New chemical and biological aspects of artemisinin-Derived trioxane dimers

Article abstract of DOI:10.1016/S0968-0896(01)00270-X

Joining two 10-deoxoartemisinin trioxane units via a p-diacetylbenzene linker produces new C-10 non-acetal dimers 3b and 3c. ¹H spectroscopy allows unambiguous assignment of the stereochemistry at C-10 in these dimers. Successful replacement of

Full text of DOI:10.1016/S0968-0896(01)00270-X

Bioorganic & Medicinal Chemistry 10 (2002) 227–232
New Chemical and Biological Aspects of Artemisinin-Derived
Trioxane Dimers
Gary H. Posner,a,* John Northrop,a Ik-Hyeon Paik,a Kristina Borstnik,a Patrick Dolan,b
Thomas W. Kensler,b Suji Xiec and Theresa A. Shapiroc
aDepartment of Chemistry, School of Arts and Sciences, The Johns Hopkins University, Baltimore, MD 21218, USA
bDivision of Toxicological Sciences and The Environmental Health Sciences Center, Department of Environmental Health Sciences,
School of Hygiene and Public Health, The Johns Hopkins University, Baltimore, MD 21205, USA
cDivision of Clinical Pharmacology, Department of Medicine, School of Medicine, The Johns Hopkins University,
Baltimore, MD 21205, USA
Received 11 June 2001; accepted 23 July 2001
Abstract—Joining two 10-deoxoartemisinin trioxane units via a p-diacetylbenzene linker produces new C-10 non-acetal dimers 3b
1
and 3c. H NMR spectroscopy allows unambiguous assignment of the stereochemistry at C-10 in these dimers. Successful replace-
ment of both carbonyl oxygen atoms in these diketone dimers by fluorine atoms produces new tetrafluorinated dimers 5a and 5b.
Each dimer was evaluated in vitro for antimalarial, antiproliferative, and antitumor activities; ketone dimers 3b and 3c, more than
fluorinated dimers 5a and 5b, are promising for chemotherapy of both malaria and cancer. # 2001 Elsevier Science Ltd. All rights
reserved.
Introduction
Scheme 1, converting artemisinin (1) into dihy-
droartemisinin a-acetate (2)6 in high yield then allowed
mild tin tetrachloride-promoted double coupling with
the bisenol trimethylsilyl ether of p-diacetylbenzene to
produce not only previously reported5 bb-dimer 3a now
reproducibly and in improved 38% yield but also, for
the first time, ab-dimer 3b (21% yield) and also aa-
dimer 3c (1% yield). Separation of these diastereomers
was achieved chromatographically via preparative
Some 1,2,4-trioxane dimers have high in vitro anti-
malarial, antiproliferative, and antitumor activities,1ꢀ4
including in vivo anticancer activity.5 We report here
several significant chemical and biological advances in
this area: (1) a new protocol (Scheme 1) for semi-synth-
esis of C-10 non-acetal dimers 3 starting with the nat-
ural 1,2,4-trioxane antimalarial aretmisinin (1) and
proceeding via C-10 ester acetal 2; (2) biological eva-
luation of new ab-dimer 3b and new aa-dimer 3c; (3)
replacement of the two carbonyl groups in ketone
dimers 3 by pharmacologically beneficial fluorine atoms
without destruction of the crucial trioxane pharmaco-
phore unit; and (4) biological evaluation of tetra-
fluorinated trioxane dimers 5.
1
HPLC. C-10 stereochemistry was assigned by H NMR
spectroscopy as has been done previously in related C-
10 substituted trioxanes.7ꢀ9 In bb-dimer 3a and aa-
dimer 3c, the aromatic protons are equivalent, produ-
cing a singlet at d 8.0, whereas in ab-dimer 3b the aro-
matic protons are non-equivalent, producing an AB
quartet. Distinction between symmetrical bb-dimer 3a
and symmetrical aa-dimer 3c relies on several char-
acteristic peaks that are at lower field in the bb- vs aa-
dimer (e.g., d 5.32 vs 5.24; 5.08 vs 4.19; 2.79 vs 2.42; 0.96
vs 0.95; 0.90 vs 0.82). Also, the doublet of the AB
quartet at d 3.20 has a smaller coupling constant in the
bb-dimer 3a than in the aa-dimer 3c. Monomers 4 are
characterized by a methyl singlet at d 2.6. On analytical
TLC plates using 20% EtOAc/hexanes, the Rf values are
diacetylbenzene > monomers > dimers.
