1,2,4-Triazine N-oxide derivatives: Studies as potential hypoxic cytotoxins. Part II

Article abstract of DOI:10.1002/ardp.200300782

New 1,2,4-Triazine N-oxide and N,N′-dioxide derivatives were synthesized in order to obtain compounds as selective hypoxic cell cytotoxins. The starting heterocycles have been prepared using a standard microwave oven in a clean and good-yielded process. The reactivity of methyl-1,2,4-triazine N ₄-oxide and N₁,N₄-dioxide with different electrophilic agents has been studied. The desired products were obtained only when iminium electrophiles were employed. The regioselectivity of this process has been studied by means of experimental and theoretical (at ab initio level) procedures. Theoretically was expected that the most stable intermediates where the benzylic-like anion from position 5. A fact which agreed with the experimental observed regioselectivity. The new compounds were tested for their cytotoxicity in oxia and hypoxia. Some of them proved to be less active in hypoxic conditions than tirapazamine, 3-amino-benzo[1,2-e]1,2,4-triazine N ₁,N₄-dioxide. Derivative 19, 6-methyl-5-[2-(5- nitrothienyl)ethenyl)-1,2,4-triazine N₄-oxide, was the most cytotoxic compound, but it was non-selective.

Full text of DOI:10.1002/ardp.200300782

Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Potential Hypoxic Cytotoxins 247
Hugo Cerecettoa,
1,2,4-Triazine N-oxide Derivatives: Studies as
Potential Hypoxic Cytotoxins. Part II.
a
´
Mercedes Gonzalez ,
Silvia Onettoa,
Patricia Saenza,
Olga Ezpeletab,
New 1,2,4-Triazine N-oxide and N,NЈ-dioxide derivatives were synthesized in
order to obtain compounds as selective hypoxic cell cytotoxins. The starting
heterocycles have been prepared using a standard microwave oven in a clean
and good-yielded process. The reactivity of methyl-1,2,4-triazine N4-oxide and
N1,N4-dioxide with different electrophilic agents has been studied. The desired
products were obtained only when iminium electrophiles were employed. The
regioselectivity of this process has been studied by means of experimental and
theoretical (at ab initio level) procedures. Theoretically was expected that the
most stable intermediates where the benzylic-like anion from position 5. A fact
which agreed with the experimental observed regioselectivity. The new com-
pounds were tested for their cytotoxicity in oxia and hypoxia. Some of them
proved to be less active in hypoxic conditions than tirapazamine, 3-amino-
benzo[1,2-e]1,2,4-triazine N1,N4-dioxide. Derivative 19, 6-methyl-5-[2-(5-nitrothi-
enyl)ethenyl)-1,2,4-triazine N4-oxide, was the most cytotoxic compound, but it
was non-selective.
b
´
´
Adela Lopez de Cerain ,
Antonio Mongeb
a
´
Departamento de Quımica
´
Organica, Facultad de
´
Quımica Ϫ Facultad de
Ciencias, Universidad de la
´
Republica,
Montevideo, Uruguay
b Unidad en I+D de
Medicamentos, CIFA,
Universidad de Navarra,
˜
Pamplona, Espana
Keywords: 1,2,4-Triazine N-oxide; Bioreductive compounds; Iminium electro-
phile
Received: January 29, 2003; Accepted: April 29, 2003 [FP782]
DOI 10.1002/ardp.200300782
of the enhanced 5-methyl-nucleophile, promoted by
the N-oxide neighbor moiety [10-13]. Regioselectivity
is an interesting target in the heterocyclic chemistry,
specially, when it is possible to modify it with an auxili-
ary moiety, i.e. N-oxide [14Ϫ17].
Introduction
Several N-oxides of aromatic heterocyclic amines
have been described as hypoxia-selective cytotoxic
agents [1]. In this sense we have synthesized and bio-
logically evaluated 1,2,5-oxadiazole N-oxide deriva-
tives [2Ϫ5]. Recently, we have described the biological
characterization of 1,2,4-triazine N4-oxide and N1,N4-
dioxide derivatives as hypoxia selective cytotoxins [6].
These derivatives showed poor activities against V79
cells in oxic and hypoxic conditions, so in attempts to
obtain more active compounds we intend to include
in the 3-position of the heterocycle an alkylaminoalkyl
substituent, a residue present in well known hypoxic-
cytotoxins [7, 8]. On this way we chose the Mannich-
iminium electrophile as reactant and triazine 1 as the
heterocyclic-nucleophile (Scheme 1) [9]. However, the
N-oxide moiety, as we thought when we designed the
synthetic route, did not increase the heterocycle nucle-
ophilicity in 3-position. In the two cases assayed, the
compounds obtained were exclusively the 5-methyl-
nucleophilic condensation products (2 and 3, Scheme
1) [6]. These compounds could be seen as the result
In this paper, we report the study of the reactivity of
some selected triazine N-oxide derivatives with differ-
ent electrophiles, with the aim to obtain a new and
more effective 1,2,4-triazine N-oxide bioreductive sys-
tem. We determine the influence of the N-oxide moiety
over the methyl nucleophilicity in 1,2,4-triazine hetero-
cycle toward different electrophilic reactants. We also
present the theoretical study of different intermediates
generated in the process, in order to explain the ob-
served reactivity. Furthermore, we analyzed the cyto-
toxic activity of these new derivatives in oxia and hy-
poxia against V79 cells.
Results and discussion
Chemistry
Our approach to study the N-oxide influence into the
methyl-reactivity involved, firstly, the use of substituted
1,2,4-triazines that could contain methyl groups in dif-
ferent positions respect to N-oxide moiety. Therefore,
we chose triazines 1, 4, and 5 as the starting heterocy-
cle for the condensation studies (Scheme 2). The tria-
zine N4-oxide rings (compounds 1, and 4) were ob-
´
Correspondence: Mercedes Gonzalez, Departamento de
´
´
´
Quımica Organica, Facultad de Quımica Ϫ Facultad de
´
´
Ciencias, Universidad de la Republica, Igua 4225, 11400
Montevideo, Uruguay. Phone: +598 2 525-86-18 (ext. 216),
Fax: +598 2 525-07-49, e-mail: megonzal@fq.edu.uy
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

´
248 Gonzalez et al.
Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Scheme 1.
