Synthesis of 1,2,4-trioxepanes via application of thiol-olefin Co-oxygenation methodology

Article abstract of DOI:10.1016/j.bmcl.2006.08.098

Thiol-olefin co-oxygenation (TOCO) of substituted allylic alcohols generates β-hydroxy peroxides that can be condensed in situ with various ketones, to afford a series of functionalised 1,2,4-trioxepanes in good yields. Manipulation of the phenylsulfenyl

Full text of DOI:10.1016/j.bmcl.2006.08.098

Bioorganic & Medicinal Chemistry Letters 16 (2006) 6124–6130
Synthesis of 1,2,4-trioxepanes via application of thiol-olefin
Co-oxygenation methodology
Richard Amewu,a Andrew V. Stachulski,a Neil G. Berry,a Stephen A. Ward,b
Jill Davies,b Gael Labat,a Jean-Francois Rossignolc and Paul M. O’Neilla,*
aDepartment of Chemistry, University of Liverpool, PO Box 147, Liverpool L69 3BX, UK
bLiverpool School of Tropical Medicine, Pembroke Place, Liverpool L3 5QA, UK
cRomark Institute for Medical Research, 6200 Courtney Campbell Causeway, Suite 200, Tampa, FL 33607, USA
Received 7 August 2006; revised 24 August 2006; accepted 25 August 2006
Available online 15 September 2006
Abstract—Thiol-olefin co-oxygenation (TOCO) of substituted allylic alcohols generates b-hydroxy peroxides that can be condensed
in situ with various ketones, to afford a series of functionalised 1,2,4-trioxepanes in good yields. Manipulation of the phenylsulfenyl
group in 8a–8c allows for convenient modification to the spiro-trioxepane substituents. Surprisingly, and in contrast to the 1,2,4-
trioxanes examined, 1,2,4-trioxepanes are inactive as antimalarials up to 1000 nM and we rationalize this observation based on
the inherent stability of these systems to ferrous mediated degradation. FMO calculations clearly show that the r* orbital of the
peroxide moiety of 1,2,4-trioxane derivatives 4a and 14b are lower in energy and more accessible to attack by Fe(II) compared
to their trioxepane analogues 8b and 9b.
Ó 2006 Published by Elsevier Ltd.
Malaria is a preventable disease caused by Plasmodium
species the most lethal of which is Plasmodium falcipa-
rum. P. falciparum malaria has developed resistance to
the most widely used regimens such as chloroquine and
sulfadoxine/pyrimethamine.1 As a result of the spread
of multi-drug resistant Plasmodia we urgently require
novel antimalarial pharmacophores.2 In the early
1970s, Chinese chemists reported isolation and structure
elucidation of the sesquiterpene 1,2,4-trioxane artemisi-
nin (qinghaosu, 1), the highly active antimalarial compo-
nent of the ancient Artemisia annua (sweet wormwood)
Chinese herbal remedy for fevers.3 This important dis-
covery represented a breakthrough in finding an effective
antimalarial that was not quinoline-based. Sodium
artesunate (2) is a succinic acid half-ester of the reduced
lactol form of artemisinin (1) that, although prone to
hydrolysis, is fast-acting, water-soluble, effective and
widely used in areas of the world where malaria is endem-
ic. Few examples of resistance to such trioxanes have
been seen in the field or in the research laboratory. In
combination with other antimalarial drugs, sodium
artesunate (2) is rapidly becoming the drug of choice in
most third-world cases of malaria.4,5 The disadvantage
of all semi-synthetic compounds is that their production
requires 1 as starting material and currently the plant
yields of artemisinin remain relatively low.
H
H
O
O
O
O
O
H
O
H
H
H
O
O
O
R
Artemisinin
(qinghaosu, (1))
Sodium artesunate, 2
R = α-OC(O)CH2CH2CO2Na
OR'
R2
ArS
O O
O
R
O
O
R3
R1
Ph
4a: R1 = Me; R2 and
3a: R = CH2SPh R' =H
3b: R = CH=CHPh R' = Ac
R
3 = (CH2)5 ; Ar=p-Cl-Ph
4b: R1 = Ph; R2 and
R3 = (CH2)5
To address the supply issue, a number of groups have
attempted to produce totally synthetic peroxide ana-
logues, some of which demonstrate remarkable antima-
larial activity.6a During the course of our recent work on
the synthesis of new antimalarial endoperoxides, we
Keywords: Artemisinin; 1,2,4-trioxane; Endoperoxide; Malaria; Mech-
anism of action.
