NO donors. Part 16: Investigations on structure-activity relationships of organic mononitrates reveal 2-nitrooxyethylammoniumnitrate as a high potent vasodilator

Article abstract of DOI:10.1016/j.bmcl.2007.08.046

The vasoactive properties of 14 organic mononitrates were investigated in vitro using PGF_2α-precontracted porcine pulmonary arteries. A surprisingly wide range of vasorelaxant potencies was observed (pD₂: 3.36-7.50). Activities showed to be highly sensitive to the molecular structure and the substituents at the molecular carrier of the nitrate group. A correlation between lipophilicity and vasorelaxant potency could not be recognized. 2-Nitrooxyethylammoniumnitrate (1) was found to be slightly superior to the high potency trinitrate GTN.

Full text of DOI:10.1016/j.bmcl.2007.08.046

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5881–5885
NO donors. Part 16: Investigations on structure–activity
relationships of organic mononitrates reveal
2-nitrooxyethylammoniumnitrate as a high potent vasodilatorq
Andreas Koenig,a Carolin Roegler,a Kathrin Lange,a Andreas Daiber,b
Erika Glusaa and Jochen Lehmanna,*
aLehrstuhl fu¨r Pharmazeutische/Medizinische Chemie, Institut fu¨r Pharmazie, Friedrich-Schiller-Universita¨t Jena,
Philosophenweg 14, D-07743 Jena, Germany
bKlinikum der Johannes Gutenberg-Universita¨t Mainz, II. Medizinische Klinik, Kardiologie, 55101 Mainz, Germany
Received 24 June 2007; revised 30 July 2007; accepted 1 August 2007
Available online 28 August 2007
Abstract—The vasoactive properties of 14 organic mononitrates were investigated in vitro using PGF2a-precontracted porcine pul-
monary arteries. A surprisingly wide range of vasorelaxant potencies was observed (pD2: 3.36–7.50). Activities showed to be highly
sensitive to the molecular structure and the substituents at the molecular carrier of the nitrate group. A correlation between lipo-
philicity and vasorelaxant potency could not be recognized. 2-Nitrooxyethylammoniumnitrate (1) was found to be slightly superior
to the high potency trinitrate GTN.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Organic nitrates are useful drugs for the treatment of
ischemic heart diseases. The vasoactive response is med-
iated by a catenary process which includes bioactivation
to nitric oxide (NO) and/or related species, and intracel-
lular accumulation of cyclic GMP causing relaxation of
vascular smooth muscle cells. Four organic nitrates
are used therapeutically: pentaerithrityl tetranitrate
(PETN), glyceryl trinitrate (GTN), isosorbide dinitrate
(ISDN), and isosorbide mononitrate (ISMN).2,3 Clini-
cally, GTN represents the most potent compound (daily
dosage about 1 mg) but it induces nitrate tolerance.
PETN, although having more nitrate groups than the
other nitrates, requires the highest daily dosage (50–
240 mg), but on the other hand it does not develop sig-
nificant tolerance.4 However, experimental in vitro stud-
ies (tension measurements) yielded different results.
Using PGF2a-precontracted porcine pulmonary arteries
we have shown that there is a clear correlation between
the number of nitrate groups in the molecule and the
vasodilator potency and PETN showed to be the most
potent nitrate (pD2: 8.18).1 Low systemic bioavailability
for PETN and its active metabolite PEtriN5,6 as well as a
limited absorption of its less active metabolites PEdiN
and PEmonoN is obviously responsible for the clinical
profile of PETN.1
Besides the number of nitrate groups, the lipophilicity
(logP) and stereochemistry of the nitrate carrier mole-
cule may influence the vasorelaxant potency.7,8 Further-
more it was reported that the vasoactivities of cyclic
organic dinitrate isomers like isoidide dinitrate, isosor-
bide dinitrate, and isomannide dinitrate differ signifi-
cantly, although their logP were similar.9,10 The
authors assumed that compounds with a nitrate group
in the endo position were less potent than exonitrates
due to the steric hindrance to endogenous bioactivating
ligands. It is supposed that the bioactivating enzyme re-
quires two attachment sites for interaction with the ni-
trate compound, one involves the oxygen in the nitrate
group, while the other seeks another electron-rich cen-
ter.11 Moreover, there are two different bioactivation
pathways for organic nitrates. Recent results revealed
that only the ‘high potency nitrates’ GTN and PETN
but not the ‘low potency nitrates’ ISDN and ISMN were
bioactivated predominantly by the mitochondrial alde-
hyde dehydrogenase (ALDH-2).12–14 At higher concen-
tration (>1 lM), GTN and PETN are also subjected
Keywords: Organic nitrates; Vasorelaxation; Structure–activity
relationships.