Scheme 2.
tained using twoϪstep classical methods depicted in
Scheme 2A [18, 19]. We also used cyclization pro-
moted by microwave energy, irradiating the mixture of
the corresponding α-hydrazono oxime (I) and trimethyl
orthoformate or trimethyl orthoacetate in presence of
catalytic amounts of p-toluenesulfonic acid (p-TsOH)
in an unmodified household microwave oven [20]. The
different conditions assayed for compound 1 are gath-
ered in Table 1. This methodology results a very con-
venient and useful method for the preparation of 1,2,4-
triazine N4-oxide. The heterocycles 1 and 4 were ob-
tained in a very clean process (decomposition pro-
moted by acid medium in the classical method was
eliminated) and an increase in the yield of isolated
products was obtained. The triazine N1,N4-dioxide, 5,
was prepared by reaction of 1 with urea-hydrogen per-
oxide complex (UHP) as oxidizing reagent (Scheme
2B) [6, 21]. Preparation of compound 5, from 1, was
tried in solid-state [22] and via microwave irradiation
under solventless conditions with UHP. In both con-
ditions the product 5 was not obtained.
Firstly, condensation of triazine 1 with different electro-
philic reactants was assayed (Scheme 3). For the imi-
nium electrophile we used the conditions previously
described [6] for the Mannich cations of morpholine
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Potential Hypoxic Cytotoxins 249
Table 1. Experimental conditions assayed in the microwaves-promoted cyclization of oxime I to obtain
compound 1.
Reagents
Catalyst
Period of irradiation (sec)†
Total time (min)
Yield (%)‡,§
I (50 mg),
HC(OMe)3 (0.5 mL)
p-TsOH
p-TsOH
p-TsOH
p-TsOH
K 10d
30
60
30
60
60
1
3
4
4
9
80
I (50 mg),
HC(OMe)3 (0.5 mL)
60
I (50 mg),
HC(OMe)3 (0.5 mL)
30
Decomposition
NR
I (50 mg),
HC(OMe)3 (0.5 mL)
I (50 mg),
HC(OMe)3 (0.5 mL)
I (50 mg),
HC(OMe)3 (0.5 mL),
Ϫ
60
5
Traces
adsorbed in SiO2
†
The mixture of reaction was irradiated during the period of time indicated, each period was separated by 1Ϫ2
min of absence of irradiation. ‡ Isolated yield of compound 1 after chromatographic purification. § Yield expressed
considering 100% of conversion of oxime stereoisomer reactive (Z-hydrazono-E-oxime). # Montmorillonite K 10.
NR Ϫ no reaction.
and piperidine, where derivatives 6 and 7 were ob-
tained as unique products (Scheme 3A). The iminium
cation was generated in situ when the corresponding
amine and formalin were stirred in ethanol at room
temperature [23]. Same results (yield and time of reac-
tion) for derivative 6 were obtained when we gener-
ated the iminium-electrophile from a mixture of par-
aformaldehyde and morpholine in ethanol at reflux, fol-
lowed by reaction with 1 at room temperature [13, 24].
The process was repeated, with the same results, be-
tween triazines 4 and 5, and all the iminium electro-
philes. The mixture of the iminium and the triazines
(1, 4, and 5) was stirred at room temperature until the
triazine was not present (24Ϫ72 h) (see Table 2). In
all cases, only condensation products were obtained,
6-15, as shown in Scheme 3A.
The same kind of results was observed when we used
phenyl isothiocyanate [26] as electrophilic species in
the synthesis of thioamide 17 (Scheme 3B). In any
case the product 17 was obtained (see assayed ex-
perimental conditions and results in Table 3).
Benzaldehyde and 5-nitrothiophenecarboxaldehyde
were the last electrophilic reactants assayed in the
condensation reactions studied (Scheme 3C) [27, 28].
The different conditions used for the preparation of de-
rivative 18 are shown in Table 4. In strongly basic me-
dium, such as the process with the other electrophilic
species, the complete decomposition of the starting
material, 1, was observed. However, when we used
catalytic amounts of piperidine or morpholine, as base,
product 18 was obtained as a trace. Using the Man-
nich protocol, such as for preparation of derivatives
6Ϫ15, the condensation products 18 and 19 were
separated as a precipitate from the reaction medium.
As another electrophilic reactant we chose acetyl chlo-
ride, in attempts to obtain the ketone 16 (Scheme 3B)
[25]. However, complete decomposition of the triazine
1 was observed in the experimental assayed con-
ditions (basic medium at reflux in methylene chloride).
All new compounds were identified by IR, MS, 1H
NMR, 13C NMR (see Table 2 and Experimental Sec-
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

´
250 Gonzalez et al.
Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Scheme 3.
tion) and HETCOR experiments, and their purity es-
tablished by TLC and microanalysis. The unequivocal
condensation site, products 6Ϫ15, and 18Ϫ19, was
determined by means of HMQC for one-bond corre-
lation and from sequences of HMBC for long distance/
carbon correlation. Thus, 2’-CH correlated (HMBC) ex-
clusively with the heterocyclic quaternary 5-carbon,
while this was determined by its correlation (HMBC)
with 3-H-heterocycle (for compounds 6Ϫ7, 12Ϫ15,
and 18-19) (Figure 1). In derivatives 8Ϫ11 the HMBC-
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Potential Hypoxic Cytotoxins 251
1
Table 2. Yields, reaction times, some representative 13C NMR and H NMR data in deuteriochloroform of 6Ϫ15.
Compd.
n
R
X
Yield
Time
(h)§
13C NMR (δ, ppm)#
1H NMR (δ, ppm)#
(%)†,‡
C-3
C-5
C-6
3-CH3
6-CH3
6
7
8
0
0
0
0
0
0
1
1
1
1
H
H
O
CH2
bond
CH3N
O
CH2
bond
CH3N
O
42
20
54
22
35
40
42
20
40
22
24
72
72
48
48
72
24
24
24
48
149.09
149.00
159.02
159.04
159.05
158.95
147.37
147.35
147.37
147.31
147.07
147.10
145.79
146.27
146.14
146.42
153.85
153.93
153.76
155.43
158.88
158.90
157.20
157.14
157.10
157.29
141.70
141.58
141.50
141.75
Ϫ
Ϫ
2.75
2.74
2.72
2.71
2.71
2.68
2.54
2.52
2.46
2.53
CH3
CH3
CH3
CH3
H
H
H
H
2.75
2.76
2.75
2.71
Ϫ
Ϫ
Ϫ
Ϫ
9
10
11
12
13
14
15
CH2
† Yield for the condensation with iminium electrophile. ‡ Isolated yield after chromatographic purification. § For the
1
condensation reactions. # Characteristic 13C and H chemical shifts for parent compounds in CDCl3, 1: 148.83
(C-3), 146.44 (C-5), 159.09 (C-6), 2.50 (5-CH3), 2.70 ppm (6-CH3); 4: 158.76 (C-3), 144.48 (C-5), 156.71 (C-6),
2.75 (3-CH3), 2.48 ppm (5-CH3), 2.67 (6-CH3); 5: 147.61 (C-3), 152.87 (C-5), 141.51 (C-6), 2.51 (5-CH3),
2.56 ppm (6-CH3).
