Synthesis of novel 1-(2,3-dihydro-5H-4,1-benzoxathiepin-3-yl)-uracil and -thymine, and their corresponding S-oxidized derivatives

Article abstract of DOI:10.1016/j.tet.2005.07.065

On the basis of molecular variations on isosteric replacements from the prototype 1-(2,3-dihydro-5H-1,4-benzodioxepin-3-yl)-5-fluorouracil a series of 3-(2,3-dihydro-5H-4,1-benzoxathiepin-3-yl)-uracil or -thymine O,N-acetals was prepared. The nature of th

Full text of DOI:10.1016/j.tet.2005.07.065

Tetrahedron 61 (2005) 10363–10369
Synthesis of novel 1-(2,3-dihydro-5H-4,1-benzoxathiepin-3-yl)-
uracil and -thymine, and their corresponding S-oxidized derivatives
a
M. del Carmen Nunez, Antonio Entrena, Fernando Rodrıguez-Serrano, Juan A. Marchal,
a
b
c
´
Antonia Aranega, Miguel A. Gallo, Antonio Espinosa and Joaquın M. Campos
´
˜
d
a
a
a,
´
*
´
´
a
´ ´ ´
Departamento de Quımica Farmaceutica y Organica, Facultad de Farmacia, c/ Campus de Cartuja s/n, 18071 Granada, Spain
b
´
Departamento de Biologıa Celular, Facultad de Ciencias, Avda. Severo Ochoa s/n, 18071 Granada, Spain
Departamento de Ciencias de la Salud, Facultad de Ciencias Experimentales y de la Salud, Paraje de las Lagunillas s/n, 23071 Jaen, Spain
c
´
d
´
Departamento de Ciencias Morfologicas, Facultad de Medicina, Avenida de Madrid s/n, 18071 Granada, Spain
Received 18 May 2005; revised 12 July 2005; accepted 14 July 2005
Available online 26 August 2005
Abstract—On the basis of molecular variations on isosteric replacements from the prototype 1-(2,3-dihydro-5H-1,4-benzodioxepin-3-yl)-5-
fluorouracil a series of 3-(2,3-dihydro-5H-4,1-benzoxathiepin-3-yl)-uracil or -thymine O,N-acetals was prepared. The nature of the cis- and
trans-sulfoxide isomers was established by means of their conformational analyses carried out with Sybyl and after comparing the theoretical
results with the 1H NMR responses of the target molecules. (RS)-3-(1,1-Dioxo-2,3-dihydro-5H-4,1-benzoxathiepin-3-yl)thymine and (1S*,
3S*)-1-(1-oxo-3,5-dihydro-2H-4,1-benzoxathiepin-3-yl)thymine were found to be inhibitors of the MCF-7 cell growth.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
thymine for 6, 8 and 10) and on the other, the oxygen atom
at position 1 of the seven-membered cycle is replaced by its
isosteric sulfur atom (5 and 6), and its oxidized states, that is
sulfoxide (7 and 8) and sulfones (9 and 10). Should these
compounds show antiproliferative activities, new avenues
of anticancer research would be opened based on the non-
toxic natural bases uracil and thymine and, very probably,
with a mechanism of action different from that of 5-FU.
O,N-Acetalic benzoxathiepins are compounds not referred
to in bibliography, and with this paper we will try to fill the
gap.
1-(2,3-Dihydro-5H-1,4-benzodioxepin-3-yl)-5-fluorouracils
(1–3) have proved to be good antiproliferative agents
against the MCF-7 human breast cancer cell line (Fig. 1).1
Such compounds accumulate the cancerous cells in the G0/G1
phase. On the other hand, 42 that shows an IC50Z5.4 mM
against the MCF-7 cell line, acts in a similar way to
Ftorafur, a known prodrug of 5-fluorouracil (5-FU) (Fig. 1),
because both gather together the cancerous cells in the
synthesis phase (S). In contrast to 5-FU,3 the benzannelated
5-FU O,N-acetals1 and corresponding open analogues2 have
proved to be non-toxic.
2. Results and discussion
2.1. Synthesis
With all this background it seems that compounds 1–3 are
drugs per se, whilst 4 is a prodrug of 5-FU. Therefore,
following our ongoing Anticancer Drug Programme we
have planned in this paper the synthesis of compounds 5–10
(Fig. 2) to be tested subsequently against the human breast
cancer cell line MCF-7.
The synthesis of the targets is depicted in Scheme 1. We
have previously reported the preparation of the cyclic acetal
(RS)-3-methoxy-2,3-dihydro-5H-1,4-benzodioxepin.1 We
applied the same procedure for the synthesis of 12, starting
from the commercially available o-(hydroxymethyl)thio-
phenol, that is selective alkylation using bromoacetaldehyde
dimethyl acetal (1 mol in the presence of 1 mol of sodium
hydride in dry dimethylformamide (DMF) under conditions
published by Crombie et al.4 gave 11 (79%). Subsequent
cyclization of the resulting hydroxyacetal using catalytic
amounts of p-toluenesulfonic acid in dry toluene for 3 h
On the one hand, all these compounds bear a natural
pyrimidine base (uracil for compounds 5, 7 and 9 and
Keywords: Antitumour compounds; Benzoxathiepins; Computer-assisted
methods; Uracils.