*
Corresponding author. E-mail: P.M.oneill01@liv.ac.uk
0960-894X/$ - see front matter Ó 2006 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2006.08.098

R. Amewu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6124–6130
6125
ArS
O2
ArS
OH
5b
ArS
OO
OH
a
OH
5a
6
Compound 8b
ArSH
ArS
OOH
OH
ArS
7
R1
R2
O
b
R1
R2
O
8a: R1 and R2 = (CH2)4; Ar = p-Cl-Ph, 72%
8b: R1 and R2 = (CH2)5; Ar = p-Cl-Ph, 76%
8c: R1 and R2 = Adamantylidene;
Ar = p-Cl-Ph, 82%
O
Compound 8c
ArS
O
8a-c
Scheme 1. Synthesis and crystal structures (ORTEP)7 of substituted 1,2,4-trioxepanes by the TOCO reaction. Reagents and conditions: (a) PhSH
(1.2 equiv), AIBN (0.07 equiv), O2 (excess), ht, 0 °C, CH3CN; (b) ketone, cat. tosic acid.
R1
R2
R1
R2
O
a
O
b
O
O
O
ArS
ArSO2
8a-c
O
O
c
9a: R1 and R2 = (CH2)4;
Ar = p-Cl-Ph, 81%
10a: R1 and R2 = (CH2)4;
Ar = p-Cl-Ph, 87%
9b: R1 and R2 = (CH2)5;
Ar = p-Cl-Ph, 82%
10b: R1 and R2 = (CH2)5;
Ar = p-Cl-Ph, 83%
9c: R1 and R2 = Adamantylidene;
Ar = p-Cl-Ph, 72%
10c: R1 and R2 = Adamantylidene;
Ar = p-Cl-Ph, 89%
R1
R2
O
O
O
H
O
11a: R1 and R2 = (CH2)4;
Ar = p-Cl-Ph, 87%
11b: R1 and R2 = (CH2)5;
Ar = p-Cl-Ph, 83%
d
e
11c: R1 and R2 = Adamantylidene;
Ar = p-Cl-Ph, 89%
R1
R2
R1
R2
O
O
O
O
O
O
MeO
N
O
O
12a: R1 and R2 = (CH2)4;
Ar = p-Cl-Ph, 35%
13a, 13b: R1 and R2 = (CH2)4;
Ar = p-Cl-Ph, 20%1:3 Z/E
12b: R1 and R2 = (CH2)5;
Ar = p-Cl-Ph, 62%
13c, 13d: R1 and R2 = (CH2)5;
Ar = p-Cl-Ph, 25% 1:2 Z/E
12c: R1 and R2 = Adamantylidene;
Ar = p-Cl-Ph, 70%
13e, 13f: R1 and R2 = Adamantylidene;
Ar = p-Cl-Ph, 18%1:3
13a, 13c,13d- Z series
13b, 13d, 13f - E series
Scheme 2. Synthesis of substituted 1,2,4-trioxepanes by functional group manipulation of the phenylsulfenyl group. Reagents and conditions: (a)
m-CPBA (2.2 equiv), CH2Cl2, room temperature, 24 h; (b) m-CPBA (1.0 equiv), CH2Cl2, room temperature, 6 h; (c) 2,6-lutidine (4.2 equiv),
trifluoroacetic acid anhydride (3.8 equiv), acetonitrile, room temperature; (d) Ph3P@CHCO2Me (1.1 equiv), CH2Cl2, room temperature, 3 h; (e)
Ph3P@CHCONEt2 (1.1 equiv), CHCl3/H2O (1:1 v/v), NaOH (1.5 equiv), room temperature, 3 h.
