Neighbouring-group participation as the key step in the reactivity of acyclic and cyclic salicyl-derived O,O-acetals with 5-fluorouracil. Antiproliferative activity, cell cycle dysregulation and apoptotic induction of new O,N-acetals against breast cancer cells

Article abstract of DOI:10.1016/j.tet.2003.08.016

The reaction between o-(hydroxymethyl)phenoxyacetaldehyde dimethyl acetals, or 3-methoxy-2,3-dihydro-5H-1,4-benzodioxepins with 5-fluorouracil (5-FU), has been studied. The intramolecular cyclization may be explained through a neighboring group attack to give a 2-(5-fluorouracil-1-yl)oxyranium ion that can be attacked by the silylated benzylic hydroxy group to yield the benzannelated seven-membered O,N-acetals. The formation of a macrocyclic trans-bis(5-FU O,N-acetal) is also reported. Such a compound arrested the human MCF-7 breast cancer cells at the G_o/G₁ phase of the cell cycle. On the contrary, the acyclic nitro O,N-acetal seems to work as a 5-FU prodrug, because it arrested cancer cells at the S phase as the well-known prodrug Ftorafur does.

Full text of DOI:10.1016/j.tet.2003.08.016

Tetrahedron 59 (2003) 8017–8026
Neighbouring-group participation as the key step
in the reactivity of acyclic and cyclic salicyl-derived O,O-acetals
with 5-fluorouracil. Antiproliferative activity,
cell cycle dysregulation and apoptotic induction of new
O,N-acetals against breast cancer cells
a
a
a
b
c
´
Estrella Saniger, Joaquın M. Campos, Antonio Entrena, Juan A. Marchal, Houria Boulaiz,
c
Antonia Aranega, Miguel A. Gallo and Antonio Espinosa
a
a
,
´
*
´
a
´
´
´
Departamento de Quımica Farmaceutica y Organica, Facultad de Farmacia, c/ Campus de Cartuja 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
b
c
´
Departamento de Ciencias Morfologicas, Facultad de Medicina, Avenida de Madrid s/n, 18071 Granada, Spain
Received 4 June 2003; revised 30 July 2003; accepted 6 August 2003
Abstract—The reaction between o-(hydroxymethyl)phenoxyacetaldehyde dimethyl acetals, or 3-methoxy-2,3-dihydro-5H-1,4-benzodiox-
epins with 5-fluorouracil (5-FU), has been studied. The intramolecular cyclization may be explained through a neighboring group attack to
give a 2-(5-fluorouracil-1-yl)oxyranium ion that can be attacked by the silylated benzylic hydroxy group to yield the benzannelated seven-
membered O,N-acetals. The formation of a macrocyclic trans-bis(5-FU O,N-acetal) is also reported. Such a compound arrested the human
MCF-7 breast cancer cells at the Go/G1 phase of the cell cycle. On the contrary, the acyclic nitro O,N-acetal seems to work as a 5-FU prodrug,
because it arrested cancer cells at the S phase as the well-known prodrug Ftorafur does.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
by the immune system and chemotherapeutic drugs in
eradicating tumour cells. Resistant tumour cells evade the
action of anticancer agents by increasing their apoptotic
threshold. This has spurred the development of novel
chemical compounds capable of inducing apoptosis in
chemo/immune-resistant tumour cells. Although breast
cancer is most often treated with conventional cytotoxic
agents it has proven difficult to induce apoptosis in breast
cancer cells using these drugs.6 By identifying therapies that
are particularly effective in activating apoptosis, improved
clinical responses may be obtained.
Breast cancer ranks just behind lung cancer as the second
leading cancer in women, and is the leading cause of cancer
death among women in the 35–54 age group and the second
cause of cancer death for women aged 55–74.1–3 The
disseminated nature of breast cancer and the development of
cross-resistant tumours are the primary causes of failure of
current therapies. By the time a tumour is detected, there is
a high probability that methastatic lesions will be present,
and many of these will already contain a resistant sub-
population of cells. Not surprisingly, there is a substantial
interest in the identification of novel anticancer agents for
the treatment of breast cancer.1–4
The non-naturally occurring base with known antitumour
activity, 5-fluorouracil (5-FU, 1), is highly toxic and this
fact has motivated important research directed to the
synthesis and biological evaluation of 5-FU derivatives
with less toxicity than the parent compound. Within the
extensive therapeutical arsenal the prodrug Ftorafur 2
occupies an outstanding position.7 As part of a programme
for the development of 5-FU derivatives, we published the
synthesis and biological activities of several acyclonucleo-
sides analogues.8,9 Later on, the synthesis of a wide range of
5-FU derivatives linked to saturated seven-membered
moieties through N-1 was carried out,10 and this method-
ology allowed us access to non-classical nucleosides with a
Apoptosis or programmed cell death is an innate mechanism
by which unwanted, defective, or damaged cells are rapidly
and selectively eliminated from the body. It occurs during
tissue remodeling, embryonic development, and immune
regulation.5 Apoptosis is the principle mechanism employed
Keywords: acetals; antitumour compounds; benzodioxepins; mechanisms;
neighbouring group effects.
*
Corresponding author. Tel.: þ34-958-243850; fax: þ34-958-243845;
e-mail: aespinos@ugr.es
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.08.016

8018
E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
3-hydroxymethyl-1,4-dioxepan-5-yl group as the ‘sugar’
moiety.11 In a recent article12 we have reported the
preparation, antiproliferative activities and apoptotic induc-
tion against the MCF-7 human breast cancer cell line of the
benzannelated seven-membered 5-FU O,N-acetals 3a–c.
On one hand, we report herein their reaction mechanism and,
on the other, the preparation of the open 5-FU O,N-acetalic
analogues, to assess their biological activities against the
breast cancer cells (Chart 1).
Table 1. Formation of cyclic and acyclic 5-FU O,N-acetals starting from
the acyclic O,O-acetals 4, the reaction time being 24 ha
Entry Starting acetal
R1
R2
Yield (%) of 3 Yield (%) of 6
1
2
3
4
5
4b
4c
4d
4e
4f
OMe
H
Cl
Br
NO2
H
OMe
H
H
H
26 (5)b
27
37
17
4 (8)b
35
–
–
–
–
a
5-FU, HMDS, TCS, SnCl4/CH2Cl2, CH3CN.
b
All the yields refer to compounds in which the 5-FU moiety is linked
through N1, except in 6eN-3 and 3bN-3, in which the link is through N3 and
whose yields are reported between brackets.
Chart 1.
of the substitution model on the 5-FU moiety. Briefly, (a) a
broad triplet due to C6-H at around d<7–8 ppm indicates
that 5-FU has a substituent at its N-1 position, and (b) a
broad triplet due to C6-H at around d<7–8 ppm indicates
that 5-FU is 3-substituted. Compound 3bN-3 (this nomen-
clature indicates that the 5-FU moiety is linked to the carbon
chain through its N-3 atom) showed a normal behaviour
2. Results and discussion
2.1. Chemistry
The synthesis of the O,N-acetals was carried out in a two-
step process: (a) preparation of the intermediate acyclic and
cyclic O,O-acetals, 4a–f and 5a–f, respectively, and (b)
reaction of these acyclic and cyclic compounds with the
pyrimidine base to give the title structures. 5-chloro-2-
hydroxybenzyl alcohols were obtained by reduction of the
corresponding salicylaldehyde.13 Hydroxyacetals 4d–f
were prepared in a similar way to the obtention of 4a–c12
by alkylation of the salicyl alcohols with bromoacetalde-
hyde dimethyl acetal, using sodium hydride as a base in
anydrous dimethylformamide (DMF). Finally, the 3-methoxy-
2,3-dihydro-5H-1,4-benzodioxepins 5d–f were obtained
under an acid-catalyzed intramolecular cyclization process
on the hydroxyacetals 4d–f, under the conditions previously
reported for 5a–c12 (Scheme 1).