*
Corresponding author. Tel.: C34 958 243849; fax: C34 958 243845;
e-mail: jmcampos@ugr.es
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.07.065

´
M. del Carmen Nun˜ez et al. / Tetrahedron 61 (2005) 10363–10369
10364
Figure 1.
oxidation of sulfide 5 (and 6) with potassium peroxymono-
sulfate (OXONEe) under previously reported conditions7
yielded 9 (and 10) with 66 (and 55%) yields.
2.2. Spectroscopic analysis of 5
The 1H and 13C NMR spectra were recorded using DMSO-
d6 as solvent. Herein we will mention only the NMR
spectroscopic data that explain the conformational pre-
ference for 5. The hydrogen atoms of the seven-membered
framework of 5 appear between d 5.98 ppm and d 3.18 ppm.
The chemical shift found for the aminalic hydrogen H-3 is d
5.98 ppm with a double doublet multiplicity (dd), and the
respective couplings to the vicinal hydrogen atoms (H-2) are
9.1 and 2.2 Hz. In fact, the two H-2 atoms appear as two
dd’s with a JgemZ14.1 Hz and the two coupling constants
observed at d 5.98 ppm (9.6 and 1.8 Hz). Finally, each of the
two benzylic H-5 atoms resonates as two doublets at d 5.02
and d 4.96 ppm with a JgemZ13.6 Hz, showing a
diastereotopic character.
Figure 2. Novel benzo-fused seven-membered sulfur-containing O,N-
acetals with natural pyrimidine nucleobases. In all the cases the aminalic
bond between the C-3 atom of the seven-membered moiety and the
nucleobase is through its N-1 atom.
produced the sulfur-containing acetal 12 (95%). The last
step was the condensation reaction between the seven-
membered acetal 12 and the natural pyrimidine bases uracil
and thymine. For this procedure we used a 1.0 M solution of
tin(IV) chloride in dichloromethane, trimethylchlorosilane
(TCS) and 1,1,1,3,3,3-hexamethyldisilazane (HMDS) in dry
acetonitrile, and the experimental conditions were
optimised (unpublished results) by warming the reaction
mixture up to 45 8C for 20–24 h. Under these conditions 5
and 6 were obtained with 73 and 58% yields, respectively.
1
The most important consequence of the H NMR study is
that it unequivocally proves through the three following data
that the linkage between the seven-membered moiety and
uracil is through N-10:
The oxidation of sulfides is one of the most important and
straightforward methods for the preparation of sulfones,
which are utilized for biologically active substances and
substrates for the synthesis of drugs.5 Sulfoxides 7 and 8
were obtained using the procedure used by Matteucci et al.6
who used catalytic scandium trifluoromethansulfonate that
greatly increases the efficiency of hydrogen peroxide
mediated monooxidation of alkyl–aryl sulfides and methyl
cysteine-containing peptides. When applied to our sulfides a
mixture of cis-7 and trans-7 (77%) was obtained, but after
using the flash 40 equipment a fast-moving spot was
isolated, that will be characterized as cis-7 (vide infra).
When the oxidation of 6 was carried out without catalyst,
trans-8 was the only product produced (50%). Finally, the
(a) At d 5.60 ppm one hydrogen resonates as a dd (JZ8.0
and 2.2 Hz), the former coupling constant being typical of
an aromatic hydrogen atom, and this resonance could be
8
assigned to H-50 of the uracil base. H-60 appears as a
doublet (7.69 ppm, d, JZ8.0 Hz) downfield in relation to
H-50. The coupling pattern for H-50 is a dd due to its
coupling with H-60 (JZ8.0 Hz) and a small one (JH5 –NH
2.2 Hz) through a 1–3 bond relationship, and a doublet for
Z
0
30
H-60. Nevertheless, had the aminalic bond been established
0
with N–H3 the coupling pattern would have been more
Scheme 1. Synthetic route for the preparation of the target molecules. Reagents and conditions: (a) BrCH2CH(OMe)2, NaH, anhydrous DMF; (b) p-TsOH,
anhydrous toluene, 3 h; (c) pyrimidine base, HMDS, TCS, SnCl4, MeCN, 45 8C; (d) H2O2, Sc(OTf)3, CH2Cl2/10% EtOH; (e) OXONEe, MeOH, H2O.

