Searching for balanced hybrid NO-donor 1,4-dihydropyridines with basic properties

Article abstract of DOI:10.1023/A:1010992412549

Purpose. Model compounds containing NO-donor furoxan moieties at the 3-positioned basic lateral chain of 1, a 1,4-dihydropyridine related to nicardipine, were synthesized in order to study their vasodilating activity as well as their basic and lipophilic

Full text of DOI:10.1023/A:1010992412549

Pharmaceutical Research, Vol. 18, No. 7, 2001
Research Paper
joined to the nitrogen in the lateral chain by a spacer selected
on the basis of the synthetic accessibility of the final models.
In this preliminary communication we report synthesis,
NO-release, and vasodilating activity of the new products 2, 3,
and 4. We also evaluated their pKa and log D, tools useful for
the next pharmacokinetic study.
Searching for Balanced Hybrid
NO-Donor 1,4-Dihydropyridines with
Basic Properties
Donatella Boschi,1 Giulia Caron,1 Sonja Visentin,1
Antonella Di Stilo,1 Barbara Rolando,1
MATERIALS AND METHODS
Synthesis
Roberta Fruttero,1 and Alberto Gasco1,2
Melting points were measured on a Bu¨chi 530 capillary
apparatus and are uncorrected. Melting points with decom-
position were determined after introducing the sample into
the bath at a temperature 10°C lower than the melting point.
A heating rate of 3°C min−1 was used. Infrared spectroscopy,
1H and 13C nuclear magnetic resonance at 200 MHz and 50
MHz, respectively, and mass spectroscopy routinely checked
the compounds. All the spectra were in accordance with the
expected structures. Column chromatography was performed
on silica gel (Merck Kieselgel 60, 230–400 mesh ASTM,
Merck, Milan) with the indicated solvent system. Petroleum
ether 40°–60°C (PE) was used. Solvent removal was achieved
under reduced pressure at room temperature. Elemental
analyses of the new compounds were performed by REDOX
(Cologno Monzese) and the results were within 0.4% of the
theoretical values. Intermediates 1 (3), 5, 6 (1), 7 (5), and 8
(4), were synthesized according to procedures described in
the literature. Compound 10, 3-nitrobenzaldehyde, and meth-
yl 3-aminocrotonate were commercial reagents (Aldrich
Chemical Co., Milwaukee, WI).
Received January 16, 2001; accepted April 5, 2001
Purpose. Model compounds containing NO-donor furoxan moieties
at the 3-positioned basic lateral chain of 1, a 1,4-dihydropyridine
related to nicardipine, were synthesized in order to study their vaso-
dilating activity as well as their basic and lipophilic behaviour.
Methods. All the compounds were obtained by a modified Hantzsch
approach. Potentiometry was used to determine pKa and lipophilicity
descriptors. The furoxan 4-aryl-1,4-dihydropyridines were assessed
for their ability to release nitrite, in the presence of a large excess of
cysteine, by the Griess reaction. Vasodilating activity of the products
in the absence and in the presence of ODQ, a well-known guanylate
cyclase inhibitor, was evaluated on rat thoracic aorta.
Results. The compounds display low basicity values and for this rea-
son their log Ds at physiological pH are identical to the log Ps of the
neutral forms. Products 2, 3 display vasodilating action principally
dependent on their Ca2+-antagonist properties, whereas 4 behaves as
a well-balanced hybrid with mixed Ca2+-channel blocker and NO-
dependent vasodilator activities.
Conclusions. Nitrogen containing lateral chain at the 3-position of 1
is a suitable molecular region to be modified in order to obtain well-
balanced furoxan NO-donor 1,4-DHPs. This manipulation produces a
decrease in the basicity. General analysis of pKa and lipophilicity
descriptors of these new DHPs suggest that molecular flexibility
could influence both their basicity and log PI.