utilized a thiol-olefin co-oxygenation (TOCO) reaction
to generate bicyclic peroxides (3a) and endoperoxide
cysteine protease pro-drugs (3b)6b structurally related
to yingzhaosu A. By replacement of the terpene with an
allylic alcohol we have recently described the one-pot syn-
thesis of some simplified 1,2,4-trioxane analogues (4a)

6126
R. Amewu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6124–6130
O
Table 1. In vitro antimalarial activity versus the 3D7 strain of
Plasmodium falciparum14,15
N
O
O
11c
a
H
O
O
O
O
R2
R2
R1
O
O
O
a
O
R1
O
O
R3
R3
O
12d
14a-14d
9a-9d
O
collapse
Compound R1
R2 and R3
IC50 (nM)
N
O
9a
9b
p-Cl–Ph–SO2–CH2– (CH2)4
p-Cl–Ph–SO2–CH2– (CH2)5
p-Cl–Ph–SO2–CH2– Adamantylidine >1000
>1000
>1000
9c16a
13
12c16b
14a6c,17
14b6c,17
14c6c,17
14d6c,17
Artemisinin
–CH@CHCO2CH3– Adamantylidine >1000
Scheme 3. Attempted synthesis of trioxane 12d by reductive amina-
tion. Reagents and conditions: (a) aldehyde 11c (1 equiv), morpholine
(1.3 equiv) NaBH(OAc)3 (1.3 equiv), CH2Cl2, 18 h, room temperature.
p-Cl–Ph–SO2–CH2– (CH2)4
p-Cl–Ph–SO2–CH2– (CH2)5
99.8
136.9
188.7
110.5
12.6
p-Cl–Ph–SO2–CH2– Adamantylidine
p-Cl–Ph–S–CH2–
(CH2)4
and (4b).6c In this communication, we report on the
TOCO mediated synthesis of the 1,2,4-trioxepane phar-
macophore, iron catalysed decomposition studies and
preliminary in vitro antimalarial assessment.
Parasites were maintained in continuous culture according to the
method of Trager and Jensen.14 IC50 values were measured according
to the methods described by Desjardins.15
The synthesis6d of target 1,2,4-trioxepanes 8a–8c in-
volves the in situ generation of a phenylthiol radical
(AIBN/hv) which attacks the double bond of the homo-
allylic alcohol 5a in a Markonikov fashion. The tertiary
radical that is generated is trapped with molecular oxy-
gen to form the peroxy radical 6; radical hydrogen
abstraction produces the a-hydroxyperoxide 7 and thi-
ophenyl radical that continues the cycle. After consump-
tion of the alcohol 5a catalytic amounts of tosic acid and
the requisite ketone are added to enable 1,2,4-trioxepane
formation. In the free radical component of this chemis-
try high dilution is essential to prevent competitive for-
mation of side products (Scheme 1).
Previous studies with bicyclic endoperoxides and 1,2,4-
trioxanes have revealed that endoperoxide sulfones
2
1
Cl
S
O O
O
Fe(II) to O1
Fe(II) to O2
4a
Fe(III)
Fe(III)
.
.
Cl
S
O O
Cl
S
O O
O
O
15a
16a
β–scission
β–scission
(III)Fe
S
Fe(III)
O
O
O
Cl
S
.
+
O
.
O
O
Cl
15c
15b 18%
-Fe(III)
TEMPO
-Fe(II)
16b
OH
O
+
O
+
O
O
H
H
O
N
S
3%
68%
Cl
16c
+
O
OH
O
Br
S
5.6%
Cl
16d
Scheme 4. Ferrous mediated degradation of trioxane 14e and TEMPO spin-trapping of primary carbon-centred radical 16b. Reagents and
conditions: (a) Trioxane 14e (1 equiv), FeBr2 (1 equiv) TEMPO (1.3 equiv), THF, 24 h, room temperature.

R. Amewu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6124–6130
6127
display potent activity both in vitro and in vivo.8–11 The
sulfides were converted into the corresponding sulfones
using excess amount of m-chloroperbenzoic acid in
dichloromethane in excellent yields. The presence of
the sulfenyl group within the trioxepane skeleton also
provided us with the opportunity to prepare the alde-
hydes 11a–11c from the corresponding sulfides by the
Pummerer reaction. The sulfides 8a–8c were converted
to the corresponding sulfoxides 10a–10c using stoichi-
ometric amount of mCPBA in dichloromethane. The
two diastereomers of the intermediate sulfoxides formed
could be separated by column chromatography but they
were used in situ for the Pummerer reaction as shown in
Scheme 2. The aldehydes could then be converted to
vinyl esters and amides by Wittig chemistry with the
appropriate ylide. The rationale for the synthesis of
12a–13c is based on the observation by Singh et al.12
that several vinyl ester 1,2,4-trioxane derivatives have
excellent in vivo activity profiles. For esters 12a–12c,
only the E-configured esters were produced [as evi-
4a was consumed during the 24 h reaction period. The
mechanistic pathway depicted in Scheme 4 can rational-
ize the formation of isolated iron degradation products;
the major product of this reaction was the TEMPO spin-
trapped adduct 16c that is produced by association of
ferrous iron with O1 to form the oxy radical intermedi-
ate that fragments by b-scission to produce the primary
carbon-centred radical. This radical species is intercept-
ed by TEMPO to produce the adduct 16c;18 in addition,
small quantities of alkyl bromide 16d are also produced.