´
M. del Carmen Nun˜ez et al. / Tetrahedron 61 (2005) 10363–10369
10365
complicated due to the simultaneous coupling of H-60 with
7 and the conformational analysis on both isomers was
carried out. The whole process was accomplished as
previously reported by us.2
H-50 and N–H1 (a dd for eac0h of the three hydrogen atoms
0
0
0
involved, that is, N–H1 , H-5 and H-6 ).
(b) H-3 exhibits connectivity through three bonds to C-60
(HMBC9).
Figure 4 shows the four most stable conformations of the
cis-7 isomer and Figure 5 shows the four most stable
conformations of the trans-7 isomer. In both isomers the
energetic differences between the different conformations
are important. These energy differences are sufficient to
consider the upper left-handed conformation of cis-7 in
Figure 4, the only one existing in solution at 25 8C, its
conformational population being calculated as 94.1%,
considering a Boltzman distribution. For the trans-7, the
most stable conformation represents 94.6% at 25 8C of the
total population [upper left conformation in Figure 5 (trans-
7)]. Therefore, it can be stated that the most stable
conformations of both cis-7 and trans-7 are practically the
only ones in the conformational equilibrium in solution, and
the molecule properties could be explained from these
conformers.
(c) The NOEDIFF effect observed between H-60 and H-2b
justifies the conformational preference of the uracil moiety
in which H-60 is near in space to H-2b. Accordingly, the
carbonyl group C-20 is far from the referred hydro0gen atom
and this obviously implies the linkage through N-1 . Had the
aminalic bond been established through C-3 and N-30, no
NOEDIFF effect would have been detected between some
hydrogen atoms of the seven-membered moiety and those of
the uracil fragment. H-3 shows a NOEDIFF effect with the
more deshielded H-5 atom (d 5.02 ppm), and with the more
shielded H-2 one (d 3.18 ppm) and these facts strongly
support a chair conformation of the seven-membered
moiety in which the aminalic hydrogen atom adopts an
axial disposition (as H-5 at d 5.02 ppm). Hence, the uracil
moiety occupies an equatorial position (Fig. 3).
Figure 6 displays the most stable conformations of cis-7 and
trans-7. In both cases the uracil moiety adopts an equatorial
disposition but the most interesting interactions are those
established between the sulfinyl group (S]O) with the
hydrogen atoms of the seven-membered fragment.
2.3. Conformational analysis of cis-7 and trans-7, and cis-
8 and trans-8
Compounds 7 and 8 can both exist as geometric isomers and
accordingly, the question is immediately posed as to which
are the cis- or the trans- ones. In the case of 7, during the
oxidation reaction of the sulfur atom, both isomers were
obtained (in a ratio of 3:1) and the pure minor compound
was isolated by flash chromatography. In the case of the
mixture, the 1H signals did not overlap and this fact allowed
us the assignment of all the protons for both isomers.
Nevertheless, the following question still remains to be
answered: Which is the cis- and which is the trans- isomer?
In the former case (cis-7) the equatorial S]O group bisects
both H-2 atoms, whilst in trans-7 the axial S]O group is
antiperiplanar in relation to H-2a and accordingly,
synperiplanar in relation to H-2b (see Fig. 6). Therefore,
the H-2 atoms of both isomers must show a completely
different behaviour in NMR. Moreover, the S]O group
interacts with the axial H-3 atom and with the axial H-5
atom trans-7. Nevertheless such interactions are not
observed in cis-7 and, therefore, such interactions must
result in different NMR responses of the hydrogen atoms of
the seven-membered moieties of both isomers. In fact, the
experimental data confirmed the conformational analysis
data in the following way: in trans-7, the differences in
chemical shifts of H-2 are higher than those of cis-7
(3.62 ppm for H-2a and 3.30 ppm for H-2b for trans-7,
whilst the chemical shifts are 3.47 ppm for H-2a and
3.34 ppm for H-2b for cis-7). The same holds true for H-5 of
both isomers (4.84 ppm for H-5a and 5.61 ppm for H-5b for
trans-7, being 4.67 ppm for H-5a and 4.83 ppm for H-5b for
cis-7). Figure 6 shows that the distance between the sulfinyl
In our case the seven-membered sulfoxides are scarcely
reported in the literature and hence the support from
bibliographic sources is null. To solve the identification of
both isomers we decided to tackle the conformational
analysis of the target molecules, trying to clarify the
conformational differences between both isomers and their
1
possible responses in their H NMR spectra.
The conformational study of cis-7 and trans-7 was carried
out in two well differentiated phases. In the first one, the
study of the base ring, without the uracil moiety, was
tackled to identify the most stable conformation of this
compound. In the second one, starting from the less
energetic conformations of this compound, the uracil
fragment was added to give rise to the cis-7 and the trans-
˚
group and the H-5a group (2.44 A) adequately explains the
chemical shift differences between H-5a and H-5b for
trans-7. Moreover, H-3 is more deshielded in trans-7 (d
6.65 ppm) than in cis-7 (d 6.40 ppm). The whole H NMR
1
spectroscopic data are collected in the Section 4.
The conformational analysis of cis-8 and trans-8 was nearly
identical to that of cis-7 and trans-7, with only a slight
variation in the energy of the several conformations (data
not shown).
We have assigned the isomerism of the unique product in
the reaction between 6 and H2O2 as being trans-8, based on
the difference between the chemical shift of its H-2 atoms
(0.36 ppm), which is closer to the 0.32 ppm difference
Figure 3. Preferred conformation of 5 showing the NOEDIFF effects by the
double arrows.

´
M. del Carmen Nun˜ez et al. / Tetrahedron 61 (2005) 10363–10369
10366
Figure 4. Representation of the four most stable conformations of compound cis-7. The relative energy values were calculated by means of molecular
mechanics (tripos).
shown by trans-7 and very different from the 0.13 ppm
value of cis-7.
The two antiproliferative compounds are 10 (IC50Z12.74G
4.79 mM) and trans-8 (IC50Z30.05G0.71 mM). The rest of
the compounds are inactive (IC50O100 mM). Therefore, it
has to be pointed out that 10 and trans-8 show an interesting
antiproliferative activity with the natural base thymine in its
structure. This is an outstanding fact that has not being
previously reported in scientific bibliography, and this
compound (or a related one) may serve as a prototype for the
development of even more potent structures, probably
endowed with a new mechanism of action. At present we are
2.4. Antiproliferative activities
The MCF-7 human breast cancer cell line had been used as
an excellent experimental model to improve the efficacy of
different therapies before its use in patients.10,11 Compounds
5–10 were assayed for their in vitro antiproliferative activity
against the MCF-7 cell line.
Figure 5. Representation of the four most stable conformations of compound trans-7. The relative energy values were calculated by means of molecular
mechanics (tripos).

´
M. del Carmen Nun˜ez et al. / Tetrahedron 61 (2005) 10363–10369
10367
˚
Figure 6. The most stable conformations of cis-7 (left) and trans-7 (right). The distance between the sulfinyl group and the H-5b group (2.44 A) is shown for
the trans-7 conformer.
studying several antiproliferative markers in order to solve
the mechanism of action at the molecular level of such a
benzannelated seven-membered O,N-acetal.
of NaH in mineral oil and 40 mL of anhydrous DMF, under
argon and at 0 8C; the resulting solution was left at rt until it
became clear (45 min). To this clear solution was added
dropwise 1.65 mL (14 mL) of bromoacetaldehyde dimethyl
acetal and the resulting solution was left overnight at rt and
then evaporated to dryness under vacuum. The residue was
treated with water (20 mL), then extracted with dichloro-
methane (4!30 mL) and treated with brine: the combined
organic phases were then dried (Na2SO4), filtered and
evaporated under reduced pressure to yield a colourless oil,
which after flash chromatography (EtOAc/hexane 1:4) gave
pure 11 (2.9 g, 92% yield) as a clear dense liquid. The
procedure previously reported used bromoacetaldehyde
diethyl acetal4 instead of bromoacetaldehyde dimethyl
acetal. 1H NMR (300 MHz, CDCl3) d 7.49 (m, 1H,
H-arom); 7.38 (m, 1H, H-arom); 7.25 (m, 2H, H-arom);
4.77 (s, 2H, CH2OH); 4.44 (t, JZ5.6 Hz, 1H, CH(OMe)2);
3.26 (s, 6H, OMe); 3.07 (d, JZ5.6 Hz, 2H, SCH2). 13C
NMR (75 MHz) d 142.33 (C-1); 134.18 (C-2); 132.17,
129.12, 128.58, 127.62 (C-arom); 102.71 (CH(OMe)2);
63.85 (CH2OH); 53.04 (OMe); 37.09 (SCH2). Anal. Calcd
for C11H16O3S: C, 57.87; H, 7.06; S, 14.05. Found: C,
57.90; H, 6.94; S, 14.26.
3. Conclusions
Six final O,N-acetals have been obtained and screened for
their anticancer activity. The modifications carried out
affect two parts of the molecule: (a) on the one hand, the
pyrimidine bases are always natural ones (uracil and
thymine), and on the other (b) they affect the nature of the
heteroatom of the seven-membered moiety that is linked
directly to the benzene ring (sulfur, and several oxidation
states of the sulfur atom: sulfinyl and sulfonyl). The
NOEDIFF data for (RS)-3-(2,3-dihydro-5H-4,1-benzoxa-
thiepin-3-yl)uracil 5 are compatible with a chair confor-
mation for the seven-membered ring and an equatorial
orientation of the uracil moiety on the C-3 position. The
nature of the cis- and trans-sulfoxide isomers of 7 and 8 has
been established by means of their conformational analyses
and after comparing the theoretical results with the 1H NMR
responses of the target molecules. (RS)-3-(1,1-Dioxo-2,3-
dihydro-5H-4,1-benzoxathiepin-3-yl)thymine 10 and
(1S*,3S*)-1-(1-oxo-2,3-dihydro-5H-4,1-benzoxathiepin-3-
yl)thymine (trans-8) were found to be inhibitors of the
MCF-7 cell growth. To our knowledge, to date there have
been no reports on nucleobase O,N-acetals with antitumour
activities. At present, studies are being carried out to
determine the mechanism of action at the molecular level of
this compound.
