Ferrocenyl, Alkyl, and Aryl-Pyrido[2,3-d]Pyrimidines as Vasorelaxant of Smooth Muscle of Rat Aorta via cAMP Conservation Through Phosphodiesterase Inhibition

Article abstract of DOI:10.1002/jhet.2380

New pyrido[2,3-d]pyrimidines 11, 12, 13, and 21 have been synthesized. The vasorelaxant effect on smooth muscle isolated from rat aorta, via PDEs inhibition, of these compounds along with other pyrido[2,3-d]pyrimidines 14, 15, 16, 17, 18, 19, 20 reported earlier by our group, has also been determined. These pyrido[2,3-d]pyrimidines 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 were synthesized by the reaction of ferrocenyl-ethynyl ketones (1, 2, 3, 4) or α-alkynyl ketones (5, 6, 7, 8, 9, 10) with 6-amino-1,3-dimethyluracil using [Ni(CN)₄]^?4as an active catalytic species, formed in situ in a Ni(CN)₂/NaOH/H₂O/CO/KCN aqueous system. Evaluation of the vasorelaxant effect of compounds 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 demonstrated that all compounds relax the tissue in a concentration-dependent manner. The structural changes do not alter the effectiveness; however, there are differences related to potency expressed as EC₅₀. Compounds 12 (7-ferrocenyl-1,3-dimethyl-5-(m-tolyl)-pyrido[2,3-d]pyrimidine) and 13 (7-ferrocenyl-1,3-dipropyl-5-(4-metoxyphenyl)-pyrido[2,3-d]pyrimidine) were the most potent compounds, even more than rolipram, reference drug; the EC₅₀was 0.41 ± 0.02 μM and 0.81 ± 0.11 μM for 12 and 13, correspondingly. The EC₅₀of compounds 15 (7-ferrocenyl-1,3-dimethyl-5-phenyl-pyrido[2,3-d]pyrimidine), 14 (7-ferrocenyl-5-(3,5-dimethoxyphenyl)-1,3-dimethylpyrido[2,3-d]pyrimidine), and 19 (5-n-butyl-7-ethyl-1,3-dimethylpyrido[2,3-d]pyrimidine) was similar to EC₅₀of rolipram. Compounds 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 significantly induce concentration-dependent vasorelaxation in endothelium-intact aortic rings. In addition, the relaxation responses to each compound in either endothelium-intact or endothelium denuded aortic rings were comparable, suggesting that removal of the functional endothelium has no significant influence on its intrinsic vasorelaxant activity. In vitro capability of conserving cyclic-AMP or cyclic-GMP (adenosine and guanosine 3′, 5′-cyclic monophosphate) via PDE inhibition for compounds 12, 13, 14, 15 and 19 was evaluated. Compounds 15 and 19 show the highest percent inhibition effect (94.83% and 83.98%, respectively) for the decomposition of c-AMP. Docking studies showed that the compound 15 was selective for the inhibition of PDE-4.

Full text of DOI:10.1002/jhet.2380

Month 2015
Ferrocenyl, Alkyl, and Aryl-Pyrido[2,3-d]Pyrimidines as Vasorelaxant of
Smooth Muscle of Rat Aorta via cAMP Conservation Through Phospho-
diesterase Inhibition
a
b
c
c
d
*
Ivonne Arellano, Fernando Rodríguez-Ramos, Martín González-Andrade, Andrés Navarrete, Manju Sharma,
a
a
*
Noé Rosas, and Pankaj Sharma
aInstituto de Química, Universidad Nacional Autónoma de México, México, Distrito Federal, 04510, México
bDepartamento de Ciencias Naturales, DCNI, Universidad Autónoma Metropolitana, Unidad Cuajimalpa, México, Distrito
Federal, 05300, México
cFacultad de Medicina, Departamento de Bioquímica, Universidad Nacional Autónoma de México, México, Distrito
Federal, 04510, México
dIngeniería Bioquímica, Instituto Tecnológico Superior de Atlixco, Atlixco, Puebla, México
*
E-mail: ivonnearj@gmail.com; pankajsh@unam.mx
Received June 2, 2014
DOI 10.1002/jhet.2380
Published online 00 Month 2015 in Wiley Online Library (wileyonlinelibrary.com).
New pyrido[2,3-d]pyrimidines 11, 12, 13, and 21 have been synthesized. The vasorelaxant effect on
smooth muscle isolated from rat aorta, via PDEs inhibition, of these compounds along with other pyrido
[2,3-d]pyrimidines 14-20 reported earlier by our group, has also been determined. These pyrido[2,3-d]py-
rimidines 11–21 were synthesized by the reaction of ferrocenyl-ethynyl ketones (1–4) or α-alkynyl ketones
(5–10) with 6-amino-1,3-dimethyluracil using [Ni(CN)4]ꢀ4 as an active catalytic species, formed in situ in a
Ni(CN)2/NaOH/H2O/CO/KCN aqueous system. Evaluation of the vasorelaxant effect of compounds 11–21
demonstrated that all compounds relax the tissue in a concentration-dependent manner. The structural
changes do not alter the effectiveness; however, there are differences related to potency expressed as
EC50. Compounds 12 (7-ferrocenyl-1,3-dimethyl-5-(m-tolyl)-pyrido[2,3-d]pyrimidine) and 13 (7-
ferrocenyl-1,3-dipropyl-5-(4-metoxyphenyl)-pyrido[2,3-d]pyrimidine) were the most potent compounds,
even more than rolipram, reference drug; the EC50 was 0.41 0.02 μM and 0.81 0.11 μM for 12 and 13,
correspondingly. The EC50 of compounds 15 (7-ferrocenyl-1,3-dimethyl-5-phenyl-pyrido[2,3-d]pyrimi-
dine), 14 (7-ferrocenyl-5-(3,5-dimethoxyphenyl)-1,3-dimethylpyrido[2,3-d]pyrimidine), and 19 (5-n-butyl-
7-ethyl-1,3-dimethylpyrido[2,3-d]pyrimidine) was similar to EC50 of rolipram. Compounds 11–21
significantly induce concentration-dependent vasorelaxation in endothelium-intact aortic rings. In addition,
the relaxation responses to each compound in either endothelium-intact or endothelium denuded aortic rings
were comparable, suggesting that removal of the functional endothelium has no significant influence on its
intrinsic vasorelaxant activity. In vitro capability of conserving cyclic-AMP or cyclic-GMP (adenosine and
guanosine 3′, 5′-cyclic monophosphate) via PDE inhibition for compounds 12–15 and 19 was evaluated.