q
See Ref. 1.
Corresponding author. Tel.: +49 3641 949803; fax: +49 3641
949802.; e-mail: j.lehmann@uni-jena.de
*
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.08.046

Table 1. Vasodilatory effects (pD2-value and EC50) of organic mononitrates, ISDN and GTN, on PGF2a-precontracted porcine pulmonary arteries with intact endothelium
Compound
Structure
pD2
EC50 (M)
n
LogP
Compound
Structure
pD2
EC50 (M)
n
LogP
À0.74
+
HO3S
H3N
1a
7.50 ( 0.02)
3.15 · 10À8
6
0.06a
9
4.48 ( 0.03)
3.35 · 10À5
4
ONO2
ONO2
HO
HO
OH
Br
2
3
4
5
6
7
6.42 ( 0.04)
6.34 ( 0.04)
5.99 ( 0.05)
5.81 ( 0.03)
5.02 ( 0.06)
4.83 ( 0.03)
3.77 · 10À7
4.12 · 10À7
1.03 · 10À6
1.54 · 10À6
9.56 · 10À6
1.47 · 10À5
3
5
6
5
7
9
1.59
2.37
0.90
2.48
0.44
1.14
10
4.14 ( 0.02)
4.01 ( 0.03)
3.85 ( 0.02)
3.36 ( 0.04)
<DMSO
7.25 · 10À5
8.77 · 10À5
1.41 · 10À4
4.35 · 10À4
>DMSO
7
4
5
4
8
6
À0.37
1.97
2.15
1.43
4.09
2.21
ONO2
ONO2
CH3
H3C
11
ONO2
ONO2
ONO2
H3COOC
HOOC
12
H3C
H3C
ONO2
O
ONO2
13
ONO2
H3C
H3C
CH3
14
ONO2
ONO2
ONO2
ONO2
H3C CH3
GTN
7.44 ( 0.02)
3.56 · 10À8
O2NO
ONO2
ONO2
HOOC
ONO2
O
O
8
4.48 ( 0.03)
3.32 · 10À5
9
À0.30
ISDN
6.37 ( 0.03)
4.24 · 10À7
7
0.90
O
O
O2NO
HO
The pD2-value presents the negative logarithm of the molar concentration producing 50% of the maximum relaxation (EC50). Values are means SEM from n separate experiments.
a Applied as 2-nitrooxyethylammoniumnitrate.

A. Koenig et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5881–5885
5883
additionally to the ‘low potency nitrates’ pathway. Most
probably, the low potency mononitrates are bioactivat-
ed exclusively by cytochrome P450 in the endoplasmic
reticulum for all concentrations.13
the inner surface with a rough plastic rod. The rings
were suspended between two L-shaped platinum hooks
and mounted in 10 mL organ baths filled with modified
Krebs–Henseleit solution. The solution was kept at
37 ꢁC and aerated with 95% O2/5% CO2. The vasorelax-
ant effects were measured using the PGF2a (3 lM)-pre-
contracted rings with a resting tension of 20 mN. The
test compounds were added to the organ bath at the
contraction plateau in a cumulative manner. Endothelial
integrity was assessed by bradykinin (10 nM)-induced
relaxation of PGF2a-precontracted vessels. In mechani-
cally endothelium-denuded arterial rings, pretreated
with 0.2 mM L-NAME, the relaxation induced by bra-
dykinin was less than 10%. Relaxation responses were
expressed as a percentage of the PGF2a (3 lM)-induced
contraction in each tissue.
The correlation between the number of nitrate functions
in the molecule and the vasorelaxant potency1,14,15
prompted us to perform further investigations of struc-
ture–activity relationships (SAR) in this class of bioac-
tive compounds, but this time by structural alterations
with a fixed number of nitrate groups. Especially organ-
ic mononitrates are appropriate for this study. They are
less explosive, easy to handle, and the break down to
active metabolites can be excluded. In addition, the
chemical variation of the organic nitrate carrier mole-
cule is easier than by using di- or trinitrates.
Several methods have been applied for the synthesis of
organic nitrates.16 In this study, the preparation of the
mononitrates (Table 1) was accomplished by esterifica-
tion of the corresponding alkyl alcohols with fuming
nitric acid or from the corresponding alkyl bromides
by reaction with silver nitrate in acetonitrile, respectively
(Scheme 1). The following nitrates were prepared
according to procedures reported in the literature:
117,18 from 2-aminoethanol, 219 from 2-bromoethanol,
37 from benzylbromide, 420 from methyl 3-hydroxypro-
panoate, 521 from 2-hydroxy-1-phenylpropanone, 618
from 3-hydroxypropionic acid, 722 from methyl 3-hy-
droxy-2,2-dimethylpropanoate and subsequent ester
hydrolysis in methanolic sodium hydroxide, 923 from
2-hydroxyethanesulfonic acid, 10 from 2-(bromo-
methyl)-2-(hydroxymethyl)propane-1,3-diol (this syn-
thesis will be reported in detail later), and 1224 from
1-bromobutane.
All of the compounds were able to elicit a significant
vasorelaxation in a concentration-dependent manner
(Fig. 1) with exception of compound 14, which induced
a very weak relaxation in the range of the pure solvent
DMSO. The maximum of vasorelaxation has been
achieved throughout for the compounds 1–12
(Emax ꢀ 100%), compound 13 only performed ꢀ70%
relaxation after adding the selected maximum concen-
tration of 1 mM. The sigmoidal concentration–response
curves (Fig. 1) were consistent with the Hill slope for
these mononitrates. The concentration–response curves
resulting from experiments with endothelium-denuded
vessels did not shift significantly, compared to the ves-
sels with intact endothelium (data not shown).
In order to recognize any relationship between vasodila-
tor potency and lipophilicity the pD2-values were plot-
ted against the calculated logP-values (ACD, Toronto,
Canada) (Table 1 and Fig. 2).
The following nitrates were purchased: 11, 13, and 14
(Sigma–Aldrich, Steinheim, Germany), GTN (Merck,
Darmstadt, Germany). Compound 8 and ISDN were
obtained as gift from Schwarz-Pharma AG (Monheim,
Germany).
According to previous investigations, it had to be
expected that the vasorelaxant potency of the organic ni-
trates investigated should increase with their lipophilic-
ity.8,9 We have found the calculated logP-values for the
bioactive nitrates in a wide range of three orders of mag-
nitude (À0.74 to 2.48), but no clear correlation with the
vasodilator activity could be recognized (Fig. 2).
To study SAR without the influence of any pharmacoki-
netic parameters, tension experiments were performed in
organ baths using PGF2a-precontracted porcine pul-
monary arteries and the vasorelaxant responses for the
nitrates were measured and are given in Table 1.
According to a previously described protocol,1 lungs
from adult pigs were obtained from the local slaughter-
house. Small branches from pulmonary arteries were
prepared and cut into rings (length 2–3 mm, diameter
1.5–2 mm). In experiments with endothelium-denuded
rings, the endothelium was removed by gently rubbing
Alkyl
OH
Alkyl
b
Br
a
Alkyl
ONO2
Scheme 1. Synthesis of mononitrates from alkyl alcohols and alkyl
bromides, respectively: Reagents and conditions: (a) fuming nitric acid,
2 h, <10 ꢁC for 1, 3, 7, and 9; (b) silver nitrate/acetonitrile, 2–28 d, rt
50 ꢁC (depending on the compound) for 2, 4, 5, 6, 10, and 12.