1
when its H NMR spectrum was recorded in a DMSO-d6
solution. Nevertheless, the situation is different for 6eN-3
:
C6-H resonated as a sharp doublet when the spectrum was
recorded in CD3OD due the rapid exchange of the hydrogen
atom of the N1-H group with the deuterium atom of the
solvent. Nevertheless, the aspect of the H-10 casts light on
the subject: this proton resonates as a pseudotriplet whereas
the same proton of 6e resonates as a double doublet of
doublet (ddd), the latter being due to the long-range
coupling with the fluoro atom,8,10,11 which showed
unambiguously that the 5-FU is linked through N-1 to the
o-(hydroxymethyl)phenoxyethyl-1-methoxy moiety. In the
former case (6eN-3), the long-range coupling with the F atom
was not observed because its electronic effect was not
transmitted through the carbonyl group at position 4 of the
uracil ring.
The substitution of the acetalic OMe group by the 5-FU
moiety in the O,O-acetals 4a–f and 5d–f has been carried
out by reaction with 5-FU in the presence of the silylating
agents 1,1,1,3,3,3-hexamethyldisilazane (HMDS) and tri-
methylchlorosilane (TCS), under acid catalysis of stannic
chloride. The aminalic bond is established through N-1 of
the 5-FU moiety unless otherwise stated. In spite of the fact
that the starting materials (4, 5) are dissimilar, in both cases
the same final compound type was obtained, with different
yields.
In relation with the condensation reaction between the
acyclic O,O-acetals 4 and 5-FU, the following can be
stated:
(1) The nature and position of substituents R1 and R2
showed clear influence on both the yields and
regioselectivity of the process. Thus, the substitution
in position 5 of 4 (equivalent to 7 of 5) affected in the
following manner (only R2¼H is considered):
1.1 The electron-withdrawing substituents (Cl, Br and
NO2) induced the preferent formation of acyclic
O,N-acetals 6 (entries 3–5).
2.1.1. Condensation between o-(hydroxymethyl)phenoxy-
acetaldehyde dimethyl acetals 4 and 5-FU. The conden-
sation between the acyclic O,O-acetals 4 and 5-FU, under
the previously described conditions,12 gave rise to the cyclic
3 and/or the acyclic O,N-acetals 6, respectively through a
process whose regioselectivity depended on the presence
and nature of the substituents on the benzene ring (Table 1),
named R1 and R2. In general, in the majority of cases the
attachment of the 5-FU moiety occurred at N-1, and rarely at
the N-3 position of the uracil ring. These facts were
confirmed through 1H NMR data. According to Ozaki et al.14
the H6 coupling pattern of the pyrimidine ring is indicative
1.2 The only electron-donating substituent used, the
OMe group, halted the regioselectivity of the
reaction forming both the cyclic 3 and acyclic 6
O,N-acetals and, moreover, in an approximately
1/1 ratio (entry 1).
1.3 In some cases the presence of substituents in one
or other type of O,O-acetals permitted the
isolation of derivatives in which the 5-FU moiety

E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
8019
Scheme 1. Reagents: (a) BrCH2CH(OMe)2, HNa, anhydrous DMF. (b) BF3·OEt2 in anhydrous Et2O.
was linked through N-3 of the pyrimidine base
(entries 1 and 4).
(2) Finally, the substitution in C-3 of 4 by the only group
studied (OMe) clearly favoured the formation of the
acyclic O,N-acetal (entry 2).
1. The better yield was obtained when R1 and R2 were
hydrogen atoms (entry 1). The process is regioselective
in the cyclic O,N-acetal 3a and the 5-FU moiety was
linked through its N-1 atom, no traces of the N-3
substituted compound having being observed.
2. Substituents R1 and R2 similarly influenced the regio-
selectivity of the previously described reaction; thus,
the R1 electron-withdrawing substituents (Cl, Br and
NO2) induced the regioselective formation of the acyclic
O,N-acetals (entries 5–7).
The nitro group of 6f has been reduced with tin(II) chloride
dihydrate in refluxing ethanol to yield compound 6g
(Scheme 2).
2.1.3. Experimental study of the condensation reaction
between the acyclic O,O-acetals 4, and the cyclic O,O-
acetals 5 with 5-FU: influence of other factors on the
progress of the reaction. As has been explained pre-
viously, the nature of the substituent on the aromatic ring
is determinant for the obtention of the cyclic or acyclic O,N-
acetals. Now the influence of the reaction time and the type of
the Lewis acid will be described. For this study the most simple
acetals have been selected. The reaction between the cyclic
O,O-acetal 5a and 5-FU is strongly affected by the reaction
time (Table 3). Thus, a short reaction time (4 h) favours the
formation of the acyclic O,N-acetal 6a, whereas its increase
(24 h) provokes the regioselective synthesis of the cyclic O,N-
acetal 3a(entries 2 and 3), aswas described previously.12 From
these experimental data it can be deduced that compound 3a is
formed from the structure 6a, as will be depicted in Scheme 4.
Scheme 2. Reagents: (a) SnCl2·2H2O, EtOH.
2.1.2. Condensation between 3-methoxy-7-substituted-
2,3-dihydro-5H-1,4-benzodioxepins 5 and 5-FU. The
results obtained when the cyclic O,O-acetals were used as
starting materials in the reaction with 5-FU are shown in
Table 2, which reproduces the reaction time and the yields
obtained. In this study, the following may be generalized:
Table 2. Formation of cyclic and acyclic 5-FU O,N-acetalsa starting from
cyclic O,O-acetals 5, the reaction time being 24 hb
Table 3. Several conditions in the condensation reaction between the cyclic
O,O-acetal 5a and 5-FU
Entry
Lewis acid
Reaction time
(h)
Yield
(%) of 3a
Yield
(%) of 6a
1
2
3
SnCl4/CH2Cl2
SnCl4/CH2Cl2
BF3·OEt2
4
24
24
4
43
36
15
–
–
Entry Starting acetal
R1
R2
Yield (%) of 3c Yield (%) of 6c
All the yields refer to compounds in which the 5-FU moiety is linked
through N1.
1
2
3
4
5
6
7
5a
5b
5b
5c
5d
5e
5f
H
H
H
H
43
27 (26)
40d
–
OMe
OMe
H
Cl
Br
21 (27)
–
5
77 (17)
33 (4)
5 (35)
Nevertheless, in the reaction between the acyclic O,O-acetal
4a and 5-FU with a reaction time of 24 h and using SnCl4/
CH2Cl2 as the Lewis acid, the two following reaction
products have been identified (Scheme 3): 3a (10%) and its
corresponding dimer 7a† (7%) that shows a 14-crown-4
ether structure with two pendant 5-FU fragments. Although
OMe 15.5
H
H
H
–
–
–
NO2
a
All the yields refer to compounds in which the 5-FU moiety is linked
through N1.
5-FU, HMDS, TCS, SnCl4/CH2Cl2, CH3CN.
The yields obtained from the acyclic O,O-acetals 4 are shown between
brackets.
b
c
†
This compound will be assigned to have a trans isomerism (Section
2.2.2).
d
Reaction time: 144 h (Ref. 12).

8020
E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
5000 conformations were compared and those which were
geometric and energetically equal were removed. By
means of the procedure were identified 307 different con-
formations whose relative energies range between 0.00 and
14.84 kcal/mol.
Figure 1 shows the four most stable conformations. It can be
observed that there is a trend in the molecule to arrange the
aromatic rings parallel to each other. Moreover, in three of
these conformations both aromatic rings are oriented
towards the same side of the macrocycle, probably due to
a p-p stacking between them. From these conformations,
the most stable one was selected for the following phase of
this study.
a
Scheme 3. Reagents:
reaction time 24 h.
5-FU, HMDS, TCS, SnCl4/CH2Cl2, CH3CN,
the yield for the obtention of 7a was very low, the structure
is very attractive and will give it a long pedigree as a result
of its notable biological properties (see Section 2.4).