Compounds 15 and 19 show the highest percent inhibition effect (94.83% and 83.98%, respectively) for
the decomposition of c-AMP. Docking studies showed that the compound 15 was selective for the inhibition
of PDE-4.
J. Heterocyclic Chem., 00, 00 (2015).
INTRODUCTION
antiviral [5], antibacterial [6], antihistaminic [7], as protein
kinases [8], tyrosine kinases [9], anti-proliferative CDK2
[10], and phosphodiesterases (PDEs) enzymes inhibitors,
specifically, types 4 (PDE-4) [11,12] and 7 (PDE-7) [13].
PDEs are enzymes in charge for the hydrolysis of
Pyrido[2,3-d]pyrimidines are versatile compounds with
pharmacological effects as anti-hypertensive [1], analgesic,
anti-inflammatory [2], vasorelaxant [3], cytotoxic [4],
© 2015 HeteroCorporation

I. Arellano, F. Rodríguez-Ramos, M. González-Andrade, A. Navarrete,
M. Sharma, N. Rosas, and P. Sharma
Vol 000
Table 1
Ethynyl ketones obtained to synthesize pyrido[2,3-d] pyrimidines.
R
R1
Compound
Ref.*
Figure 1. Rolipram.
1
2
3
4
5
6
7
8
9
10
Ferrocenyl
Ferrocenyl
Ferrocenyl
Ferrocenyl
Phenyl
Phenyl
Phenyl
Butyl
Butyl
p-methoxyphenyl
m-tolyl
3,5-dimethoxyphenyl
Phenyl
Phenyl
Ethyl
Propyl
Ethyl
Propyl
Propyl
26
26
19
19
27
18
18
18
18
28
intracellular cAMP or cGMP to AMP and GMP, respec-
tively. Inhibition of these enzymes causes the increase of
mentioned second messengers, inducing a vasorelaxant ef-
fect in a variety of blood vessels [14] and plays an important
role for controlling the blood pressure [15] and is used for
the development of new anti-hypertensive drugs [16,17].
Previously, we have reported the synthesis of some
pyrido[2,3-d]pyrimidines analogs by an efficient and
simple way through heterocyclization of α-ketoalkynes or
ferrocenyl ethynyl ketones with 6-amino-1,3-dime-
thyluracil using [Ni(CN)4]ꢀ4 as an active catalytic species,
formed in situ by a Ni(CN)2/NaOH/H2O/CO/KCN aqueous
system [18,19].
On the other side, many ferrocenyl compounds have dem-
onstrated biological activities as cytotoxic, antitumor,
antimalarial, antifungal, antioxidant, and antineoplastic
[20–23]. It is also known that ferrocenyl group has been
used widely in the design or redesign of drugs as phenyl
bioisostere so it is worthwhile to functionalize pyrido[2,3-
d]pyrimidines with a ferrocenyl group. In this work, several
analogs of these compounds (some of them contain
ferrocenyl group in the molecule) were synthesized to
(CH3)3Si
*Ref., reference.
To obtain pyrido[2,3-d]pyrimidines 11 and 13, compound 1 was used.
explore their activity as vasorelaxant through the inhibition
of PDEs and to analyze the structure–activity relationship.
Rolipram (Figure 1), a selective PDE-4 inhibitor that has
shown to conserve cAMP levels in vascular smooth muscle
[24], was used as a reference drug.
RESULTS AND DISCUSSION
Chemistry.
through
Pyrido[2,3-d]pyrimidines were obtained
heterocyclization reaction between α-
a
Scheme 1. Synthesis of ethynyl ketones. This figure is available in colour online at wileyonlinelibrary.com/journal/jhet
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2015
Ferrocenyl, Alkyl, and Aryl-Pyrido[2,3-d]Pyrimidines as Vasorelaxant of Smooth
Muscle of Rat Aorta via cAMP Conservation Through Phosphodiesterase Inhibition
Scheme 2. Synthesis of pyrido[2,3-d]pyrimidines.
(11, 12, 14, and 15) or 6-amino-1,3-dipropyluracil (13)
using a nickel aqueous catalytic system, as reported earlier
[19]. Compounds that are substituted at the C-5 and C-7 po-
sitions by aryl or alkyl groups (16–21) were obtained by
using α-ketoalkynes and 6-amino-1,3-dimethyluracil [18].
Compound 21 was obtained from the heterocyclization of
silyl-α-ketoalkyne and 6-amino-1,3-dimethyluracil. Com-
pounds 11, 12, 13, and 21 are new and have not been re-
ported earlier (Scheme 2) and were characterized by
various physico-chemical methods. This method allows
the easy preparation of several pyrido[2,3-d]pyrimidines-
5,7 disubstituted in water under mild conditions.
Vasorelaxant effect of ferrocenyl, alkyl, and aryl-pyrido
[2,3-d]pyrimidines
on
rat
aorta.
Pyrido[2,3-d]
pyrimidines 11–21 (Table 2) were tested for their
vasorelaxant activity in rat aortic rings pre-contracted
with norepinephrine (NE) (0.1 μM). Figure 2 shows the
concentration–response curves of these compounds.