Figure 1. Concentration–response curves for the relaxation of PGF2a-
precontracted porcine pulmonary arteries with intact endothelium
induced by the mononitrates given in Table 1. GTN (-•-) and ISDN
(
). Mean SEM.

5884
A. Koenig et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5881–5885
The 2-amino-substituted ethylmononitrate 1 was found
to be the most potent mononitrate with pD2 = 7.50. This
surprising result indicates that in addition to previous
observations1 the number of nitrate moieties in the vaso-
dilator compound is not necessarily the leading param-
eter for vasodilator potency. Recently, we speculated
that the reactivity of oligonitrates, which is directly cor-
related with the number of nitrate groups, determines
the vasodilator potency of organic nitrates and whether
they are bioactivated by ALDH-2 or not.12 Probably,
the reactivity can also be increased by insertion of other
functional groups (Table 1). Compound 1 showed to be
more potent than all of the dinitrates investigated up to
now and furthermore it reaches the efficacy of the most
potent trinitrate GTN (pD2 = 7.44).
Figure 2. Relationship between lipophilicity (logP) and the concen-
trations for half maximal relaxation (expressed as pD2-values) of the
organic mononitrates. *Calculated for the uncharged molecule.
Contradictory to the idea that the structure of the or-
ganic carrier is more or less negligible with regard to
the biological activity, the present results reveal that
the carrier molecule and its substituents are highly
important. The vasoactivity of nitrovasodilators is
definitively more sensitive to the functional groups
rather than to the lipophilicity of the compounds.
Therefore, we conclude that this is due to the significant
differences in the binding affinities for the nitrate bioac-
tivating enzymes ALDH-2 and cytochrome P450. It is
suggested that these affinities for the binding areas of
the enzymes are determined not only by the stereochem-
istry and the physicochemical properties of the com-
pounds but in first line by polar substituents which are
able to perform additional binding options. In particu-
lar, the primary protonated amino group in compound
1 is capable of improving the binding to the target
enzyme.
Nevertheless there is a significant influence of the nitrate
carrier structures on the vasodilator potency. Com-
pounds bearing the nitrate function attached to bulky
and lipophilic carrier molecules without any further
functional groups (11–14) displayed the lowest activities,
particularly the most lipophilic compound 14 which
even failed to outperform the solvent DMSO.
Polarization of the nitrate carriers by hydroxylation (8
and 10), introduction of hydrophilic cyclic ether (8), car-
boxy (6 and 7) or sulfonic acid moieties (9) enhances the
vasodilator activities significantly. Acidity in general
seems to be less crucial. The sulfonic acid 9 is more
acidic but less active than the carboxylic acids 6 and 7.
Remarkably, the lipophilic ester 4 is more active than
the corresponding less lipophilic acid 6. It can be specu-
lated that chemically reactive functions such as a car-
boxylic ester (4) or an alkyl bromide (2) are able to
support the bioactivation of the attached nitrate. Com-
pounds 3 and 5 extremely drop out of any potency/lipo-
philicity correlation (Fig. 2). We suggest that these
unexpected high potencies may be based on a lability
of the nitrate functions due to the specific chemical
properties of these compounds which are not present
in all of the other structures. Both, benzylnitrates and
a-nitratophenones are capable to decompose in terms
of a disproportionation.21 Scheme 2 demonstrates that
both structures may be considered as ‘self-reducing ni-
trates’. Independently whether disproportionations as
outlined in Scheme 2 will be performed under physiolog-
ical conditions, we suggest that these properties support
the bioactivation of the organic nitrate to break-down
species.
Acknowledgment
A Ph.D. scholarship for A.K. by Graduiertenfo¨rderung
(Thuringia, Germany) is gratefully acknowledged.
References and notes
1. Koenig, A.; Lange, K.; Konter, J.; Daiber, A.; Stalleicken,
D.; Glusa, E.; Lehmann, J. J. Cardiovasc. Pharmacol.
2007, 50, 68.
2. Ahlner, J.; Andersson, R. G.; Torfgard, K.; Axelsson, K.
L. Pharmacol. Rev. 1991, 43, 351.
3. Parker, J. D.; Parker, J. O. N. Engl. J. Med. 1998, 338,
520.
4. Jurt, U.; Gori, T.; Ravandi, A.; Babaei, S.; Zeman, P.;
Parker, J. D. J. Am. Coll. Cardiol. 2001, 38, 854.
5. Wenzel, P.; Oelze, M.; Coldewey, M.; Hortmann, M.;
Seeling, A.; Hink, U.; Mollnau, H.; Stalleilken, D.;
Weiner, H.; Lehmann, J.; Li, H.; Forstermann, U.;
Munzel, T.; Daiber, A. Arterioscler. Thromb. Vasc. Biol.
2007, 27, 1729.
+5
ONO2
CHO
+3
NO2
-
+
6. Seeling, A.; Lehmann, J. J. Pharm. Biomed. Anal. 2006,
40, 1131.
7. Wessler, C.; Homann, A.; Fricke, U.; Lehmann, J. Eur. J.
Med. Chem. 2003, 38, 581.
O
O
+5
ONO2
+3
NO2
-
+
CHO
8. Noack, E. Methods Find. Exp. Clin. Pharmacol. 1984, 6,
583.
Scheme 2. Disproportionation of benzylnitrates and a-nitratophenones.21

A. Koenig et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5881–5885
5885
9. Bron, J.; Sterk, G. J.; Vanderwerf, J. F.; Timmerman, H.
Pharm. World Sci. 1995, 17, 120.
16. Boschan, R.; Merrow, R. T.; Vandolah, R. W. Chem. Rev.
1955, 55, 485.
10. Bogaert, M. G.; Rosseel, M. T. Naunyn Schmiedebergs
Arch. Pharmacol. 1972, 275, 339.
11. Tzeng, T. B.; Fung, H. L. Med. Hypotheses 1992, 37,
58.
12. Chen, Z. Q.; Zhang, J.; Stamler, J. S. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 8306.
17. Berthmann, A.; Ratz, H.. In Houben-Weyl Methoden der
Organischen Chemie; Bayer, E., Berthmann, A., Hauswe-
iler, A., Eds.; Thieme: Stuttgart, 1963; Vol. 6, p 329.
18. Ishihara, S.; Saito, F.; Ohhata, Y.; Kanai, M.; Mizuno,
H.; Fujisawa, M.; Yorikane, R.; Koike, H. Bioorg. Med.
Chem. Lett. 2003, 13, 1527.
13. Munzel, T.; Daiber, A.; Mulsch, A. Circ. Res. 2005, 97,
618.
19. Kartasasmita, R. E.; Laufer, S.; Lehmann, J. Arch. Pharm.
2002, 335, 363.
14. Wenzel, P.; Hink, U.; Oelze, M.; Seeling, A.; Isse, T.;
Bruns, K.; Steinhoff, L.; Brandt, M.; Kleschyov, A. L.;
Schulz, E.; Lange, K.; Weiner, H.; Lehmann, J.; Lackner,
K. J.; Kawamoto, T.; Munzel, T.; Daiber, A. Br. J.
Pharmacol. 2007, 150, 526.
15. Daiber, A.; Oelze, M.; Coldewey, M.; Bachschmid, M.;
Wenzel, P.; Sydow, K.; Wendt, M.; Kleschyov, A. L.;
Stalleicken, D.; Ullrich, V.; Mulsch, A.; Munzel, T. Mol.
Pharmacol. 2004, 66, 1372.
20. Mccallum, K. S.; Emmons, W. D. J. Org. Chem. 1956, 21,
367.
21. Emmons, W. D.; Freeman, J. P. J. Am. Chem. Soc. 1955,
77, 4415.
22. Decker, M.; Koenig, A.; Glusa, E.; Lehmann, J. J. Bioorg.
Med. Chem. Lett. 2004, 14, 4995.
23. Whewell, C. U.S. Patent 20,060,293,384, 2006.
24. Muthuramu, K.; Shepson, P. B.; Obrien, J. M. Environ.
Sci. Technol. 1993, 27, 1117.