Results and Discussion
Synthesis of C-10 non-acetal dimer 3a was achieved
previously5 in variable yield (maximum 26%) from
artemether (the methyl ether of dihydroartemisinin)
using titanium tetrachloride as promoter. As shown in
Introduction of one or more fluorine atoms into a bio-
logically active compound often produces an analogue
*Corresponding author. Tel.: +1-410-516-4670; fax: +1-410-516-
8420; e-mail: ghp@jhunix.hcf.jhu.edu
0968-0896/02/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00270-X

228
G. H. Posner et al. / Bioorg. Med. Chem. 10 (2002) 227–232
having more desirable pharmacological properties.10,11
Fluorination of diketones 3 with diethylaminosulfur
trifluoride (DAST), however, was unsuccessful. Fluor-
ination at a higher temperature using the more robust,
commercially available fluorinating agent bis(2-meth-
oxyethyl)aminosulfur trifluoride (BAST) successfully
produced tetrafluorinated dimers 5a and 5b, albeit in
only modest yields. For comparison of the biological
activity of fluorinated dimers 5 with that of a fluori-
nated monomer, BAST fluorination of 10b-benzoyl-
methyl-10-deoxoartemisinin (6) was performed to
provide difluorinated monomer 7 [eq (1)]).12
than artemisinin (1), whereas tetrafluorinated dimers 5a
and 5b are 2–4 times less potent than artemisinin (1).
Also, both dicarbonyl dimers 3a and 3b are more potent
than trioxane monomers 4 and 6, and monomer 6 is
comparable in potency to its difluorinated version 7.
Table 1. Antimalarial activities in vitro
Compd
IC50 (nM)a
3a
3b
1.9
1.7
3c
4a
3.9
4.4
4b
5a
5b
6
3.0
28
15
5.2
5.1
7.6ꢁ1.4
7
Artemisinin
aAntimalarial activity was determined against the chloroquine-sensi-
tive NF54 strain of P. falciparum as reported previously.13 The stan-
dard deviation for each set of quadruplicates was an average of 11%
(ꢂ 58%) of the mean. R2 values for the fitted curves were ꢂ 0.982.
Artemisinin activity is meanꢁstandard deviation of concurrent con-
trol (n=9).
Although these trioxane dimers do not have enormously
increased antimalarial potency over that of the corre-
sponding trioxane monomers, these trioxane dimers
(but not the monomers) do have potent antiproliferative
properties. Antiproliferative activities, measured in vitro
using murine keratinocytes as described previously,14
are shown in Figure 1 for new non-fluorinated dimers
3b and 3c and for new fluorinated dimers 5a and 5b. It is
noteworthy that these dimers are more effective at 1 mM
concentration than calcitriol (1a,25-dihydroxyvitamin
D3), the hormonally active form of vitamin D3 that is
used clinically as a drug to treat psoriasis,15 a skin dis-
order characterized by uncontrolled cell proliferation.
Even at 0.1 mM concentrations, dimers 3b and 3c are
still more antiproliferative than calcitriol. The fluori-
nated dimers 5 are no more potent in this assay than the
non-fluorinated dimers 3.
To test whether a corresponding dimer lacking the two
peroxide units has any antiproliferative activity, triox-
ane dimer 3a was doubly deoxygenated to form bis-
deoxy dimer 3a0 [eq (2)]. Even at 1 mM concentration,
however, bisdeoxy dimer 3a0 was not antiproliferative
(data not shown). Dioxolanes (like this bisdeoxy com-
pound 3a0, antimalarial IC50=1030 nM) are known to
have weak or no antimalarial activity.16 The structural
requirement of two trioxane units in order for dimers
like 3 to have both high antimalarial and high anti-
proliferative activities does not necessarily mean, how-
ever, that the same biological mechanism is operative in
both cases. At this time, although the mechanistic
details underlying a trioxane’s antimalarial activity are
becoming clear,16,17 the biological mechanism of a
trioxane dimer0s antiproliferative action remains to be
clarified.