Rolipram, a PDE-4 inhibitor, was used as a reference
drug. It is noted that all compounds relax the tissues in a
concentration-dependent manner. All compounds present
vasodilator effect, which indicates that the pyrido[2,3-d]
pyrimidine ring retains its effect and does not lose it even
if it is substituted at N-1 and N-3 positions with a methyl
group (compounds 11, 12, and 14–21) or a propyl group
as in compound 13.
ketoalkynes or ferrocenyl ethynyl ketones and 6-amino-
1,3-dimethyluracil. Ferrocenyl ethynyl ketones were
obtained from ethynylferrocene and acyl chlorides via a
palladium-catalyzed coupling reaction as reported earlier
by our group (1–4) [19]. α-ketoalkynes were obtained
from an alkynide, formed by an alkyne, n-BuLi, and
BF.3OEt2, and an anhydride in an inert atmosphere at
ꢀ78°C (5–10) (Scheme 1) [25]. The compounds obtained
in this work are shown in Table 1.
Pyrido[2,3-d]pyrimidines substituted at the C-7 position
by a ferrocenyl group (11–15) were obtained from
ferrocenyl ethynyl ketones and 6-amino-1,3-dimethyluracil
EC50 values for compounds 11–21 along with the
EC50 value for rolipram are shown in Table 2. These
values are in the following descendent potency order
Table 2
Pyrido[2,3-d]pyrimidines obtained and evaluated by their vasodilator effect in rat aortic rings (results expressed as EC50).
EC50 [μM]
Compound
Yield or reference
R1
R2
With endothelium
Without endothelium
11
12
13*
14
15
16
17
18
19
p-methoxyphenyl
m-tolyl
p-methoxyphenyl
3,5-dimethoxyphenyl
Phenyl
Phenyl
Phenyl
Phenyl
Butyl
Butyl
H
Ferrocenyl
Ferrocenyl
Ferrocenyl
Ferrocenyl
Ferrocenyl
Phenyl
Ethyl
Propyl
Ethyl
Propyl
70%
70%
70%
9.83 2.28
0.41 0.02
0.81 0.11
1.45 0.39
2.00 0.71
3.13 0.16
3.61 0.74
2.31 0.77
1.60 0.36
5.00 0.44
5.63 1.16
1.821 0.72
10.84 2.50
0.36 0.06
0.80 0.01
1.78 0.80
1.81 0.45
3.52 0.25
3.63 0.83
2.57 0.79
1.80 0.26
4.14 0.33
4.31 0.59
NE
Ref. 18
Ref. 18
65%
Ref. 17
Ref. 17
Ref. 17
Ref. 17
70%
20
21
Rolipram
Propyl
NA
NA
NA
*For compound 13, R3 = n-Propyl. For compounds 11, 12, 14–21, R3 = Me.
EC50 [μM], half maximal effective concentration. Each value represents the mean SEM of at least six experiments.
NA, do not apply; NE, was not evaluated; Ref., reference.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

I. Arellano, F. Rodríguez-Ramos, M. González-Andrade, A. Navarrete,
M. Sharma, N. Rosas, and P. Sharma
Vol 000
position C-7 is substituted with a propyl group as in com-
pounds 18, 20, and 21, the relaxing power is slightly higher
for compound 20, which is substituted by an n-butyl group
at C-5 than compound 21, which is unsubstituted at C-5.
This indicates that the substitution at C-5 position of the
ring has a plus effect on the activity and it is better if sub-
stituent is a phenyl group (compound 18). A similar ten-
dency is observed in the relaxing effects for compounds
16 and 17, which is greater than the relaxing effects for
compounds 20 and 21, confirming that the presence of
aromatic rings at C-5 increases the relaxing potency.
In the case of compounds 15 and 18, which have
aromatic ring at C-5, whereas C-7 position is substituted
with a ferrocenyl group and propyl group, respectively,
an increased relaxing effect was observed. When the
aromatic ring at C-5 of compound 15 was replaced with a
3,5-dimethoxyphenyl group as in compound 14, an
improvement of vasorelaxant effect was observed, which
indicates that substitution on phenyl group at position
C-5 is preferred. Compounds 12 and 13 are the most
potent, which are ferrocenyl-substituted at C-7 position.
Compounds 15 and 19 show a similar potency as shown
by the reference drug rolipram.
Inhibition of degradation of cAMP or cGMP via
Figure 2. Concentration–response curves of pyrido[2,3-d]pyrimidines
12–15, 19, and rolipram. Each point represents the mean SEM (n = 6).
*p < 0.05, significant difference compared with the respective control
(Dunnett’s t-test after analysis of variance).
phosphodiesterases inhibition.
Compounds 12–15 and
19 were evaluated as inhibitors of PDEs, as these
compounds presented the most potent vasorelaxant effect.
A colorimetric enzymatic assay was carried out, in which
these compounds were capable to avoid the degradation
of secondary messengers (cAMP and cGMP) via PDEs
inhibition. Rolipram was used as a reference compound,
which is a selective inhibitor of PDE-4. Table 3 presents
the percentage inhibition of degradation of cAMP or
cGMP by compounds 12–15 and 19. All the tested
compounds inhibit the action of PDE, and the observed
changes depend on the substrate used, and in fact, the
compounds inhibit the degradation of cAMP. When
cAMP was used as a substrate, all compounds presented
inhibitory effect from 30% to 95%, unlike when cGMP
12 > 13 >14 > 19 >rolipram >15 > 18 > 16 >17 > 20 -
21 > 11. It is observed that 15 and 19 have a similar
potency as of rolipram. From the statistical analysis of
maximum effect for compounds 11–21, it is concluded
that structural changes do not alter the effectiveness;
however, there are difference related to potency
expressed as EC50. It was observed that compounds
11–21 significantly induce concentration-dependent vaso-
relaxation in endothelium-intact aortic rings (Table 2).
Figure 2 shows concentration–response curves of com-
pounds 12–15, 19, and rolipram for a better understand-
ing. In addition, the relaxation responses to each
compound in either endothelium-intact or endothelium
denuded aortic rings were comparable (Table 2), suggest-
ing that removal of the functional endothelium does not
influence significantly on its intrinsic vasorelaxing
activity.
Table 3
Percentage inhibition of degradation of cAMP or cGMP via PDE
inhibition.