The alternative pathway proceeds through the forma-
tion of the alternative oxyl radical species 15a by associ-
ation of O2 with ferrous iron (Scheme 4). Fragmentation
produces ketone 18 and cyclohexanone (from the car-
bon-centred radical species 15c). Our results are consis-
tent with the recent iron degradation studies on
cyclohexyl functionalized 1,2,4-trioxanes where both
products of the O1 and O2 pathways were observed.19
The reaction of the corresponding 1,2,4-trioxepane 8b
under the same conditions led to poor turnover of
substrate (<15%) (Scheme 5). No products of the O2
denced by the large vinylic coupling constant (JH–H
=
16.2 Hz)]. For the vinyl amides 13a–13c, a mixture of
products was obtained with isomer ratios varying from
1:2 to 1:3 Z/E. The isomers could be readily separated
by flash column chromatography.
Cl
2
O
1
Attempts to enhance water solubility of the trioxepanes
by either reductive amination to produce 12d or
oxidation of the aldehyde 11c were unsuccessful. In
the former case we observed decomposition of the
trioxepane ring system to 2-adamantanone. For sub-
strate 11c the major product of the reaction was 13,
the reductive amination product of 2-adamantanone
and morpholine (Scheme 3).
O
S
O
8b
Fe(II) to O1
Less than 15% turnover
Fe (III)
.
O
O
S
Cl
O
Selected 1,2,4-trioxepanes depicted in Scheme 2 were
subjected to in vitro antimalarial assessment versus the
3D7 strain of P. falciparum according to a published
procedure and the data are recorded in Table 1. For
comparison several 1,2,4-trioxanes were also included
in the screen. Remarkably, all of the 1,2,4-trioxepanes
synthesized were inactive as antimalarials up to a con-
centration of 1000 nM. This is stark contrast to the cor-
responding spiro 1,2,4-trioxanes where activities as low
as 99.8 nM were recorded.
17a
β–scission
(III) Fe
S
O
.
O
O
Cl
TEMPO
Recent studies in the Dussault group have described the
use of the 1,2,4-trioxepanes as a carbonyl protecting
group due to the exceptional stability of this ring system
under a range of different reaction conditions.13a Since
the interaction of endoperoxide antimalarials with iron
is key to their biological mechanism of action13b we rea-
soned that the poor activity of this series may be down
to inherent lack of reactivity and enhanced stability of
the 1,2,4-trioxepane ring compared with the 1,2,4-triox-
ane heterocycle. Thus, we decided to the compare the
ferrous mediated degradation of a selected 1,2,4-triox-
ane with the corresponding 1,2,4-trioxepane.
OH
+
O
N
S
O
O
Cl
11%
17b
OH
O
S
Br
O
Cl
17c
2%
Exposure of 1,2,4-trioxane 4a to 1 equivalent of ferrous
bromide in the presence of the spin-trapping agent
TEMPO produced several products that were character-
ized by standard techniques (Scheme 4). Notably, all of
Scheme 5. Ferrous mediated degradation of trioxepane 8b and
TEMPO spin-trapping. Reagents and conditions: (a) Trioxepane 8b
(1 equiv), FeBr2 (1 equiv) TEMPO (1.3 equiv), THF, 24 h, room
temperature.

6128
R. Amewu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6124–6130
Figure 1. Low energy conformations of 1,2,4-trioxane 4a and 1,2,4-trioxepane 8b with the r* orbital mapped onto the electron density molecular
surface.
pathway were observed; only products of the O1 path-
way were detected with 11% of the spin-trapped adduct
17b constituting the major product of the reaction.
Acknowledgments
The authors thank the BBSRC (SAW, PON Grants BB/
C006321/1 and 26/B13581) and Romark (AVS, NB,
RA) for generous funding of this work.