Scheme 1.
Using our standard assay,13 the in vitro antimalarial
potencies of non-fluorinated dimers 3 and of fluorinated
dimers 5 against chloroquine-sensitive Plasmodium fal-
ciparum (NF54) parasites were established (Table 1).
Dicarbonyl dimers 3a–3c are 2–5 times more potent

G. H. Posner et al. / Bioorg. Med. Chem. 10 (2002) 227–232
229
mmol) and the bisenol trimethylsilyl ether of p-diace-
tylbenzene (750 mg, 2.45 mmol) in CH2Cl2 (32 mL) then
cooled to ꢀ78 ꢃC (the optimal temperature for this
coupling reaction). SnCl4 (5.30 mL, 1.0 M in CH2Cl2)
was added dropwise over 20 min via gas-tight syringe
through a needle against which a piece of dry ice was
held during the entire 20 min. TLC immediately after
full addition of SnCl4 showed the complete consump-
tion of starting material and formation of a mixture of
products. The reaction was quenched with satd.
NaHCO3 solution, then let warm to room temperature.
The aqueous layer was extracted (3ꢄ) with CH2Cl2, the
organic layer was combined, dried over MgSO4, filtered
and concentrated. Further purification by column chro-
matography (30% ethyl acetate/hexane) afforded a
fraction containing an elimination product (147.0 mg,
0.552 mmol, 11%) and another fraction containing the
bb-dimer, ab-dimer, aa-dimer, b-monomer, and a-
monomer. This fraction was purified again by column
chromatography [(100% CH2Cl2 pack; 5% ethyl ace-
tate/CH2Cl2(monomer eluted); 15% ethyl acetate/
CH2Cl2(dimer eluted)]. Finally the isomers of the dimers
and monomers were separated by HPLC. Dimers: (silica
preparative column 20% ethyl acetate/hexanes) bb-
dimer, 3a (649 mg, 0.934 mmol, 38% Rt=39.1 min); ab-
dimer, 3b (361 mg, 0.520 mmol, 21% Rt=37.1 min); aa-
dimer, 3c (20.1 mg, 0.029 mmol, 1.2%, Rt=35.5 min).
Monomers: (silica preparative column 95:4:1 hexanes/
CH2Cl2/EtOH) b-monomer, 4a (288 mg, 0.670 mmol,
14%, Rt=62.2 min); a-monomer, 4b (61.7 mg, 0.140
mmol, 3%, Rt=55.8 min).
Growth inhibitory activities at nanomolar to micromolar
concentrations, measured in vitro as described previously
using a diverse panel of 60 human cancer cell lines in the
National Cancer Institute (NCI’s) Developmental and
Therapeutics Program,18 indicate that our non-fluori-
nated dimers 3b and 3c are particularly inhibitory to leu-
kemia cells, and these dimers are very selectively potent in
a few other cancer cell lines (e.g., colon 205, ovarian can-
cer OVCAR-4, non-small cell lung EKVX, Fig. 2); in
examining total growth inhibition (TGI), bars extending
to the right of the centerline indicate cells more sensitive
than average to a particular dimer, whereas bars to the left
indicate less sensitive cells. Disappointingly, the fluori-
nated dimers 5 are less active in these assays than the non-
fluorinated dimers 3. The highly selective and powerful
anticancer activities of dimers 3b and 3c, coupled with
their lack of cytotoxicity, make these promising lead
compounds for further preclinical study in dual action
chemotherapy of both malaria and cancer.19ꢀ21
Experimental
Synthesis of dimers 3 and monomers 4
ꢀꢁ-Dimer 3b. 1H NMR (400 MHz, CDCl3) d 8.04
(ABq, JAB=8.4 Hz, ÁnAB=30.9 Hz, 4H), 5.32 (s, 1H),
5.24 (s, 1H), 5.08 (m, 1H), 4.18 (m, 1H), 3.20 (dABq,
Jd=5.8 Hz, JAB=15.8 Hz, ÁnAB=153.8 Hz, 2H), 3.20
A flame dried 100 mL round bottom flask was charged
with the dihydroartemisinin acetate 2 (1.60 g, 4.90
Figure 1. Dose–response effect of analogues on keratinocyte proliferation (96 h).

230
G. H. Posner et al. / Bioorg. Med. Chem. 10 (2002) 227–232
Figure 2.

G. H. Posner et al. / Bioorg. Med. Chem. 10 (2002) 227–232
231
(dABq, Jd=6.6 Hz, JAB=16.0 Hz, ÁnAB=84.4 Hz,
2H), 2.79 (m, 1H), 2.42 (m, 1H), 2.32 (m, 2H), 2.2–1.4
(m, 20H), 1.33 (s, 3H), 1.30 (s, 3H), 0.97 (d, J=6.0 Hz
3H), 0.95 (d, J=6.8 Hz 3H), 0.89 (d, J=7.6 Hz 3H),
0.82 (d, J=7.2 Hz 3H); 13C NMR (100 MHz, CDCl3) d
198.8, 197.9, 140.9, 140.1, 129.0, 128.4, 104.2, 103.1,
92.0, 89.8, 81.0, 81.0, 71.7, 70.2, 52.2, 46.3, 44.2, 43.6,
40.7, 37.7, 37.5, 36.7, 36.5, 34.6, 34.3, 33.0, 30.1, 26.2,
26.0, 25.0, 24.9, 21.7, 20.5, 20.3, 14.2, 13.1; HRMS(ES)
m/z calcd for C40H55O10(M+H) 695.3795, found
695.3810; mp 124–127 ꢃC.
reaction mixture was diluted with 40 mL dichloro-
methane (EM Science, Gibbstown, NJ, USA) and
quenched with 40 mL ice-cold satd NaHCO3 solution.
The organic layer was separated and washed with 40
mL brine, dried over MgSO4, filtered and concentrated
to give the fluorinated product as a yellow oil.
Tetrafluorinated ꢁꢁ-dimer 5a. The crude mixture was
purified by column chromatography (silicagel, 14%
ethyl acetate/hexanes). Further purification by HPLC
(silica semi-preparative column, 12% ethyl acetate/hex-
anes, Rt=16.4 min) provided 5a (10.4 mg, 0.014 mmol,
28%) as a white solid: mp 148–150 ꢃC; 1H NMR
(400 MHz, CDCl3) d 7.57 (s, 4H), 5.26 (s, 2H), 4.49–4.45
(m, 2H), 2.73–2.68 (m, 2H), 2.57–2.44 (m, 2H), 2.35–
2.16 (m, 4H), 1.93–1.87 (m, 2H), 1.82–1.76 (m, 2H),
1.68–1.63 (m, 2H), 1.61–1.50 (m, 3H), 1.48–1.18 (m,
9H), 1.39 (s, 6H), 1.05–0.90 (m, 2H), 0.95 (d, J=6.4 Hz
6H), 0.83 (d, J=7.6 Hz 6H); 13C NMR (100 MHz,
CDCl3) d 138.5 (t, J=26.6 Hz), 125.4 (t, J=6.1 Hz),
122.2 (t, J=243 Hz), 103.3, 88.7, 80.7, 69.5, 52.3, 44.3,
39.0 (t, J=26.6 Hz), 37.5, 36.5, 34.4, 30.0, 26.0, 24.6,
24.6, 20.2, 13.1; 19F NMR (CDCl3, CFCl3 as a refer-
ence) d ꢀ90.0 (ddd, J=248, 16.9, 13.0 Hz) ꢀ95.4 (ddd,
J=248, 16.9, 16.9 Hz); IR (CHCl3) 2940, 2876, 1453,
1
ꢀꢀ-Dimer 3c. H NMR (400 MHz, CDCl3) d 8.06 (s,
4H), 5.24 (s, 2H), 4.19 (ddd J=11.2, 7.2, 4 Hz, 2H), 3.20
(dABq, Jd=5.8 Hz, JAB=15.8 Hz, ÁnAB=153.8 Hz,
4H), 2.42 (m, 2H), 2.32 (m, 2H), 2.02–1.96 (m, 2H),
1.90–1.87 (m, 2H), 1.75–1.70 (m, 4H), 1.59–1.43 (m,
4H), 1.35–1.20 (m, 3H), 1.34 (s, 6H), 1.10–1.00 (m, 3H),
0.98–0.83 (m, 2H), 0.95 (d, J=6.0 Hz 6H), 0.82 (d,
J=7.2 Hz 6H); 13C NMR (100 MHz, CDCl3) d 198.7,
140.6, 128.5, 104.0, 91.8, 80.8, 71.5, 52.0, 46.1, 43.4, 37.3,
36.2, 34.1, 32.8, 26.0, 24.7, 21.4, 20.3, 14.0; IR (CHCl3)
2926, 2874, 1685, 1455, 1378, 1265, 1206, 1132, 1090, 1061,
1044, 731 cmꢀ1; HRMS (ES) m/z calcd for C40H54NaO10
(M+Na) 717.3615, found 717.3605; mp 107–108ꢃC.
1409, 1377, 1322, 1096, 1060, 1013, 876, 846, 732 cmꢀ1
;
HRMS(E)S m/z calcd for C40H54F4NaO8 (M+Na)
761.3653, found 761.3643.
1
ꢁ-Monomer 4a. H NMR (400 MHz, CDCl3) d 8.02 (s,
4H), 5.32 (s, 1H), 5.08 (m, 1H), 3.20 (dABq, Jd=6.8 Hz,
JAB=16.0 Hz, ÁnAB=87.3 Hz, 2H), 2.78 (m, 1H), 2.63
(s, 3H), 2.29 (ddd,J=14.4, 13.4, 3.8 Hz, 1H), 2.02–1.96
(m, 1H), 1.95–1.88 (m, 1H), 1.85–1.79 (m, 1H), 1.74–
1.65 (m, 2H), 1.50–1.20 (m, 5H), 1.28 (s, 3H), 0.96 (d,
J=6.0 Hz 3H), 0.90 (d, J=7.2 Hz 3H); 13C NMR
(100 MHz, CDCl3) d 197.9, 197.7, 140.4, 140.2, 128.7,
103.1, 89.8, 81.0, 70.2, 52.2, 44.2, 40.8, 37.7, 36.7, 34.6,
30.1, 27.1, 25.9, 25.0, 20.3, 13.1; HRMS(ES) m/z calcd
for C25H33O6 (M+H) 429.2277, found 429.2286; mp
51-52 ꢃC.
Tetrafluorinated ꢀꢁ-dimer 5b. The crude mixture was
purified by column chromatography (silicagel, 14%
ethyl acetate/hexanes). Further purification by HPLC
(silica semi-preparative column, 10% ethyl acetate/hex-
anes, Rt=21.0 min) provided 5b (6.0 mg, 0.008 mmol,
16%) as a white solid: mp 146–148 ꢃC; 1H NMR
(400 MHz, CDCl3) d 7.57 (d, J=0.8 Hz 4H), 5.27 (s,
1H), 5.08 (s, 1H), 4.49–4.45 (m, 1H), 3.68–3.62 (m, 1H),
2.71–2.67 (m, 1H), 2.52–2.40 (m, 2H), 2.37–2.18 (m,
5H), 2.05–1.76 (m, 5H), 1.70–1.61 (m, 3H), 1.51–1.16
(m, 10H), 1.39 (s, 3H), 1.31 (s, 3H), 1.04–0.87 (m, 2H),
0.95 (d, J=6.0 Hz 3H), 0.93 (d, J=6.4 Hz, 3H), 0.81 (d,
J=7.6 Hz, 3H), 0.80 (d, J=7.2 Hz, 3H); 13C NMR
(100 MHz, CDCl3) d 146.6 (t, J=26.6 Hz), 140.4 (t,
J=26.6 Hz), 125.4 (t, J=6.1 Hz), 125.2 (t, J=6.1 Hz),
122.2 (t, J=243 Hz), 122.0 (t, J=243 Hz), 103.9, 103.3,
91,6, 88.8, 80.7, 80.5, 69.6, 69.5, 52.3, 51.9, 45.9, 44.3,
42.5 (t, J=26.6 Hz), 39.0 (t, J=26.6 Hz), 37.5, 37.3,
36.5, 36.2, 34.4, 34.1, 32.2, 30.0, 26.0, 25.9, 24.7, 24.7,
24.5, 21.3, 20.3, 20.2, 14.0, 13.1; 19F NMR (CDCl3,
CFCl3 as a reference) d ꢀ90.0 (ddd, J=252, 17.7, 14.5
Hz, 2F),-93.2 (ddd, J=247, 16.7, 11.3 Hz, 2F), ꢀ95.8 to
ꢀ-Monomer 4b. 1H NMR (400 MHz, CDCl3) d 8.04
(tABq, Jt=2.0 Hz, JAB=8.6 Hz, ÁnAB=32.7 Hz, 4H),
5.23 (s, 1H), 4.17 (m, 1H), 3.19 (dABq, Jd=5.8 Hz,
JAB=15.6 Hz, ÁnAB=162.40 Hz, 2H), 2.62 (s, 3H), 2.42
(m, 1H), 2.33 (ddd,J=14.4, 13.6, 4.0 Hz, 1H), 2.00–1.94
(m, 1H), 1.88–1.81 (m, 1H), 1.74–1.69 (m, 3H), 1.53–
1.40 (m, 3H), 1.31 (s, 3H), 1.32–1.19 (m, 2H), 0.94 (d,
J=6.0 Hz 3H), 0.82 (d, J=7.2 Hz 3H); 13C NMR
(100 MHz, CDCl3) d 198.9, 197.8, 141.1, 140.1, 129.0,
128.4, 104.2, 92.0, 80.9,71.8, 52.1, 46.3, 43.5, 37.5, 36.4,
34.3, 32.9, 27.1, 26.1, 24.9, 21.6, 20.5, 14.2; HRMS(ES)
m/z calcd for C25H33O6 (M+H) 429.2277, found
429.2294; mp 56-57 ꢃC.
ꢀ96.5 (m, 4F); HRMS(E) S
m/z calcd for
C40H54F4NaO8 (M+Na) 761.3653, found 761.3645.
Fluorination of dimers: general procedure
Difluorinated monomer 7
To a 7 mL Teflon vial (purchased from Berghof/Amer-
ica Inc., Coral Springs, FL, USA) the dicarbonyl dimer
(0.050 mmol) and BAST (bis[2-methoxyethyl]amino
sulfur trifluoride) (0.40 mL, 2.2 mmol, 43 equiv, pur-
chased from SynQuest Lab. Inc. (Alachua, FL, USA)
and distilled prior to use) were added and the vial
capped with a Teflon cap and sealed with Teflon tape.
The resulting mixture was stirred at 80 ꢃC for 24 h. The
BAST fluorination of ketone monomer 6 gave a crude
product mixture that was purified by column chroma-
tography (silicagel, 10% ethyl acetate/hexanes). Further
purification by HPLC (silica semi-preparative column,
5% ethyl acetate/hexanes, Rt=24.3 min) provided
difluorinated monomer 7 (4.5 mg, 0.011 mmol, 22%) as
a white solid: mp 134–135 ꢃC; 1H NMR (400 MHz,

232
G. H. Posner et al. / Bioorg. Med. Chem. 10 (2002) 227–232
CDCl3) d 7.53–7.51 (m, 2H), 7.42–7.41 (m, 3H), 4.50
(ddd, J=8.6, 6.0, 2.8 Hz, 1H), 2.75–2.66 (m, 1H), 2.56–
2.43 (m, 1H), 2.35–2.27 (m, 1H), 2.28–2.19 (m, 1H), 2.02
(ddd, J=14.4, 4.8, 2.8 Hz, 1H), 1.93–1.86 (m, 1H),
1.81–1.75 (m, 1H), 1.68–1.62 (m, 1H), 1.61–1.55 (m,
1H), 1.49–1.41 (m, 1H), 1.39 (s, 3H), 1.32–1.19 (m, 4H),
1.00–0.87 (m, 1H), 0.95 (d, J=6.0 Hz, 3H), 0.82 (d,
J=7.6 Hz, 3H); 13C NMR (100 MHz, CDCl3) d 137.1
(t, J=28.0 Hz), 129.6, 128.3, 125.2 (t, J=6.5 Hz), 122.4
(t, J=243 Hz), 103.3, 88.8, 80.7, 69.6 (t, J=6.0 Hz),
52.3, 44.4, 39.0 (t, J=27.0 Hz), 37.5, 36.5, 34.4, 30.0,
26.0, 24.7, 24.5, 20.2, 13.1; 19F NMR (CDCl3, CFCl3 as
a reference) d 93.2 (ddd, J=247, 18.0, 11.0 Hz), ꢀ95.8
(ddd, J=247, 17.6, 17.6 Hz); IR (CHCl3) 3010, 2921,
2973, 1490, 1377, 1321, 1260, 1098, 1056, 1016, 876,
financial support, the NCI0s Developmental and Ther-
apeutics program for evaluation of these dimers, the
Pfizer Pharmaceutical Co. for a summer research
fellowship to J.N., and Dr. Paul O0Neill (Liverpool,
UK) for helpful suggestions.
References and Notes
1. Beekman, A. C.; Barenstsen, A. R. W.; Woerdenbag, H. J.;
Van Uden, W.; Pras, N. J. Nat. Prod. 1997, 60, 325.
2. Galal, A. M.; Ahmad, M. S.; El-Feraly, F. S. J. Nat. Prod
1996, 59, 917.
3. Venugopalan, B.; Bapat, C. P.; Karnik, P. J.; Catterjee,
D. K.; Iyer, N.; Lepcha, D. J. Med. Chem. 1995, 38, 1922.
4. Posner, G. H.; Ploypradith, P.; Hapangama, W.; Wang, D.;
Cumming, J. N.; Klinedinst, D.; Shapiro, T. A.; Dolan, P.;
Kensler, T. W.; Daughenbaugh, R.; Murray, C. K.; Zheng,
Q. Y. Bioorg. Med. Chem. 1997, 5, 1257.
5. Posner, G. H.; Ploypradith, P.; Parker, M. H.; O0Dowd,
H.; Woo, S. H.; Northrop, J.; Krasavin, M.; Dolan, P.; Kens-
ler, T. W.; Xie, S.; Shapiro, T. A. J. Med. Chem. 1999, 42,
4275.
6. Ying, L.; Peilin, Y.; Yixin, C.; Liangquan, L.; Yuanzhu, G.;
Desheng, W.; Yaping, Z. Acta Pharm. Sinica 1981, 26, 429.
7. Ma, J.; Katz, E.; Ziffer, H. Tetrahedron Lett. 1999, 40,
8543.
8. Woo, S. H.; Parker, M. H.; Ploypradith, P.; Northrop, J.;
Posner, G. H. Tetrahedron Lett. 1998, 39, 1533.
9. Jung, M.; Lee, S. Heterocycles 1997, 45, 1055.
10. Banks, R. E.; Smart, B. E.; Tatlow, J. C. Organofluorine
Chemistry: Principles and Commercial Applications; Plenum:
New York, 1994.
11. Park, B. K.; Kitteringham, N. R.; O0Neill, P. M. Annu.
Rev. Pharm. Tox. 2001, 41, 443.
12. Related difluorinated trioxanes have recently been pre-
pared using tin tetrachloride-promoted coupling: Chorki, F.,
Crouse, B.; Bonnet-Delpon, D.; Begue, J.-P.; Brigand, T.;
Portella, C. Tetrahedron Lett. 2001, 42, 1487
13. For experimental details of the testing protocol, including
references to development of the assay, see: Posner, G. H;
´
Gonzalez, L.; Cumming, J. N.; Klinedinst, D.; Shapiro, T. A.