Substrate
Compound
cAMP
cGMP
The structural analysis showed that compound 11,
substituted with a ferrocenyl group at C-7 and p-
methoxyphenyl at C-5, was less potent as vasorelaxant;
however, its effectiveness is not very different from the
other compounds, except from compound 12. On compar-
ing compound 13 analogous to 11, it was inferred that a
longer chain at positions N-1 and N-3 promotes the
vasodilatory effect of pyrido[2,3-d]pyrimidines. When the
12
13
14
15
19
Rolipram
31.24
30.00
50.83
94.83
83.98
70.43
20.13
17.53
16.73
5.68
6.38
16.56
Each percentage inhibition shows the maximum inhibition effect of
compounds to 100 μM.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2015
Ferrocenyl, Alkyl, and Aryl-Pyrido[2,3-d]Pyrimidines as Vasorelaxant of Smooth
Muscle of Rat Aorta via cAMP Conservation Through Phosphodiesterase Inhibition
was used as a substrate presenting inhibitory effect of
approximately 6% and 20%. Compounds 12 and 13
present the best vasorelaxant effect, inhibiting the
degradation of cAMP (31.24% and 30.00%, respectively),
which indicates that their relaxant effect occurs through
another mechanism.
It has been well documented that the presence of PDE-1,
PDE-3, PDE-4, and PDE-5 in vascular smooth muscle tis-
sue [14,29] and an increase in the concentration of cAMP
causes relaxation of vascular smooth muscle [29] tissue,
where the pyrido[2,3-d]pyrimidines 15 and 19 showed a
similar activity than rolipram. This result suggests that 15
and 19 have a vasorelaxant effect on smooth muscle isolated
from rat aorta via inhibition of cAMP-dependent PDEs.
rolipram. Besides, both compounds have a minor Ki value
for PDE-3 (0.77 and 0.99 fold less, respectively, than for
PDE-4). Compound 19 has affinity for the catalytic site
of PDE-3. These facts demonstrate a good correlation with
the in vitro assays. Thus, compound 15, which is
substituted by a ferrocenyl group at C-7 position and by a
six-membered aromatic ring at C-5 position, is the most ef-
fective inhibitor of degradation of cAMP (Table 3) and
shows a similar vasorelaxant effect as rolipram (Table 2)
and the best Ki value for PDE-4. In the same way, com-
pound 19, which is substituted by alkyl chains at C-5 and
C-7 positions, shows a best Ki value for PDE-3 than for
PDE-4. This explains the fact that in in vitro PDE assay, 19
is more effective than rolipram and 15 is more effective than
19. Figure 3 shows the results of docking study. Compound
15 and rolipram bound to the same catalytic site formed by
the amino acids Tyr233, His234, His238, Asp392, Leu393,
Asn395, Pro396, Tyr403, Trp406, Thr407, Ile410, Phe414,
Met431, Ser442, Gln443, Phe445, and Phe446.
Molecular docking.
According to the results of the
enzyme inhibition, it was observed that compounds
12–15 and 19 are more efficient to inhibit degradation of
cAMP, and its binding to the cAMP-dependent PDEs
was predicted using a docking study. Docking studies
were carried out with the AUTODOCK 4.0.2 program,
and compounds 12–15 and 19 were evaluated by their
capability to bind PDE-3 and PDE-4 cleavage to obtain
the theory Ki. These enzymes were used as they are
cAMP-dependent and present in a considerable extent in
rat aorta.
The structures of compounds 12–15 and 19 were built
using the program HyperChem 8 release and optimized
using the Gaussian software. Initially, the ligands were
docked to the entire protein, and then the best conformations
were docked in a smaller area (grid) in order to refine the re-
sults. The compounds analyzed bound to the pocket corre-
sponding to one or more of the sites of PDE-3 and PDE-4
(Ki values are indicated in Table 4). All compounds exhib-
ited affinity to the proteins. In relation to the reference drugs,
enoximone (PDE-3 Inhibitor) and rolipram (PDE-4 inhibi-
tor) showed the proven affinity for their respective PDE.
In silico studies showed that compound 15 binds to the
catalytic site of PDE-4, in a similar way as shown by
CONCLUSION
New pyrido[2,3-d]pyrimidines 11, 12, 13, and 21 have
been synthesized and characterized. Vasorelaxant effect
via PDEs inhibition of these compounds along with known
pyrido[2,3-d]pyrimidines 14–20 has also been determined.
Apparently, there is no clear trends in potency from the dif-
ferent substitution patterns, compounds 15 and 19 are the
most potent PDE inhibitors and are more effective than
rolipram. Docking study demonstrates that compound 15
has an affinity for the catalytic site of PDE-4, and com-
pound 19 has an affinity for the catalytic site of PDE-3,
and these results have a correlation with their vasorelaxant
effects. Even though compounds 12 and 13 were presented
as the most potent vasorelaxant substances, nevertheless,
they are not so effective to inhibit degradation of cAMP
or cGMP via PDE-inhibition suggesting that 12 and 13
present their relaxant effect through another pharmacolog-
ical action mechanism.
Table 4
Finally, because hypertension is one of the most com-
mon diseases related with vascular smooth muscle contrac-
tion, and the therapeutic strategies to combat the
consequent damage to the vascular endothelium are gener-
ally aimed at modulating the molecular and biochemical
mechanisms underlying this dysfunction, such as vascular
smooth muscle contraction, the pyrido[2,3-d]pyrimidines
are suggested as possible effective antihypertensive agents
as vasodilators via PDE inhibition.
Theoretical parameters of phosphodiesterase inhibitors. Predicted activity
values of the studied compounds.
PDE-3
PDE-4
Ki
(nM)
EFEB
(kcal/mol)
Ki
(nM)
EFEB
(kcal/mol)
Compound
12
13
14
15
90.04
131.84
1110
127.36
894.61
784.38
3840
ꢀ9.61
ꢀ9.39
ꢀ8.12
ꢀ9.41
ꢀ8.25
ꢀ8.33
ꢀ7.39
151.32
87.72
444.26
71.56
1240
ꢀ9.30
ꢀ9.63
ꢀ8.67
ꢀ9.75
ꢀ8.06
ꢀ8.74
ꢀ7.24
EXPERIMENTAL
19
Rolipram
Enoximone
392.28
4920
Chemistry. 6-amino-1,3-dimethyluracil and 6-amino-1,3-
1
diprophyluracil were purchased from Aldrich. H and 13C spectra
were recorded on a JEOL GX-300 instrument (Tokyo, Japan),
Ki, estimate inhibition constant (μM). EFEB, estimating free energy of
binding (Kcal/Mol).