In order to rationalize this difference in reactivity molec-
ular modelling studies were performed. A conformation-
al search using a Monte-Carlo method using the
MMFF94 forcefield20 was performed on molecules 4a,
14b and 8b, 9b. Each conformer generated was subjected
to a single point energy calculation at a semi-empirical
level using PM3 parameters and the energy of the r*
orbital of the peroxide bond was calculated. The Boltz-
mann weighted average energy of the orbital was com-
pared for the trioxane and trioxepane molecular pairs
4a/14b and 8b/9b. The Boltzmann weighted average of
the exposed surface area of the oxygen atoms in the per-
oxide bond was also calculated for each compound in
order to assess the accessibility of the peroxide to attack
by Fe(II). Interestingly, the energy of the r* orbital of
the peroxide bond for the trioxane compounds was
markedly lower than that of the corresponding trioxe-
panes for both the sulfide (ꢀ0.44 kcal/mol difference)
and sulfone (ꢀ3 kcal/mol difference). Figure 1 displays
a low energy conformation of 4a and 8b with the r*
orbital mapped onto the molecular electron density sur-
face. It is noteworthy that the accessibility to r* for the
trioxane would appear to be much greater than that of
r* of the trioxepane. Additionally, the trioxane mole-
References and notes
1. Krishna, S.; Uhlemann, A. C. Curr. Top. Microbiol.
Immunol. 2006, 295, 39.
2. Biagini, G. A.; O’Neill, P. M.; Nzila, A.; Ward, S. A.;
Bray, P. G. Trends Parasitol. 2003, 19, 479.
3. Klayman, D. L. Science 1985, 228, 1049.
4. Barradell, L. B.; Fitton, A. Drugs 1995, 50, 714.
5. White, N. J.; Olliaro, P. L. Parasitol. Today 1996, 12, 399.
6. (a) O’Neill, P. M.; Posner, G. H. J. Med. Chem. 2004, 47,
2945; (b) O’Neill, P. M.; Stocks, P. A.; Pugh, M. D.;
Araujo, N. C.; Korshin, E. E.; Bickley, J. F.; Ward, S. A.;
Bray, P. G.; Pasini, E.; Davies, J.; Verissimo, E.; Bachi, M.
D. Angew. Chem., Int. Ed. 2004, 43, 4193; (c) O’Neill, P.
M.; Mukhtar, A.; Ward, S. A.; Bickley, J. F.; Davies, J.;
Bachi, M. D.; O’Neill, P. M. Org. Lett. 2004, 6, 3035; (d)
Although we encountered no difficulties in working with
these 1,2,4-trioxepanes, routine precautions such as the
use of shields, fume hoods and avoidance of metal salts
should be observed whenever possible.
7. Crystallographic data (excluding structure factors) for the
structures 8b and 8c have been deposited with the
Cambridge Crystallographic Data Center (CCDC) as
supplementary publication numbers CCDC616381 and
CCDC616382. Copies of the data can be obtained, free of
charge, on application to CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (fax: +44 1223 336033 or
e-mail: deposit@ccdc.cam.ca.uk).
8. Posner, G. H.; O’Dowd, H.; Caferro, T.; Cumming, J. N.;
Ploypradith, P.; Xie, S. J.; Shapiro, T. A. Tetrahedron
Lett. 1998, 39, 2273.
9. Bachi, M. D.; Korshin, E. E.; Ploypradith, P.; Cumming,
J. N.; Xie, S. J.; Shapiro, T. A.; Posner, G. H. Bioorg.
Med. Chem. Lett. 1998, 8, 903.
10. Bachi, M. D.; Korshin, E. E.; Hoos, R.; Szpilman, A. M.;
Ploypradith, P.; Xie, S. J.; Shapiro, T. A.; Posner, G. H.
J. Med. Chem. 2003, 46, 2516.
11. O’Neill, P. M.; Verissimo, E.; Ward, S. A.; Davies, J.;
Korshin, E. E.; Araujo, N.; Pugh, M. D.; Cristiano, M. L.
S.; Stocks, P. A.; Bachi, M. D. Bioorg. Med. Chem. Lett.
2006, 16, 2991.
2
˚
cules have a larger exposed surface area (ꢀ18 A ) of
the peroxide oxygen atoms compared to the correspond-
2
˚
ing trioxepanes (ꢀ16 A ). Thus, the two factors of the
accessibility and energy of the r* orbital of the peroxide
bond could account for the surprisingly low biological
activity observed and very poor turnover in the spin-
trapping experiments of the 1,2,4-trioxepane com-
pounds compared to the 1,2,4-trioxanes.
In summary, thiol-olefin co-oxygenation (TOCO) of
substituted allylic alcohols generates b-hydroxy perox-
ides that can be condensed in situ with various ketones,
to afford a series of functionalised 1,2,4-trioxepanes in
good yields. Surprisingly, endoperoxides in this class
are inactive up to 1000 nM and we rationalize this
observation based on the inherent stability of these sys-
tems to ferrous mediated degradation.21 FMO calcula-
tions support the degradation studies in the sense that
the r* orbital of the peroxide bridge in 1,2,4-trioxanes
is lower in energy and more accessible to attack by
Fe(II) compared to their trioxepane analogues.
12. Singh, C.; Malik, H.; Puri, S. K. J. Med. Chem. 2006, 49,
2794–2803.
13. (a) Ahmed, A.; Dussault, P. H. Org. Lett. 2004, 6, 3609; In
collaboration with S. Krishna, we have recently reported
on studies implicating PfATP6 as a potential target for the

R. Amewu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6124–6130
6129
endoperoxide class of drug. The paper supports the idea of
a non-haem iron mediated mechanism of bioactivation for
the artemisinins see; (b) Eckstein-Ludwig, U.; Webb, R. J.;
Van Goethem, I. D.; East, J. M.; Lee, A. G.; Kimura, M.;
O’Neill, P. M.; Bray, P. G.; Ward, S. A.; Krishna, S.
Nature 2003, 424, 957.
calculated for 449.1165/451.1136, C21H27NO4NaS35Cl/
C21H27NO4NaS37Cl. Found: 449.1169/451.1152, respec-
tively.; (b) Preparation of 12c; to a solution of the
sulfoxide 10c (1.17 g, 2.9 mmol) at 0 °C in CH3CN
(12 ml), 2,6-lutidine (1.30 g, 12.3 mmol) and trifluoro
acetic anhydride (TFAA) (2.40 g, 11.2 mmol), in CH3CN
(12 ml) were added. The mixture was stirred at room
temperature for 3 h and extracted with ethyl acetate. The
organic layer was dried in MgSO4 and the solvent
removed under reduced pressure. Purification by column
14. Trager, W.; Jensen, J. B. Science 1976, 193, 673.
15. Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay,
J. D. Antimicrob. Agents Chemother. 1979, 16, 710.
16. (a) Procedure for the synthesis of compounds 8c and 9c; A
2-necked 500 ml round-bottomed flask was charged with a
solution of 3-methyl-but-3-en-1-ol (0.5 g, 5.8 mmol) and
AIBN (77.5 mg, 4.72 mmol) in acetonitrile (115 ml). The
reaction vessel was flushed with oxygen for several minutes
at 0 °C then stoppered and kept under a positive pressure
of pure oxygen, with the aid of two oxygen balloons. The
reaction mixture was vigorously stirred and UV irradiated
at 0 °C using an externally mounted 100 W BLACK-RAY
UV lamp at a distance of 5–7 cm, with the simultaneous
addition of 4-chlorothiophenol (1250 mg, 8.64 mol) solu-
tion in acetonitrile (32 ml) over a period of 30 min. After
completion of the addition, the reaction was left to
continue stirring at 0 °C, for 4–6 h or until consumption
of starting materials (monitored by tlc). The reaction
vessel was then allowed to warm to À10 °C, flushed with
nitrogen and a solution of 2-adamantanone (2.61 mg,
17.35 mmol) in dichloromethane (32 ml) was added fol-
lowed by catalytic amount of tosic acid (25 mg). The
mixture was left stirring at À10 °C and allowed to cool
slowly to room temperature overnight. The solvent was
removed by rotary evaporation and column chromatog-
raphy on the crude mixture gave the product 8c in 80% as
a colourless solid; mp 62–64 °C; IR; Vmax (CHCl3)/
cmÀ11011.2, 1090.2, 1112.2, 1450.1, 1472.0, 2840.6,
2901.3, 2980.3; 1H NMR (400 MHz, CDCl3): d 1.20 (s,
3H, CH3), 1.60 (m, 6H, adamantylidene), 1.76 (m, 3H,
adamantylidene), 1.94 (m, 5H, adamantylidene), 2.20 (s,
1H, CH2), 2.40 (s, 1H, CH2), 3.20 (d, 1H, J = 13.18 Hz,
SCH2), 3.45 (d, 1H, J = 13.18 Hz, SCH2), 3.65–3.85 (m,
2H, OCH2), 7.20 (d, 2H, J = 8.46 Hz, Ar), 7.35 (d, 2H,
J = 8.31 Hz, Ar); 13C NMR (100 MHz, CDCl3): d 134.15,
130.14, 128.89, 127.07, 109.64, 106.38, 70.32, 58.15, 56.39,
51.74, 40.13, 40.05, 35.59, 32.65, 32.19, 32.15, 31.73, 31.36
25.36, 21.80, 20.78; MS (ES+) [M+Na]+ (100), 417/419,
[2M+Na]+ 811/814, HRMS calculated for 417.1267
C21H27O3NaSCl. Found: 417.1280 (Caution): since
vapours of organic solvents may form explosive mixtures
with oxygen in closed systems, all such reactions should be
conducted behind safety shields. A solution of 8c (0.43 g,
1.1 mmol) and m-CPBA (0.56 g, 3.3 mmol) in CH2Cl2
(17 ml) was stirred for 4–6 h at room temperature. After
consumption of the more polar intermediate (monitored
by tlc), the mixture was poured into a saturated solution of
5% K2CO3 solution. The mixture was then extracted with
dichloromethane, the organic layer separated, dried over
MgSO4 and evaporated. Purification of the residue by
column chromatography gave the desired sulfone 9c
compound in 72% yield; mp 100–102 °C; IR;
Vmax(CHCl3)/cmÀ1821.4, 912.3, 1010.8, 1090.3, 1113.1,
1143.4, 1272.2, 1317.6, 1374.4, 1442.6, 1472.9, 1579.0,
2847.8, 2908.4, 2999.3; 1H NMR (400 MHz, CDCl3): d
1.30–2.00 (m, 14H, adamantylidene), 1.55 (s, 3H, CH3),
2.10 (s, 1H, CH2), 2.25 (m, 1H, CH2), 3.45 (d, 1H,
J = 14.66 Hz, SO2CH2), 3.75 (m, 2H, OCH2), 3.82 (d, 1H,
J = 14.66 Hz, SO2CH2), 7.50 (d, 2H, J = 8.57 Hz, Ar), 7.95
(d, 2H, J = 8.55 Hz, Ar); 13C NMR (100 MHz, CDCl3): d
23.92, 27.63, 35.32, 37.59, 44.21, 58.19, 61.93, 81.73,
108.87, 129.60, 130.40, 139.77, 140.56. MS (ES+)
[M+Na]+ (100), 449/451, [2M+Na]+ (<5%) 875 HRMS
1
chromatography gave the product 11c in 89%; H NMR
(400 MHz, CDCl3): d 1.14 (s, 3H, CH3), 1.53–1.86 (m, 4H,
adamantyl), 1.90–3.13 (m, 10H, adamantyl), 2.55 (br s,
2H, CH2), 2.74 (t, 1H, J = 7.31 Hz, OCH2), 3.11 (t, 1H,
J = 7.16 Hz, OCH2), 9.57 (s, 1H, CHO); (100 MHz,
CDCl3): d 27.86, 36.71, 39.64, 43.30, 47.37, 129.51,
131.29, 218.57. To a solution of the aldehyde 11c (0.37 g,
1.4 mmol) in CH2Cl2(12 ml) was added Ph3P@CHCO2Me
(0.5 g, 1.5 mmol) at room temperature and the solution
was allowed to stir at this temperature for 3 h. The
reaction mixture was concentrated and chromatographed
on a silica gel to give the desired product 12c in 70% yield
as a colourless oil; IRVmax (neat) cmÀ1 1108.7, 1161.3,
1
1319.3, 1446.6, 1653.3, 1722.2, 2854.6, 2919.5; H NMR
(400MHz, CDCl3): d 1.28 (s, 3H, CH3), 1.53–1.75 (m, 6H,
adamantylidene), 1.80 (br s, 3H, adamantylidene), 1.86–
2.20 (m, 5H, adamantylidine), 2.36 (s, 1H, CH2), 2.44 (s,
1H, CH2), 3.60–4.00 (m, 2H, CH2O), 3.79 (s, 2H, OCH3),
5.98 (d, 1H, CH, J = 16.21 Hz), 7.20 (d, 1H, J = 16.19 Hz,
CH); 13C NMR (100 MHz, CDCl3): d 26.11, 27.97, 34.33,
38.24, 43.10, 52.51, 59.16, 84.07, 109.35, 120.63, 152.62,
167.72. MS (ES+) [M+Na]+ (100) 345, HRMS calculated
for 345.1678 C18H26NO5Na. Found: 345.1675.
17. All additional new compounds in Table 1 provided
satisfactory 1H and 13C NMR and elemental analysis
data. Details can be found in: O’ Neill, P.M.; Amewu, R.;
Mukhtar, A.; Ward, S.A.; Publication number
WO2006016903; PCT/US2005/012236.
18. To a solution of 4a (0.3 g, 0.91 mmol) in THF (15 ml)
ferrous bromide (0.40 g, 1.82 mmol) and TEMPO (0.3 g,
1.82 mmol) were added and the reaction mixture was
allowed to stir at ambient temperature under nitrogen
atmosphere for more than 16 h. Following rotary evapo-
ration of THF the crude product was dissolved in ethyl
acetate, washed with water and brine, dried over MgSO4,
filtered and concentrated. The crude product was purified
by flash chromatography to afford 16c as the major
1
product in 58%; H NMR (400 MHz, CDCl3): d1.09(br s,
6H, CH3), 1.14 (br s, 6H, CH3), 1.23–1.27 (m, 2H, CH2),
1.29 (s, 3H, CH3), 1.35–1.57 (m, 8H, CH2), 1.59–1.70 (m,
2H, CH2), 2.31 (t, 2H, J = 7.4 Hz, COCH2), 3.07 (d, 1H,
J = 13.47 Hz, SCH2), 3.17 (d, 1H, J = 13.48 Hz, SCH2),
3.72 (t, 2H, J = 6.45 Hz, CH2O), 4.02 (d, 1H,
J = 11.39 Hz, CH2O), 4.09 (d, 1H, J = 11.20 Hz, CH2O),
7.24 (d, 2H, J = 8.54 Hz, Ar), 7.34 (d, 2H, J = 8.54 Hz,
Ar); 13C NMR (100 MHz, CDCl3): d 17.57, 24.53, 25.40,
26.51, 28.81, 32.75, 34.50, 40.05, 44.75, 69.74, 72.32, 76.90,
129.58, 131.71, 133.05, 135.58, 173.80 MS (ES+), [M+H]+
(100) 486.1 and [M+Na]+ 508.2, HRMS calculated for
486.2445 C25H41O4NSCl. Found: 86.2459.The minor
fraction was identified as 16d MS (ES+), (100) [M+Na]+
431/433/435. HRMS calculated for 431.0058/433.0039/
435.0009
C16H22O3NS35ClBr/
C16H22O3NS35ClBr/
C16H22O3NS37ClBr. Found: 431.0053/433.0013/435.0000.
19. Tang, Y. Q.; Dong, Y. X.; Wang, X. F.; Sriraghavan, K.;
Wood, J. K.; Vennerstrom, J. L. J. Org. Chem. 2005, 70,
5103.
20. Spartan’04, Wavefunction Inc, Irvine, CA, <http://
www.wavefun.com//>.

6130
R. Amewu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6124–6130
21. For studies on the iron mediated degradation of
diastereomers of 10-(2-hydroxy-1-naphthyl)-deo-
related to the ease with which the peroxide bond is
cleaved by iron. For detailed studies to the contrary,
see Haynes, R. K.; Ho, W. Y.; Chan, H.-W.;
Fugmann, B.; Stetter, J.; Croft, S. L.; Vivas, L.; Peters,
W.; Robinson, B. L. Angew. Chem., Int. Ed. 2004, 43,
1381.
xoqinghaosu, (-deoxoartemisinin) see Wang, D.-Y.;
Wu, Y.; Wu, Y.-L.; Li, Y.; Shan, F. J. Chem. Soc.,
Perkin Trans. 1 1999, 1827; this paper suggests that the
antimalarial potency of a given trioxane is directly