General Procedure for the Synthesis of 1,4-DHPs 2, 3
The appropriate 4-bromomethylfuroxan 5 or 6 (4.8
mmol) was added to a solution of 1 (1.86 g, 4.8 mmol) in a
mixture of acetone (30 ml) and 0.5 N KHCO3 (20 ml). The
solution was stirred for 6 h and then was neutralized with 1N
HCl; the acetone was evaporated and the aqueous solution
was extracted with ethyl acetate. The dried organic phases
were evaporated, and the residue was purified by flash-
chromatography. The purified fractions were dissolved in
HCl saturated methanolic solution. Solvent removal gave the
final product as a hydrochloride. Chromatographic solvents,
yields, melting points, and analytical data were as follows:
Methyl 2-(N-methyl-N-(3-methylfuroxan-4-yl)methyl)-
aminoethyl 1,4-dihydro-2,6-dimethyl-4-(3-nitrophenyl)-3,5-
pyridinedicarboxylate hydrochloride (2). Eluent: CH2Cl2/
MeOH 9.7/0.3; yield 44%; softening at 79°C decomposed at
110°C. Anal. (C23H27N5O8⅐HCl⅐0.5H2O) C, H, N.
KEY WORDS: 1,4-Dihydropyridines; nitric oxide (NO); NO do-
nors; hybrid drugs; furoxans.
INTRODUCTION
Recently a research program was started in our labora-
tory aimed at obtaining new “hybrid” 1,4-dihydropyridines
(1,4-DHPs) endowed with mixed nitric oxide (NO)-like and
Ca2+-channel antagonist vasodilating properties. Nifedipine
was first selected as a model for the development of these new
drugs. We substituted the o-nitro group with o- and m-
positioned furoxan moieties at the 4-phenyl ring, gifted with
different NO-releasing property (1). In a further study these
same substructures were introduced at the 3-positioned chain
of the lead (2).
We have now undertaken the study of basic NO-donor
1,4-DHPs obtained by introducing NO-donor furoxan sys-
tems on the nitrogen lateral chain of product 1, whose struc-
ture is related to nicardipine. The furoxan ring bears suitable
substituents chosen for modulating its ability to release NO to
obtain well-balanced hybrids. The NO-donor moieties were
Methyl 2-(N-methyl-N-(3-carbamoylfuroxan-4-yl)-
methyl)aminoethyl 1,4-dihydro-2,6-dimethyl-4-(3-nitro-
phenyl)-3,5-pyridindicarboxylate hydrochloride (3). Eluent:
CH2Cl2/MeOH 9.9/0.1→9.7/0.3; yield 84%; softened at 135°C
and then slowly decomposed until 151°C. Anal.(C23H26N6O9⅐
HCl⅐H2O) C, H, N.
4-(2-(N-(2-Hydroxyethyl)-N-tert-butoxycarbonylamino)-
ethyloxy)-3-phenylsulfonylfuroxan (9)
1 Dipartimento di Scienza e Tecnologia del Farmaco, Universita` degli
Studi di Torino, Via P. Giuria 9, I-10125 Torino, Italy.
To whom correspondence should be addressed. (e-mail:
Fifty percent NaOH (3.28 g, 41 mmol) was added drop-
wise to a solution of 7 (12.44 g, 34 mmol) and 8 (7.02 g, 34
mmol) in THF (100 ml) at −5°C. The mixture was stirred at
2
gasco@pharm.unito.it)
987
0724-8741/01/0700-0987$19.50/0 © 2001 Plenum Publishing Corporation

988
Boschi et al.
−15°C for 2.5 h, then phosphate buffer (pH 5.4, 0.5 M, 100 ml) mmol) in 2-propanol (80 ml) was refluxed for 10 h. Solvent
was added and the solution obtained was extracted with removal gave a residue, which was purified by flash chroma-
CH2Cl2. The dried organic phases were evaporated and the tography (eluent: PE/EtOAc/MeOH 7/2.5/0.5). The product
residue was purified by flash-chromatography (eluent: PE/ obtained was a yellow solid. Yield 51%, softened at 79°C
EtOAc 8/2). The purified product was an unstable oil that was decomposed at 105°C. Anal.(C33H37N5O13S) C, H, N.
immediately used for the preparation of the intermediate 11.
The yield was quantitative.
Methyl 2-(N-(2-(3-phenylsulfonylfuroxan-4-
yloxy)ethyl)aminoethyl 1,4-dihydro-2,6-dimethyl-4-(3-
nitrophenyl)-3,5-pyridinedicarboxylate hydrochloride (4)
N-(2-(3-Phenylsulfonylfuroxan-4-yloxy)ethyl)-N-tert-
butoxycarbonyl-2-aminoethyl acetoacetate (11)
HCl saturated diethyl ether was added to a solution of 12
in dry CH2Cl2 and the mixture was stirred for 48 h. Solvent
removal gave a residue that was triturated with dry diethyl
ether, filtered, and dried. Yield 66%; darkening at 115°C,
decomposed at 130°C. Anal. (C28H29N5O11S⅐HCl⅐0.5H2O) C,
H, N.
A solution of 9 (2.06 g, 4.8 mmol) and 2,2,6-trimethyl-
4H-1,3-dioxin-4-one (10) (0.68 ml, 4.8 mmol) in toluene (20
ml) was refluxed for 1 h. Solvent removal gave a residue,
which was purified by flash chromatography (eluent: PE/
EtOAc 7/3). Yield 60%, oil. Anal. (C21H27N3O10S) C, H, N.
Methyl N-(2-(3-phenylsulfonylfuroxan-4-yloxy)ethyl)-N-
tert-butoxycarbonyl-2-aminoethyl 1,4-dihydro-2,6-dimethyl-4-
(3-nitrophenyl)-3,5-pyridinedicarboxylate (12)
pH-Metric Approach to Obtain pKa and Log P
Potentiometric titrations of compounds (1–4) were per-
formed with the GlpKa apparatus (6,7) (Sirius Analytical In-
struments Ltd, Forrest Row, East Sussex, UK). Ionization
A solution of 11 (3.50 g, 6.8 mmol), 3-nitrobenzaldehyde
(1.03 g, 6.8 mmol), and methyl 3-aminocrotonate (0.78 g, 6.8
Fig. 1. Schemes illustrating the synthesis of the compounds 2, 3 (A), 4 (B). a) 5 or 6, KHCO3 0.5N, acetone; b) HCl
saturated methanol; c) 8, 50% NaOH, THF, −15° C; d) 10, refluxed toluene; e) modified Hantzsch procedure: 3-nitro-
benzaldehyde, methyl 3-aminocrotonate, refluxed iPrOH; f) HCl saturated diethyl ether.

Balanced Hybrid NO-Donor DHPs
989
constants were determined according to Ref. (8). The low the lateral chain of the strong electron withdrawing furoxan
aqueous solubility of the compounds required pKa measure- moieties (11), with the only exception derivative 3. Simple
ments in the presence of methanol as a cosolvent.
pKa calculations confirm this assumption (12). Indeed, intra-
To obtain lipophilicity data, at least four separate titra- molecular effects connected with molecular flexibility could
tions of ca. 0.5 mM for each compound, containing various also be important in determining the basicity behavior of
volumes of octan-1-ol (from 0.5 ml of organic solvent/20 ml of these compounds.
H2O to 13 ml of organic solvent/7 ml of H2O), were per-
Analysis of lipophilicity data shows that, for all the in-
formed in the pH range 1.8 to 9. The titrations were carried vestigated compounds, there is a rather good agreement be-
out under N2 at 25.0 0.1°C (9). Final data were obtained by tween log PN, namely the experimental lipophilicity value of
the Multiset approach as described elsewhere (10) and vali- the neutral form, and the corresponding CLOGP value (see
dated by shake-flask when necessary.
RESULTS AND DISCUSSION
Chemistry
Table I). This means that the lipophilicity behavior of the
neutral forms is not heavily influenced by intramolecular ef-
fects, which depend on the molecular flexibility not codified
in the CLOGP algorithm. All the products display quite simi-
lar log PN values with the only exception the more lipophilic
phenylsulfonyl 4. The distribution coefficient at the physi-
ological pH (log D7.4) of 1, is different from the correspond-
ing log PN, due to the high pKa value of this compound,
whereas the low basicity of the furoxan derivatives 2 and 3
Syntheses of the 1,4-DHPs 2–4 are reported in Fig. 1.
1,4-DHPs 2 and 3 were easily obtained by stirring in acetone
in the presence of potassium bicarbonate 2-(N-
methylamino)ethyl methyl 2,6-dimethyl-4-(m-nitrophenyl)-
1,4-dihydropyridine-3,5-dicarboxylate (1) with 4-bromometh-
ylfuroxan 5 and 6, respectively (Fig. 1A). Synthesis of the
1,4-DHP 4 required preliminary preparation of acetylacetic
ester 11 (Fig. 1B). This intermediate was synthesized by treat-
ing 3,4-bis(phenylsulfonyl)furoxan (7) with N-Boc diethanol-
amine (8), in THF in the presence of 50% sodium hydroxide,
at −15°C. The reaction of 9 with the synthon of acetylketene
10 in boiling toluene afforded the acetylacetate 11. 1,4-DHP
12 was obtained by the reaction of 11 with m-nitrobenzalde-
hyde and methyl 3-aminocrotonate in refluxing isopropanol.
Deprotection of the nitrogen at the lateral chain with hydro-
chloric acid in diethyl ether gave the expected 1,4-DHP 4.
All the 1,4-DHPs were obtained and tested as racemic
renders their log PN ס log D7.4
.
Recently, the diff(log PN-I), namely the difference be-
tween the lipophilicity of neutral and ionized species of a
given compound, was successfully used (8) to investigate the
lipophilic behavior of ionized species of compounds. As a
rule, the diff(log PN-I) covers a range from 3 to 4 in the
octanol/water system. Among the diff(log PN-I) entered in
Table I, only that of 3 is in keeping. This suggests that mo-
lecular flexibility should influence the lipophilicity behavior
of the remaining cations.
NO-Release
The capacity of all the final 1,4-DHPs to produce nitrite
mixtures as previously described for other NO-donor 1,4- in physiological conditions, in the presence of an excess of
DHPs (2 and references therein). cysteine, was evaluated by the procedure previously de-
All furoxan 1,4-DHPs discussed in the present work bear scribed (13). The results are entered in Table II. The forma-
a basic nitrogen chain. The pKas of compounds 2–4 (Table I) tion of nitrite is relevant for sulfonyl derivative 4, intermedi-
are quite a bit lower than the pKa of the reference 1. As a ate for carbamoyl compound 3, and negligible for methyl de-
consequence, all the hybrid products, at physiological pH, rivative 2. Detection of nitrite can be used to infer the
exist largely in the neutral form, unlike the lead. Such low previous presence of NO. It can be used as a measure of the
basicity properties are not fully explained by the presence in original concentration of nitric oxide but only in a first ap-
Table I. Ionisation Constants and Lipophilic Descriptors in Octanol/water of the Compounds Under Study
a
Compound
pKa
CLOGPb
log PN c
diff(log Pexp-calc d
)
log PI e
diff(log PN-I f
)
log D7.4 k
1
2
3
4
8.78 0.04
3.83 0.04
5.07 0.02
4.59 0.01
3.06
2.84
2.18
5.30
2.86 0.01
2.86 0.06g
2.85 0.02
−0.20
0.02
0.67
—
0.35 0.02
0.85 0.09h
−0.08 0.02i
—
2.51
2.01
2.93
—
1.50
2.86
2.85
—
j
—
a Determined by potentiometry; MeOH as cosolvent was used in percentage ranging from 30 to 65. The extrapolation
to zero was obtained by the Yasuda-Shedlovsky procedure (15).
b Calculated log P values by CLOGP programme v. 4.0, BioByte Corp.
c Log P of the neutral form of the compound determined by potentiometry.
d Log PN minus CLOGP.
e Log P of the cationic form of the compound determined by potentiometry otherwise specified.
f Log PN − log PI
g Obtained by mixed procedure: potentiometry combined with shake-flask.
h Validated by shake-flask.
i Obtained by shake-flask.
j Insoluble in either water or octanol.
a
1
10pK −pH
D = PN и
+ PI и
k Calculated from the following equation:
ͩ ͪ ͩ ͪ
1 + 10pK −pH
1 + 10pK −pH
a
a

990
Boschi et al.
Table II. Nitrite Formation and Pharmacological Results
ODQ
%NO−2
SEa
EC50
EC50
Compound
SE (nM)b
SE (nM)c
Nicardipine
—
—
1.0 0.1
65 0.6
2.5 0.6
—
—
2.5 0.4
1
2
3
4
0.20 0.11
3.8 0.1
34.3 1.7
38
30
5
2
35
1
100 10***d
a Nitrite (%NO−2 mol/mol) was determined according to the Griess
reaction in the presence of L-cysteine.
b Vasodilating activity.
c Vasodilating activity performed in the presence of 1 ␮M ODQ.
d P < 0.0001 when compared to EC50 value in the absence of ODQ;
Student’s t-test for unpaired values.
proximation because, in these conditions, NO can undergo a
variety of reactions besides oxidation (14).
Pharmacology
Fig. 2. Concentration-response curves for vasodilator activity of com-
pound 4, in the presence and in the absence of ODQ. All points are
mean values SE from independent experiments.
Vasodilator activities of 1,4-DHPs were evaluated on K+-
depolarized rat thoracic aorta strips by the procedure previ-
ously described (1). EC50 values, calculated from concentra-
tion-response curves, are reported in Table II, as are the val-
ues in the presence of 1 ␮M 1H-1,2,4-oxadiazolo[4,3-
a]quinoxalin-1-one (ODQ), a well-known inhibitor of the
soluble guanylate cyclase (sGC) (EC50ODQ). Analysis of the
vasodilating properties (EC50, Table II) shows that introduc-
tion of furoxan substructures into the basic 3-lateral chain
affords vasodilating agents slightly more potent than the lead
1 when 3-carbamoyl (derivative 3) and 4-phenylsulfonyl (de-
rivative 4) substituted furoxan moieties are used. By using the
3-methylfuroxan substructure, the rather more potent 1,4-
DHP 2 is obtained. This last compound displays an activity
comparable to that of nicardipine.
For compounds 2 and 3, vasodilating potencies evaluated
in the presence of 1 ␮M ODQ are the same as those deter-
mined in its absence (Table II, EC50ODQ). This means that in
the vasodilator response only Ca2+-channel antagonism is in-
volved. By contrast 4, which is a more potent NO-donor, in
the presence of ODQ decreases its potency by about three
times. This indicates that NO is involved in the response. No
further shift of the concentration-response curve was ob-
served on increasing the inhibitor concentration to 2 ␮M
ODQ. It is reasonable to assume that the vasodilating activity
retained in the presence of ODQ is principally due to its
Ca2+-channel antagonist properties (1,2). Thus, the introduc-
tion in 1 of the 3-methylfuroxan moiety and, to a lesser extent,
of the 3-carbamoyl one, is beneficial for blocking the 1,4-DHP
receptor, whereas the introduction of the 3-phenylsulfonylfu-
roxan substructure is detrimental. Analysis of the rightward
shift of the concentration-response curve of 4 (Fig. 2) shows
that the compound behaves as a well-balanced hybrid. In fact,
in a large part of the tested concentration range, it triggers
vasodilation, which depends on its NO-donor as well as its
Ca2+-blocker properties. The decrease in the ability to block
the 1,4-DHP receptor and the increase of NO-donor proper-
ties are responsible for the balance of this hybrid with respect
to unbalanced hybrids 2, 3.
nicardipine analogue 1 as a reference model to build “hybrid”
1,4-DHPs endowed with mixed NO-like and Ca2+-channel
antagonist vasodilating properties can lead to the achieve-
ment of well-balanced hybrids when an appropriate substi-
tuted NO-donor furoxan moiety is selected. General analysis
of pKa and lipophilicity descriptors of these new DHPs sug-
gests that molecular flexibility could influence both their ba-
sicity and log PI.
ACKNOWLEDGMENTS
This work was supported by a grant from MURST Studi
e Ricerche Finalizzate 40%, Roma.
REFERENCES
1. A. Di Stilo, S. Visentin, C. Cena, A. M. Gasco, G. Ermondi, and
A. Gasco. New 1,4-dihydropyridines conjugated to furoxanyl
moieties, endowed with both nitric oxide-like and calcium chan-
nel antagonist vasodilator activities. J. Med. Chem. 41:5393–5401
(1998).
2. C. Cena, S. Visentin, A. Di Stilo, D. Boschi, R. Fruttero, and A.
Gasco. Studies on agents with mixed NO-dependent and calcium
channel antagonist vasodilating activities. Pharm. Res. 18:157–
165 (2001).
3. T. Shibanuma, M. Iwanami, M. Fujimoto, T. Takenaka, and M.
Murakami. Synthesis of the metabolites of 2-(N-benzyl-N-
methylamino)ethyl methyl 2,6-dimethyl-4-(m-nitrophenyl)-1,4-
dihydropyridine-3,5-dicarboxylate hydrochloride (nicardipine hy-
drochloride, YC-93). Chem. Pharm. Bull. 28:2609–2613 (1980).
4. S. C. Bergmeier and S. L. Fundy. Synthesis of oligo(5-amino-
pentanoic acid)-nucleobases (APN): Potential antisense agents.
Bioorg. Med. Chem. Lett. 7:3155–3138 (1997).
5. G. Sorba, G. Ermondi, R. Fruttero, U. Galli, and A. Gasco. Un-
symmetrically substituted furoxans. Part 16. Reaction of benzen-
esulfonyl substituted furoxans with ethanol and ethanethiol in
basic medium. J. Heterocyclic Chem. 33:327–334 (1996).
6. A. Avdeef. pH-metric log P. Part 1. Difference plots for deter-
mining ion-pair octanol-water partition coefficients of multiprotic
substances. Quant. Struct.-Act. Relat. 11:510–517 (1992).
7. G. Caron, P. Gaillard, P. A. Carrupt, and B. Testa. Lipophilicity
In conclusion, these findings indicate that the use of

Balanced Hybrid NO-Donor DHPs
991
behavior of model and medicinal compounds containing a sulfide,
sulfoxide, or sulfone moiety. Helv. Chim. Acta 80:449–462 (1997).
8. R. Fruttero, G. Caron, E. Fornatto, D. Boschi, G. Ermondi, A. 12. D. D. Perrin, B. Dempsey, and E. P. Serjeant. pKa Prediction for
O. Exner. Electronic substituent effects of furoxan and furazan
system. J. Chem. Res. (S): 495–496 (1998).
Gasco, P. A. Carrupt, and B. Testa. Mechanisms of liposomes/
water partitioning of (p-methylbenzyl)alkylamines. Pharm. Res.
15:1407–1413 (1998).
9. A. Avdeef. Assessment of distribution-pH profiles. In V. Pliska,
B. Testa, and H. van de Waterbeemd (eds.), Lipophilicity in Drug
Action and Toxicology, VCH Publishers, Weinheim, 1996 pp.
109–139.
10. A. Avdeef, K. J. Box, J. E. A. Comer, C. Hibbert, and K. Y. Tam.
pH-metric log P. 10. Determination of vesicle membrane-water
partition coefficients of ionizable drugs. Pharm. Res. 15:209–215
(1998).
Organic Acids and Bases, Chapman and Hall, London, 1981;
PALLAS for Windows, v. 1.3, pKa Prediction Module: pKalc 3.2,
1995.
13. C. Medana, A. Di Stilo, S. Visentin, R. Fruttero, A. Gasco, D.
Ghigo, and A. Bosia. NO donor and biological properties of dif-
ferent benzofuroxans. Pharm. Res. 16:956–960 (1999).
14. M. Feelisch and J. S. Stamler (eds.). Methods in Nitric Oxide
Research, John Wiley and Sons, Chichester, UK, 1996.
15. A. Avdeef, J. E. A. Comer, and S. J. Thomson. pH-Metric log P.
3. Glass electrode calibration in methanol-water, applied to pKa
determination of water-insoluble substances. Anal. Chem. 65:42–
49 (1993).
11. R. Fruttero, D. Boschi, E. Fornatto, A. Serafino, A. Gasco, and