1
300 MHz for H and 75 MHz for 13C using CDCl3 as solvent. IR
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

I. Arellano, F. Rodríguez-Ramos, M. González-Andrade, A. Navarrete,
M. Sharma, N. Rosas, and P. Sharma
Vol 000
Figure 3. Structural models of PDE4-15 and PDE4 = rolipram complex. On the right shows the complex in magenta cartoon, yellow, and red stick the
PDE4, 15, and rolipram, respectively. The 2D representation of PDE4-15 and PDE4-rolipram complexes resulting from the docking is shown. The images
were made with PyMOL and LIGPLOT.
spectra were recorded in film on a Nicolet FT 5SX spectrophotometer
(Madison, Wisconsin, USA). Mass spectra were obtained using a
JEOL JMSAX505HA spectrometer (Tokyo, Japan). Ferrocenyl-
ethynyl ketones were prepared according to the previous report
[19]. α-alkynyl ketones were obtained as follows.
5.00 (t, 2H, J = 1.9 Hz, cp-ring), 7.00 (d, 2H, J = 10.8 Hz, 3,5-
C6H4), 7.30 (d, 2H, J = 8.6 Hz, 2,6-C6H4), 8.13 (s, 1H, C¼CH);
13C NMR (75 MHz, CDCl3, ppm): 29.1 (N(2)CH3), 31.7 (N(1)
CH3), 55.6 (OCH3), 69.6 (C–C, cp-ring), 70.4 (C–H, cp′-ring),
70.9 (C–H, cp-ring), 79.4 (C–H, cp-ring), 113.8 (C6), 119.4 (C
3,5-C6H4), 130.6 (C 2,6-C6H4) 132.5 (C1-C6H4) 150.9 (C2),
151.3 (C7), 154.7 (C5), 157.3 (C9), 160.1 (C4).
7-Ferrocenyl-1,3-dimethyl-5-(m-tolyl)-pyrido[2,3-d]pyrimidine-
2,4-dione (12). The product was obtained as orange powder
in 70% yield (mp 249.5–250°C); mass spectrum IE: m/z(%):
465 (97) M+., 400 (100); IR (film, cmꢀ1): 1700, 1657
(C¼O), 1548 (C¼N); 1H NMR (300 MHz, CDCl3, δ in
ppm): 2.50 (s, 3H, Ph-CH3), 3.40 (s, 3H,CH3N(2)), 3.84 (s,
3H, CH3N(1)), 4.16 (s, 5H, cp′-ring), 4.47 (t, 2H, J = 1.8 Hz,
cp-ring), 4.72 (t, 2H, J = 1.8 Hz, cp-ring), 7.34 (d, 1H,
J = 7.3 Hz, 4-C6H4), 7.45 (t, 1H, J = 7.5 Hz, 3-C6H4), 7.95 (d,
2H, J = 8.8 Hz, 2,6-C6H4), 8.15 (s, 1H, C¼CH);13C NMR
(75 MHz, CDCl3, ppm): 21.9 (PhCH3), 28.6 (N(2)CH3), 30.5
(N(1)CH3),68.5 (C–C, cp-ring), 70.5 (C–H, cp′-ring), 72.7
(C–H, cp-ring), 84.2 (C–H, cp-ring), 107.6 (C¼CH), 120.6
(C2-C6H4), 124.6 (C3-C6H4), 128.2 (C4-C6H4), 129.3 (C6-C6H4),
131.6 (C5-C6H4), 138.9 (C1-C6H4), 151.6 (C2), 151.9 (C7),
154.4 (C5), 157.9 (C9), 160.7 (C4).
7-Ferrocenyl-1,3-dipropyl-5-(4-methoxyphenyl)-pyrido[2,3-d]
pyrimidine-2,4-dione (13). The product was obtained as orange
powder in 70% yield (mp 166–167°C); mass spectrum IE: m/z (%):
537 (32) M+., 472 (23), 132 (100); IR (KBr, cmꢀ1): 1703, 1660
(C¼O), 1542 (C¼N); 1H NMR (300 MHz, CDCl3, δ in ppm):
0.92 (t, 3H, J = 7.4 Hz, N(2)(CH2CH2CH3), 1.05 (t, 3H,
J = 7.4 Hz, N(1)(CH2CH2CH3), 1.65 (sex, 2H, J = 7.5 Hz, N(2)
(CH2CH2CH3), 1.85 (sex, 2H, J = 7.4 Hz, N(1)(CH2CH2CH3),
3.91 (s, 3H, OCH3), 3.97 (m, 2H, N(2)(CH2CH2CH3), 4.16 (s, 5H
cp′-ring), 4.43 (m, 2H, N(1)(CH2CH2CH3), 4.45 (t, 2H,
J = 1.9 Hz, cpring), 4.68 (t, 2H, J = 1.8 Hz, cp-ring), 7.07 (d, 2H,
J = 9.1, 3,5-C6H4), 8.08 (s, 1H, C¼CH), 8.12 (d, 2H, J = 8.9, 2,6-
C6H4); 13C NMR (75 MHz, CDCl3, δ in ppm): 11.4 (N(2)
CH2CH2CH3), 11.6 (N(1)CH2CH2CH3), 21.2 (N(2)CH2CH2CH3),
In N2 atmosphere, 40 mL of dried THF was added to a flask
which was kept in a bath of acetone-dried ice to ꢀ78°C. About
30 mmol of the corresponding alkyl compound and 30 mmol of
Bu-Li were then added. After 30 min, 30 mmol of BF.3OEt2 and
45 mmol of the corresponding anhydride were added. After stir-
ring for 15 min, reaction was quenching using 2 N NaOH solution
(50 mL). The crude product was purified by flash chromatography
using ethyl acetate/hexane (90:10) as eluent.