4.1.1.2. (RS)-3-Methoxy-2,3-dihydro-5H-4,1-benzox-
athiepin (12). A mixture of 1.90 g (8.33 mmol) of 11,
75 mg (0.436 mmol) of p-toluenesulfonic acid and 100 mL
of anhydrous toluene was heated to reflux (110 8C) during
2 h; the solution was then brought back to rt and evaporated
to dryness. The residue was taken up in diethyl ether
(50 mL) and the resulting solution was dried (K2CO3): it
was then filtered and evaporated to dryness, yielding an oil,
which upon flash chromatography (EtOAc/hexane 1:2) gave
1.55 g of pure 12 (95% yield) as a clear oil. 1H NMR
(300 MHz, CDCl3) d 7.51 (m, 1H, H-arom); 7.23 (m, 3H,
H-arom); 5.22 (d, JZ13.5 Hz, 1H, H-5); 4.85 (t, JZ4.1 Hz,
1H, H-3); 4.50 (d, JZ13.5 Hz, 1H, H-5); 3.52 (s, 3H, OMe).
13C NMR (75 MHz) d 142.26 (C-9a); 136.61, 131.78,
129.59, 128.09, 127.58 (C-arom); 102.74 (C-3); 66.01
(C-5); 55.60 (OMe); 37.93 (C-2). HR LSIMS (NOBA
matrix) calcd for C10H12O2S: 196.0558; found: 196.0558.
Anal. Calcd for C10H12O2S: C, 61.20; H, 6.16; S, 16.34.
Found: C, 61.56; H, 6.15; S, 16.68.
4. Experimental
4.1. Chemistry
The general methods were the same as those previously
described.12,13
4.1.1. Starting materials.
4.1.1.1. o-Hydroxymethylphenylthioacetaldehyde
dimethyl acetal (11). 2-Mercaptobenzyl alcohol
(1.62 mL, 14 mmol) was added dropwise to a stirred
mixture of 547 mg (14 mmol) of a 60% (w/w) suspension

´
M. del Carmen Nun˜ez et al. / Tetrahedron 61 (2005) 10363–10369
10368
4.1.2. Final compounds.
was left during 3.5 h more. The mixture was diluted with
CH2Cl2 (20 mL) and washed with H2O (2!15 mL), the
organic layer was dried (Na2SO4), concentrated and purified
by flash chromatography using a mixture of CH2Cl2/MeOH
9.5:0.5 to yield 164 mg, which were characterized as a
mixture of cis-7 and trans-7 (164 mg, 77%). 130 mg of this
mixture were purified by flash 40 chromatography using a
gradient elution (EtOAc/hexane 3:2/EtOAc/hexane
4.1.2.1. (RS)-1-(2,3-Dihydro-5H-4,1-benzoxathiepin-
3-yl)uracil (5). A 1.0 M solution of SnCl4/CH2Cl2
(2.45 mL, 2.44 mmol) was added dropwise with stirring
under argon to a suspension of 12 (400 mg, 2.04 mmol),
uracil (250 mg, 2.23 mmol), which contained TCS
(0.26 mL, 2.04 mmol) and HMDS (0.43 mL, 2.04 mmol)
in dry acetonitrile (10 mL) at K25 8C. After 10 min at
K20 8C, the suspension was left to reach rt and then
warmed to 45 8C for 24 h. After cooling, the reaction was
quenched by the addition of a concentrated aqueous solution
of Na2CO3. The solvent was removed with a rotary
evaporator and the residue was triturated with CH2Cl2
(5 mL), filtered and 5 was obtained as a creamy solid
(410 mg, 73%); mp: 223–224 8C. 1H NMR (300 MHz,
DMSO-d6) d 11.48 (s, 1H, NH); 7.69 (d, 1H, JZ8.0 Hz,
H-60); 7.66 (m, 1H, H-arom); 7.52 (m, 1H, H-arom); 7.39
(m, 1H, H-arom); 5.98 (dd, 1H, JZ9.1, 2.2 Hz, H-3a); 5.60
(dd, 1H, JZ8.0, 2.2 Hz, H-50); 5.02 (d, 1H, JZ13.6 Hz,
H-5a); 4.96 (d, 1H, JZ13.6 Hz, H-5b); 3.27 (dd, 1H, JZ
14.1, 9.1 Hz, H-2a); 3.18 (dd, 1H, JZ14.1, 2.3 Hz, H-2b).
13C NMR (750 MHz) d 162.93 (C-40); 149.85 (C-20); 142.14;
141.07 (C-6 ); 135.76, 132.12, 130.13, 128.73, 128.22 (C-
arom); 101.66 (C-50); 86.86 (C-3); 72.90 (C-5); 36.74 (C-2).
Anal. Calcd for C13H12N2O3S C, 56.51; H, 4.38; N, 10.14;
S, 11.60. Found: C, 56.36; H, 4.20; N, 9.86; S, 11.72.
1
4:1/EtOAc) and cis-7 was obtained pure. cis-7 H NMR
(300 MHz, CDCl3 C2 drops of CD3OD) d 7.86 (dd, JZ7.8,
1.3 Hz, 1H, H-arom); 7.63 (dt, JZ7.7, 1.3 Hz, 1H, H-arom);
7.50 (dt, JZ7.4, 1.4 Hz, 1H, H-arom); 7.32 (dd, JZ7.4,
1.1 Hz, 1H, H-arom); 7.21 (d, JZ8.2 Hz, 1H, H-60); 6.40
(dd,0 JZ10.6, 1.8 Hz, 1H, H-3); 5.67 (d, JZ8.2 Hz, 1H,
H-5 ); 4.83 (d, JZ14.3 Hz, 1H, H-5b); 4.67 (d, JZ14.3 Hz,
1H, H-5a); 3.47 (dd, JZ12.1, 1.8 Hz, 1H, H-2a); 3.34 (dd,
1
JZ12.0, 10.6 Hz, 1H, H-2b). trans-7 H NMR (300 MHz,
CDCl3 C2 drops of CD3OD) d 7.86 (dd, JZ7.1, 1.8 Hz, 1H,
H-arom); 7.50 (dt, JZ7.6, 1.5 Hz, 1H, H-arom); 7.50 (dt,
JZ7.6, 1.5 Hz, 1H, H-arom); 7.40 (d, JZ8.2 Hz, 1H, H-60);
6.65 (dd, JZ9.5, 2.4 Hz, 1H, H-3); 5.67 (d, JZ8.2 Hz, 1H,
H-50); 5.61 (d, JZ14.0 Hz, 1H, H-5b); 04.84 (d, JZ14.0 Hz,
1H, H-5a); 4.84 (d, JZ8.2 Hz, 1H, H-5 ); 3.62 (dd, JZ13.4,
2.4 Hz, 1H, H-2); 3.30 (dd, JZ13.4, 6.7 Hz, 1H, H-2). Anal.
Calcd for C13H12N2O4S: C, 53.42; H, 4.14; N, 9.58; S,
10.97. Found: C, 53.40; H, 4.20; N, 9.46; S, 10.72.
4.1.2.2. (RS)-1-(2,3-Dihydro-5H-4,1-benzoxathiepin-
3-yl)thymine (6). A 1.0 M solution of SnCl4/CH2Cl2
(1.22 mL, 1.22 mmol) was added dropwise with stirring
under argon to a suspension of 12 (200 mg, 1.02 mmol),
thymine (140 mg, 1.15 mmol), which contained TCS
(0.13 mL, 1.02 mmol) and HMDS (0.22 mL, 1.02 mmol)
in dry acetonitrile (6 mL) at K25 8C. After 10 min at
K20 8C, the suspension was left to reach rt and then
warmed to 45 8C for 40 h. After cooling, the reaction was
quenched by the addition of a concentrated aqueous solution
of Na2CO3. The solvent was removed with a rotary
evaporator and the residue was purified by flash chroma-
tography (CH2Cl2/MeOH 9.99:0.01) and 6 was obtained as
4.1.2.4. (1S*,3S*)-1-(1-Oxo-2,3-dihydro-5H-4,1-ben-
6
zoxathiepin-3-yl)thymine (trans-8). Compound
(0.06 mL, 1.04 mmol) was dissolved in a H2O2 50 wt%
solution in water, and the resulting solution was warmed
under stirring for 2 h. After cooling, the mixture was diluted
with H2O (10 mL) and was extracted with EtOAc (3!
10 mL), the organic layer was dried (Na2SO4), concentrated
and purified by flash chromatography using a gradient
elution (CH2Cl2/MeOH 9.8:0.2/9.5:0.5/9:1) to yield
1
trans-8 (104 mg, 50%); mp: 271–273 8C. trans-8 H NMR
(300 MHz, CDCl3CDMSO-d6) d 10.85 (br s, 1H, NH); 7.66
(dd, JZ7.6, 1.5 Hz, 1H, H-arom); 7.25 (dt, JZundet.,
1.5 Hz, 1H, H-arom); 7.19 (dt, JZ7.6, 1.5 Hz, 1H, H-arom);
7.06 (d,0JZ7.3, 1.3 Hz, 1H, H-arom); 6.88 (d, JZ1.2 Hz,
1H, H-6 ); 6.03 (dd, JZ10.5, 1.6 Hz, 1H, H-3); 4.95 (d, JZ
14.0 Hz, 1H, H-5); 4.51 (d, JZ14.0 Hz, 1H, H-5); 3.57 (dd,
JZ14.3, 10.4 Hz, 1H, H-2); 3.21 (dd, JZ14.3, 1.6 Hz, 1H,
H-2); 1.40 (d, JZ1.2 Hz, 1H, Me). 13C NMR (75 MHz) d
165.02 (C-40); 151.13 (C-20); 140.71 (C-arom); 136.45,
135.31, 132.28, 130.25, 128.75 (C-aromCC-60); 112.27
(C-50); 83.58 (C-3); 72.59 (C-5); 60.66 (C-2); 13.38 (Me).
LSIMS (NOBA matrix) calcd for C14H14N2O4SNa (MC
Na)C: 329.0572; found: 329.0572. Anal. Calcd for
C14H14N2O5S: C, 52.17; H, 4.38; N, 8.69; S, 9.95. Found:
C, 52.40; H, 4.40; N, 8.36; S, 9.72.
1
a white solid (343 mg, 58%); mp: 203–204 8C. H NMR
(400 MHz, DMSO-d6) d 11.35 (s, 1H, NH); 7.59 (m, 1H,
H-arom); 7.49 (s, 1H, H-60); 7.44 (m, 1H, H-arom); 7.31 (m,
2H, H-arom); 5.89 (dd, JZ9.7, 1.4 Hz, 1H, H-3); 4.89 (d,
JZ13.5 Hz, 1H, H-5b); 4.86 (d, JZ13.5 Hz, 1H, H-5a);
3.20 (dd, JZ14.1, 9.8 Hz, 1H, H-2a); 3.05 (dd, JZ14.1,
1.4 Hz, 1H, H-2b); 1.68 (s, 3H, Me). 13C NMR (100 MHz) d
163.63 (C-40); 149.84 (C-20); 142.32 (C-arom); 136.59
(C-60); 136.83, 132.18, 130.15, 128.72, 128.26 (C-arom);
109.40 (C-50); 86.59 (C-3); 72.89 (C-5); 35.79 (C-2); 11.85
(Me). HR LSIMS (NOBA matrix) calcd for C14H14N2O3SNa
(MC Na)C: 313.0623; found: 313.0624. Anal. Calcd for
C14H14N2O3S: C, 57.92; H, 4.86; N, 9.65; S, 11.04. Found:
C, 57.83; H, 5.10; N, 9.46; S, 11.32.
4.1.2.5. (RS)-3-(1,1-Dioxo-2,3-dihydro-5H-4,1-ben-
zoxathiepin-3-yl)uracil (9). Potassium peroxymonosulfate
(OXONEe, 709 mg, 1.15 mmol) in H2O (3 mL) was added
to a solution of 5 in MeOH (10 mL) and the resulting
suspension was left at rt for 2 h. After filtration and washing
with H2O and CH2Cl2, the residue was recrystallized from
acetone to yield 9 (118 mg, 66%) as a white solid; mp: 245–
4.1.2.3. (1R*,3S*)- and (1S*,3S*)-1-(2,3-Dihydro-5H-
4,1-benzoxathiepin-3-yl)uracil (cis-7 and trans-7).
Hydrogen peroxide, 50 wt% solution in water (0.05 mL,
0.087 mmol) was added to a solution of Sc(OTf)3 (71 mg,
0.145 mmol) in a mixture of 0.7 mL of EtOH and 6.3 mL of
CH2Cl2. Compound 5 (200 mg, 0.724 mmol) was added
after 5 min and the resulting solution was left at rt for 3.5 h.
After this, more H2O2 was added (0.05 mL) and the reaction
1
247 8C. H NMR (300 MHz, DMSO-d6) d 7.98 (d, JZ7.6,
1.5 Hz, 1H, H-arom); 7.66 (d, JZ8.2 Hz, 1H, H-60); 7.61
(m, 3H, H-arom); 6.28 (d, JZ10.6 Hz, 1H, H-50); 5.58 (d,

´
M. del Carmen Nun˜ez et al. / Tetrahedron 61 (2005) 10363–10369
10369
JZ8.0 Hz, 1H, H-3); 5.10 (d, JZ14.1 Hz, 1H, H-5); 4.99
References and notes
(d, JZ14.1 Hz, 1H, H-5); 4.37 (dd, JZ14.3, 10.7 Hz, 1H,
H-2); 4.04 (d,0 JZ14.3 Hz, 1H, 0H-2). 13C NMR (075 MHz) d
162.85 (C-4 ); 149.90 (C-2 ); 141.00 (C-6 ); 139.60
(C-arom); 136.13, 136.29, 131.41, 129.29, 127.00
(C-arom); 102.20 (C-50); 82.18 (C-3); 70.57 (C-5); 58.16
(C-2). Anal. Calcd for C13H12N2O5S: C, 50.64; H, 3.92; N,
9.09; S, 10.40. Found: C, 50.60; H, 3.80; N, 9.34; S, 10.75.
´
1. Saniger, E.; Campos, J.; Entrena, A.; Marchal, J. A.; Suarez, I.;
´ ´
Aranega, A.; Choquesillo, D.; Niclos, J.; Gallo, M. A.;
Espinosa, A. Tetrahedron 2003, 59, 5457–5467.
2. Saniger, E.; Campos, J.; Entrena, A.; Marchal, J. A.; Houria,
´
B.; Aranega, A.; Gallo, M. A.; Espinosa, A. Tetrahedron 2003,
59, 8017–8026.
3. Campos, J.; Domınguez, J. F.; Gallo, M. A.; Espinosa, A.
´
Curr. Pharm. Des. 2000, 6, 1797–1810.
4. Crombie, L.; Josephs, J. L.; Larkin, J.; Weston, J. B. Chem.
Soc., Chem. Commun. 1991, 14, 972–973.
4.1.2.6. (RS)-3-(1,1-Dioxo-2,3-dihydro-5H-4,1-ben-
zoxathiepin-3-yl)thymine (10). The procedure was similar
to Section 4.1.2.5, but starting from 6. Yield (58%) as a
white solid; mp: 272–274 8C. 1H NMR (300 MHz, DMSO-
d6) d 11.55 (s, 1H, NH); 8.07 (dd, JZ7.6, 1.4 Hz, 1H, H-60);
7.74 (m, 1H, H-arom); 6.37 (dd, JZ10.6, 1.4 Hz, 1H, H-3);
5.20 (d, JZ14.1 Hz, 1H, H-5); 5.07 (d, JZ14.1 Hz, 1H, H-
5); 4.42 (dd, JZ14.5, 10.7 Hz, 1H, H-2); 4.11 (dd, JZ14.5,
1.4 Hz, 1H, H-2); 1.76 (s, 3H, Me). 13C NMR (75 MHz) d
163.34 (C-40); 149.72 (C-20); 139.48 (C-arom); 136.36,
136.02, 134.08, 131.22, 129.08, 126.79 (C-60CC-arom);
109.66 (C-50); 81.73 (C-3); 70.32 (C-5); 58.22 (C-2); 11.67
(Me). Anal. Calcd for C14H14N2O5S: C, 52.17; H, 4.38; N,
8.69; S, 9.95. Found: C, 52.40; H, 4.58; N, 8.44; S, 9.75.
5. For reviews, see: (a) Ward, R. S.; Diaper, R. Sulfur Rep. 2001,
22, 251. (b) Jackson, D. A.; Rawlinson, H.; Barba, O. Spec.
Publ.-R. Soc. Chem. 2001, 260, 47–53. (c) Mashkina, A. V.
Sulfur Rep. 1991, 10, 279–388.
6. Matteucci, M.; Bhalay, G.; Bradley, M. Org. Lett. 2003, 5,
235–237.
7. Khanapure, S. P.; Garvey, D. S.; Young, D. V.; Ezawa, M.;
Earl, R. A.; Gaston, R. D.; Fang, X.; Murty, M.; Martino, A.;
Shumway, M.; Trocha, M.; Marek, P.; Tam, S. W.; Janero,
D. R.; Letts, L. G. J. Med. Chem. 2003, 46, 5484–5504.
8. From now on, the primes (0) will accompany the numbering of
the pyrimidine bases.
9. Bax, A.; Summers, M. F. J. Am. Chem. Soc. 1986, 108,
2093–2094.
4.2. Biological activity
10. Matsuo, S.; Tanako, S.; Yamashita, J.; Ogawa, M. Anticancer
Res. 2000, 12, 1575–1579.
´
11. Trouet, A.; Passioukov, A.; Derpooten, K. V.; Fernandez,
The biological methods were the same as those previously
described.12
A. M.; Abarca-Quin˜ones, J.; Baurain, R.; Lobl, T. J.; Oliyai,
C.; Dubois, V. Cancer Res. 2001, 61, 2843–2846.
´
´
´ ´
12. Dıaz-Gavilan, M.; Rodrıguez-Serrano, F.; Gomez-Vidal, J. A.;
´
Marchal, J. A.; Aranega, A.; Gallo, M. A.; Espinosa, A.;
Acknowledgements
Campos, J. M. Tetrahedron 2004, 60, 11547–11557.
´ ´
13. Saniger, E.; Dıaz-Gavilan, M.; Delgado, B.; Choquesillo, D.;
This study was supported by the Instituto Carlos III (Fondo
de Investigaciones Sanitarias, Projects PI03225 and
PI041206).
´ ´
Gonzalez-Perez, J. M.; Aiello, S.; Gallo, M. A.; Espinosa, A.;
Campos, J. M. Tetrahedron 2004, 60, 11453–11464.