Tetrahedron 1997, 53, 37.
14. Posner, G. H.; Nelson, T. D.; Guyton, K. Z.; Kensler,
T. W. J. Med. Chem. 1992, 35, 3280.
15. Holick, M. F. Vitamin D: Molecular Biology, Physiology,
and Clinical Applications; Humana: Totowa, 1999.
16. Cumming, J. N.; Ploypradith, P.; Posner, G. H. Adv.
Pharmacol. 1997, 37, 253.
17. Posner, G. H., Cumming, J. N.; Krasvin, M. In Biomedi-
cal Chemistry: Applying Chemical Principles to the Under-
standing and Treatment of Disease. Torrence, P., Ed., John
Wiley, 2000, p 284.
802, 699 cmꢀ1
;
C23H30F2NaO4 (M+Na) 431.2004, found 431.2010.
HRMS(E) S
m/z calcd for
ꢁꢁ- Bisdeoxydimer 3a0. Immediately prior to use, Zn
dust was washed with 5% HCl (3ꢄ), de-ionized water
(3ꢄ), EtOH (3ꢄ), diethyl ether (3ꢄ), then dried in
vacuo and placed under Ar. A flamed dried 25 mL
round bottomed flask was charged with the bb-dimer 3a
(35.0 mg, 0.050 mmol), glacial acetic acid (15.0 mL, 26.0
mmol), and washed Zn dust (8.2 mg, 0.13 mmol). The
reaction was monitored by TLC and during 8 h addi-
tional Zn dust (25 mg, 0.40 mmol) was added. The
reaction was transferred with CHCl3 to a large flask
through a glass frit to remove the Zn dust. The solution
was slowly neutralized with satd NaHCO3 until evolu-
tion of gas ceased. The aqueous layer was extracted with
CHCl3, the organic layers combined, dried over MgSO4,
and concentrated. Purification by column chromato-
graphy (silicagel, 0–25% ethyl acetate/petroleum ether)
afforded a white powder that was a mixture of the bb-
bisdeoxydimer 3a0 and starting material 3a. Further
purification by HPLC (silica preparative column; 95:4:1
hexanes CH2Cl2 EtOH) provided bb-bisdeoxydimer 3a0
(28.1 mg, 0.042 mmol, 85%, Rt=18.7 min) as a white
solid: mp 185–186 ꢃC; H NMR (400 MHz, CDCl3) d
1
8.00 (s, 4H), 5.21 (s, 2H), 4.78 (q J=7.2 Hz, 2H), 3.14
(dABq, Jd=7.0 Hz, JAB=16.2 Hz, ÁnAB=41.8 Hz,
4H), 2.46 (m, 2H), 1.98 (m, 2H), 1.86–1.76 (m, 6H),
1.72–1.54 (m, 10H), 1.44 (s, 6H), 1.30–1.16 (m, 4H),
0.88 (d, J=7.2 Hz, 6H), 0.86 (d, J=9.2 Hz, 6H); 13C
NMR (100 MHz, CDCl3) d 198.5, 140.3, 128.5, 107.4,
97.2, 82.6, 65.9, 45.4, 41.8, 40.5, 35.8, 34.7, 34.6, 29.2,
25.4, 23.7, 22.4, 19.0, 12.9; IR (neat) 2952, 2874, 1688,
1384, 1208, 1097, 1000 cmꢀ1; HRMS(ES) m/z calcd for
C40H55O8 (M+H) 663.3897, found 663.3874.
18. Boyd, M. R.; Paull, K. D. Drug Dev. Res. 1995, 34, 91.
19. Lai, H.; Singh, N. P. Cancer Lett. 1995, 91, 41.
20. Singh, N.P.; Lai, H. Life Sciences, in press
21. Efferth, T.; Dunstan, H.; Sauerbrey, A.; Miyachi, H.;
Chitambar, C. R. Int. J. Oncol. 2001, 18, 767.
Acknowledgements
We thank the NIH (AI-34885, CA-44530, and RR-
00052) and the Burroughs Wellcome Fund for generous