General procedure.
Pyrido[2,3‐d]pyrimidines.
A 5N
NaOH solution (50 mL) was degassed and saturated with CO
under atmospheric pressure for 30 min, 2 mmol of Ni(CN)
2.4H2O was added to the solution, the mixture was kept at room
temperature overnight, with stirring and slow bubbling of CO,
until a pale yellow solution was obtained. Addition of 15 mmol
of KCN resulted in a color change to orange. The catalytic
species [Ni(CN)4]ꢀ4 is obtained when an excess of KCN is
added to an alkaline solution of Ni(CN)2 in CO atmosphere.
After stirring for 0.5 h, the corresponding ketoalkyne (10 mmol)
and the corresponding 6-amino-uracil (10 mmol) were added.
The evolution of the reaction was followed by TLC. At the end
of the reaction, ethyl acetate was used to extract the product.
After evaporation of the solvent followed by drying over
MgSO4, the crude product was purified by flash chromatography
using ethyl acetate/hexane (80:20) as eluent.
7-Ferrocenyl-1,3-dimethyl-5-(4-methoxyphenyl)-pyrido[2,3-
d]pyrimidine-2,4-dione (11).
The product was obtained as red
crystals in a 70% yield (mp 243–244°C); mass spectrum IE: m/z
(%): 481 (6) M+., 416 (3), 91 (100); IR (film, cmꢀ1): 1702, 1656
(C¼O), 1545 (C¼N); 1H NMR (300 MHz, CDCl3, δ in ppm):
3.38 (s, 3H,CH3N(2)), 3.81 (s, 3H, CH3N(1)), 3.88 (s, 3H,
OCH3), 4.07 (s, 5H, cp′-ring), 4.52 (t, 2H, J = 1.9 Hz, cp-ring),
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2015
Ferrocenyl, Alkyl, and Aryl-Pyrido[2,3-d]Pyrimidines as Vasorelaxant of Smooth
Muscle of Rat Aorta via cAMP Conservation Through Phosphodiesterase Inhibition
for 60 min, and they were washed with fresh Krebs solution at
15-min interval before starting the experiments. After
stabilization period, the rings were contracted with NE (0.1 μM)
two times at 30-min interval. The integrity of the endothelium
was verified by the relaxant response to ACh (10 μM); intact
endothelium showed relaxation over 84%, the endothelium
absence (denuded aorta) did not relax with Ach. About 30 min
after the tissues were contracted with NE (0.1 μM), cumulative
concentrations of synthesized compounds or reference drugs
were added to the bath to yield the required aorta relaxant
effects and allowed to reach a steady state at each concentration.
The concentration required to relax 50% of the aorta pre-
contracted with NE that was expressed as EC50.
21.4 (N(1)CH2CH2CH3), 43.4 (N(2)CH2CH2CH3), 44.8 (N(1)
CH2CH2CH3), 55.6 (OCH3), 68.8 (C–C, cp-ring), 70.2 (C–H, cp′-
ring), 72.2 (C–H, cp-ring), 84.9 (C–H, cp-ring), 114.6 (C6), 119.3
(C 3,5-C6H4), 128.7 (C 2,6-C6H4), 130.1 (C1-C6H4), 151.6 (C2),
154.0 (C7), 154.9 (C5), 157.2 (C9), 160.5 (C4), 161.7 (C4-C6H4).
7-propyl-1,3-dimethylpyrido[2,3-d]pyrimidine-2,4-dione (21).
The product was obtained as white needles in 70% yield (mp
86–86.5°C); mass spectrum IE: m/z (%): 233 (14) M+., 205 (100);
1H NMR (300 MHz, CDCl3, δ in ppm): 0.91 (t, 3H, CH3), 1.72
(m, 2H, CH3CH2CH2), 2.74 (t, 2H, CH3CH2CH2), 3.40 (s, 3H, N
(1)CH3), 3.65 (s, 3H, N(2)CH3), 6.96 (d, 1H, CH¼CH) 8.26 (d,
1H, CH¼CH); 13C NMR (75 MHz, CDCl3, δ in ppm): 13.8
(CH3CH2CH2), 22.3 (CH3CH2CH2), 28.3 (N(2)CH3), 29.3 (N(1)
CH3), 40.6 (CH3CH2CH2), 108.2 (C7), 118.2 (C6), 137.5 (C5),
150.5 (C10), 151.0 (C9), 161.6 [N(C¼O)N], 168.4 [N(C¼O)C].
Phosphodiesterase inhibition procedure.
For the PDE
assay, the drugs were dissolved in a binary mixture of
acetonitrile (Burdick & Jackson, MI, USA)/water (Direct-Q
Millipore water purification system) (1:1). PDE enzyme from
bovine brain, 5′nucleotidase from Crotalusatrox venom, 3′,5′c-
AMP, 3′,5′-cGMP, buffer (10-mM Tris-HCl, pH 7.4), BIOMOL
GREENTM reagent and IBMX were purchased from BIOMOL
International, Inc. (PA, USA).The enzyme inhibition assay was
performed using BIOMOL cyclic nucleotide PDE assay kit,
consisting in a colorimetric non-radioactive assay designed in a
microplate format. The basis for this assay is the cleavage of
cAMP or cGMP by a cyclic nucleotide PDE. The 5′-nucleotide
released is further cleaved into the nucleoside and phosphate by
the enzyme 5′-nucleotidase. The phosphate released because of
enzymatic cleavage is quantified using BIOMOL GREEN
reagent in a modified malachite green assay [31]. It may be
used to screen inhibitors and modulators of cyclic nucleotide
PDE activity. The compounds and reference drugs were tested
at different concentrations: ferrocenyl, alkyl and aryl-pyrido[2,3-
d]pyrimidines (1.0–300 μM), rolipram (3.0–300 μM), and
sildenafil (0.001–300 μM).
Other pyrido[2,3-d]pyrimidines 14, 15, 16, 17, 18, 19, and 20
were synthesized according to our earlier reports. 7-Ferrocenyl-
5-(3,5-dimethoxyphenyl)-1,3-dimethylpyrido[2,3-d]pyrimidine-2,4-
dione (14) and 7-Ferrocenyl-1,3-dimethyl-5-phenyl-pyrido[2,3-d]
pyrimidine-2,4-dione (15) [19]. 1,3-dimethyl-5,7-diphenylpyrido
[2,3-d]pyrimidine-2,4-dione (16); [30] 7-Ethyl-1,3-dimethyl-5-phen-
ylpyrido[2,3-d]pyrimidine-2,4-dione (17); 1,3-Dimethyl-5-phenyl-7-
propylpyrido[2,3-d]pyrimidine-2,4-dione (18); 5-n-Butyl-7-ethyl-
1,3-dimethylpyrido[2,3-d]pyrimidine-2,4-dione (19) and 5-n-Bu-
tyl-1,3-dimethyl-7-n-propylpyrido[2,3-d]pyrimidine-2,4-dione
(20) [18].
Vasorelaxant procedure.
Acetylcholine chloride (ACh),
(ꢀ) NE, rolipram, and enoximone were purchased from Sigma
Chemical Co. (St. Louis, MO, USA). Sildenafil citrate was used
as USP reference standard. The drugs were suspended in 0.05%
Tween 80 in distilled water. The final concentration of Tween
80 was in trace (less than 0.0005%) and did not affect the
tracheal or vascular response. The drug solution or suspensions
was freshly prepared each time few minutes before the
experimentation. Male rat (Wistar) weighting 200–300 g was
obtained from Harlan México (México). Animals were
maintained at constant room temperature (22 2°C) and
submitted to 12 h light/dark cycle with free access to food and
water. Procedures involving animal care were conducted in
conformity with the Mexican Official Norm for Animal Care
and Handing (NOM-062-ZOO-1999; Especificaciones Técnicas
para la Producción, Cuidado y Uso de Animales de Laboratorio)
and in compliance with international rules on the care and use
of laboratory animals. Furthermore, clearance for conducting the
studies was obtained from the Ethics Committee for the Use of
Animals in Pharmacological and Toxicological Testing,
Facultad de Química, UNAM. Male Wistar rats were euthanized
in a CO2 chamber, and thoracic aorta was removed and
immediately immersed in a Krebs solution at 37°C. After
removal the excess of connective tissue and fat, the aorta was
divided into eight small rings of about 2 mm in length. Each
aorta ring was hung between two nichrome hooks inserted into
the lumen and placed in a 10-mL organ bath containing Krebs
solution (composition in mM: NaCl 118, KCl 4.7, NaH2PO4
1.2, MgSO4°7H2O 1.2, CaCl2°2H2O 2.5, NaHCO3 25, glucose
11.1) at 37°C and bubbled constantly with 5% CO2–95% O2.
Isometric tension was recorded through an eight-channel
Biopack System polygraph MP100 via a Grass FT 03E force
transducer. The data were digitalized and analyzed by mean of
software for data acquisition (Acknowledge 3.9.0). Tissues were
placed under a resting tension of 4.0 g and allowed to stabilize
Data analysis.
The EC50 values for relaxant effect were
calculated by linear regression and are shown as mean SEM of
at least six experiments. The differences among these obtained
values were statistically calculated by one-way analysis of
variance and then determined by Dunnett’s t-test [16]. p < 0.05
was considered as statistically significant. The inhibitory
activities of PDE by ferrocenyl, alkyl, and aryl-pyrido[2,3-d]
pyrimidines or reference drugs are reported as IC50 SEM or
maximal percent of inhibition of at least three experiments.
Computational method. Docking was carried out using the
X-ray structure of PDE3B in complex with a dihydropyridazine
inhibitor (1SO2.pdb) and PDE4B complexed with rolipram
(1RO6). The ligands 12–15, 19, rolipram, and enoximone were
built using the program HyperChem 8.0 and optimized using the
density functional method, at the B3LYP/6-31G* level [32], with
the program Gaussian 09, revision A.02 (Gaussian Inc.,
Wallingford, CT, USA). The protein complex and the ligand
were further prepared using the utilities implemented by
AutoDockTools 1.5.4, revision 30 (http://mgltools.scripps.edu/).
The protein complex was adding polar hydrogen atoms, Kollman
united-atom partial charges and to the ligands computing
Gasteiger–Marsilli formalism charges, rotatable groups which
were assigned automatically as were the active torsions. The
initial grid box size was 60Ǻ × 60Ǻ × 60Ǻ in the x, y, and z-
dimensions. Blind docking was carried out using AutoDock4
version 4.2 software http://autodock.scripps.edu/ [33,34] using
the default parameters, the Lamarkian genetic algorithm with
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

I. Arellano, F. Rodríguez-Ramos, M. González-Andrade, A. Navarrete,
M. Sharma, N. Rosas, and P. Sharma
Vol 000
[12] Draheim, R.; Egerland, U.; Rundfeldt, C. J Pharmacol Exp
Therapeut 2003, 308, 555.
[13] Pitts, W. J.; Barbosa, J. Quinazoline and pyrido[2,3-d]py-
rimidine inhibitors of phosphodiesterase (PDE) 7. WO/2002/102315,
Dec 27, 2002.
[14] Lugnier, C. Pharmacol Ther 2006, 109, 366.
[15] Garcia-Sáinz, J. A.; Villalobos-Molina, R. Eur J Pharmacol
2004, 500, 113.
[16] Kümmerle, A. E.; Raimundo, J. M.; Leal, C. M.; da Silva, G. S.;
Balliano, T. L.; Pereira, M. A.; de Simone, C. A.; Sudo, R. T.; Zapata-Sudo, G.;
Fraga, C. A. M.; Barreiro, E. Eur J Med Chem 2009, 44, 4004.
[17] Harris, M. D.; Chen, X.; Pesant, S.; Cohn, I. H.; MacDonell,
M. S.; Boucher, M.; Vinge, E. L.; Raake, P.; Moraca, R. S.; Li, D.;
Most, P.; Houser, R. S.; Koch, J. W.; Eckhart, D. A. J Mol Cell
Cardiol 2009, 46, 100.
local search, number of individuals in population (150), maximum
number of energy evaluations (2.5 million), maximum number of
generations (27,000), rate of gene mutation (0.02), rate of
crossover (0.8), and 200 runs for docking. Electrostatic grid
maps were generated for each atom type in the ligands using the
auxiliary program AutoGrid4 part of the software AutoDock4.
All calculations were made using a parallel distributed memory
supercomputer (Kanbalam, Dirección General de Cómputo y de
Tecnologías de Información y Comunicación, UNAM), which
contains 1368 processors AMD Opteron, around 3 terabyte of
memory and 160 terabyte of storage (http://www.super.unam.mx/
). The analysis of the docking was made with AutoDockTools
using cluster analysis and program LIGPLOT and PyMOL [35].
[18] Rosas, N.; Sharma, P.; Álvarez, C.; Cabrera, A.; Ramírez, R.;
Delgado, A.; Arzoumanian, H. J. J Chem Soc Perkin Trans 1 2001, 2341.
[19] Arellano, I.; Sharma, P.; Arias, J. L.; Toscano, A.; Cabrera, A.;
Rosas, N. J Mol Catal Chem 2007, 266, 294.
[20] Fouda, M. F. R.; Abd-Elzaher, M. M.; Abdelsamaia, R. A.;
Labib, A. A. Appl Organomet Chem 2007, 21, 613.
[21] Top, S.; Vessieres, A.; Cabestaing, C.; Laios, I.; Leclereq, G.
J Organomet Chem 2001, 637-639, 500.
Acknowledgments. I. Arellano acknowledges fellowship (No.
1715, 200409) from CONACyT.
REFERENCES AND NOTES
[22] Hocek, M.; Stepnicka, P.; Ludvik, J.; Císarová, I.; Votruba, I.;
Reha, D.; Hobza, P. Chem Eur J 2004, 10, 2058.
[23] van Staveren, D. R.; Metzler-Nolte, N. Chem Rev 2004,
104, 5931.
[1] Chhabria, M. T.; Srinivas, S.; Rajan, K. S.; Ravikumar, T.;
Rathnam, S. Arzneimittelforschung 2002, 52, 792.
[2] El-Gazzar, A. B. A.; Hafez, H. N. Bioorg Med Chem Lett
2009, 19, 3392.
[24] Marivet, M. C.; Bourguignon, J. J.; Lugnier, C.; Mann, A.;
Stoclet, J. C.; Wermuth, C. G. J Med Chem 1989, 7, 1450.
[25] Brown, H. C.; Racherla, U. S.; Singh, S. M. Tetrahedron Lett
1984, 25, 2411.
[26] Arellano, I.; Sharma, P.; Cabrera, A.; Pérez, D.; Arias, J. L.;
Rosas, N. J Organomet Chem 2009, 694, 3823.
[27] Wu, X. F.; Sundararaju, B.; Neumann, H.; Dixneuf, P. H.;
Beller, M. Chem Eur J 2011, 17, 106.
[28] Suárez-Ortiz, G. A.; Sharma, P.; Amézquita-Valencia, M.;
Arellano, I.; Cabrera, A.; Rosas, N. Tetrahedron Lett 2011, 52, 1641.
[29] Zypp, A.; Rechtman, M.; Majeweski, H. Gen Pharmacol 2000,
34, 245.
[30] Wawzonek, S. J Org Chem 1976, 41, 3149.
[31] Beers, S. A. Bioorg Med Chem 1997, 5, 2203.
[32] da Silva, C. H.; Del Ponte, G.; Neto, A. F.; Taft, C. A. Bioorg
Chem 2005, 33, 274.
[33] Morris, G. M.; Goodsell, D. S. J Comput Chem 1998, 19,
1639.
[3] Pastor, A.; Alajarin, R.; Vaquero, J. J.; Alvarez-Builla, J.;
Casa-Juana, M. F.; Sunkel, C.; Priego, J. G.; Fonseca, I.; Sanz-Aparicio, J.
Tetrahedron, 1994, 50, 8085.
[4] Moreno, E.; Plano, D.; Lamberto, I.; Font, M.; Encío, I.; Palop,
J. A.; Sanmartín, C. Eur J Med Chem 2012, 47, 283.
[5] Nasr, M. N.; Gineinah, M. M. Arch Pharm Pharm Med Chem
2002, 6, 289.
[6] Matsumoto, J.; Minami, S. J Med Chem 1975, 18, 74.
[7] Quintela, J. M.; Peinador, C.; Botana, L.; Estevez, M.;
Riguera, R. Bioorg Med Chem Lett 1997, 5, 1543.
[8] Cowart, M.; Lee, C. H.; Gfesser, G. A.; Bayburt, E. K.;
Bhagwat, S. S.; Stewart, A. O.; Yu, H.; Kohlhaas, K. L.; McGaraughty, S.;
Wismer, C. T.; Mikusa, J.; Zhu, C.; Alexander, K. M.; Jarvis, M. F.;
Kowaluk, E. A. Bioorg Med Chem Lett 2001, 11, 83.
[9] Hamby, J. M.; Connolly, C. J. C.; Schroeder, M. C.; Winters,
R. T.; Showalter, H. D. H.; Panek, R. L. J Med Chem 1997, 40, 2296.
[10] Ibrahim, D. A.; Ismail, N. S. M. Eur J Med Chem 2011,
46, 5825.
[34] Huey, R.; Morris, G. M.; et al. J Comput Chem 2007, 28, 1145.
[35] DeLano, W. L. Abstr Paper Am Chem Soc 2004, 228, U313.
[11] Nam, G.; Yoon, C. M.; Kim, E.; Rhee, C. K.; Kim, J. H.; Shin,
J. H.; Kim, S. H. Bioorg Med Chem Lett 2001, 11, 611.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet