 277, 250, 221, 203, 161, 143, 129, 108.
F igu r e 1. Docking interaction of hemin with artemisinin
analog 37 in Sybyl showing hydrogen bonding between hemin
side chains and the C-3 ester group of analog 37.
C-9, we thought that dual substitution would provide
additional interesting SAR information. The activities
of analogs 32-35 were unimpressive, while the activity
of 31 and 36 were good to excellent. The following
trends were observed. For increasing alkyl bulk at C-3
a drop in antimalarial efficacy was noted (14, relative
activity of 1.1; 16, relative activity of 0.2). Upon butyl
substitution at C-9, the corresponding dual substituted
analogs (3-alkyl, 9-butyl) showed a doubling of activity
(31, relative activity of 2.6; 32, relative activity of 0.33).
For the C-3 arylalkyl-substituted analogs alone, an
increase in activity was observed with increasing chain
length between ring system and the aryl ring (two
carbons for 19, relative activity of 0.01; three carbons
for 20, relative activity of 1.3; four carbons for 21,
relative activity of 2.8). Dual substituted arylalkyl
analogs (3-arylalkyl, 9-butyl) were generally less active
than 3-substituted arylalkyl analogs alone (e.g., 20,
relative activity of 1.3; 34, relative activity of 0.53).
Thus, a number of opposing trends were evident.
If 1.5 mL of acid was used, small amounts of byproduct 3
were also formed (isolated by flash chromatography, 10:90
MeOH/CHCl3). 1H NMR: δ 4.67 (br s, 1H), 4.61 (br s, 1H),
2.62 (m, 2H), 2.39 (m, 3H), 2.23 (m, 1H), 2.13 (s, 3H), 1.53-
2.07 (m, 5H), 1.01-1.31 (m, 3H), 0.97 (d, 3H, J ) 5.8 Hz). 13C
(75 MHz): δ 209.3, 178.8, 152.5, 103.1, 50.1, 41.4, 40.9, 39.2,
38.0, 34.9, 34.4, 30.0, 22.0, 20.4. IR (CHCl3): 3480, 3089, 2921,
2871, 1708, 1641, 1409, 1288, 1172, 896 cm-1. DCIMS (NH3):
m/z 256 (M + NH4), 238 (M+), 221, 119.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxobu tyl)-2′(E)-[(tr i-
m et h ylsilyl)m et h ylen e]cycloh exyl]et h a n oic Acid Di-
m eth ylh yd r a zon e (5). To keto acid 4 (0.9 g, 2.9 mmol) was
added 1,1-dimethylhydrazine (4 mL). The reaction mixture
was refluxed for 5 h and cooled, and the excess dimethylhy-
drazine was evaporated. The reaction mixture was then
dissolved in dry benzene, dried over anhydrous MgSO4,
filtered, evaporated, and dried overnight in vacuo to give 0.938
g (92%) of pure 5 as a white crystalline solid, mp 98-99 °C.
1H NMR: δ 5.51 (s, 1H), 2.92 (br s, 1H), 2.41-2.50 (m, 8H),
2.24 (m, 4H), 2.00 (s, 3H), 1.84 (m, 4H), 1.60 (m, 1H), 1.42 (m,
1H), 1.12 (m, 1H), 0.94 (d, 3H, J ) 7.0 Hz), 0.08 (s, 9H). IR:
2949, 2861, 2813, 1631, 1600, 1414, 1245, 846 cm-1. DCIMS
(isobutane): m/z 353 (M + H), 295, 221, 161.
Gen er a l P r oced u r e for Alk yla tion of th e Dim eth ylh y-
d r a zon es 5 a n d 24: Syn th esis of Keton es 6-13 a n d 25-
30. To an ice-cold solution of diisopropylamine (2.25 mmol)
in dry THF (4 mL/mmol) was added n-butyllithium (2.25 equiv
of a 2.5 M solution). The mixture was stirred at 0 °C for 15
min and then cooled to -78 °C. The hydrazone 5 or 24 (1
mmol) in dry THF (4 mL/mmol) was added via cannula, and
the mixture was stirred at -78 °C for 10 min. Dry HMPA
(0.5 mL/mmol) was then added to the reaction mixture via
syringe, and the mixture was stirred at -78 °C for 15 min.
After warming to 0 °C, alkylating agent (2.5 mmol) was added
The high potency of the dual substituted analog 36,
on the other hand, is somewhat of a surprise in light of
the foregoing discussion. By comparison to 18 (ester
side chain), 36 should be roughly no better than 2-fold
more potent than artemisinin. However, one is faced
with a 14-23-fold enhancement in activity for 36 rela-
tive to artemisinin.
These findings suggest the following qualitative
statements: (a) a binding site or acceptor for artemisi-
nin and analogs exists with limited dimensions at both
C-9 and C-3, more tolerant of aryl and ester substitution
on n-alkyl chains than of branched alkanes of any
length; b) dual substitution at C-3 and C-9, explored for
only 9-butyl analogs, was on the whole detrimental to
activity with one interesting exception, 36. As it hap-
pens, 36 would be predicted to dock with hemin better
than other members of this series. Perhaps the efficacy
of hydrogen bonding of 36 to hemin (putative mecha-
nistic bottleneck in the mode of action of these antima-
larials) is reflected in the enhanced activity of this
analog. The interaction energies of artemisinin and
dihydroartemisinin with hemin have been calculated in
Sybyl using the dock routine.29 We have applied this
technique to the interaction of 36 with hemin in the Fe-
(II) oxidation state. The lowest E alignment of 36 with
hemin places the ester group in the vicinity of the
carboxyl groups of hemin, as shown in Figure 1, and
allows for hydrogen bonding between hemin and 36. If
this hypothesis is true, then other hydrogen bond

2904 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Avery et al.
to the reaction mixture which was then stirred at ambient
temperature for 6 h, poured onto ice-cold saturated aqueous
NH4Cl, extracted with EtOAc, dried (MgSO4), and evaporated
in vacuo to give a crude product which was purified by flash
chromatography. Elution with EtOAc/hexane containing 1%
HOAc gave pure alkylated products 6-13 and 25-30.
Gen er a l P r oced u r e for P r ep a r a tion of Ta r get Ar te-
m isin in An a logs: Ozon olysis a n d Acid -Ca ta lyzed Cy-
cliza tion of Keto Acid s 6-13 a n d 25-30. Into a solution
of keto acid (1 mmol) in dry CH2Cl2 (125 mL) at -78 °C was
bubbled a stream of O3/O2 (4 PSI, 0.04 L/min, 80 V) until the
reaction mixture had turned faint blue. After the reaction
mixture was purged with O2 followed by N2, silica gel (750 mg)
and 15% aqueous H2SO4 (75 µL) were added. The mixture
was allowed to come to ambient temperature and stirred
overnight, the solids were filtered and washed with CH2Cl2 (2
× 10 mL) and then EtOAc (2 × 10 mL). The resulting filtrate
was washed with saturated aqueous NaHCO3, dried over Na2-
SO4, and evaporated in vacuo to give crude products which
were easily purified by preparative thin-layer chromatography
(PTLC) on silica gel.
hexane provided the pure product 11 in 62% yield. 1H NMR:
δ 7.18-7.26 (m, 5H), 5.41 (s, 1H), 2.88 (dd, 2H, J ) 7.5, 7.5
Hz), 2.76 (dd, 2H, J ) 7.8, 7.8 Hz), 2.57 (dd, 1H, J ) 9.5, 14.9
Hz), 2.32-2.42 (m, 3H), 2.12 (m, 1H), 1.72-1.88 (m, 4H), 1.62
(m, 1H), 1.40 (m, 1H), 1.14 (m, 1H) 0.91 (d, 3H, J ) 6.8 Hz),
0.07 (s, 9H). IR: 3400-2500, 1715, 1708, 1603, 1495, 1452,
1400, 1247, 873, 839 cm-1. DCIMS (NH3): m/z 418 (M + NH4),
401 (M + H), 383, 311, 195.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-[6′′-(p-ch lor op h en yl)-3′′-
oxoh exyl]-2′(E)-[(tr im eth ylsilyl)m eth ylen e]cycloh exyl]-
eth a n oic Acid (12). Chromatography of the crude material
with 20% EtOAc/hexane provided the pure product 12 in 59%
yield. 1H NMR: δ 7.24 (d, 2H, J ) 8.5 Hz), 7.09 (d, 2H, J )
8.6 Hz), 5.42 (s, 1H), 2.8 (br s, 1H), 2.54-2.62 (m, 3H), 2.34-
2.42 (m, 4H), 2.12 (m, 1H), 1.72-1.92 (m, 6H), 1.54-1.68 (m,
2H), 1.42 (m, 1H), 1.12 (m, 1H), 0.92 (d, 3H, J ) 6.9 Hz), 0.07
(s, 9H). IR: 3380-2600, 1708, 1601, 1491, 1452, 1407, 1368,
1246, 1092, 867, 847 cm-1. DCIMS (NH3): m/z 449 (M + H),
448 (M+), 388, 373, 359, 229, 252.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(7′′-ph en yl-3′′-oxoh eptyl)-
2′(E)-[(tr im eth ylsilyl)m eth ylen e]cycloh exyl]eth an oic Acid
(13). Chromatography of the crude material with 20% EtOAc/
hexane provided the pure product 13 in 62% yield. 1H NMR:
δ 7.12-7.28 (m, 5H), 5.42 (s, 1H), 2.80 (br s, 1H), 2.54-2.66
(m, 3H), 2.36-2.48 (m, 4H), 2.12 (m, 1H), 1.74-1.90 (m, 3H),
1.58-1.64 (m, 4H), 1.42 (m, 1H), 1.14 (m, 1H), 0.92 (d, 3H, J
) 7.0 Hz), 0.09 (s, 9H). IR: 3400-2500, 1735, 1710, 1602,
1496, 1453, 1294, 1247, 1157, 868, 850 cm-1. DCIMS (NH3):
m/z 446 (M + NH4), 429 (M + H), 411, 339, 278, 261.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxobu tyl)-2′(E)-[(tr i-
m eth ylsilyl)m eth ylen e]cycloh exyl]h exa n oic Acid (23).
To a vigorously stirred suspension of silica gel (14.25 g) in dry
CH2Cl2 (180 mL) was added 10% aqueous oxalic acid (6.8 mL).
After 15 min, a solution of the ketal acid 22 (3.2 g, 7.0 mmol)
in CH2Cl2 was added. The reaction mixture was stirred at
ambient temperature for 40 h and then heated at 40 °C for 8
h, filtered, and washed successively with CH2Cl2 and then
EtOAc. The organic layer was washed with water, dried
(MgSO4), and evaporated in vacuo to give 2.2 g (85% yield) of
pure keto acid 23 as a viscous liquid. 1H NMR: δ 5.35 (s, 1H),
2.65 (m, 1H), 2.35-2.52 (m, 3H), 2.14 (s, 3H), 1.82 (m, 4H),
1.12-1.62 (m, 10H), 0.93 (d, 3H, J ) 5.0 Hz), 0.89 (dd, 3H, J
) 6.7, 6.7 Hz), 0.09 (s, 9H). IR: 3420-2400, 1730, 1693, 1598,
1461, 1408, 1361, 1246, 1165, 1099, 952, 822 cm-1. DCIMS
(NH3): m/z 384 (M + NH4), 367 (M + H), 349, 231, 143.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxobu tyl)-2′(E)-[(tr i-
m et h ylsilyl)m et h ylen e]cycloh exyl]h exa n oic Acid Di-
m eth ylh yd r a zon e (24). To the keto acid 23 (2.20 g, 6.0
mmol) under N2 was added N,N-dimethylhydrazine (7.0 mL).
The reaction mixture was refluxed for 4 h, excess dimethyl-
hydrazine was evaporated, and the reaction mixture was
dissolved in dry benzene, dried over MgSO4, filtered, evapo-
rated, and dried overnight in vacuo to give 2.28 g (93%) of the
pure hydrazone 24 as a viscous liquid. 1H NMR: δ 5.30 (s,
1H), 2.30-2.46 (m, 9H), 2.02-2.20 (m, 2H), 1.90 (m, 4H), 1.75
(m, 2H), 1.08-1.52 (m, 10H), 0.92 (d, 3H, J ) 8.3 Hz), 0.82
(dd, 3H, J ) 7.0, 7.0 Hz), 0.10 (s, 9H). IR: 2953, 2947, 2896,
1704, 1600, 1452, 1359, 1246, 1164, 854, 846 cm-1. DCIMS
(isobutane): m/z 409 (M + H), 366, 319, 256, 221, 164.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxopen tyl)-2′(E)-[(tr i-
m eth ylsilyl)m eth ylen e]cycloh exyl]h exa n oic Acid (25).
Flash chromatography of the crude product was accomplished
eluting with 20% EtOAc/hexane and gave pure 25 (56%) as a
yellow viscous liquid. 1H NMR: δ 5.34 (s, 1H), 2.68 (m, 1H),
2.30-2.50 (m, 6H), 2.14 (m, 1H), 1.86 (m, 4H), 1.42-1.68 (m,
2H), 1.12-1.66 (m, 6H), 1.06 (dd, 3H, J ) 7.3, 7.3 Hz), 0.93
(d, 3H, J ) 7.1 Hz), 0.89 (dd, 3H, J ) 6.8, 6.8 Hz), 0.10 (s,
9H). IR: 3400-2400, 1735, 1697, 1599, 1453, 1414, 1290,
1246, 1160, 1108, 861 cm-1. DCIMS (isobutane): m/z 381 (M
+ H), 363 (M - H2O), 291, 264, 157.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxop en tyl)-2′(E)-[(tr i-
m eth ylsilyl)m eth ylen e]cycloh exyl]eth an oic Acid (6). Chro-
matography of the crude material with 20% EtOAc/hexane
provided the pure product 6 in 93% yield. 1H NMR: δ 5.42
(s, 1H), 2.80 (br s, 1H), 2.60 (dd, 1H, J ) 9.5, 14.8 Hz), 2.38-
2.46 (m, 6H), 2.14 (m, 1H), 1.76-1.90 (m, 4H), 1.68 (m, 1H),
1.42 (m, 1H), 1.04 (dd, 3H, J ) 7.3, 7.3 Hz), 0.92 (d, 3H, J )
6.9 Hz), 0.08 (s, 9H). IR: 3280-2810, 1734, 1710, 1601, 1453,
1410, 1291, 1247, 848 cm-1
NH4), 325 (M + H), 307, 235, 175.
. DCIMS (NH3): m/z 342 (M +
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxoh exyl)-2′(E)-[(tr i-
m eth ylsilyl)m eth ylen e]cycloh exyl]eth an oic Acid (7). Chro-
matography of the crude material with 20% EtOAc/hexane
provided the pure product 7 in 83% yield. 1H NMR: δ 5.42
(s, 1H), 2.82 (br s, 1H), 2.60 (dd, 1H, J ) 9.4, 14.7 Hz), 2.36-
2.44 (m, 6H), 2.20 (m, 1H), 1.84 (m, 4H), 1.54-1.68 (m, 3H),
1.42 (m, 1H), 1.14 (m, 1H), 0.91 (dd, 3H, J ) 7.4, 7.4 Hz), 0.93
(d, 3H, J ) 6.9 Hz), 0.08 (s, 9H), IR: 3300-2800, 1731, 1709,
1601, 1454, 1409, 1247, 867, 837 cm-1
338 (M+), 174, 189, 129, 108, 73.
. DCIMS (NH3): m/z
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxoh ep tyl)-2′(E)-[(tr i-
m eth ylsilyl)m eth ylen e]cycloh exyl]eth an oic Acid (8). Chro-
matography of the crude material with 20% EtOAc/hexane
provided the pure product 8 in 58% yield. 1H NMR: δ 5.42
(s, 1H), 2.82 (br s, 1H), 2.56-2.64 (m, 1H), 2.37-2.42 (m, 6H),
2.16 (m, 1H), 1.82 (m, 1H), 1.66 (m, 1H), 1.52 (m, 1H), 1.42
(m, 1H), 1.22-1.30 (m, 4H), 0.87-0.94 (m, 6H), 0.09 (s, 9H).
IR: 3380-2600, 1712, 1601, 1461, 1409, 1289, 1247, 863, 849
cm-1. DCIMS (NH3): m/z 370 (M + NH4), 353 (M + H), 335,
292, 263, 195.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(5′′-m eth yl-3′′-oxoh exyl)-
2′(E)-[(tr im eth ylsilyl)m eth ylen e]cycloh exyl]eth an oic Acid
(9). Chromatography of the crude material with 20% EtOAc/
hexane provided the pure product 9 in 63% yield. 1H NMR:
δ 5.42 (s, 1H), 2.82 (br s, 1H), 2.59 (dd, 1H, J ) 9.5, 14.8 Hz),
2.34-2.48 (m, 2H), 2.28 (d, 1H, J ) 6.6 Hz), 2.10-2.15 (m,
2H), 1.82 (m, 4H), 1.64 (m, 1H), 1.42 (m, 1H), 1.24 (m, 1H),
0.93 (d, 3H, J ) 5.6 Hz), 0.92 (d, 6H, J ) 6.6 Hz), 0.08 (s, 9H).
IR: 3380-2520, 1731, 1708, 1601, 1463, 1452, 1409, 1249, 849
cm-1. DCIMS (NH3): m/z 370 (M + NH4), 353 (M + H), 335,
295, 263, 195.
(2R,1′S,3′S,4′S)-2-[4′-Met h yl-3′-(5′′-ca r b et h oxy-3′′-oxo-
p e n t yl)-2′(E )-[(t r im e t h ylsilyl)m e t h yle n e ]cycloh e xyl]-
eth a n oic Acid (10). Chromatography of the crude material
with 20% EtOAc/hexane provided the pure product 10 in 63%
yield. 1H NMR: δ 5.43 (s, 1H), 4.11 (dd, 2H, J ) 7.1, 14.2
Hz), 2.70-2.84 (m, 2H), 2.52-2.62 (m, 3H), 2.40-2.56 (m, 2H),
2.12-2.20 (m, 1H), 1.78-1.88 (m, 4H), 1.25 (dd, 3H, J ) 7.1,
7.1 Hz), 0.93 (d, 3H, J ) 7.0 Hz), 0.08 (s, 9H). IR: 3200-
2800, 1737, 1712, 1601, 1461, 1411, 1247, 1097, 867, 849 cm-1
.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(3′′-oxoh eptyl)-2′(E)-[(tr i-
m eth ylsilyl)m eth ylen e]cycloh exyl]h exa n oic Acid (26).
Flash chromatography over silica gel with 20% EtOAc/hexane
gave pure 26 (82%) as a thick oil. 1H NMR: δ 5.34 (s, 1H),
2.68 (m, 1H), 2.30-2.52 (m, 6H), 2.12 (m, 1H), 1.82 (m, 4H),
1.52-1.60 (m, 4H), 1.12-1.42 (m, 8H), 0.90 (dd, 3H, J ) 7.3,
DCIMS (NH3): m/z 414 (M + NH4), 397 (M + H), 379, 307,
289, 261.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(5′′-ph en yl-3′′-oxopen tyl)-
2′(E)-[(tr im eth ylsilyl)m eth ylen e]cycloh exyl]eth an oic Acid
(11). Chromatography of the crude material with 20% EtOAc/

3-Substituted Antimalarial Artemisinin Analogs
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15 2905
7.3 Hz), 0.87 (dd, 3H, J ) 6.8, 6.8 Hz), 0.93 (d, 3H, J ) 7.0
Hz), 0.10 (s, 9H). IR: 3400-2400, 1735, 1704, 1599, 1463,
1454, 1290, 1246, 1161, 1102, 859 cm-1. DCIMS (isobutane):
m/z 409 (M + H), 391 (M - H2O), 319, 292, 273, 203.
Oct a h yd r o-3-(2-m et h ylp r op yl)-6-m et h yl-3,12-ep oxy-
12H-pyr an o[4,3-j]-1,2-ben zodioxepin -10(3H)-on e (17). PTLC
of the crude product on silica gel, eluting with 20% EtOAc/
hexane, gave 17 (49%) as an oil. 1H NMR: δ 5.92 (s, 1H), 3.19
(dd, 1H, J ) 6.9, 8.3 Hz), 2.33-2.42 (m, 1H), 2.27 (dd, 1H J )
1.1, 8.2 Hz), 1.82-2.08 (m, 4H), 1.66-1.76 (m, 2H), 1.54-1.64
(m, 6H), 1.82-1.98 (m, 2H), 0.99 (d, 3H, J ) 5.9 Hz), 0.93 (dd,
3H, J ) 6.7, 8.4 Hz). IR: 2959, 2926, 2895, 1749, 1465, 1365,
1208, 1106, 1032, 994, 948 cm-1. DCIMS (isobutane): m/z 310
(M+), 296, 294, 281, 212, 197.
Octah ydr o-3-(3′-car beth oxyeth yl)-6-m eth yl-3,12-epoxy-
12H-p yr a n o[4,3-j]-1,2-ben zod ioxep in -10(3H)-on e (18). The
crude product was purified on a silica gel PTLC plate using
50% EtOAc/hexane to give pure 18 (26%) as an oil. 1H NMR:
δ 5.91 (s, 1H), 4.11 (dd, 2H, J ) 7.1, 14.2 Hz), 3.17 (dd, 1H, J
) 6.9, 18.2 Hz), 2.30-2.52 (m, 4H), 2.26 (dd, 1H, J ) 1.1, 18.2
Hz), 1.96-2.09 (m, 4H), 1.84-1.94 (m, 1H), 1.70-1.78 (m, 2H),
1.38-1.48 (m, 4H), 1.23 (dd, 3H, J ) 7.1, 7.1 Hz), 0.99 (d, 1H,
J ) 5.9 Hz). IR: 2952, 2926, 1744, 1209, 1191, 1091, 1001,
940, 895 cm-1. FABMS: m/z 355 (M + H), 337, 307, 237, 154,
136.
Octa h yd r o-3-(2-p h en yleth yl)-6-m eth yl-3,12-ep oxy-12H-
pyr an o[4,3-j]-1,2-ben zodioxepin -10(3H)-on e (19). The crude
product was purified by flash chromatography eluting with
20% EtOAc/hexane to give pure 19 as a white crystalline solid,
mp 114 °C. 1H NMR: δ 7.14-7.28 (m, 5H), 5.96 (s, 1H), 3.21
(dd, 1H, J ) 6.2, 18.9 Hz), 2.65-2.84 (m, 2H), 2.35-2.48 (m,
1H), 2.30 (dd, 1H, J ) 1.1, 18.2 Hz), 1.98-2.12 (m, 4H), 1.88-
1.98 (m, 1H), 1.72-1.80 (m, 2H), 1.39-1.58 (m, 5H), 1.01 (d,
3H, J ) 5.7 Hz). IR: 2967, 2926, 2852, 1741, 1453, 1208, 1030,
995 cm-1. FABMS: m/z 359 (M + H), 313, 307, 289, 279.
Octa h yd r o-3-[3-(4-ch lor op h en yl)p r op yl]-6-m eth yl-3,12-
ep oxy-12H -p yr a n o[4,3-j]-1,2-b en zod ioxep in -10(3H )-on e
(20). The crude product was purified by flash chromatography
eluting with 20% EtOAc/hexane to give 20 (63%) as an oil. 1H
NMR: δ 7.24 (d, 2H, J ) 8.5 Hz), 7.08 (d, 2H, J ) 8.6 Hz),
5.91 (s, 1H), 3.18 (dd, 1H, J ) 6.9, 18.2 Hz), 2.52 (m, 2H), 2.26-
2.38 (m, 1H), 2.27 (dd, 1H, J ) 1.1, 18.2 Hz), 1.96-2.02 (m,
2H), 1.84-1.94 (m, 1H), 1.69-1.78 (m, 6H), 1.40 (m, 4H), 0.98
(d, 3H, J ) 5.8 Hz). IR: 2955, 2924, 2848, 1743, 1491, 1209,
1089, 999 cm-1. DCIMS (NH3): m/z 407, (M + H), 406 (M+),
392, 377, 362, 358.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(5′′-ph en yl-3′′-oxopen tyl)-
2′(E)-[(tr im eth ylsilyl)m eth ylen e]cycloh exyl]h exan oic Acid
(27). Chromatography of the crude material with 20% EtOAc/
hexane provided the pure product 26 in 42% yield. 1H NMR:
δ 7.18-7.30 (m, 5H), 5.34 (s, 1H), 2.90 (dd, 1H, J ) 7.6, 7.6
Hz), 2.62-2.74 (m, 3H), 2.26-2.52 (m, 6H), 2.12 (m, 1H), 1.72-
1.88 (m, 4H), 1.12-1.74 (m, 8H), 0.92 (d, 3H, J ) 7.1 Hz), 0.87
(dd, 3H, J ) 6.8, 6.8 Hz), 0.09 (s, 9H). IR: 3380-2500, 1729,
1703, 1600, 1452, 1408, 1290, 1246, 857, 835 cm-1. DCIMS
(isobutane): m/z 457 (M + H), 439, 367, 340, 251, 233.
(2R,1′S,3′S,4′S)-2-[4′Meth yl-3′-[6′′-(p-ch lor op h en yl)-3′′-
oxoh exyl]-2′(E)-[(tr im eth ylsilyl)m eth ylen e]cycloh exyl]-
h exa n oic Acid (28). Chromatography of the crude material
with 20% EtOAc/hexane provided the pure product 28 in 60%
yield. 1H NMR: δ 7.26 (d, 2H, J ) 8.5 Hz), 7.08 (d, 2H, J )
8.6 Hz), 5.34 (s, 1H), 2.54-2.68 (m, 2H), 2.32-2.48 (m, 4H),
2.08-2.14 (m, 1H), 1.74-1.92 (m, 6H), 1.42-1.72 (m, 2H),
1.12-1.40 (m, 9H), 0.92 (d, 3H, J ) 7.0 Hz), 0.85 (dd, 3H, J )
6.8, 6.8 Hz), 0.09 (s, 9H). IR: 3400-2329, 1712, 1703, 1598,
1491, 1454, 1406, 1365, 1290, 1246, 1216, 1091, 858, 834 cm-1
.
DCIMS (isobutane): m/z 505 (M + H), 487, 415, 369, 299, 281.
(2R,1′S,3′S,4′S)-2-[4′-Meth yl-3′-(6′′-ph en yl-3′′-oxoh eptyl)-
2′(E)-[(tr im eth ylsilyl)m eth ylen e]cycloh exyl]h exan oic Acid
(29). Chromatography of the crude material with 20% EtOAc/
hexane provided the pure product 29 in 66% yield. 1H NMR:
δ 7.16-7.28 (m, 5H), 5.34 (s, 1H), 2.62 (m, 3H), 2.28-2.52 (m,
4H), 2.12 (m, 1H), 1.82 (m, 4H), 1.61 (m, 6H), 1.12-1.15 (m,
8H), 0.93 (d, 3H, J ) 7.0 Hz), 0.92 (dd, 3H, J ) 6.9, 6.9 Hz),
0.10 (s, 9H). IR: 3400-2480, 1730, 1703, 1600, 1494, 1452,
1408, 1373, 1291, 1246, 1216, 1104, 962, 858, 835 cm-1
.
DCIMS (isobutane): m/z 485 (M + H), 467 (M - H2O), 395,
368, 349, 279, 261.
(2R,1′S,3′S,4′S)-2-[4′-Met h yl-3′-(5′′-ca r b et h oxy-3′′-oxo-
p e n t yl)-2′(E )-[(t r im e t h ylsilyl)m e t h yle n e ]cycloh e xyl]-
h exa n oic Acid (30). Chromatography of the crude material
with 20% EtOAc/hexane provided the pure product 30 in 40%
yield. 1H NMR: δ 5.34 (s, 1H), 4.12 (dd, 2H, J ) 7.1, 14.2
Hz), 2.44-2.72 (m, 4H), 2.40 (m, 3H), 2.14 (m, 3H), 1.82-1.92
(m, 4H), 1.12-1.64 (m, 13H), 0.93 (d, 3H, J ) 7.0 Hz), 0.89
(dd, 3H, J ) 6.9, 6.9 Hz), 0.10 (s, 9H). IR: 3400-2400, 1738,
1707, 1599, 1463, 1452, 1411, 1367, 1247, 1099, 1029, 858, 835
cm-1. DCIMS (isobutane): m/z 453 (M + H), 435, 336, 317,
229, 194.
Oct a h yd r o-3-et h yl-6-m et h yl-3,12-ep oxy-12H -p yr a n o-
[4,3-j]-1,2-ben zod ioxep in -10(3H)-on e (14). The crude prod-
uct was purified by PTLC eluting with 20% EtOAc/hexane to
give 14 (62%) as an oil. 1H NMR: δ 5.92 (s, 1H), 3.17 (dd,
1H, J ) 6.9, 18.2 Hz), 2.27-2.38 (m, 1H), 2.27 (dd, 1H, J )
1.2, 8.2 Hz), 1.94-2.08 (m, 2H), 1.84-1.94 (m, 1H), 1.68-1.80
(m, 4H), 1.38-1.52 (m, 4H), 0.99 (d, 3H, J ) 5.8 Hz), 0.95 (dd,
3H, J ) 7.5, 7.5 Hz). IR: 2978, 2920, 2862, 1744, 1460, 1208,
1031, 998, 895 cm-1. DCIMS (isobutane): m/z 283 (M + H),
282 (M+), 247.
Octa h ydr o-3-(4-p h en ylbu tyl)-6-m eth yl-3,12-epoxy-12H-
pyr an o[4,3-j]-1,2-ben zodioxepin -10(3H)-on e (21). The crude
product was purified by flash chromatography eluting with
20% EtOAc/hexane to give 21 (44%) as an oil. 1H NMR: δ
7.10-7.28 (m, 5H), 5.91 (s, 1H), 3.18 (dd, 1H, J ) 6.9, 18.2
Hz), 2.59 (m, 2H), 2.33-2.40 (m, 1H), 2.27 (d, 1H, J ) 18.2
Hz), 1.81-2.19 (m, 3H), 1.70-1.80 (m, 3H), 0.99 (d, 3H, J )
5.8 Hz). IR: 2959, 2917, 2857, 1744, 1210, 1031, 995 cm-1
.
FABMS: m/z 387 (M + H), 386, 341, 313, 154, 136, 117.
Oct a h yd r o-3-et h yl-6-m et h yl-9-b u t yl-3,12-ep oxy-12H -
pyr an o[4,3-j]-1,2-ben zodioxepin -10(3H)-on e (31). The crude
product was purified by silica gel PTLC eluting with 20%
hexane/CHCl3 to give 31 (26%) as the pure product. Crystal-
lization from hexane gave white crystals, mp 86-88 °C. 1H
NMR: δ 5.83 (s, 1H), 3.18 (m, 1H), 2.28-2.40 (m, 1H), 1.96-
2.08 (m, 3H), 1.68-1.82 (m, 6H), 1.20-1.48 (m, 8H), 1.12 (m,
1H), 1.00 (d, 3H, J ) 5.6 Hz), 0.93 (dd, 3H, J ) 7.5, 7.5 Hz),
0.88-0.92 (m, 3H). IR: 2954, 2928, 2869, 1741, 1461, 1392,
Octa h yd r o-3-p r op yl-6-m eth yl-3,12-ep oxy-12H-p yr a n o-
[4,3-j]-1,2-ben zod ioxep in -(3H)-on e (15). Flash chromatog-
raphy over silica gel with 20% EtOAc/hexane gave 15 (58%)
as an oil. 1H NMR: δ 5.91 (s, 1H), 3.17 (dd, 1H, J ) 6.9, 18.2
Hz), 2.33-2.40 (m, 1H), 2.27 (dd, 1H, J ) 1.1, 18.2 Hz), 1.94-
2.10 (m, 2H), 1.62-1.78 (m, 4H), 1.36-1.52 (m, 6H), 0.99 (d,
3H, J ) 6.9 Hz), 0.89 (dd, 3H, J ) 7.3, 7.3 Hz). IR: 2959,
1346, 1286, 1216, 1182, 1113, 1035, 1002, 946, 895 cm-1
.
DCIMS (NH3): m/z 356 (M + NH4), 339 (M + H), 323, 293,
275, 265.
Oct a h yd r o-3-b u t yl-6-m et h yl-9-b u t yl-3,12-ep oxy-12H -
pyr an o[4,3-j]-1,2-ben zodioxepin -10(3H)-on e (32). The crude
product was purified by silica gel PTLC eluting with 20%
hexane/CHCl3 to give pure 32 (29%) as an oil. 1H NMR: δ
5.84 (s, 1H), 3.20 (m, 1H), 2.30-2.40 (m, 1H), 1.96-2.08 (m,
3H), 1.64-1.84 (m, 6H), 1.20-1.48 (m, 12H), 1.12 (m, 1H), 1.00
(d, 3H, J ) 5.9 Hz), 0.85-0.95 (m, 6H). IR: 2953, 2926, 2870,
1747, 1456, 1392, 1344, 1270, 1178, 1112, 1033, 1004, 953
cm-1. FABMS: m/z 367 (M + H), 349 (M - H2O), 293, 221,
193.
2926, 2871, 1747, 1208, 1031, 994, 949 cm-1
. FABMS: m/z
297 (M + H), 289, 279, 251.
Oct a h yd r o-3-b u t yl-6-m et h yl-3,12-ep oxy-12H -p yr a n o-
[4,3-j]-1,2-ben zod ioxep in -10(3H)-on e (16). PTLC on silica
gel, eluting with 20% EtOAc/hexane, gave 16 (55%) as an oil.
1H NMR: δ 5.91 (s, 1H), 3.19 (dd, 1H, J ) 6.8, 18.2 Hz), 2.33-
2.40 (m, 1H), 2.28 (dd, 1H, J ) 1.1, 18.2 Hz), 1.94-2.10 (m,
2H), 1.82-1.94 (m, 1H), 1.64-1.78 (m, 4H), 1.24-1.48 (m, 6H),
0.99 (d, 3H, J ) 5.7 Hz), 0.88 (dd, 3H, J ) 7.1, 7.1 Hz). IR:
2954, 2937, 2914, 2872, 1743, 1458, 1210, 1106, 1032, 1002,
993 cm-1. FABMS: m/z 311 (M + H), 297, 265, 237, 165, 154,
137.
Octa h yd r o-3-(2-p h en yleth yl)-6-m eth yl-9-bu tyl-3,12-ep -
oxy-12H-pyr an o[4,3-j]-1,2-ben zodioxepin -10(3H)-on e (33).
The crude product was purified by silica gel PTLC eluting with

2906 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 15
Avery et al.
(9) Meshnick, S. R.; Thomas, A.; Ranz, A.; Xu, C. M.; Pan, H.
Artemisinin (Qinghaosu), the Role of Intracellular Hemin in its
Mechanism of Antimalarial Action. Mol. Biochem. Parasitol.
1991, 49, 181-190.
20% hexane/CHCl3 to give pure 33 (25%) as a white solid, mp
103-105 °C. 1H NMR: δ 7.22-7.36 (m, 5H), 5.92 (s, 1H), 3.30
(dddd, 1H, J ) 2.5, 5.0, 5.0, 8.0, Hz), 2.72-2.92 (m, 2H), 2.44-
2.58 (m, 1H), 2.02-2.22 (m, 5H), 1.82-1.96 (m, 3H), 1.34-
1.64 (m, 9H), 1.14 (m, 1H), 1.10 (d, 3H, J ) 5.7 Hz), 0.99 (dd,
3H, J ) 6.7, 6.7 Hz). IR: 2951, 2869, 1739, 1453, 1217, 1195,
1183, 1112, 1033, 1002, 898 cm-1. DCIMS (NH3): m/z 432 (M
+ NH4), 415 (M + H), 414 (M+), 397, 386, 369, 341, 105, 91.
Octa h yd r o-3-[2-(4-ch lor op h en yl)eth yl]-6-m eth yl-9-bu -
t yl-3,12-ep oxy-12H -p yr a n o[4,3-j]-1,2-b en zod ioxep in -10-
(3H)-on e (34). The crude product was purified by silica gel
PTLC eluting with 20% hexane/CHCl3 to give pure 34 (27%)
as an oil. 1H NMR: δ 7.24 (d, 2H, J ) 8.5 Hz), 7.08 (d, 2H, J
) 8.6 Hz), 5.8 (s, 1H), 3.14 (m, 1H), 2.54 (m, 2H), 2.28-2.38
(m, 1H), 1.92-2.08 (m, 2H), 1.62-1.84 (m, 6H), 1.22-1.44 (m,
9H), 1.02 (m, 2H), 0.98 (d, 3H, J ) 5.7 Hz), 0.90 (dd, 3H, J )
6.7, 6.7 Hz). IR: 2960, 2955, 2920, 2868, 1743, 1492, 1456,
(10) Zhang, Y.; Merali, S.; Meshnick, S. R. p-Aminobenzoic Acid
Transport by Normal and Plasmodium falciparum-infected
Erythrocytes. Biochem. Parasitol. 1992, 52, 185-194.
(11) Posner, G.; Oh, C. A. Regiospecifically Oxygen-18 Labeled
1,2,4-Trioxane: A Simple Chemical Model System to Probe the
Mechanism(s) for the Antimalarial Activity of Artemisinin
(Qinghaosu). J . Am. Chem. Soc. 1992, 114, 8328-8329.
(12) Asawamahasakda, W.; Ittarat, I.; Pu, Y.-M.; Ziffer, H.; Meshnick,
S. R. Alkylation of Parasite-specific Proteins by Endoperoxide
Antimalarials. Antimicrob. Agents Chemother. 1994, 38, 1854-
1858.
(13) Posner, G. H.; Oh, C. H.; Wang, D.; Gerena, L.; Milhous, W. K.;
Meshnick, S. R.; Asawamahasadka, W. Mechanism-Based De-
sign, Synthesis, and In Vitro Antimalarial Testing of New
4-Methylated Trioxanes Structurally Related to Artemisinin:
The Importance of a Carbon-centered Radical for Antimalarial
Activity. J . Med. Chem. 1994, 37, 1256-1258.
1405, 1272, 1217, 1189, 1112, 1035, 1003, 901, 833 cm-1
.
(14) J efford, C. W. Bicyclic Artemisinin Substitutes: Looking for the
Pharmacophore. Presented at the 43rd Annual Meeting of the
American Society of Tropical Medicine and Hygiene, Cincinnati,
OH, 1994.
FABMS: m/z 463 (M + H), 445, 417, 389, 309, 251, 221.
Octa h yd r o-3-(4-p h en ylbu tyl)-6-m eth yl-9-bu tyl-3,12-ep -
oxy-12H-pyr an o[4,3-j]-1,2-ben zodioxepin -10(3H)-on e (35).
The crude product was purified by silica gel PTLC eluting with
20% hexane/CHCl3 to give pure 35 (27%) as an oil. 1H NMR:
δ 7.14-7.26 (m, 5H), 5.85 (s, 1H), 3.28 (m, 1H), 2.60 (m, 2H),
2.32 (m, 1H), 1.94-2.08 (m, 3H), 1.65-2.84 (m, 4H), 1.56-
1.62 (m, 3H), 1.20-1.50 (m, 9H), 1.02-1.12 (m, 2H), 0.99 (d,
3H, J ) 5.7 Hz), 0.91 (dd, 3H, J ) 6.8, 6.8 Hz). IR: 2950,
2926, 2867, 1741, 1494, 1453, 1392, 1350, 1193, 1111, 1033,
999, 954 cm-1. FABMS: m/z 443 (M + H), 425 (M - H2O),
397, 369, 221, 193.
(15) Avery, M. A.; Fan, P.-C.; Karle, J .; Miller, B.; Goins, D. K.
Replacement of the Nonperoxidic Trioxane Oxygen Atom of
Artemisinin by Carbon: Total Synthesis of (+)-Carbaartemisinin
and Related Structures. Tetrahedron Lett. 1995, 36, 3965-3968.
(16) Wu, J .; J i, R. A Quantitative Hansch Structure-Activity Study
on Artemisinin Derivatives. Acta Pharm. Sin. 1982, 3, 55-60.
(17) Avery, M. A.; Gao, F.; Chong, W. K.; Mehrotra, S.; Milhous, W.
K. Structure-Activity Relationships of the Antimalarial Agent
Artemisinin. 1. Synthesis and Comparative Molecular Field
Analysis of C-9 Analogs of Artemisinin and 10-Deoxoartemisinin.
J . Med. Chem. 1993, 36, 4264-4275.
Octa h yd r o-3-(3-eth oxyp r op ion yl)-6-m eth yl-9-bu tyl-3,-
12-ep oxy-12H -p yr a n o[4,3-j]-1,2-b en zod ioxep in -10(3H )-
on e (36). The crude product was purified by silica gel PTLC
eluting with 20% hexane/CHCl3 to give pure 36 (21%) as an
oil. 1H NMR: δ 5.84 (s, 1H), 4.11 (dd, 2H, J ) 7.1, 14.2 Hz),
3.12 (m, 1H), 2.34-2.60 (m, 4H), 1.96-2.12 (m, 5H), 1.80 (m,
3H), 1.20-1.42 (m, 1H), 1.10 (m, 1H), 1.00 (d, 3H, J ) 5.9 Hz),
0.91 (dd, 3H, J ) 6.8, 6.8 Hz). IR: 2959, 2921, 2869, 2851,
1737, 1459, 1376, 1262, 1103, 1033, 895, 796 cm-1. DCIMS
(isobutane): m/z 411 (M + H), 395, 377, 378, 301, 221, 129.
(18) Avery, M. A.; Bonk, J . D.; Mehrotra, S.; Chong, W. K.; Miller,
R.; Milhous, W.; Goins, D. K.; Venkatesan, S.; Wyandt, C.; Khan,
I.; Avery, B. A. Structure-Activity Relationships of the Antima-
larial Agent Artemisinin. 2. Effect of Heteroatom Substitution
at O-11: Synthesis of N-Alkyl-11-Azaartemisinins. J . Med.
Chem. 1995, 38, 5038-5044.
(19) For an example of tricyclic artemisinin analogs substituted at
C-3, see: Posner, G. H.; Park, S. B.; Gonzalez, L.; Wang, D.;
Cummings, J . N.; Klinedinst, D.; Shapiro, T. A.; Bachi, M. D.
Evidence for the Importance of High-Valent FedO and of a
Diketone in the Molecular Mechanism of Action of Antimalarial
Trioxane Analogs of Artemisinin. J . Am. Chem. Soc. 1996, 118,
3537-3538.
(20) Avery, M. A.; Chong, W. K. M.; J ennings-White, C. Stereoselec-
tive Total Synthesis of (+)-Artemisinin, the Antimalarial Con-
stituent of Artemisia annua L. J . Am. Chem. Soc. 1992, 114,
974-979.
Ack n ow led gm en t. We wish to thank the U.S. Army
Medical Research and Development Command for sup-
port of this work under Contract No. DAMD17-91-C-
1099. This investigation also received financial support
from the UNDP/World Bank/WHO Special Programme
for Research and Training in Tropical Diseases (TDR).
(21) Bergbreiter, D. E.; Newcomb, M. Regioselectivity in Deprotona-
tion of Ketone Dimethylhydrazones with Lithium Amide Bases.
Tetrahedron Lett. 1979, 43, 4145-4148.
(22) Yamashita, M.; Matsumiya, K.; Tanabe, M.; Suemitsu, R. The
Facile Synthesis of Unsymmetrical Ketones Using Acetone
Dimethylhydrazone. Bull. Chem. Soc. J pn. 1985, 58, 407-408.
(23) Avery, M. A.; J ennings-White, C.; Chong, W. K. M. Synthesis of
a C,D-Ring Fragment of Artemisinin. J . Org. Chem. 1989, 54,
1789-1792.
(24) Avery, M. A.; J ennings-White, C.; Chong, W. K. M. Simplified
Analogues of the Antimalarial Artemisinin: Synthesis of 6,9-
Desmethylartemisinin. J . Org. Chem. 1989, 54, 1792-1795.
(25) Avery, M. A.; Chong, W. K. M.; J ennings-White, C. The Total
Synthesis of (+)-Artemisinin and (+)-9-Desmethylartemisinin.
Tetrahedron Lett. 1987, 28, 4629-4632.
(26) Still, W. C. Macromodel, v. 4.5; Columbia Innovation Enter-
prise: New York, 1994.
(27) Desjardins, R. E.; Canfield, C. J .; Haynes, D. E.; Chulay, J . D.
Quantitative Assessment of Antimalarial Activity In Vitro by a
Semiautomated Microdilution Technique. Antimicrob. Agents
Chemother. 1979, 16, 710-718.
(28) Milhous, W. K.; Weatherley, N. F.; Bowdre, J . H.; Desjardins,
R. E. In Vitro Activities of and Mechanisms of Resistance to
Antifolate Antimalarial Drugs. Antimicrob. Agents Chemother.
1985, 27, 525-530.
(29) Shukla, K. L.; Gund, T. M.; Meshnick, S. R. Molecular Modeling
Studies of the Artemisinin (Qinghaosu)-Hemin Interaction:
Docking Between the Antimalarial Agent and its Putative
Receptor. J . Mol. Graph. 1995, 13, 215-222.
Refer en ces
(1) Klayman, D. L. Qinghaosu (Artemisinin): an antimalarial drug
from China. Science 1985, 228, 1049-1055.
(2) Luo, X.-D.; Shen, C.-C. The chemistry, pharmacology, and clinical
applications of qinghaosu (artemisinin) and its derivatives. Med.
Res. Rev. 1987, 7, 29-52.
(3) Zhou, W.-S.; Xu, X.-X Total Synthesis of the Antimalarial
Sesquiterpene Peroxide Qinghaosu and Yingzhaosu A. Acc.
Chem. Res. 1994, 27, 211-216.
(4) Zaman, S. S.; Sharma, R. P. Some Aspects of the Chemistry and
Biological Activity of Artemisinin and Related Antimalarials.
Heterocycles 1991, 32, 1593-1638.
(5) J ung, M. Current Developments in the Chemistry of Artemisinin
and Related Compounds. Curr. Med. Chem. 1994, 1, 35-49.
(6) Avery, M. A.; Gao, F.; Mehrotra, S.; Chong, W. K.; J ennings-
White, C. The Organic and Medicinal Chemistry of Artemsinin
and Analogs; Research Trends Trivandrum: India, 1993; pp
413-468.
(7) Butler, A. R.; Wu, Y. L. Artemisinin (Qinghaosu): A New Type
of Antimalarial Drug. Chem. Soc. Rev. 1992, 21, 85-90.
(8) Meshnick, S. R.; Yang, Y.; Lima, V.; Kuypers, F.; Kamchon-
wongpaisan, S.; Yuthavong, Y. Iron-dependent Free Radical
Generation and the Antimalarial Artemisinin (Qinghaosu).
Antimicrob. Agents Chemother. 1993, 37, 1108-1114.
J M960200E