2.2. Conformational analysis
The conformational study of the macrocyclic compound 7a
was carried out in two well differentiated phases. In the first
one, the study of the base ring, without the two 5-FU
moieties, was tackled to identify the most stable confor-
mation of this compound. In the second one, from the less
energetic conformation of this compound, the two 5-FU
fragments were added to give rise to the cis- and trans-7a
and the conformational analysis on both isomers was carried
out.
2.2.1. Conformational analysis of the 4,10-dimethoxy-
2,5,8,11-tetraoxa-1,7(1,2)dibenzenadodecaphane.15,16
The initial geometry of the non-substituted macrocycle
was generated from standard fragments from the Sybyl
libraries.17 Once the structure was built up, its opti-
mization was undertaken using the Tripos force field,18
and the charges were generated by means of the Gasteiger–
Hu¨ckel methodology.19 The minimization of the struc-
ture was carried out by means of the Powell method,18 up
to the point where the energy gradient is less than
Figure 1. The four most stable conformations of the base ring, together
with their relative energy values (kcal/mol) calculated by means of
molecular mechanics (Tripos). It is worth noting the nearly parallel
arrangement of the aromatic rings in all of them.
2
˚
0.005 kcal/mol A .
Once the initial molecule was optimized a conformational
search was begun to find the most stable conformation. The
large size of the saturated ring involves an important
restriction on carrying out the conformational search by
means of random search or systematic techniques, because
the number of bonds that would have to be investigated
using these two methodologies is very high. Accordingly, a
conformational search by means of molecular dynamics was
undertaken using the ‘simulated annealing’ technique. This
technique simulates a heating of the molecule to an elevated
temperature allowing conformational changes to take place.
Once a certain time has elapsed the molecule is cooled down
to a low temperature, that makes the last conformation
‘freeze’. Next, such a conformation is saved and the
procedure is repeated many times. In our case, a heating
to 1000 K was simulated for 1000 ps and, after this time had
elapsed such a temperature was kept for further 1000 ps
before saving the last conformations. The cycle was
repeated 5000 times, which allowed the identification of
5000 conformations of the base ring. Once minimized, the
2.2.2. Conformational analysis of cis- and trans-4,10-
bis[5-(fluorouracil-1-yl)]-2,5,8,11-tetraoxa-1,7(1,2)di-
benzenadodecaphane. Once the most stable conformation
of the macrocycle was identified, the two 5-FU cis and trans
isomers were built up. For this the adequate hydrogen atoms
on positions 4 and 10 of the base ring were substituted by
standard fragments from the Sybyl libraries. The partial
atomic charges for each isomers were calculated by
means of the Gasteiger–Hu¨ckel method and their optimiz-
ation was carried out until the gradient was less than
2
˚
0.005 kcal/mol A . Following the same procedure explained
for the base ring, in the cis isomer were identified 392
conformations whose relative energies ranged between 0.00
and 23.13 kcal/mol, whereas for the trans isomer were
identified 427 conformations whose relative energies ranged
between 0.00 and 22.67 kcal/mol. On each compound the
most stable conformations were studied.

E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
8021
Figure 2 shows the four most stable conformations of the cis
isomer, in which it can be observed that at least two parallel
rings exist, possibly due to p stacking interactions between
them. The differences in energy between such confor-
mations is not very high and accordingly, the molecule
could adopt either of them without any difficulty. Although
only four conformations were represented, our data show
that in this isomer a very complex conformational
equilibrium exists with the participation of many confor-
mations of similar energy.
Figure 3. Representation of the four most stable conformations of compound
trans-7a. The relative energy values calculated by means of molecular
mechanics (Tripos) are expressed in kcal/mol. It is worth noting the parallel
arrangement of the aromatic rings in the most stable conformation, in which
the symmetry centre of the molecule can be observed.
substitution of the acetalic OMe group by the 5-FU takes
places as we have previously reported for bis-acetals20
leading to the intermediate 8. The intramolecular attack of
the silylated benzylic hydroxy group might explain the
formation of 3 (route 2). Nevertheless, although the mech-
anism is simple and logical, it does not justify the influence
of the R1 substituent over the course of the reaction. A
possible solution might involve a different pathway for the
intramolecular cyclization in which the phenolic oxygen
atom should intervene as a neighboring group, whose
nucleophilicity may be strongly influenced by the electronic
character of the substituent R1. In fact, route 3 shows a
possible mechanism for the intramolecular cyclization.
According to this route, the intermediate 9 suffers the
neighboring group attack to give the oxyranium ion 10,
much more reactive than its predecessor and that can be
attacked by the silylated benzylic hydroxy group. Finally,
the aqueous work-up renders 3. Accordingly, the intra-
molecular cyclization product will depend on the stability of
the intermediate 10 which will in turn be influenced by the
electronic character of the substituent R1 (and R2); in fact,
electron-withdrawing groups destabilize the positive charge
of the phenolic oxygen atom on generating an electronic
deficiency in the carbon atom that carries the oxygen atom,
making the intramolecular closing impossible. The contrary
holds true for the electron-donating groups such as the
methoxy moiety.
Figure 2. Representation of the four most stable conformations of
compound cis-7a. The relative energy values calculated by means of
molecular mechanics (Tripos) are expressed in kcal/mol. It is worth noting
the nearly parallel arrangement of at least two aromatic rings in all of them.
Figure 3 shows the four most stable conformations of trans
isomer. In this compound has been found a great
predominance of a conformation in which each benzene
ring is located parallel to a 5-FU ring. The great stability of
this conformation is probably due to the two p-stacking
interactions present in it. Far from isomer cis, the trans one
showed a more simple behaviour from a conformational
point of view because the energetic differences between the
different conformations are very important. It can be stated
that the most stable conformation (upper left conformation
in Figure 3) is practically the only one in the conformational
equilibrium and, accordingly, the molecule properties
would be explained from this conformer. This conformation
has an inversion centre, which makes the symmetrical atoms
equivalent and, hence, its NMR behaviour should be very
simple and should give rise to spectra that contain only half
the molecule signals. In the 1H and 13C NMR spectra of this
compound this kind of behaviour was found and conse-
quently the trans derivative might be the isolated isomer,
instead of the cis one that should identify each proton or
carbon of the molecule. Moreover, if it were a mixture of
both isomers, the NMR spectra should have been even more
complex.
In relation with the condensation reaction between the
cyclic O,O-acetals 5 and 5-FU the results seem to suggest
that, in the presence of SnCl4, the rest of reaction conditions
and in accordance with our previous findings,8 8 was
formed. This intermediate is the precursor of the two types
of target compounds according to Scheme 4. Consequently,
the formation of the seven-membered ring reveals that the
2.3. Mechanism of the condensation reaction from O,O-
acetals and 5-FU: neighbouring-group participation
The mechanism represented in Scheme 4 (routes 1 and 3),
might explain the nature of the reaction products. The

8022
E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
Scheme 4.
attack of the benzylic trimethylsilyloxy group leads to a
favoured 7-exo-Tet process, in accordance with Baldwin’s
rules.21 It does not follow that because a process is ‘favored’
it will necessarily occur readily in every case. The other
factor such as the presence of an electron-withdrawing
group at position para in relation to the phenolic ethereal
atom exerts a negative influence and, accordingly, the
formation of the seven-membered ring does not take place.
This mechanism is supported by the following fact: when
the reaction was carried out in a short time (4 h, Table 3,
entry 1) the preferential formation of the acyclic O,N-acetal
6a over the cyclic one 3a was observed.
actors regulating the cell cycle and from the better
understanding of the connections between cell cycle and
apoptosis. As more and more ‘cell cycle drugs’ are being
discovered, their use as anticancer drugs is being exten-
sively investigated.23 To study the mechanisms of the
antitumour and antiproliferative activities of the com-
pounds, the effects on the cell cycle distribution were
analyzed by flow cytometry. DMSO-treated cell cultures
contained 68.39% G0/G1-phase cells, 12.04% G2/M-phase
cells and 19.57% S-phase cells. In contrast, MCF-7 cells
treated during 48 h with the IC50 concentrations of 6a,c–g
and 7a showed important differences in cell cycle
progression compared with DMSO-treated control cells.
Ftorafur, 2 treatment showed a decrease of the G0/G1-phase
cells and a corresponding accumulation of S-phase cells
(45.62% G0/G1-phase cells and 54.38% S-phase cells).
Moreover, there was an almost total disappearance in the
G2/M population of the cells treated with this drug. In
general the cell cycle regulatory activities for the newly
synthesized compounds can be divided into the following
three groups: (a) the exposure of MCF-7 cells during 48 h to
the novel derivative 7a, caused strong differences in the cell
cycle progression in comparison with Ftorafur, 2: the breast
cancer cells treated with the IC50 doses showed a significant
accumulation in the G0/G1-phase, up to 82.48% of the cells,
mainly at the expense of the G2/M-phase population that
decreased to a percentage of 5.13% of the cells; (b)
compounds 6d and 6e accumulated the cancerous cells in
the G2/M-phase, in the former compound at the expense of
the S-phase cells, and (c) compound 6f induced a S-phase
cell cycle arrest (50.24%) in a similar percentage to that
caused by Ftorafur, 2 (54.28%) and, accordingly, it can be
affirmed that the nitro derivative (6f) may act as a 5-FU
prodrug. Nevertheless, this hypothesis needs to be corrobor-
ated by further assays.
2.4. Antiproliferative activities, cell cycle distribution
and apoptosis induction in the human breast cancer
MCF-7 cell line
The IC50 values of compounds are shown in Table 4. The
most active compounds are 6f (5.42^0.26 mM) and 7a
(5.5^0.58 mM), with antiproliferative activities in the same
order as that of Ftorafur (3^0.11 mM).
Cell cycle regulation has attracted a great deal of attention
as a promising target for cancer research and treatment.22,23
The use of cell-cycle-specific treatments in cancer therapy
has greatly benefited from the major advances that have
been recently made in the identification of the molecular
Table 4. Antiproliferative activities, cell cycle dysregulation, and apoptosis
induction in the MCF-7 human breast cancer cell line after treatment for 24
and 48 h for the compounds
Compound
IC50 (mM)
Cell cycle (48 h)
Apoptosis (h)
24 48
1.24 1.24
G0/G1
G2/M
S
Control
Ftorafur, 2
68.39
45.62
67.18
62.72
71.01
51.45
46.92
67.32
82.48
12.04
0
19.57
54.38
28.16
35.69
0
27.88
50.24
23.28
12.40
3^0.11
18.5^0.95
29^1.63
18^0.85
16^1.18
5.42^0.26
21^1.02
5.5^0.58
52.2
59.9
58.06
40.23
37.87
50.64
36.37
48.22
37.81
19.05
In response to 6a, the percentage of apoptotic cells
increased, from 1.24% in control cells to a maximum of
59.9% apoptotic cells (24 h) at a concentration equal to its
IC50 against the MCF-7 cell line. This is a remarkable
property because the demonstration of apoptosis in MCF-7
breast cancer cells by known apoptosis-inducing agents has
proved to be difficult and only few cytotoxic agents act
6a
6c
6d
6e
6f
4.67
1.59
28.99
20.66
2.84
9.40
5.13
33.35
44.36
42.24
40.73
41.15
14.37
6g
7a

E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
8023
preferentially through an apoptotic mechanism in human
breast cancer cells.6,24,25 Finally, a fact that is worth
emphasizing is that 6f and 7a (the only two compounds
tested) induce neither toxicity nor death in mice after one
month’s treatment when administered intravenously twice a
week, with a 50 mg/kg dose each time (results not shown).
1H, H-3, J¼8.6 Hz); 4.69 (t, 1H, H-10, J¼5.2 Hz); 4.59 (d,
2H, CH2OH, J¼6.2 Hz); 4.01 (d, 2H, H-20, J¼5.2 Hz); 3.41
(s, 6H, (OCMe)2). 13C NMR (75 MHz, CDCl3) d 155.68
(C-2); 132.30 (C-1); 131.72 (C-4)0; 131.49 (C-6); 113.87
(C-3); 113.67 (C-5); 101.68 (C-1 ); 67.94 (C-20); 61.41
(CH2OH); 54.17 ((OMe)2). HR LSIMS calcd for C11H15-
O4NaBr (MþNa)þ 313.0051, found: 313.0057. Anal. for
C11H15O4Br: calcd: C 45.38; H 5.19. Found: C 45.40; H
5.22.
3. Conclusion
1-(Hydroxymethyl)-5-nitro-2-phenoxyacetaldehyde
dimethyl acetal 4f. Yield: 26%. Rf (2/1, diethyl ether/
hexane): 0.2. H NMR (300 MHz, CD3COCD3) d 8.34 (d,
We have completed the synthesis and biological evaluation
of novel 1-[o-(hydroxymethyl)phenoxyethyl-1-methoxy]-5-
fluorouracil compounds, starting from several salicyl
alcohols. In this study we have investigated the Lewis
acid-promoted transformation of the acyclic O,N-acetals to
the corresponding cyclic O,N-acetals when the R1 group is
either a hydrogen atom or a methoxy group. All the
compounds exhibited an IC50 against the MCF-7 human
breast cancer cell line in the micromolar range, among
which the nitro derivative 6f was found to be the most
cytotoxic. We have demonstrated that compound 6a
induced apoptosis and G0/G1 cell cycle arrest in the
MCF-7 human breast cancer cell line, whereas 6f seems to
act as the prodrug Ftorafur on the basis that both compounds
accumulate the tumoural cells in the S phase of the cell
cycle. Taken together, the experimental findings provide
evidence of specific antitumour activity of these new
substances and warrant further evaluation in in vivo models
of breast cancer to future clinical applications.
1
1H, H-6, J¼2.9 Hz); 8.13 (dd, 1H, H-4, J¼0 2.9, 9.0 Hz); 7.17
(d, 1H, H-3, J¼9.0 Hz); 4.76 (t, 1H, H-1 , J¼5.2 Hz); 4.70
(s, 2H, CH2OH); 4.16 (d, 2H, H-20, J¼5.2 Hz); 3.41 (s, 6H,
(OMe)2). 13C NMR (75 MHz, CD3COCD3) d 161.11 (C-1);
142.49 (C-5); 133.21 (C-1); 124.74 (C-4); 122.84 (C-6);
111.94 (C-3); 102.79 (C-10); 69.26 (C-20); 58.98 (CH2OH);
54.48 ((OMe)2). HR LSIMS calcd for C11H15NO6Na
(MþNa)þ 378.0808, found: 378.0882. Anal. for
C11H15NO6: calcd: C 51.36; H 5.88; N 5.45. Found: C
52.21; H 6.22; N 5.46.
Synthesis of (RS)-3-methoxy-2,3-dihydro-5H-1,4-benzo-
dioxepins 5. The preparation of 5a–c was previously
reported by us12 and we proceeded with the same method
to obtain 5d–f.
(RS)-7-Chloro-3-methoxy-2,3-dihydro-5H-1,4-benzodioxe-
1
pin 5d. Yield: 44%. Rf (4/1, diethyl ether/hexane): 0.7. H
NMR (400 MHz, CDCl3) d 7.14 (dd, 1H, H-8, J¼2.4,
8.5 Hz); 7.07 (d, 1H, H-6, J¼2.4 Hz); 6.91 (d, 1H, H-9,
J¼8.5 Hz); 5.17 (d, 1H, H-5b, Jgem¼14.5 Hz); 4.88 (dd, 1H,
H-3a, J¼3.9, 5.9 Hz); 4.33 (d, 1H, H-5a, Jgem¼14.5 Hz);
4.09 (m, 2H, H-2a, H-2b); 3.50 (s, 3H, OMe). 13C NMR
(100 MHz, CDCl3) d 157.49 (C-10); 131.10 (C-7); 128.57
(C-8); 128.23 (C-6); 127.53 (C-11); 121.41 (C-9); 101.08
(C-3); 72.23 (C-2); 62.34 (C-5); 55.57 (OMe). HR LSIMS
calcd for C10H11O3NaCl (MþNa)þ 237.0294, found:
237.0298. Anal. for C10H11O3Cl: calcd: C 55.96; H 5.17.
Found: C 55.94; H 5.19.
4. Experimental
4.1. Chemistry
For a description of general information, see Ref. 8. 5-Chloro-
2-hydroxybenzyl alcohol was obtained by reduction of the
corresponding salicylaldehyde.13
4.1.1. Starting materials
4.1.1.1. Synthesis of o-(hydroxymethyl)phenoxyacet-
aldehyde dimethyl acetals 4. The preparation of 4a–c was
previously reported12 and we proceeded with the same
method to obtain 4d–f.
(RS)-7-Bromo-3-methoxy-2,3-dihydro-5H-1,4-benzodioxe-
1
pin 5e. Yield: 35%. Rf (3/1, diethyl ether/hexane): 0.6. H
NMR (300 MHz, CDCl3) d 7.26 (dd, 1H, H-8, J¼2.4,
8.5 Hz); 7.20 (d, 1H, H-6, J¼2.4 Hz); 6.84 (d, 1H, H-9,
J¼8.5 Hz); 5.16 (d, 1H, H-5b, Jgem¼14.5 Hz); 4.87 (dd, 1H,
H-3a, J¼4.1, 5.8 Hz); 4.31 (d, 1H, H-5a, Jgem¼14.5 Hz);
4.07 (m, 2H, H-2a, H-2b); 3.48 (s, 3H, OMe). 13C NMR
(75 MHz, CDCl3) d 158.01 (C-10); 131.57 (C-8); 131.52
(C-11); 131.15 (C-6); 121.85 (C-9); 114.98 (C-7); 101.01
(C-3); 72.16 (C-2); 62.24 (C-5); 55.61 (OMe). HR LSIMS
calcd for C10H11O3NaBr (MþNa)þ 280.9789, found:
280.9794. Anal. for C10H11BrO3: calcd: C 46.36; H 4.28.
Found: C 46.40; H 4.19.
1-(Hydroxymethyl)-5-chloro-2-phenoxyacetaldehyde
dimethyl acetal 4d. Yield: 34%. Rf (2/1, diethyl ether/
hexane): 0.4. H NMR (400 MHz, CDCl3) d 7.27 (d, 1H,
1
H-6; J¼2.5 Hz); 7.21 (dd, 1H, H-4; J¼0 2.5, 8.6 Hz); 6.80 (d,
1H, H-3, J¼8.6 Hz); 4.72 (t, 1H0, H-1 , J¼5.2 Hz); 4.63 (s,
2H, CH2OH); 4.04 (d, 2H, H-2 , J¼5.2 Hz); 3.45 (s, 6H,
(OMe)2); 3.12 (bs, 1H, CH2OH). 13C NMR (100 MHz,
CDCl3) d 155.13 (C-2); 131.91 (C-5); 128.89 (C-4); 128.44
(C-60); 126.37 (C-1); 113.37 (C-3); 101.73 (C-10); 68.04
(C-2 ); 61.38 (CH2OH); 54.15 ((OMe)2). HR LSIMS calcd
for, C11H15O4NaCl (MþNa)þ 269.0556, found 269.0555.
Anal. for C11H15O4Cl: calcd: C 53.56; H 6.13. Found: C
54.00; H 6.15.
(RS)-3-Methoxy-7-nitro-2,3-dihydro-5H-1,4-benzodioxepin
5f. Yield: 24%. Rf (2/1, diethyl ether/hexane): 0.6. 1H NMR
(300 MHz, CDCl3) d 7.99 (dd, 1H, H-8, J¼2.7, 8.8 Hz);
7.92 (d, 1H, H-6, J¼2.7 Hz); 6.96 (d, 1H, H-9, J¼8.8 Hz);
5.25 (d, 1H, H-5b, Jgem¼15.0 Hz); 4.94 (t, 1H, H-3a,
J¼5.8 Hz); 4.37 (d, 1H, H-5a, Jgem¼15.0 Hz); 4.17 (m,
1-(Hydroxymethyl)-5-bromo-2-phenoxyacetaldehyde
dimethyl acetal 4e. Yield: 44%. Rf (2/1, diethyl ether/
1
hexane): 0.37. H NMR (300 MHz, CDCl3) d 7.37 (d, 1H,
H-6, J¼2.5 Hz); 7.33 (dd, 1H, H-4, J¼2.5, 8.6 Hz); 6.73 (d,

8024
E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
2-H, H-2a, H-2b); 3.43 (s, 3H, OMe). 13C NMR (75 MHz,
CDCl3) d 163.78 (C-10); 142.54 (C-7); 129.05 (C-11);
124.51 (C-8); 120.58 (C-6); 100.50 (C-9); 100.48 (C-3);
71.28 (C-2); 61.91 (C-5); 55.64 (C-3-OMe). HR LSIMS
calcd for C10H11NO5Na (MþNa)þ 248.0534, found:
248.0538. Anal. for C10H11NO5: calcd: C 53.33; H 4.92;
N 6.22. Found: C 53.52; H 5.31; N 6.34.
331.0705. Anal for C14H13N2O5F: C, 54.55; H, 4.25; N,
9.09. Found: C 54.21; H 3.98; N 9.28.
(RS)-1-[2-(2-Hydroxymethyl-6-methoxyphenoxy)-1-methoxy-
ethyl]-5-fluorouracil 6c. See Table 1, entry 2. White solid,
mp 130–1338C. Rf (9.5/0.5, CH2Cl2/MeOH): 0.4. 1H NMR
(300 MHz, CDCl3) d 10.46 (s, 1H, NH); 7.5 (d, 1H, H5-FU
,
JH–F¼5.8 Hz); 6.96 (d, 1H, H-5, J¼7.7 Hz); 6.86 (dd, 1H,
H-6, J¼1.6, 7.7 Hz); 6.79 (dd, 1H, H-4, J¼1.6, 8.1 Hz);
5.88 (dt, 1H, H-10, J¼1.6, 4.7 Hz); 4.59 (d, 2H, CH2OH,
J¼1.5 Hz); 4.19 (q, 2H, H-20, J¼5.3 Hz); 3.75 (s, 3H, C-3-
OMe); 3.42 (s, 3H, OMe). 13C NMR (75 MHz, CDCl3) d
157.33 (d, C-45-FU; 34.9); 151.88 (s, C-25-FU); 150.32 (C-3);
144.94 (C-2); 140.93 (d, C-55-FU; J¼315 Hz); 134.35 (C-1);
124.56 (C-5); 124.12 (C-65-FU; 44.6); 121.16 (C-6); 112.04
(C-4); 85.03 (C-10); 71.77 (C-20); 60.76 (CH2OH); 57.19
(C-3-OMe); 55.56 (OMe). HR LSIMS calcd for C15H17N2-
O6FNa (MþNa)þ 363.0968, found 363.0972. Anal for
C15H17N2O6F: C 52.94; H 5.04; N 8.23. Found: C 53.22; H
5.33; N 8.44.
4.1.2. Final products
4.1.2.1. Reaction between the o-(hydroxymethyl)
phenoxyacetaldehyde dimethyl acetals 4 and 5-Fluoro-
uracil. General procedure. A 1.0 M solution of SnCl4/
CH2Cl2 (1.25 mmol) was added dropwise to a suspension of
4 (1 mmol) and 5-fluorouracil (1 mmol), which contains
trimethylchlorosilane (TCS, 0.8 mmol) and 1,1,1,3,3,3-
hexamethyldisilazane (HMDS, 0.8 mmol) in dry aceto-
nitrile (10 ml/mmol of 4), with stirring under argon at rt.
After stirring, the reaction was quenched by the addition of a
concentrated aqueous solution of Na2CO3. The solvent was
removed with a rotary evaporator. The sticky residue was
dissolved in MeOH, and silica gel was added. The solvent
was removed with a rotary evaporator, and the residue was
applied to the top of a flash chromatography column packed
with CH2Cl2/MeOH (100/1). Elution with mixtures of
CH2Cl2/MeOH by gradient elution (100/1!100/5) and
concentration with a rotary evaporator gave the target
molecules 3 and 6. The yields of the different compounds
are shown in Table 1 and Scheme 3.
(RS)-1-[2-(2-Hydroxymethylphenoxy)-1-methoxyethyl]-5-
fluorouracil 6a. See Table 3, entry 1. White solid, mp 100–
1058C. Rf (9.5/0.5, CH2Cl2/MeOH): 0.57. 1H NMR
(300 MHz, CDCl3) d 10.61 (s, 1H, NH); 7.51 (d, 1H, H5-FU
,
JH–F¼5.8 Hz); 7.23 (dd, 1H, HAr, J¼1.6, 7.4 Hz); 7.16 (dd,
1H, HAr, J¼1.7, 7.8 Hz); 6.88 (t, 1H, HAr, J¼7.3 Hz); 6.74
(d, 1H, HAr, J¼8.1 Hz); 5.90 (dt, 1H, H-10, J¼1.5, 4.3 Hz);
4.54 (s, 2H, CH2OH); 4.15 (dd, 1H, H-20, J¼4.2, 10.2 Hz);
C
(RS)-1-[2-(2-Hydroxymethyl-4-methoxyphenoxy)-1-methoxy-
ethyl]-5-fluorouracil 6b. See Table 1, entry 1. White
solid, mp 858C. Rf (9/1, CH2Cl2/MeOH): 0.45. H NMR
(300 MHz, CDCl3) d 10.49 (s, 1H, NH); 7.5 (d, 1H, H5-FU
4.03 (dd, 1H, H-20, J¼5.8, 10.2 Hz); 3.41 (s, 3H, OMe). 13
NMR (75 MHz, CDCl3) d 157.20 (d, C-45-FU; J¼26.45 Hz);
1
155.42 (C-1); 150.30 (s, C-25-FU); 141.10 (d, C-55-FU
;
,
J¼239.16 Hz); 129.42 (C-2); 129.02, 128.73 (C-5 y C-3);
123.16 (d, C-65-FU; J¼33.72 Hz); 121.55 (C-4); 111.13
(C-6); 84.66 (C-10); 67.54 (C-20); 60.72 (PhCH2OH); 57.26
(OMe). HR LSIMS calcd for C14H15N2O5FNa (MþNa)þ
310.1034, found 310.0965. Anal. for C14H15N2O5F: calcd:
C 54.19; H 4.87; N 9.03. Found: C 54.02; H 4.93; N 9.23.
JH–F¼5.75 Hz); 6.85 (d, 1H, H-6, J¼1.6 Hz); 6.69 (d,
2H, H-4, H-3, J¼1.6 Hz); 5.89 (dt, 1H, H-10, J¼1.5,
5.6 Hz); 4.55 (s, 2H, CH2OH); 4.12 (dd, 2H, H-20,
J¼5.6, 10.2 Hz); 3.69 (s, 3H, C-5-OMe); 3.42 (s, 3H,
OMe). 13C NMR (75 MHz, CDCl3) d 157.19 (d, C-45-FU
;
35.1); 154.3 (s, C-25-FU); 150.32 (C-5); 149.5 (C-2);
141.13 (d, C-55-FU; J¼316 Hz); 130.81 (C-1); 123.21
(C-65-FU; 44.5); 114.85 (C-4); 112.63 (C-6); 84.76 (C-10);
68.40 (C-20); 60.78 (CH2OH); 57.30 (OMe); 55.62 (C-5-
OMe). HR LSIMS calcd for C15H17N2O6FNa (MþNa)þ
363.0968, found 363.0969. Anal. for C15H17N2O6F:
calcd: C 52.94; H 5.04; N 8.23. Found: C 51.78; H
4.69; N 8.44.
(RS)-trans-4,10-Bis[5-(fluorouracil-1-yl)]-2,5,8,11-tetraoxa-
1,7(1,2)dibenzenadodecaphane 7a. See Scheme 3. Yellow-
ish solid, mp 80–838C. Rf (9.5/0.5, CH2Cl2/MeOH): 0.6; Rf.
1H NMR (300 MHz, DMSO-d6) d 7.85 (d, 1H, H5-FU
,
JH–F¼5.5 Hz); 7.28 (t, 2H, H-3 and H-12, J¼6.5 Hz); 7.27
(d, 2H, H-1 and H-10, J¼5.9 Hz); 7.06 (t, 2H, H-2 and H-11,
J¼6.6 Hz); 7.04 (d, 2H, H-4 and H-13); 6.12 (d, 2H, H-7
and H-16, J¼7.9 Hz); 4.95 (d, 2H, H-9 and H-18, Jgem¼13.9
Hz); 4.76 (dd, 2H, H-6 and H-15, J¼9, 12.5 Hz); 4.68 (d,
2H, H-9 and H-18, J¼13.9 Hz); 4.41 (dd, 2H, H-6 and H-15,
J¼1.4, 12.6 Hz). 13C NMR (100 MHz, DMSO-d6) d 158.90
(C-20 and C-22); 157.17 (d, C-45-FU; J¼25.41 Hz); 149.40
(s, C-25-FU); 139.2 (d, C-55-FU; J¼224.67 Hz); 131.08 (C-19
and C-21); 129.53, 129.49 (C-1, C-10, C-3 and C-12);
125.92 (d, C-65-FU; 31.51); 123.39 (C-4 and C-13); 120.31
(C-2 and C-11); 84.50 (C-7 and C-16); 73.04 (C-9 and
C-18); 70.05 (C-6 and C-15). HR LSIMS calcd for
C26H22N4O8F2Na (MþNa)þ 579.1303, found 579.1305.
Anal. for C26H22N4O8F2: calcd: C 56.12; H 3.98; N 10.07.
Found: C 55.93; H 4.12; N 10.02.
(RS)-3-(7-Methoxy-2,3-dihydro-5H-1,4-benzodioxepin-3-
yl)-5-fluorouracil 3bN-3. See Table 1, entry 1. White solid,
1
mp 235–2378C. Rf (9/1, CH2Cl2/MeOH): 0.48. H NMR
(400 MHz, DMSO-d6) d 11.16 (d, 1H, NH, J¼5.3 Hz); 7.85
(t, 1H, H5-FU, JH–F¼5.6 Hz); 6.97 (d, 1H, H-9, J¼8.7 Hz);
6.88 (d, 1H, H-6, J¼3.1 Hz); 6.80 (dd, 1H, H-8, J¼3.1,
8.7 Hz); 6.05 (d, 1H, H-3a, J¼8.1 Hz); 4.81 (d, 1H, H-5a,
J¼13.7 Hz); 4.65 (d, 1H, H-5b, J¼13.7 Hz); 4.60 (dd, 1H,
H-2a, J¼8.1, 12.4 Hz); 4.35 (dd, 1H, H-2b, J¼1.5,
12.4 Hz); 3.7 (s, 3H, C-7-OMe). 13C NMR (100 MHz,
DMSO-d6) d 157.10 (d, C-45-FU; 25.5); 154.99 (C-7);
152.74 (s, C-25-FU); 149.34 (C-10); 139.18 (d, C-55-FU
;
J¼224.5 Hz); 132.43 (C-11); 125.90 (C-65-FU; 31.4);
121.17 (C-8); 114.5 (C-9); 114.16 (C-6); 84.91 (C-3);
73.52 (C-2); 70.32 (C-5); 55.43 (C-7-OMe). HR LSIMS
calcd for C14H13N2O5FNa (MþNa)þ 331.0706, found
(RS)-1-[2-(4-Chloro-2-hydroxymethylphenoxy)-1-methoxy-
ethyl]-5-fluorouracil 6d. See Table 1, entry 3. White solid,
mp 115–1108C. Rf (9/1, CH2Cl2/MeOH): 0.73. H NMR
1

E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
8025
(300 MHz, DMSO-d6) d 11.93 (s, 1H, NH); 8.00 (d, 1H,
H5-FU, JH–F¼6.8 Hz); 7.32 (d, 1H, H-6, J¼2.6 Hz); 7.23
(dd, 1H, H-4, J¼2.6, 9.4 Hz); 6.97 (d, 1H, H-3, J¼9.4 Hz);
5.79 (dt, 1H, H-10, J¼1.3, 4.7 Hz); 5.18 (t, 1H, CH2OH,
J¼5.4 Hz); 4.35 (d, 1H, CH2OH, J¼5.4 Hz); 4.29 (dd, 1H,
H-20, J¼5.2, 10.5 Hz); 4.18 (dd, 1H, H-20, J¼6.2, 10.5 Hz);
3.33 (s, 3H, OMe). 13C NMR (75 MHz, DMSO-d6) d 157.02
(d, C-45-FU; 34.7); 152.86 (s, C-25-FU); 149.79 (C-2); 143.39
(d, C-55-FU; J¼307 Hz); 133.02 (C-5); 126.79 (C-4); 126.12
(C-6); 124.61 (C-1); 124.39 (C-65-FU; 44.9); 113.01 (C-3);
83.47 (C-10); 66.79 (C-20); 57.05 (CH2OH); 56.26 (OMe).
HR LSIMS calcd for C14H14N2O5FNaCl (MþNa)þ
367.0473, found 367.0472. Anal. for C14H14N2O5FCl:
calcd: C 48.78; H 4.09; N 8.13. Found: C 48.53; H 3.84;
N 7.99.
calcd for C14H14N3O7FNa (MþNa)þ 378.0713, found
378.0708. Anal. for C14H14N3O7F: calcd: C 47.33; H
3.97; N 11.83. Found: C 46.98; H 4.02; N 11.98.
4.1.2.2. Reaction between the benzoannelated seven-
membered compounds 5 and 5-Fluorouracil. General
procedure. The procedure was similar to the one described
in Section 4.1.2.1 but changing 4 for 5. The yields and the
nature of the compounds are shown in Table 2. Compounds
3a–c have been previously reported by us.12
4.1.2.3. Reduction of 6f: obtention of (RS)-3-[2-(4-
amino-2-hydroxymethylphenoxy)-1-methoxyethyl]-5-
fluorouracil 6g. After dissolving 6f (1.7 g, 4.8 mmol) in
ethanol (24 mL) and adding SnCl2 2H2O (5.41 g, 24 mmol),
the suspension was refluxed for 2 h. After cooling, the
mixture was neutralized, rotaevaporated off and the
resulting crude was purified by flash chromatography (10/
1, CH2Cl2/MeOH) to give 6g (1 g, 65%). Brown solid, mp
100–1038C. Rf (9/1, CH2Cl2/MeOH): 0.3. 1H NMR
(RS)-1-[2-(4-Bromo-2-hydroxymethylphenoxy)-1-methoxy-
ethyl]-5-fluorouracil 6e. See Table 1, entry 4. White solid,
mp 748C. Rf (10/0.25, CH2Cl2/MeOH): 0.16. 1H NMR
(300 MHz, CDCl3) d 7.47 (d, 1H, H5-FU, JH–F¼5.6 Hz);
7.37 (d, 1H, H-6, J¼2.5 Hz); 7.31 (dd, 1H, H-4, J¼2.5,
8.7 Hz); 6.64 (d, 1H, H-3, J¼8.7 Hz); 5.94 (dt, 1H, 0H-10,
J¼1.6, 4.0 Hz); 4.5 (s, 2H, CH2OH); 4.18 (dd, 1H, H-2 , 4.0,
10.0); 4.01 (dd, 1H, H-20, J¼6.3, 10.0 Hz); 3.44 (s, 3H,
(400 MHz, CD3COCD3) d 7.69 (d, 1H, H5FU, JH–F
¼
5.2 Hz); 6.86 (d, 1H, H-3, J¼8.4 Hz); 6.76 (d, 1H, H-6,
J¼2.5 Hz); 6.48 (dd, 1H, H-4, J¼2.5, 8.4 0Hz); 6.22 (t,
1H, H-10, J¼5.8 Hz); 4.55 (m, 4H, CH2OH, H-2 ); 3.41 (s, 3H,
OMe). 13C NMR (100 MHz, CD3COCD3) d 157.12 (s,
C-45-FU); 152.09 (s, C-25-FU); 146.37 (C-2); 140.45 (d,
C-55-FU; J¼230.2 Hz); 132.27 (C-5); 132.20 (C-1);
125.40 (C-65-FU; J¼32.2 Hz); 119.71 (C-4); 119.02 (C-6);
112.94 (C-3); 84.88 (C-10); 68.95 (C-20); 59.69 (CH2OH);
57.36 (OMe). HR LSIMS calcd for C14H16N3O5FNa
(MþNa)þ 348.0971, found 348.0980. Anal. for
C14H16N3O5F: calcd: C 51.69; H 4.96; N 12.92. Found: C
51.88; H 5.32; N 13.21.
OMe). 13C NMR (75 MHz, CDCl3) d 157.15 (d, C-45-FU
;
J¼35 Hz); 154.44 (s, C-25-FU); 150.40 (C-2); 141.31 (d,
C-55-FU; J¼316.9 Hz); 131.75 (C-4); 131.70 (C-6); 122.91
(C-65-FU; 44.5); 114.08 (C-1); 112.83 (C-3); 84.67 (C-10);
67.82 (C-20); 60.29 (CH2OH); 57.54 (OMe). HR LSIMS
calcd for C14H14N2O5FNaBr (MþNa)þ 410.9956, found
410.9941. Anal. for C14H14N2O5FBr: calcd: C 43.21; H
3.63; N 7.20. Found: C 43.09; H 3.26; N 7.48.
4.2. Biological activity
(RS)-3-[2-(4-Bromo-2-hydroxymethylphenoxy)-1-methoxy-
ethyl]-5-fluorouracil 6eN-3. See Table 1, entry 4. White
solid, mp 105–1088C. Rf (10/0.25, CH2Cl2/MeOH): 0.13.
The antiproliferative activities26 and apoptosis induction25
in the MCF-7 human breast cancer cell line were followed
in accordance with the protocols previously reported.
1H NMR (300 MHz, CD3OD) d 7.61 (d, 1H, H5-FU
,
JH–F¼5.1 Hz); 7.49 (d, 1H, H-6, J¼2.5 Hz); 7.33 (dd, 1H,
H-4, J¼2.5, 8.7 Hz); 6.86 (d, 1H, H-3, J¼8.7 Hz); 6.25 (t,
1H, H-10, J¼5.9 Hz); 4.89 (s, 2H, CH2OH); 4.55 (dd,
2H, H-20, J¼5.9, 18.5 Hz); 3.44 (s, 3H, OMe). 13C NMR
(75 MHz, CD3OD) d 155.66 (s, C-45-FU); 139.93 (C-2);
4.2.1. Cell cycle distribution analysis. Cells at 70%
confluence were treated with either DMSO alone or with
concentrations of the compounds determined by their IC50
values. After 48 h of treatment the medium was aspirated,
the cells were harvested, washed twice with sample buffer
(1 g glucose, 1 L PBS without Ca2þ or Mg2þ) and fixed in
70% (vol/vol) cold ethanol for up to one week. Cells were
pelleted, washed once with sample buffer and resuspended
in PI solution (50 mg/mL PI, 0.5 mg/mL RNase in sample
buffer, pH 7.4) for 30 min in the dark. Then, fluorescence-
activated cell sorter analysis was performed and the data
from 10,000 cells per sample were collected and analyzed
using a VANTAGE flow cytometer (Becton Dickinson, San
Jose, CA, USA) by CellFIT cell cycle analysis software. All
experiments were performed in duplicate and yielded
similar results.
133.87 (s, C-25-FU); 132.39 (C-1); 131.86 (d, C-55-FU
;
;
J¼401.1 Hz); 131.76 (C-4); 131.19 (C-6); 126.57 (C-65-FU
32.1); 114.50 (C-3); 85.45 (C-5); 85.45 (C-10); 68.70 (C-20);
59.40 (CH2OH); 57.78 (OMe). HR LSIMS calcd for
C14H14N2O5FNaBr (MþNa)þ 410.9967, found 410.9965.
Anal. for C14H14N2O5FBr: calcd: C 43.21; H 3.63; N 7.20.
Found: C 43.34; H 3.86; N 6.98.
(RS)-1-[2-(2-Hydroxymethyl-4-nitrophenoxy)-1-methoxy-
ethyl]-5-fluorouracil 6f. See Table 1, entry 5. Yellow solid,
mp 230–2338C. Rf (9.5/0.5, CH2Cl2/MeOH): 0.8. 1H NMR
(400 MHz, DMSO-d6) d 11.6 (bs, 1H, NH); 8.23 (d, 1H,
H-6, J¼2.8 Hz); 8.16 (dd, 1H, H-4, J¼2.8, 7.5 Hz); 8.03 (d,
1H, H-3, J¼7.5 Hz); 7.76 (d, 1H, H5FU, JH–F¼6.08 Hz);
5.89 (t, 1H, H-10, J¼5.0 Hz); 5.45 (bs, 1H, CH2OH); 4.43
(m, 4H, CH2OH, H-20); 3.36 (s, 3H, OMe). 13C NMR
(100 MHz, DMSO-d6) d 157.12 (d, C-45-FU; 26); 150.06 (s,
C-25-FU); 149.86 (C-2); 140.45 (d, C-55-FU; J¼230.2 Hz);
141.21 (C-5); 132.32 (C-1); 124.38 (C-65-FU; J¼33.5 Hz);
123.99 (C-04); 121.69 (C-6); 111.60 (C-3); 83.37 (C-10);
67.33 (C-2 ); 57.01 (CH2OH); 56.44 (OMe). HR LSIMS
The results of the antiproliferative activities, cell cycle
distribution and apoptosis induction are recorded in
Table 4.
Acknowledgements
This study was supported by the Instituto Carlos III (Fondo

8026
E. Saniger et al. / Tetrahedron 59 (2003) 8017–8026
13. Meier, C.; De Clercq, E.; Balzarini, J. Eur. J. Org. Chem.
1998, 837–846.
de Investigaciones Sanitarias, 01/0928). E.S. was supported
´
by a fellowship from the Ministerio de Educacion, Cultura y
Deporte.
14. Ozaki, S.; Watanabe, Y.; Hoshiko, T.; Nagase, T.; Ogasawara,
T.; Furukawa, H.; Uemura, A.; Ishikawa, K.; Mori, H.; Hoshi,
A.; Iigo, M.; Tokuzen, R. Chem. Pharm. Bull. 1986, 34,
150–157.
References
15. Phane Nomenclature. Part I: Phane Parent Names. IUPAC
Recommendations, 1998. http://www.chem.qmw.ac.uk/iupac/
phane.
1. Richard, M. A.; Westcombe, A. M.; Love, S. B.; Littlejonhs,
P.; Ramirez, A. J. Lancet 1999, 353, 1119–1126.
2. Norton, L. Semin. Oncol. 1999, 26, 1–4.
16. Favre, H. A.; Powell, W. H. Phane Nomenclature. Part II.
Modification of the Degree of Hydrogenation and Substitution
Derivatives of Phane Parent Hydrides. IUPAC Recommen-
dations 2002. http://www.chem.qmw.ac.uk/iupac/phane.
17. SYBYL Molecular Modeling Software; Tripos Inc.:1699
S. Hanley Road, St Louis, MO 63144-2913. www.tripos.com.
18. Powell, M. J. D. Math. Program. 1997, 12, 241–254.
19. (a) Gasteiger, J.; Marsili, M. Tetrahedron 1980, 36,
3219–3228. (b) Marsili, M.; Gasteiger, J. Croat. Chem. Acta
1981, 53, 601–614. (c) Gasteiger, J.; Marsili, M. Org. Magn.
Reson. 1981, 15, 353–360.
3. Morrow, M.; Jordan, V. C.; Takei, H.; Gradishar, W. J.; Pierce,
L. J. Curr. Probl. Surg. 1999, 36, 163–216.
4. Ruppert, J. M.; Wright, M.; Rosenfeld, M.; Grushcow, G.;
Bilbao, G.; Curiel, D. T.; Strong, T. V. Breast Cancer Res.
Treat. 1997, 44, 93–114.
5. Coultas, L.; Strasser, A. Apoptosis 2000, 5, 491–507.
6. Rasbridge, S. A.; Gillet, C. E.; Seymour, A. M.; Patel, K.;
Richards, M. A.; Rubens, R. D.; Millis, R. R. Brit. J. Cancer
1994, 70, 335–341.
´
20. Domınguez, J. F.; Marchal, J. A.; Correa, A.; Carrillo, E.;
´
7. Malet-Martino, M.; Jolimaitre, P.; Martino, R. Curr. Med.
Chem.-Anti-Cancer Agents 2002, 2, 267–310.
´
8. Campos, J.; Pineda, M. J.; Gomez, J. A.; Entrena, A.; Trujillo,
´
Boulaiz, H.; Aranega, A.; Gallo, M.A.; Espinosa, A. Bioorg.
Med. Chem. 2003, 11, 315–323.
´
M.A.; Gallo, M.A.; Espinosa, A. Tetrahedron 1996, 52,
´
21. Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734–736.
22. Sampath, D.; Plunkett, W. Curr. Opin. Oncol. 2001, 13,
484–490.
8907–8924.
´
9. Gomez, J. A.; Campos, J.; Marchal, J. A.; Trujillo, M.A.;
´
´
Melguizo, C.; Prados, J.; Gallo, M.A.; Aranega, A.; Espinosa,
´
23. Gali-Muhtasib, H.; Bakkar, N. Curr. Cancer Drugs Targets
2002, 2, 309–336.
A. Tetrahedron 1997, 53, 7319–7334.
´
10. Gomez, J. A.; Trujillo, M.A.; Campos, J.; Gallo, M.A.;
´
´
24. Saunders, D. E.; Lawrence, W. D.; Christensen, C.; Wappler,
N. L.; Ruan, H.; Deppe, G. Int. J. Cancer 1997, 70, 214–220.
25. Chadderton, A.; Villeneuve, D. J.; Gluck, S.; Kirwan-Rhude,
A. F.; Gannon, B. R.; Blais, D. E.; Parissenti, A. M. Breast
Cancer Res. Tr. 2000, 59, 231–244.
Espinosa, A. Tetrahedron 1998, 54, 13295–13312.
´
´
´
11. Trujillo, M.A.; Gomez, J. A.; Campos, J.; Gallo, M.A.;
Espinosa, A. Tetrahedron 2001, 57, 3951–3961.
12. Saniger, E.; Campos, J. M.; Entrena, A.; Marchal, J. A.;
´
26. Campos, J.; Nun˜ez, M. C.; Rodrıguez, V.; Gallo, M.A.;
´ ´ ´
Suarez, I.; Aranega, A.; Choquesillo, D.; Niclos, J.; Gallo,
´
´
´
M.A.; Espinosa, A. Tetrahedron 2003, 59, 5457–5467.
Espinosa, A. Bioorg. Med. Chem. Lett. 2000, 10, 767–770.