Table 3. Experimental conditions assayed in the experiments to obtain compound 17.
Conditions
THF (dry)
Temperature
Results
NR
r. t.
Reflux
K2CO3 (1 equiv.)/THF (dry)
(i-Pr)2NEt (1 equiv.)/THF (dry)
r. t.
r. t.
NR
NR
p-TsOH (catalytic amounts)/THF (dry)
Reflux
Decomposition
NR Ϫ no reaction.
correlations and the characteristic protonic chemical
shifts of 3-, 5-, and 6-methyl in heterocycle 4 allowed
us to assign the reaction site (see Table 2).
The mechanism of this process could be studied by
quantum-chemical calculations, thus the stability of the
different intermediates generate in each nucleophilic
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

´
252 Gonzalez et al.
Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Table 4. Experimental conditions assayed to obtain compound 18.
Conditions†
Time
Results
p-TsOH (catalytic amounts)/EtOH
5 d
Decomposition
Decomposition
MeONa/MeOH
30 min
Traces of 18
(complex crude of reaction)
piperidine (1 equiv.)/EtOH
5 d
5d
morpholine (1 equiv.)/EtOH
(Mannich protocol)
18 (35%)
†
All reactions were carried out at room temperature.
5) and isosurfaces were examined. Figure 3 shows,
as example, the displayed electron density and HOMO
isosurfaces of the two anionic intermediates of com-
pound 5. Both, electron density and LUMO isosur-
faces, were calculated for starting heterocycles 1, 4,
and 5 (data not shown) in order to identify the acidic
hydrogen in the corresponding 3-, 5- or 6-methyl moi-
eties. These data were totally in agreement with the
anions HOMO calculations.
Figure 1. Correlations observed in HMBC experiment.
Biological studies
center of the starting heterocycles, 3-, 5- or 6-methyl
moieties [29-33]. We theoretically studied in terms of
energy, in gas-phase and in water as polar solvent,
the anionic intermediates of this process for each com-
pound (1, 4, and 5, Figure 2) using molecular orbital
calculations following the ab initio method at HF/6-
31G* [34]. The geometries of these intermediates
were fully optimized at the ab initio level indicated. The
pair of electrons in the carbanion forms was con-
sidered as totally delocalized. A single point calcu-
lation was applied over each optimized structure. In
order to compare the stability of the two or three pos-
sible intermediates, the relative energy (∆E) was de-
termined (∆E was defined as the difference between
the energy of the most stable intermediate for each
compound and the energy of the studied intermedi-
ate). The relative energy values are composed in
Table 5. The highest-occupied molecular orbitals
(HOMO) of the different anions were also studied in
order to know the nucleophilic capacity of possible in-
termediates, in this sense, energetic terms (see Table
Compounds were subjected to preliminary cytotoxic
evaluation on V79 cells under hypoxic and aerobic
conditions using a cloning assay as previously de-
scribed [4, 8]. All the compounds were tested at
20 µM. As well, derivative 19 was tested at 10 µM,
5 µM, 1 µM and 0.1 µM. The survival fraction under
the two conditions (SFair and SFhipox) was deter-
mined. The results obtained are summarized in Table
6. Derivatives 6Ϫ15 and 18 resulted less cytotoxic
than the parent compounds (1, 4 and 5) [6] and un-
selective. However derivative 19 was the most cyto-
toxic compound under both conditions, even more
than the parent compounds. Derivative 19 is potent
but not selectivity.
Discussion
Based on the experimental results, we can infer that
the 1,2,4-triazine N4-oxides react mainly with good
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Potential Hypoxic Cytotoxins 253
Figure 2. Intermediates studied theoretically.
Table 5. Energetic terms of proposed intermediates of condensation process.
Compound No
Intermediate
∆(HOMO energy) (kcal molϪ1 †
)
∆E (kcal molϪ1 ‡
)
in vacuo
in water
1.5
1.6
5.1
0.0
0.0
Ϫ15.0
0.0
Ϫ21.8
1
4
5
4.3
4.5
4.6
5.3
6.7
0.0
Ϫ18.0
0.0
Ϫ18.0
Ϫ19.7
0.0
Ϫ24.3
5.5
5.6
3.7
0.0
0.0
Ϫ15.2
0.0
Ϫ15.6
†
Relative HOMO energies of intermediates of condensation process, ∆(HOMO energy) = E(intermediate with
high HOMO energy) Ϫ E(intermediate with low HOMO energy).
Relative energies of intermediates of condensation process, ∆E = E(intermediate most stable for each com-
pound) Ϫ E(intermediate in study for each compound).
‡
electrophilic species, such as iminium cations. The
The regioselectivity in the 5-methyl position for 1,2,4-
triazine N4-oxide and N1,N4-dioxide was demonstrated
experimentally and, in contrast with a previous report
for the reaction between 6-methyl-3-phenyl-1,2,4-tria-
zine N4-oxide and 1-ethoxy-N,N-dimethylvinylamine
other electrophiles assayed in the present work (acyl
chloride, isothiocyanate and aldehyde) resulted less
reactive, so they conduct to the decomposition of het-
erocycle.
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

´
254 Gonzalez et al.
Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Figure 3. Electron density and HOMO isosurfaces. A. Electron density (transparent, isovalue 0.002) and HOMO
(solid, isovalue 0.02). B. Another view of the electron density (transparent, isovalue 0.002) and HOMO (transpar-
ent, isovalue 0.019).
[32], we did not observe condensation products at the
6-methyl position with iminium electrophile.
compound 5, 5.5 and 5.6, it was possible to note that
the first intermediate, 5.5, presented a higher HOMO
contribution than the intermediate 5.6. These obser-
vations are in agreement with the observed reactivity.
Concerning the influence of the N-oxide moiety in the
α-methyl nucleophilicity we observed experimentally,
that the process doesn’t depend on the N-oxide
neighbor presence: 4 should react throughout the N-
oxide neighboring 3-methyl substituent and 5 should
react throughout the N-oxide neighboring 6-methyl
substituent, however, these phenomena were not ob-
served under the experimental conditions. From the
theoretical calculations, the thermodynamic stabilities
of the proposed intermediates are in agreement with
the products experimentally obtained. With the theo-
retical relative energy of the anionic intermediates cal-
culated, some conclusions can be drawn. In all cases,
the 5-intermediates were found to be the most relative
stable species. The electrostatic stabilization of 5-in-
termediates in water, a polar solvent such as the sol-
vent used experimentally (ethanol), increase the ∆E
between the latter and the other compounds. The 5-
anion brought the most favorable intermediates, more
than 15 kcal mol-1 to each other possible anionic inter-
mediates (3-anion and 6-anion) in vacuo and in water.
Furthermore, the anion’s HOMO energies indicated
that the 5-intermediates presented, in all cases, the
highest values. The relative 5-intermediates contri-
bution in the HOMO could be graphically observed
when both electron density and HOMO isosurfaces
were displayed together, such as is shown for com-
pound 5 in Figure 3. For both anionic intermediates of
Twelve new 1,2,4-triazine N-oxide derivatives were
prepared and evaluated as selective hypoxic cytotox-
ins. Some compounds were less active, under hypoxic
conditions at the dose assayed (20 µM), than tirapaza-
mine. Compound 19 showed good cytotoxic activity,
displaying excellent V79 cell-growth inhibition even
though only at 5 µM, and thus had the same cytotoxic
activity as the reference compound, but was non-
selective.
This fact could be explained, considering that the pro-
cess involved bio-reduction, whereas the reduction of
5-nitrothienyl is more facile to obtain than that of the
N-oxide pharmacophore, so cytotoxic free radical-ev-
ents will be produced in major proportion. It is well
known that the nitrothienyl group possesses a re-
duction potential [35, 36] comparable to benzo-1,2,4-
triazine N1,N4-dioxide moiety [8] and less negative
than 1,2,4-triazine N4-oxide system [6].
The information obtained with these derivatives, es-
pecially with 19, allowed us to re-design new struc-
tures with potential desired activity. Further studies
such as electrochemistry and EPR spectroscopy, to
determine the ability to produce radicals, as well as
SAR, are currently in progress.
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Potential Hypoxic Cytotoxins 255
up included washing with brine and drying the organic layer
with sodium sulfate before concentration. Melting points were
determined using a Leitz Microscope Heating Stage Model
350 apparatus (Wetzlar, Germany) and are uncorrected. El-
emental analyses were obtained from vacuum-dried samples
(over phosphorous pentoxide at 3Ϫ4 mM Hg, 24 h at room
temperature) and performed on a Fisons EA 1108 CHNS-O
analyzer (Valencia, USA), and were within 0.4% of theo-
retical values. Infrared spectra were recorded on a Perkin
Elmer 1310 (Jügesheim, Germany) apparatus, using potas-
Table 6. Biological characterization of 1,2,4-triazine
derivatives.
Compound
No.
Doses
(µM)
SFair
(%)†,‡
SFhipox
(%)‡,§
1#
2#
3#
4#
5#
6
7
8
9
10
11
12
13
14
15
18
20
20
20
20
20
20
Ϫ
20
20
20
Ϫ
20
20
20
20
20
20
10
5
75 16
74
85
67
88 15
73 11
8
4
5
86
86
0
8
sium bromide tablets; the frequencies are expressed in cmϪ1
.
1H NMR, 13C NMR spectra and HETCOR experiments were
recorded on a Bruker DPX-400 instrument (Rehinstetten,
Germany (at 400 and 100 MHz)), with tetramethylsilane as
the internal reference and in CDCl3 solvent; the chemical
shifts are reported in ppm and J values are reported in Hz.
Mass spectra were recorded on a Shimadzu GC-MS QP 1100
EX instrument (Kyoto, Japan) using electron impact (EI) ion-
ization at 70 eV, or chemical ionization (CI) using CH4 as
carrier gas. The α-hydrazono oxime I and compound 5 were
prepared as reported previously [6].
100 0††
96
98
6
7
76
7
ND
ND
95
4
83 12
100 0††
86 10
100 0‡‡
100 0††
100 0††
100 0††
100 0††
74
99
91
86 11
0
0
3,5,6-Trimethyl-1,2,4-triazine N4-oxide 4
4
2
4
In an open vessel a mixture of I (0.15 g, 1.30 mM), trimethyl
orthoacetate (0.5 mL, 3.00 mM) and catalytic amounts of p-
toluenesulfonic acid was irradiated in an unmodified house-
hold microwave oven (high potency, 220 W-2450 MHz) during
15 min, in 1-min intervals separated by 1-2 min of no ir-
radiation. The excess of solvent was removed in vacuo. The
mixture was purified by column chromatography (SiO2-
EtOAc) to give 4 in 60% (expressed considering 100% of
conversion of oxime stereoisomer reactive (Z-hydrazono-E-
oxime)).
93
9
88 10
100 0††
0
0
0
0
0
0
0
0
0
0
0
0
0
19
0
1
0.1
5
69 10
100
100
0
92
100
100
6
0
0
Reference
Tirapazamine
General procedure for the synthesis of derivatives 6Ϫ15
20
0
A solution of triazine (1, 4 or 5, 1.0 equiv.) in EtOH as solvent
was added drop wise to a stirred mixture of amine (1.0 equiv.)
and formaldehyde (37 wt-% solution in water) (1.0 equiv.) in
EtOH as solvent. The mixture was stirred at room tempera-
ture until the heterocycle was not present (SiO2-CH2Cl2:
MeOH, 99:1). The EtOH was evaporated in vacuo and the
residue was dissolved in EtOAc. After the work-up process
the residue was purified as indicated.
†
Survival fraction in air.
‡
The tests were carried in duplicate.
Survival fraction in hypoxia.
From reference 6.
§
#
††
Highest concentration tested: 20 µM.
ND Ϫ not determined.
ReferenceϪ7-Chloro-3-[3-(N,N-dimethylamine)propyl-
amino]-2-quinoxalinecarbonitrile hydrochloride.
6-Methyl-5-[2-(morpholine-4-yl)ethyl]-1,2,4-triazine N4-oxide 6
Purified by column chromatography (SiO2-EtOAc), brown oil.
1
IR ν: CH 3050, CH 1450, NO 1288 cmϪ1. Ϫ H NMR δ: 2.53
(m, 4H, N-CH2), 2.71 (t, 2H, -CH2, JCH2CH2 = 6.8), 2.75 (s,
3H, -CH3), 3.08 (t, 2H, -CH2, JCH2CH2 = 6.9), 3.68 (m, 4H, O-
CH2), 9.15 ppm (s, 1H, -C3-H).- 13C NMR δ: 19.95 (-CH3),
24.15 (-C1’H2), 53.49 (-C2’H2), 54.04 (-CH2-morpholine),
67.26 (-CH2-morpholine), 147.07 (C5), 149.09 (C3), 158.88
ppm (C6). Ϫ MS (EI); m/z (%): 207 (18) [M•+-17], 124 (3),
100 (100). Anal. Calcd. for C10H16N4O2: C, 53.57; H, 7.14; N,
25.00. Found: C, 53.80; H, 7.03; N, 25.22.
Acknowledgments
´
This research has been supported by Comision Hono-
raria de Lucha Contra el Cancer (Uruguay) and
CYTED. We thank a master grant for P. Saenz from
Programa de Desarrollo de Ciencias Basicas (PEDE-
´
´
CIBA-Uruguay).
6-Methyl-5-[2-(piperidine-1-yl)ethyl]-1,2,4-triazine N4-oxide 7
Experimental
Purified by column chromatography (SiO2-EtOAc), colorless
oil. IR ν: CH 2935, CH 1440, NO 1280 cmϪ1. Ϫ 1H NMR δ:
1.42 (m, 6H, -CH2), 2.45 (m, 4H, N-CH2), 2.67 (t, 2H, -CH2,
JCH2CH2 = 6.7), 2.74 (s, 3H, -CH3), 3.08 (t, 2H, -CH2,
JCH2CH2 = 6.8), 9.14 ppm (s, 1H, -C3-H). Ϫ 13C NMR δ: 19.91
(-CH3), 24.20 (-C1ЈH2), 24.40 (-CH2-piperidine), 26.20 (-CH2-
piperidine), 53.50 (-C2ЈH2), 54.50 (-CH2-piperidine), 147.10
Chemistry
All initial materials were commercially available research-
grade chemicals and used without further purification. All sol-
vents were dried and distilled prior to use. All the reactions
were carried out in a nitrogen atmosphere. The typical work-
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

´
256 Gonzalez et al.
Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
(C5), 149.00 (C3), 158.90 (C6).- MS (EI); m/z (%): 205 (2)
[M•+-17], 123 (1), 98 (100). Anal. Calcd. for C11H18N4O: C,
59.46; H, 8.11; N, 25.23. Found: C, 59.31; H, 7.98; N, 25.42.
cmϪ1. Ϫ 1H NMR δ: 1.80 (m, 4H, -CH2), 2.54 (s, 3H, -CH3),
2.62 (m, 4H, N-CH2), 2.86 (t, 2H, -CH2, JCH2CH2 = 6.9), 3.19
(t, 2H, -CH2, JCH2CH2 = 6.9), 8.74 ppm (s, 1H, -C3-H). Ϫ 13C
NMR δ: 14.85 (-CH3), 23.92 (-CH2-pirrolidine), 27.67 (-C1ЈH2),
51.47 (-C2’H2), 54.61 (-CH2-pirrolidine), 141.70 (C6), 147.37
(C3), 153.85 ppm (C5).- MS (EI); m/z (%): 208 (2) [M+-16],
207 (9), 84 (100). Anal. Calcd. for C10H16N4O2: C, 53.57; H,
7.14; N, 25.00. Found: C, 53.30; H, 7.00; N, 25.10.
3,6-Dimethyl-5-[2-(pirrolidine-1-yl)ethyl]-1,2,4-triazine N4-di-
oxide 8
Purified by column chromatography (SiO2-CH2Cl2:MeOH
(0Ϫ5%)), oil. IR ν: CH 2967, CH 1449, NO 1275 cmϪ1. Ϫ 1H
NMR δ: 1.83 (m, 4H, -CH2), 2.69 (m, 4H, N-CH2), 2.72 (s,
6-Methyl-5-[2-(4-methylpiperazine-1-yl)ethyl]-1,2,4-triazine
N1,N4-dioxide 13
3H, -CH3), 2.75 (s, 3H, -CH3), 2.84 (t, 2H, -CH2, JCH2CH2
=
7.3), 3.18 ppm (t, 2H, -CH2, JCH2CH2 = 7.3).- 13C NMR: δ
17.65 (-3-CH3), 19.86 (-6-CH3), 23.97 (-CH2-pirrolidine),
26.27 (-C1ЈH2), 50.46 (-C2’H2), 54.36 (-CH2-pirrolidine),
145.79 (C5), 157.20 (C6), 159.02 ppm (C3). Ϫ MS (CI); m/z
(%): 223 (4) [M•++ H], 206 (12) [M•+-16], 205 (79), 84 (100).
Anal. Calcd. for C11H18N4O: C, 59.46; H, 8.11; N, 25.23.
Found: C, 59.41; H, 7.90; N, 25.00.
Purified by column chromatography (SiO2-CH2Cl2:MeOH
(0Ϫ5%)), oil. IR ν: CH 3090, CH 1433, CH 1360, NO
1
1287 cmϪ1. Ϫ H NMR δ: 2.27 (s, 3H, N-CH3), 2.42 (bs, 4H,
N-CH2), 2.52 (s, 3H, -CH3), 2.57 (bs, 4H, N-CH2), 2.74 (t,
2H, -CH2, JCH2CH2 = 6.8), 3.13 (t, 2H, -CH2, JCH2CH2 = 6.8),
8.77 ppm (s, 1H, -C3-H). Ϫ 13C NMR δ: 14.95 (-CH3), 25.77
(-C1’H2), 46.25 (CH3-N), 53.08 (-CH2-piperazine), 53.65
(-C2’H2), 55.28 (-CH2-piperazine), 141.58 (C6), 147.35 (C3),
153.93 ppm (C5). Ϫ MS (EI); m/z (%): 236 (23) [M+-17], 113
(13), 58 (100). Anal. Calcd. for C11H19N5O2: C, 52.17; H,
7.51; N, 27.67. Found: C, 51.98; H, 7.33; N, 27.37.
3,6-Dimethyl-5-[2-(4-methylpiperazine-1-yl)ethyl]-1,2,4-tri-
azine N4-oxide 9
Purified by column chromatography (SiO2-CH2Cl2:MeOH
(0Ϫ5%)), oil. IR ν: CH 2939, CH 1460, NO 1287 cmϪ1. Ϫ 1H
NMR δ: 2.32 (s, 3H, N-CH3), 2.49 (bs, 4H, N-CH2), 2.62 (bs,
4H, N-CH2), 2.71 (s + m, 5H, -CH3 + -CH2), 2.76 (s, 3H,
-CH3), 3.10 ppm (t, 2H, -CH2, JCH2CH2 = 6.8). Ϫ 13C NMR δ:
17.68 (-3-CH3), 19.90 (-6-CH3), 24.63 (-C1ЈH2), 46.14 (CH3-
N), 52.90 (-C2ЈH2), 53.17 (-CH2-piperazine), 55.34 (-CH2-pip-
erazine), 146.27 (C6), 157.14 (C5), 159.04 ppm (C3). Ϫ MS
(EI); m/z (%): 235 (2) [M•+-16], 234 (19), 113 (100). Anal.
Calcd. for C12H21N5O: C, 57.37; H, 8.37; N, 27.89. Found: C,
57.12; H, 8.42; N, 27.69.
6-Methyl-5-[2-(morpholine-4-yl)ethyl]-1,2,4-triazine N1,N4-di-
oxide 14
Purified by column chromatography (SiO2-CH2Cl2:MeOH
(0Ϫ5%)), oil. IR ν: CH 3060, CH 1433, CH 1358, NO 1279
1
cmϪ1. Ϫ H NMR δ: 2.45 (s + m, 7H, -CH3 + N-CH2), 2.67 (t,
2H, -CH2, JCH2CH2 = 6.5), 3.07 (t, 2H, -CH2, JCH2CH2 = 6.2),
3.60 (m, 4H, O-CH2), 8.71 ppm (s, 1H, -C3-H).- 13C NMR δ:
14.89 (-CH3), 25.59 (-C1’H2), 53.98 (-CH2-morpholine), 54.19
(-C2’H2), 67.20 (-CH2-morpholine), 141.50 (C6), 147.37 (C3),
153.76 ppm (C5). Ϫ MS (CI); m/z (%): 241 (9) [M++ H], 223
(45) [M•+-17], 207 (3), 100 (77), 98 (100). Anal. Calcd. for
3,6-Dimethyl-5-[2-(morpholine-4-yl)ethyl]-1,2,4-triazine N4-
oxide 10
C10H16N4O3: C, 50.00; H, 6.67; N, 23.33. Found: C, 49.71;
Purified by column chromatography (SiO2-CH2Cl2:MeOH (0
to 1%)), oil. IR ν: CH 2950, CH 1462, NO 1279 cmϪ1. Ϫ 1H
NMR δ 2.58 (m, 4H, N-CH2), 2.71 (s + m, 5H, -CH3 + -CH2),
2.75 (s, 3H, -CH3), 3.10 (t, 2H, -CH2, JCH2CH2 = 7.4), 3.72
ppm (m, 4H, O-CH2).- 13C-NMR δ: 17.68 (-3-CH3), 19.90
(-6-CH3), 24.48 (-C1’H2), 53.45 (-C2’H2), 53.99 (-CH2-morph-
oline), 67.60 (-CH2-morpholine), 146.14 (C5), 157.10 (C6),
159.05 ppm (C3). Ϫ MS (EI); m/z (%): 222 (3) [M•+-16], 221
(16), 100 (100). Anal. Calcd. for C11H18N4O2: C, 55.46; H,
7.56; N, 23.53. Found: C, 55.21; H, 7.88; N, 23.49.
H, 6.88; N, 23.20.
6-Methyl-5-[2-(piperidine-1-yl)ethyl]-1,2,4-triazine N1,N4-di-
oxide 15
Purified by column chromatography (SiO2-CH2Cl2:MeOH
(0Ϫ5%)), oil. IR ν: CH 3065, CH 1437, CH 1356, NO 1291
cmϪ1. Ϫ 1H NMR δ: 1.56 (m, 6H, -CH2), 2.48 (m, 4H, N-CH2),
2.53 (s, 3H, -CH3), 2.70 (t, 2H, -CH2, JCH2CH2 = 7.0), 3.14 (t,
2H, -CH2, JCH2CH2 = 7.0), 8.73 ppm (s, 1H, -C3-H). Ϫ 13C
NMR δ: 14.86 (-CH3), 24.34 (-CH2-piperidine), 25.90 (-C1ЈH2),
26.27 (-CH2-piperidine), 54.45 (-CH2-piperidine), 54.97
(-C2ЈH2), 141.75 (C6), 147.31 (C3), 155.43 ppm (C5). Ϫ MS
(CI); m/z (%): 239 (9) [M•++ H], 222 (11) [M+-16], 221 (49),
98 (100). Anal. Calcd. for C11H18N4O2: C, 55.46; H, 7.56; N,
23.53. Found: C, 55.43; H, 7.60; N, 23.46.
3,6-Dimethyl-5-[2-(piperidine-1-yl)ethyl]-1,2,4-triazine
N4-
oxide 11
Purified by column chromatography (SiO2-CH2Cl2:MeOH
(0Ϫ5%)), oil. IR ν: CH 2932, CH 1449, NO 1277 cmϪ1. Ϫ 1H
NMR δ: 1.41 (m, 2H, -CH2), 1.56 (m, 4H, -CH2), 2.48 (m, 4H,
N-CH2), 2.64 (t, 2H, -CH2, JCH2CH2 = 7.6), 2.68 (s, 3H, -CH3),
2.71 (s, 3H, -CH3), 3.09 ppm (t, 2H, -CH2, JCH2CH2 = 7.6).
General procedure for the synthesis of derivatives 18 and 19
Ϫ
13C NMR δ: 17.65 (-3-CH3), 19.87 (-6-CH3), 24.45 (-CH2-
A solution of 1 (1.0 equiv.) in EtOH as solvent was added
drop wise to a stirred mixture of morpholine (1.0 equiv.) and
aldehyde (benzaldehyde or 5-nitrothiophene-2-carboxal-
dehyde, 1.0 equiv.) in EtOH as solvent. The mixture was kept
at room temperature. The precipitate was filtered off, washed
with ethanol and dried.
piperidine), 24.52 (-C1ЈH2), 26.18 (-CH2-piperidine), 53.54
(-C2’H2), 54.78 (-CH2-piperidine), 146.42 (C5), 157.29 (C6),
158.95 ppm (C3). Ϫ MS (CI); m/z (%): 237 (4) [M++ H], 220
(10) [M•+-16], 219 (74), 98 (100). Anal. Calcd. for C12H20N4O:
C, 61.02; H, 8.47; N, 23.73. Found: C, 61.15; H, 8.20; N,
23.82.
5-(2-Phenylethenyl)-6-Methyl-1,2,4-triazine N4-oxide 18
6-Methyl-5-[2-(pirrolidine-1-yl)ethyl]-1,2,4-triazine N1,N4-di-
oxide 12
Time of reaction:
5 d, green-brown solid (35%); mp
145.7Ϫ147.0°C. IR ν: CH 3076, CH 2924, C=C 1603, 1503,
NO 1273 cmϪ1. Ϫ 1H NMR δ: 2.86 (s, 3H, -CH3), 7.31 (d,
1H, =CH, J = 16.0 Hz), 7.47 (m, 3H, phenyl H), 7.76 (m, 2H,
Purified by column chromatography (SiO2-CH2Cl2: MeOH
(0Ϫ5%)), oil. IR ν: CH 3065, CH 1458, CH 1362, NO 1289
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
Potential Hypoxic Cytotoxins 257
phenyl H), 9.09 (d, 1H, =CH, J = 16.0 Hz), 9.22 ppm (s, 1H,
-C3-H).- 13C NMR δ: 19.94 (-CH3), 114.46 (=C1’H), 128.48
(phenyl-C), 129.41 (phenyl-C), 130.56 (phenyl-C), 136.89
(phenyl-C), 139.98 (C5), 143.39 (=C2’H), 150.24 (C3), 157.65
ppm (C6).- MS (EI); m/z (%): 213 (100) [M•+], 196 (41), 77
(34). Anal. Calcd. for C12H11N3O: C, 67.59; H, 5.20; N, 19.71.
Found: C, 67.81; H, 5.02; N, 19.52.
semi empirical method and finally ab initio level (HF/6-31G*)
(all implemented in PC Spartan Pro package). From each
minimized structure a single point calculation was performed
(in vacuo and water [38]) using the standard 6-31G* basis
set at HF level [39Ϫ41].
References
6-Methyl-5-[2-(5-nitrothienyl)ethenyl)-1,2,4-triazine N4-oxide
19
´
[1] H. Cerecetto, M. Gonzalez, Minirev. Med. Chem. 2001,
1, 219Ϫ231.
Time of reaction: 3 d, green solid (10%); mp 218.7Ϫ219.2°C.
´
´
[2] A. Monge, A. Lopez de Cerain, O. Ezpeleta, H. Cere-
IR ν: CH 3075, C=C 1524, 1451, NO2 1325, NO 1269 cmϪ1
.
´
cetto, E. Dias, R. Di Maio, M. Gonzalez, S. Onetto, M.
Risso, G. Seoane, F. Zinola, C. Olea-Azar, Pharmazie
1998, 53, 698Ϫ702.
1
Ϫ H NMR δ: 2.87 (s, 3H, -CH3), 6.99 (d, 1H, =CH, J = 15.7
Hz), 7.31 (d, 1H, thenyl H, J = 5.0 Hz), 7.87 (d, 1H, thenyl H,
J = 5.0 Hz), 9.18 (s, 1H, -C3-H), 9.29 ppm (d, 1H, =CH, J =
15.7 Hz).- 13C NMR δ: 20.53 (-CH3), 116.32 (=C1’H), 129.63
(thenyl-C), 129.81 (thenyl-C), 135.69 (=C2’H), 137.97 (C5),
147.77 (thenyl-C), 150.18 (C3), 153.00 (thenyl-C), 157.23
ppm (C6). Ϫ MS (EI); m/z (%): 264 (100) [M+], 248 (21), 247
(4). Anal. Calcd. for C10H8N4O3S: C, 45.45; H, 3.05; N, 21.20;
S, 12.13. Found: C, 45.08; H, 3.07; N, 21.05; S, 11.74.
´
´
[3] A. Monge, A. Lopez de Cerain, O. Ezpeleta, H. Cere-
cetto, E. Dias, R. Di Maio, M. González, S.
Onetto, G. Seoane, L. Suescun, R. Mariezcurrena, Phar-
mazie 1998, 53, 758Ϫ764.
´
´
[4] H. Cerecetto, M. Gonzalez, M. Risso, G. Seoane, A. Lo-
´
pez de Cerain, O. Ezpeleta, A. Monge, L. Suescun, A.
Mombru, A. M. Bruno, Arch. Pharm. Pharm. Med. Chem.
2000, 333, 387Ϫ393.
´
Biology
Cells: V79 cells (Chinese hamster lung fibroblasts) were ob-
tained from ECACC (European Collection of Animal Cell Cul-
tures) and maintained at logarithmic growth as sub confluent
monolayer by trypsinization and subculture to (1Ϫ2) ϫ 104
cells/cm2 twice weekly. The growth medium was EMEM (Eag-
le’s Minimal Essential Medium), containing 10% (vol/vol) fetal
bovine serum (FBS) ad penicillin/streptomycin at 100 U/100
µg/mL. Aerobic and hypoxic cytotoxicity: suspension cultures.
Monolayer of V79 cells in exponential growth was trypsinized,
and suspension cultures were set up in 50 mL glass flasks:
2 ϫ 104 cells/mL in 30 mL of EMEM, containing 10% (vol/
vol) FBS and HEPES (10 mM). The glass flasks were sub-
merged and stirred in a water bath at 37°C, where they were
gassed with humidified air or pure nitrogen. Treatment: Com-
pounds solutions were prepared just before dosing. Stock
solutions, 150-fold more concentrated, were prepared in pure
DMSO (Aldrich, Milwaukee, USA) or sterilized distilled water.
30 min after the start of gassing, 0.2 mL of the stock com-
pound solution was added to each flask, two flasks per dose.
In every assay there was one flask with 0.2 mL of DMSO
(negative control) and another with 0.06 µM nitracine (posi-
tive control). Cloning: After 2 h exposure to the compound,
the cells were centrifuged and resuspended in plating me-
dium (EMEM plus 10% (vol/vol) FBS and penicillin/strepto-
mycin). Cell numbers were determined with a hemocytometer
and 102Ϫ103 cells were plated in 6-well plates to give a final
volume of 2 mL/30 mM of well. Plates were incubated at 37°C
in 5% CO2 for 7 d and then stained with aqueous crystal
violet. Colonies with more than 64 cells were counted. The
plating efficiency (PE) was calculated by dividing the number
of colonies by the number of cells seeded. The percent of
control-cell survival for the compound-treated cultures (SFair
and SFhipox) was calculated as PEtreated/PEcontrol ϫ 100. The
compounds were initially tested at 20 µM in duplicate flasks
both in aerobic and hypoxic conditions.
´
[5] M. Boiani, H. Cerecetto, M. Gonzalez, M. Risso, C.
´
´
Olea-Azar, O. Ezpeleta, A. Lopez de Cerain, O. Ezpe-
leta, A. Monge-Vega, Eur. J. Med. Chem. 2001, 36,
771-782.
´
[6] H. Cerecetto, M. Gonzalez, S. Onetto, M. Risso, P. Sa-
enz, G. Seoane, A. M. Bruno, J. Alarcon, C. Olea-Azar,
A. Lopez de Cerain, O. Ezpeleta, A. Monge, Med. Chem.
Res. 2001, 10, 328Ϫ337.
´
´
[7] A. I. Minchinton, M. J. Lemmon, M. Tracy, D. J. Pollart,
A. P. Martínez, L. M. Tosto, J. M. Brown, Int. J.
Radiat. Oncol. Biol. Phys. 1992, 22, 701Ϫ705.
´
´
´
[8] A. Monge, F. J. Martınez-Crespo, A. Lopez de Cerain,
´
´
J. A. Palop, S. Narro, V. Senador, A. Marın, Y. Sainz, M.
Gonzalez, E. Hamilton, A. J. Barker, J. Med. Chem.
´
1995, 38, 4488Ϫ4494.
[9] S. Kamiya, G. Okusa, M. Osada, M. Kumagai, A. Naka-
mura, M. Koshinuma, Chem. Pharm. Bull. 1968, 16, 939.
[10] J. F. Vozza, J. Org. Chem. 1962, 27, 3856Ϫ3860.
[11] V. J. Traynelis, A. I. Gallagher, J. Org. Chem. 1970, 35,
2792Ϫ2796.
[12] C. H. Issidorides, M. A. Atfah, J. J. Sabounji, A. R. Si-
dani, M. J. Haddadin, Tetrahedron 1978, 34, 217Ϫ221.
[13] H. Yamanaka, S. Ogawa, S. Konno, Chem. Pharm. Bull.
1980, 28, 1526Ϫ1533.
[14] M. Hamana, M. Yamazaki, Chem. Pharm. Bull. 1961,
9, 414Ϫ418.
[15] F. Kröhnke, H. Schäfer, Chem. Ber. 1962, 95, 1098-
1100.
[16] M. van Ammers, H. J. den Hertog, B. Hasse, Tetra-
hedron 1962, 18, 227Ϫ232.
Theoretical calculations
Firstly the intermediates were built with standard bond
lengths and angles using the PC Spartan Pro (Irvine, USA)
molecular modeling package [34, 37]. The structure of each
intermediate was fully optimized by molecular mechanics
methods (SYBYL molecular mechanic force fields), then AM1
´
[17] J. Alcazar, M. Begtrup, A. de la Hoz, J. Chem. Soc.,
Perkin Trans. 1995, 1, 2467Ϫ2470.
[18] H. Neunhoeffer, F. Weischedel, V. Böhnisch, Justus Lie-
bigs Ann. Chem. 1971, 750, 12Ϫ20.
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

´
258 Gonzalez et al.
Arch. Pharm. Pharm. Med. Chem. 2004, 337, 247−258
[19] V. Böhnisch, G. Burzer, H. Neunhoeffer, Justus Liebigs
Ann. Chem. 1977, 1713Ϫ1717.
[32] J. Adler, V. Böhnisch, H. Neunhoeffer, Chem. Ber. 1978,
111, 240Ϫ247.
[33] H. Neunhoeffer in Comprehensive Heterocyclic Chemis-
try (Eds.: A.R. Katritzky, C.W. Rees), Pergamon Press,
Oxford, 1984, pp 401Ϫ402.
¨
[20] F. Louerat, K. Bougrin, A. Loupy, A. M. Ochoa de Re-
tana, J. Pagalday, F. Palacios, Heterocycles 1998, 48,
161Ϫ170.
[34] Wavefunction, Inc. 18401 Von Karman Avenue, Suite
370. Irvine, California 92612 USA.
[21] M. Cooper, H. Heaney, A. Newbold, W. Sanderson, Syn-
lett 1990, 9, 533Ϫ535.
´
[35] C. Olea-Azar, A.M. Atria, F. Mendizabal, R. Di Maio, G.
[22] R. S. Varma, K. P. Naicker, Org. Lett. 1999, 1, 189Ϫ192.
[23] F. Blicke, Organic Reactions 1942, 1, 303Ϫ341.
Seoane, H. Cerecetto, Spectroscopy Letters 1998,
31(4), 849Ϫ857.
´
[36] H. Cerecetto, R. Di Maio, M. Gonzalez, M. Risso, G.
[24] T. J. Delia, J. P. Scovill, W. D. Munslow, J. H. Burck-
Sagrera, G. Seoane, A. Denicola, G. Peluffo, C. Quijano,
M. A. Basombrıo, A.O. M. Stoppani, M. Paulino, C. Olea-
Azar, Eur. J. Med. Chem. 2000, 35, 343Ϫ350.
halter, J. Med. Chem. 1976, 19, 344Ϫ346.
´
[25] Vogel’s, Textbook of Practical Organic Chemistry (4th
Ed.), Longman Group, London, 1978, pp. 902Ϫ904.
[37] PC Spartan Pro User’s Guide, Wavefunction Inc., Cali-
fornia, 1999.
[26] K. Mitsuhashi, E. Itho, T. Kawahara, K. Tanaka, J. Het-
erocyclic Chem. 1983, 20, 1103Ϫ1105.
[38] C. C. Chambers, G. D. Hawkins, C. J. Cramer, D. G.
[27] M. J. Weiss, C. R. Hauser, J. Amer. Chem. Soc. 1949,
71, 2023Ϫ2026.
Truhlar, J. Phys. Chem. 1996, 100, 16385Ϫ16398.
[39] P. C. Hariharan, J. A. Pople, Chem. Phys. Lett. 1972,
16, 217Ϫ220.
[28] S. Skidmore, E. Tidd, J. Chem. Soc. 1959, 20, 1641.
[40] W. J. Hehre, L. Radom, P. v. R. Schleyer, J. A. Pople in
Ab initio Molecular Orbital Theory, Wiley, New York,
1986.
[29] W. E. Taft, R. G. Shepherd, J. Med. Chem. 1967, 10,
883-887.
[30] J. Adams, R. G. Shepherd, Tetrahedron Lett. 1968, 23,
2747Ϫ2750.
[41] W. J. Hehre, A. J. Shusterman, W. W. Huang in A Labo-
ratory Book of Computational Organic Chemistry, Wave-
function Inc., California, 1996.
[31] W. W. Paudler, J. Lee, J. Org. Chem. 1971, 36,
3921Ϫ3925.
 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim