2-Aminonicotinic acid 1-oxides are chemically stable inhibitors of quinolinic acid synthesis in the mammalian brain: A step toward new antiexcitotoxic agents

Article abstract of DOI:10.1021/jm401249c

3-Hydroxyanthranilic acid 3,4-dioxygenase (3-HAO) is the enzyme responsible for the production of the neurotoxic tryptophan metabolite quinolinic acid (QUIN). Elevated brain levels of QUIN are observed in several neurodegenerative diseases, but pharmacological investigation on its role in the pathogenesis of these conditions is difficult because only one class of substrate-analogue 3-HAO inhibitors, with poor chemical stability, has been reported so far. Here we describe the design, synthesis, and biological evaluation of a novel class of chemically stable inhibitors based on the 2-aminonicotinic acid 1-oxide nucleus. After the preliminary in vitro evaluation of newly synthesized compounds using brain tissue homogenate, we selected the most active inhibitor and showed its ability to acutely reduce the production of QUIN in the rat brain in vivo. These findings provide a novel pharmacological tool for the study of the mechanisms underlying the onset and propagation of neurodegenerative diseases.

Full text of DOI:10.1021/jm401249c

Article
pubs.acs.org/jmc
2‑Aminonicotinic Acid 1‑Oxides Are Chemically Stable Inhibitors of
Quinolinic Acid Synthesis in the Mammalian Brain: A Step toward
New Antiexcitotoxic Agents
Gian Paolo Vallerini,† Laura Amori,† Claudia Beato,† Margarita Tararina,‡ Xiao-Dan Wang,‡,§
Robert Schwarcz,‡ and Gabriele Costantino*,†
†Dipartimento di Farmacia, Universita
̀
degli Studi di Parma, Parco Area delle Scienze 27/A, 43124 Parma, Italy
‡Maryland Psychiatric Research Center, Department of Psychiatry, University of Maryland School of Medicine, Baltimore, Maryland
21228, United States
S
* Supporting Information
ABSTRACT: 3-Hydroxyanthranilic acid 3,4-dioxygenase (3-
HAO) is the enzyme responsible for the production of the
neurotoxic tryptophan metabolite quinolinic acid (QUIN).
Elevated brain levels of QUIN are observed in several
neurodegenerative diseases, but pharmacological investigation
on its role in the pathogenesis of these conditions is difficult
because only one class of substrate-analogue 3-HAO inhibitors,
with poor chemical stability, has been reported so far. Here we
describe the design, synthesis, and biological evaluation of a
novel class of chemically stable inhibitors based on the 2-aminonicotinic acid 1-oxide nucleus. After the preliminary in vitro
evaluation of newly synthesized compounds using brain tissue homogenate, we selected the most active inhibitor and showed its
ability to acutely reduce the production of QUIN in the rat brain in vivo. These findings provide a novel pharmacological tool for
the study of the mechanisms underlying the onset and propagation of neurodegenerative diseases.
metabolic arms are still far from being fully understood. The
dearth of selective, potent pharmacological agents is, in fact,
one of the main impediments to unraveling KP function and
dysfunction in biological systems.
INTRODUCTION
■
Both in the periphery and in the brain, mammals degrade most
nonproteinogenic L-tryptophan through the so-called kynur-
enine pathway (KP, Scheme 1).1 This branched metabolic
cascade is now recognized as a key component of an intricate
network of immune and inflammatory processes.2 As several
KP metabolites have distinct neuroactive properties and may be
causally involved in the etiology of acute and chronic brain
diseases, the pathway has also attracted substantial attention
from investigators in the neurosciences.3 In particular, an
imbalance between neuroprotective [kynurenic acid (KYNA)]
and neurotoxic [3-hydroxykynurenine (3-HK) and quinolinic
acid (QUIN)] KP metabolites has been tentatively linked to
the onset and propagation of a number of severe neurological
disorders, including Huntington’s disease (HD),4 Alzheimer’s
disease,5 epilepsy,6 amyotrophic lateral sclerosis,7 and multiple
sclerosis.8 Moreover, there is increasing evidence implicating
abnormal KP metabolism in schizophrenia,9 major depressive
disorders,10 and other psychiatric diseases.
Individual KP enzymes are widely expressed in peripheral
tissues but are also present in the central nervous system, albeit
with much lower activity.11 In the brain, the enzymes of the two
pathway branches are preferentially localized in astrocytes and
microglial cells, respectively.12 Targeting of astro- and micro-
glial KP metabolism by genetic tools or enzyme inhibitors has
recently begun to clarify the biological relevance of this
compartmentalization, but the respective roles of the two
Because of our long-standing interest in the neurotoxic
effects of the free radical generator and NMDA receptor agonist
QUIN, and because of the growing realization that elevated
brain QUIN likely plays a pathogenic role in neurodegenerative
diseases, we and others have recently renewed efforts to reduce
QUIN levels by pharmacological intervention. In the brain as in
the periphery, QUIN is formed by 3-hydroxyanthranilic acid
3,4-dioxygenase (3-HAO), which catalyzes the oxidative ring-
opening of the KP metabolite 3-hydroxyanthranilic acid (3-
HANA), leading to the reactive intermediate 2-amino-3-
carboxymuconic-6-semialdehyde (ACMS).13 ACMS, in turn,
spontaneously cyclizes to QUIN, which is then further
degraded to NAD+ (Scheme 2). As the production of QUIN
is more effective and faster than its transformation into NAD+,
brain QUIN concentrations are about 50 times higher than 3-
HANA levels.14 Notably, only one series of 3-HAO inhibitors,
including a number of mono-, di-, and trisubstituted 3-HANA
analogues, has been reported so far (see Figure 1, 1).15 Most of
these substrate derivatives are extremely potent in vitro, with
IC50 values in the low nanomolar range (the KM for 3-HANA is
Received: May 30, 2013
Published: November 23, 2013
© 2013 American Chemical Society
9482
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
Scheme 1. The Kynurenine Pathway of Tryptophan
Metabolism
Scheme 2. 3-HAO Catalyzed Reaction
a
a
Trp, L-tryptophan; IDO, indoleamine 2,3-dioxygenase; TDO,
tryptophan 2,3-dioxygenase; KYN, L-kynurenine; KATs, kynurenine
aminotransferases; KYNA, kynurenic acid; KMO, kynurenine 3-
monooxygenase; ANA, anthranilic acid; 3-HK, 3-hydroxykynurenine;
XA, xanthurenic acid; 3-HANA, 3-hydroxyanthranilic acid; CINA,
cinnabarinic acid; 3-HAO, 3-hydroxyanthranilic acid 3,4-dioxygenase;
ACMS, 2-amino-3-carboxymuconic-6-semialdehyde; ACMSD, 2-
amino-3-carboxymuconic-6-semialdehyde decarboxylase; PA, picolinic
acid; QUIN, quinolinic acid; QPRT, quinolinic acid phosphoribosyl-
transferase.
Figure 1. (a) Comparison between the 3-hydroxyanthranilate core of
previously reported 3-HAO inhibitors (1) and the 2-aminonicotinate
1-oxide, one of compounds described in this work (2). (b) Structural
and electronic analogies between 3-HAO/3-HANA adduct 3, one of
the transition states proposed for 3-HAO catalysis,17 and the
resonance contributing structures of 2-aminonicotinate 1-oxide.
Residues numbering refers to 3-HAO from Ralstonia metallidurans.
between 1 and 3 μM).16 This corresponds to unusually high
ligand efficiency values, which in some cases are higher than 1.
Unfortunately, however, the experimental and possible clinical
use of this series of compounds is severely limited by their lack
of stability under physiological conditions due to the tendency
of the o-aminophenol nucleus to auto-oxidize and generate
reactive radical species.
3-HAO is a nonheme, Fe(II)-dependent enzyme, which
belongs to the class of extradiol dioxygenases. In 2005, Zhang
et al. reported the overexpression and purification of 3-HAO
from Ralstonia metallidurans and the crystal structure of the
unbound enzyme. The authors also described the enzyme in
complex with its natural substrate, 3-HANA, and with the first
discovered 3-HAO inhibitor, 4-chloro-3-hydroxyanthranilic acid
(4-Cl-3-HANA).17,18 Subsequently, the structure of a mamma-
lian (bovine) 3-HAO was published in its apo form.19 These
studies revealed the presence of a Fe2+ ion in the active site,
which is responsible for catalytic activity, coordinated by fully
conserved residues among the species, namely His51, His95,
and Glu57 (bacterial numbering).
In their detailed biochemical study, which involved the use of
several X-ray structures of 3-HAO in the apo form, or
cocrystallized with the substrate or 4-Cl-3-HANA, and mass
spectrometry analyses, Zhang et al. proposed a mechanism for
the 3-HAO-mediated transformation of 3-HANA to QUIN,
along with a hypothesis for the mechanism of inhibition by 4-
Cl-3-HANA. In particular, the authors used mass spectrometry
to rule out that 4-Cl-3-HANA covalently inactivates the
9483
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
enzyme. Although difficult to reconcile with the exceptionally
high ligand efficiency, they proposed that inhibition was due to
an unproductive positioning of the inhibitor inside the active
site.
Table 1. Results of the in Vitro 3-HAO Inhibition Test for
All the Synthesized Compounds
a
To overcome some of the shortcomings displayed by, and
associated with, the in vivo use of substituted 3-HANA
analogues, we recently identified new chemotypes that were
capable of functioning as 3-HAO inhibitors. We report here the
design, synthesis, and preliminary evaluation of a new series of
3-HAO inhibitors based on the 2-aminonicotinic acid 1-oxide
nucleus.
Also on the basis of the proposed catalytic mechanism of 3-
HAO, we pictured the 2-aminonicotinic acid 1-oxide nucleus
(2, Figure 1a) as a suitable scaffold to escape from the unstable
o-aminophenol moiety of reported inhibitors (1) while
retaining most of its relevant structural and electronic features.
In particular, the 2-aminopyridine-N-oxide motif is expected to
keep the Fe(II) chelating ability of the o-aminophenol ring; 2-
aminonicotinic acid 1-oxide owns the needed carboxylic group
for interaction with a conserved arginine residue (Arg99 in the
bacterial numbering, or Arg95 in the bovine isoform; see also
the Molecular Modeling Studies below); it has an electron
deficiency at position 1 (the oxidized pyridine nitrogen),
corresponding to position 3 of 3-HANA. Furthermore, the 2-
aminonicotinic acid 1-oxide nucleus has a partial imine
character, which is reminiscent of adduct 3 in the proposed
mechanism scheme (Figure 1b).
We envisaged that the 2-aminonicotinic acid 1-oxide may
trap the enzyme in a catalytically inert disposition and that the
potency of inhibition can be modulated by substituents on the
aromatic nucleus. Thus, 2-aminonicotinic acid 1-oxide (4, Table
1) and its derivatives 5−22 were synthesized and tested as 3-
HAO inhibitors. The most interesting inhibitor was then
characterized in awake rats to assess its ability to block QUIN
synthesis under in vivo conditions.
CHEMISTRY
■
Most of the compounds described in this study (4−22, Table
1) were synthesized starting from commercially available
pyridine derivatives, exploiting few quite classical synthetic
procedures, among which the selective oxidation of the pyridine
nitrogen appeared to be the key step.
Compound 4 was obtained from commercially available 2-
aminonicotinic acid (23a), via prior conversion into its methyl
ester (24a) by reaction with (trimethylsilyl)diazomethane in
toluene/methanol,20 followed by methyltrioxorhenium(VII)
(MTO)-catalyzed oxidation of the pyridine nitrogen21 and
alkaline ester hydrolysis (Scheme 3). The same scheme was
followed for the synthesis of 5 and 8, starting from 6-chloro-2-
aminonicotinic acid (23b) and 6-chloronicotinic acid (23c),
respectively.
The reaction pathway followed for the synthesis of 6 allowed
us to also isolate compounds 7 and 9−11 (Scheme 4). It
started with aromatic amination of commercial 2-chloro-6-
methylnicotinonitrile (26) with ammonia saturated ethanol,22
affording 27. Direct hydrolysis of the cyano group of 27 gave 7.
Oxidation of intermediate 27 provided 11, which, when
completely hydrolyzed, led to the synthesis of the most active
compound of the series, 6. Hydrolysis conditions of 11 were
attenuated in order to synthesize amide 10. This was obtained
by simply treating the crude from the oxidative step with
saturated aqueous NaHCO3, at room temperature, exploiting
the known conversion of nitriles into amides by alkaline
a
Biological tests were performed in both rat and human brain
homogenate, measuring the production of 14C-QUIN after incubation
with 14C-3-HANA and different concentrations of test compounds.
Activities are expressed as percentage of inhibition of 3-HAO, with
standard errors of triplicate experiments.
hydrogen peroxide (present in the reaction medium).34
Compound 9 was obtained by esterification of 6 following
the procedure described for 24a−c.
9484
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
a
Scheme 3. Synthesis of Compounds 4, 5, and 8
a
Reagents and conditions: (a) TMS-diazomethane, toluene/MeOH, 0 °C, 30 min, 90−95% yield; (b) MTO, 35% aqueous H2O2, EtOH, rt, 3−24 h,
68−88% yield; (c) 10% NaOH in MeOH, reflux, 1 h; (d) 3 N aqueous HCl, 0 °C, pH 4−5, quantitative yield.
a
Compound 12 was obtained from 2-aminonicotinic acid
(23a) by reaction with formamide under microwave con-
ditions,23 followed by MTO-catalyzed oxidation of pyridine
nitrogen (Scheme 6). Oxidation was selective for the pyridine
Scheme 4. Synthesis of Compounds 6, 7, and 9−11
a
Scheme 6. Synthesis of Compound 12
a
Reagents and conditions: (a) formamide, microwave irradiation, 60
W, 125 °C, 20 min, 47% yield; (b) MTO, 35% aqueous H2O2, EtOH,
rt, 4 h, 71% yield.
1
nitrogen, as was demonstrated by both H- and 13C NMR
analyses: the common shielding effect brought by N-oxide
function was found only on signals relative to pyridine protons
and carbon atoms, and no effect was observed on signals
belonging to the pyrimidine ring (see Experimental Section for
details).
In some cases, due to the lack of useful pyridine-based
starting materials from readily available commercial sources, the
de novo construction of the pyridine ring, with appropriate
substituents, was necessary. A reported microwave-accelerated,
one-pot, multicomponent reaction24 (MCR) was exploited for
the synthesis of compounds 15−18, with slight modifications to
the original procedure (Scheme 7). Irradiating a microwave
tube containing an appropriate aldehyde (aromatic or aliphatic,
35), malononitrile (36), an enolizable, aliphatic ketone (37),
and ammonium acetate (38) as an ammonia source, with
neither solvent nor catalyst, for 15 min at 100 °C, provided key
a
Reagents and conditions: (a) NH3/EtOH, 160 °C, 24 h, 85% yield;
(b) 10% aqueous KOH, reflux, 3 h; (c) 3 N aqueous HCl, 0 °C, pH
4−5, quantitative yield; (d) MTO, 35% aqueous H2O2, EtOH, rt, 3 h,
78% yield; (e) saturated aqueous NaHCO3, rt, 30 min, 90% yield; (f)
TMS-diazomethane, toluene/MeOH, 0 °C, 30 min, 75% yield.
Compounds 14 and 19 were synthesized in the same manner
as 6, but 2-chloro-4,6-dimethylnicotinonitrile (28a) and 2-
chloro-5,6-dimethylnicotinonitrile (28b) were used as the
starting materials, respectively (Scheme 5). Preparation of 13
was similar except for the additional step at the beginning:
elongation of the 6-methyl substituent of 26 by reaction with
iodomethane in DMF, in presence of sodium hydride (Scheme
5).22
a
Scheme 5. Synthesis of Compounds 13, 14, and 19
a
Reagents and conditions: (a) NH3/EtOH, 160 °C, 24 h, 73−83% yield; (b) MTO, 35% aqueous H2O2, EtOH, rt, 3 h, 71−77% yield; (c) 10%
aqueous KOH, reflux, 3 h; (d) 3 N aqueous HCl, 0 °C, pH 4−5, quantitative yield; (e) NaH, CH3I, DMF, from 0 °C to rt, 4 h, 69% yield.
9485
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
a
Scheme 7. Synthesis of Compounds 15−18
a
Reagents and conditions: (a) microwave irradiation, 100 °C, 15 min, 37−65% yield; (b) MTO, 35% aqueous H2O2, EtOH, rt, 3 h, 70−76% yield;
(c) 10% aqueous KOH, reflux, 3 h; (d) 3 N aqueous HCl, 0 °C, pH 4−5, quantitative yield.
a
Scheme 8. Synthesis of Compounds 20−22
a
Reagents and conditions: (a) K4[Fe(CN)6]·3H2O, Pd(PPh3)4, DBU, tBuOH/H2O, 85 °C, 18 h, 12% yield; (b) MTO, 35% aqueous H2O2, EtOH,
rt, 3 h, 71−92% yield; (c) 10% aqueous KOH, reflux, 3 h; (d) 3 N aqueous HCl, 0 °C, pH 4−5, quantitative yield; (e) K4[Fe(CN)6]·3H2O,
t
Pd(PPh3)4, DBU, BuOH/H2O, microwave irradiation, 120 °C, 10 min, 79% yield.
pyridine products 39a−d in 37−65% isolated yield. Oxidation
of pyridine nitrogen of MCR-derived intermediates, followed
by hydrolysis of the cyano group, afforded compounds 15−18.
20 was obtained from commercially available 3-bromo-2-
amino-5-methylpyridine (41), via Pd/DBU-catalyzed cyanation
with potassium hexacyanoferrate trihydrate in tert-butanol/
water,25 followed by oxidation of pyridine nitrogen of 42 and
nitrile hydrolysis of 43 (Scheme 8). The first step of this
synthesis presented the problems of poor conversion of the
starting material and hydrolysis of newly formed nitrile to
amide due to the presence of water in the reaction medium.
The same pathway was utilized for the synthesis of 21, starting
from 3-bromo-5-chloro-2-aminopyridine (44). The first
cyanation step was stressed and performed under microwave
conditions in this case in order to achieve complete starting
material consumption and to improve yield of amide 45, a
useful intermediate (Scheme 8). Finally, 22 was obtained via
direct oxidation of 5-bromo-2-aminonicotinic acid (47),
affording a 92% pure product (by HPLC/MS analysis) in
one step (Scheme 8).
(4), the first compound to be tested, inhibited QUIN
production in rat and human brain homogenates by 47.6
8.4% and 74.0 7.3% at 10 μM, respectively. This result was an
apparent proof of concept of the bioisosteric equivalence
between the o-aminophenol ring and the 2-aminopyridine N-
oxide nucleus. Compound 4, in fact, inhibited 3-HAO at a
concentration comparable with the KM of 3-HANA.16 More
interestingly, when evaluated under the same conditions where
4-Cl-3-HANA decomposed by 62% in 24 h,15 compound 4 was
completely resistant to auto-oxidative degradation. Indeed, as it
can be noted in Figure 2, HPLC/UV−vis analysis was not able
to detect any sort of decomposition after 24 h incubation of
compound 4 in PBS buffer.
Next, encouraged by these results, we embarked on a
program of systematic modification of 4, aimed at enucleating
the pharmacophore requirements for 3-HAO inhibition. These
modifications were intended to assess the role of the oxidized
pyridine nitrogen, the amino group, the carboxylic acid group,
and the decoration of the aromatic ring. As a starting point, we
decided to decorate the 6-position of the 2-aminopyridine ring
with a chlorine atom. The corresponding 6-chloro-2-amino-
nicotinic acid 1-oxide (5) is topologically equivalent to the
known 3-HAO inhibitor 4-Cl-3-HANA. Unexpectedly, 5 was
almost equipotent to the parent compound 4 (see Table 1).
RESULTS AND DISCUSSION
■
In Vitro Inhibition of 3-HAO Activity by Newly
Synthesized Compounds. 2-Aminonicotinic acid-1-oxide
9486
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
Figure 2. Results of the stability test for compounds 4 (a) and 6 (b). The test was carried out under the exact conditions described by Linderberg et
al.15 In detail, compounds were dissolved in PBS buffer at pH 7.5 (two samples for each compound, at two concentrations). The solution was
maintained at 37 °C in the dark. Aliquots were taken at the beginning of the experiment and after 24 h analyzed by HPLC/UV−vis (phosphate
buffer pH 3/methanol/acetic acid, 50/50/1). Detected chromatographic peaks are reported, and quantitative analysis is described below the
corresponding plot. Each experiment was performed in triplicate.
This marks a clear difference to the 3-HANA series, where the
4-chloro derivative shows an affinity to the enzyme 3 orders of
magnitude higher than that of 3-HANA. When the substituent
at the 6-position was changed to a methyl group (6), the
potency significantly increased. It can be appreciated from
Table 1 that compound 6 inhibited 3-HAO by 76.9 11.0%
and 85.5
6.3% in rat and human brain homogenates,
respectively, already at 10 μM. Full characterization of 6 yielded
an IC50 of 2.8 and 1.1 μM in rat and human brain homogenates,
respectively (Figure 3). Additionally, compound 6 showed the
same chemical stability held by compound 4 (Figure 2b).
Thus, these initial data seemed to question the isosteric
equivalence of the 2-aminonicotinate 1-oxide and 3-HANA
series. To obtain further insights, we evaluated the role of the
N-oxide and the 2-amino groups in the interaction with the
enzyme. Accordingly, for the sake of synthetic accessibility, we
prepared and tested 2-amino-6-methylnicotinic acid (7) and 6-
chloronicotinic acid 1-oxide (8) to examine the actual need of
the N-oxide and 2-amino moieties, respectively. The two
derivatives were completely inactive up to 100 μM, showing
only partial inhibition (<50%) at 1 mM. These data, hence,
indicated that the 1-oxide-2-amino core is a pharmacophore
requirement for interaction with 3-HAO, thus confirming the
initial design assumptions. Then, we proceeded to investigate
the role of the carboxylic group at position C3. In the 3-HANA
series, according to the available X-ray structures, the
carboxylate moiety is involved in a ionic interaction with an
arginine residue.17 By using the scaffold of 6, the carboxylic acid
was thus esterified, amidated, and replaced with a cyano group,
to give derivatives 9, 10, and 11, respectively. None of these
compounds showed any appreciable enzyme inhibition up to 1
mM. Furthermore, the carboxylic acid and the 2-amino group
were inserted into the pyrido[2,3-d]pyrimidine scaffold of 12.
Figure 3. Inhibition curves for compound 6 in rat (A) and human (B)
brain homogenates. Each experiment was performed in triplicate.
Also in this case, the reduction of potency is evident although
50% inhibition is retained at 1 mM. Taken together, these data
confirm the need of a free carboxylic group at the 3-position,
adding another piece of evidence to the original idea that the 2-
aminonicotinic acid 1-oxide nucleus is a bioisosteric replace-
ment for the 3-HANA nucleus.
9487
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
After establishing that the 1-oxide and 2-amino groups, along
with C3 free carboxylate, are irreplaceable functionalities, we
introduced different substituents at positions C4, C5, and C6
and examined their effect on enzyme activity. The first change
to the structure of 6 was the one carbon elongation of the
methyl group at position C6, replacing it with an ethyl group;
compound 13 indeed produced substantial inhibition already at
10 μM, comparable with compounds 4−6. The screening then
proceeded via the application, one by one, of the following
modifications: (I) introduction of methyl (14), ethyl (15),
isopropyl (16), and phenyl (17) groups at position C4,
maintaining C6 methyl (18, di-isopropyl substituted at C4 and
C6, was chosen due to synthetic accessibility); (II) introduction
of a methyl group at position C5, maintaining C6 methyl (19);
and (III) elimination of C6 methyl and introduction of methyl
(20) or halogen substituents (21−22) at position C5. Among
structures 14−22, only compounds 19−22 showed some
activity up to 100 μM, with percentage of inhibition between
30% and 76%. All the results from in vitro tests are summarized
in Table 1.
rubredoxin-like FeS4 site present in bacterial and yeast 3-
HAO.17,27 Furthermore, even though key residues are
conserved between species, some mutations occurred surround-
ing the active site (see sequence alignment in Figure S1,
Supporting Information). The choice of the crystal structure
was therefore limited between the two mammalian 3-HAO
crystal structures so far available (human and bovine). Although
human 3-HAO has been crystallized (PDB code 2QNK), its
structure presents an open conformation of the small loop
above the binding site. Preliminary docking experiments
performed with this crystal structure were entirely unable to
explain the lack of activity that some compounds showed in the
experimental assays. Moreover, the human crystal structure has
so far not been corroborated by further experimentation (for
example by restoring activity after purification). Human and
bovine 3-HAO share the same secondary structure, and
residues are almost fully conserved (86% and 93% of identical
and similar residues, respectively). We therefore decided to
build a homology model of the human structure based on the
bovine one by point-mutating nonconserved residues between
the two structures. In practice, in the region spanning 4 Å from
the active site, this only involves mutation of Val37 into an
isoleucine. Our docking studies performed on the structure
thus generated suggested a binding mode of compound 6,
which is consistent with the crystallographic binding mode of 4-
Cl-3-HANA to bacterial 3-HAO (Figure 5), i.e., the 1-oxide
portion coordinates the Fe2+, and the carboxylic acid moiety at
position C3 interacts with Arg95 (referring to the numbering of
the bovine crystal structure; the same numbering is followed in
the next paragraphs). On the other hand, the 2-amino portion
interacts with Glu53 and is probably not involved in iron
coordination, as also reported in the case of 4-Cl-3-HANA.17
The 2-aminonicotinic acid 1-oxide scaffold is well accom-
modated in the 3-HAO binding site, making further hydro-
phobic interactions with residues shaping the binding cavity,
Met35, Iso37, Phe55, and Val103. It is of interest to note that in
the crystal structures of both the bacterial and the mammalian
enzyme, Arg43 is posed in such a way as to point toward the
aromatic ring of the inhibitor, thus suggesting the presence of a
π−cation interaction. Given the high degree of conservation in
the active site, Arg43 keeps the same positioning in our
homology structure. In this case, however, it is pointing toward
the electron deficient pyridine N-oxide ring, thus questioning
whether the proposed interaction is actually significant in
stabilizing the complex. The Glide Score per residue, for some
of the active site residues, including Arg43, was calculated for
compound 6, 4-Cl-3-HANA, and 3-HANA (data shown in
Figure S2, Supporting Information). As can be seen, the relative
interaction of compound 6 with Arg43 is lower than that for the
other two compounds. However, the Glide score per residue
only provides a qualitative trend and, in our opinion, cannot be
used to draw quantitative conclusions. Small substituents at
position C6 of the aromatic ring (compounds 5, 6, and 13) are
generally tolerated because they can be accommodated in the
small subpocket delineated by Arg108, Cys134, Leu137,
Leu141, and His147.
In Vivo Test of Compound 6. As described previously,26
3H-kynurenine in the excitotoxically lesioned rat striatum was
3
3
rapidly converted to H-QUIN and H-KYNA (by 3.4 0.5%
and 3.0 0.5%, respectively; n = 4). Co-infusion of 6 (n = 5)
3
significantly reduced the de novo formation of H-QUIN (to
3
0.9 0.2%; p < 0.01 vs H-kynurenine alone) and at the same
time increased the proportion of newly produced 3H-KYNA (to
3
5.6 0.7%; p < 0.05 vs H-kynurenine alone) (Figure 4). The
3-HAO inhibitor 6 was thus shown to acutely shift brain KP
metabolism away from the excitotoxic metabolite QUIN and
toward the neuroprotective metabolite KYNA.
3
3
3
Figure 4. Acute conversion of H-kynurenine to H-QUIN and H-
KYNA in the excitotoxically lesioned rat striatum in vivo. Data are the
mean
SEM of four (control) and five (compound 6) rats,
respectively. **p < 0.01, *p < 0.05 vs controls (unpaired Student’s t
test). See text for experimental details.
Molecular Modeling. To gain further insights into the
mode of action of the 2-aminonicotinic acid 1-oxide derivatives,
a molecular modeling study was performed, following the
procedure reported in the Experimental Section. For this study,
we selected the crystallographic structure of bovine 3-HAO in
its apo form.19 The reasons for this choice are multiple. Indeed,
even though the crystal structures of the bacterial enzyme
cocrystallized with the known inhibitor 4-Cl-3-HANA and the
substrate are available,17 there are substantial differences
between the mammalian and the bacterial 3-HAO enzymes:
first, mammalian 3-HAO is no longer a dimer but evolved to a
monomeric enzyme constituted by two β-barrel domains, of
which only one is functionally relevant, containing the iron ion
and the catalytic residues; moreover, there is no trace of the
Docking poses generated for compounds with disubstitution
at positions C4 and C6 of the pyridine ring (compounds 14−
18) present steric clashes with residues surrounding sub-
stituents at position C4 (Val22, Arg43), as shown in Figure 6,
whereas the bicyclic compound 12 failed to satisfy all the
criteria shown to be crucial for inhibitory activity (amino group
9488
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
Figure 5. (A) Crystallographic pose of 4-Cl-3-HANA within the bacterial 3-HAO enzyme (PDB code 1YFW). (B) Docking pose of compound 6 in
the bovine 3-HAO crystal structure. The main interactions occurring between ligand and protein structure are conserved. The iron atom is colored in
orange. Figures are prepared with Pymol.28
Scheme 9. Schematic Representation of One of the
Reactions Involved in the Proposed 3-HAO Catalytic
Mechanism17
the iron ion. Halogen substituents, in particular, can lower the
HOMO energy of the o-aminophenol moiety, thus blocking
this oxidation reaction. In contrast, our 2-aminonicotinate 1-
oxide derivatives are inherently unsusceptible to oxidation at
the pyridine nitrogen (formally equivalent to carbon 3 of the o-
aminophenol moiety) and thus less sensitive to electronic
effects.
Figure 6. Docking pose of compound 15. Steric clashes between the
ligand and the surrounding residues are shown in black. Hydrophobic
residues shaping the binding pocket and creating the hydrophobic
environment for the ligand are highlighted in violet. The iron atom is
colored in orange. Figure is prepared with Pymol.28
The steric clashes found in docking studies of the 4-
substituted 2-aminonicotinate 1-oxide derivatives contrast with
the activity of the 6-substituted anthranilates. This could be
explained by protein flexibility. Thus, like the bacterial enzyme,
mammalian 3-HAO has a small flexible loop that moves from
open to closed position above the binding site, especially upon
ligand binding.17 In our docking studies, this loop was partially
closed, as in the bovine 3-HAO crystal.19 Reactive molecules,
such as halogenated anthranilic derivatives, may chelate the
catalytic iron with high efficiency, thereby shifting the
equilibrium toward the open conformation and allowing the
accommodation of bulkier substituents.
interaction with Glu53, N-oxide moiety coordinating the iron
ion, carboxylate interaction with Arg95).
Compounds 19−22 (substituted at position C5 or
disubstituted at positions C5 and C6) generate binding poses
with small steric clashes, which can justify their reduced activity
only in a qualitative way.
The previously reported anthranilic derivatives apparently
have a completely different SAR because halogen and small
alkyl substituents are well tolerated at positions C6 and C5 of
the o-aminophenol ring,15 equivalent to positions C4 and C5 of
the 2-aminonicotinic acid 1-oxide nucleus. This diverse
behavior may be ascribed to a different mechanism of inhibition
between our inhibitors and the anthranilic series. In particular,
also on the basis of the mechanism of dioxygenation proposed
by Zhang et al.,17 one can speculate that the introduction of
electron-withdrawing substituents in the anthranilic acid
nucleus has the effect of blocking the single electron transfer
from the electron-rich substrate to the catalytic iron, resulting
in the production of a radical at position C3 and in the
restoration of the oxidation state of the cofactor (Scheme 9),17
thus trapping the system in an unproductive oxidized state of
CONCLUSIONS
■
In this study, we describe a series of 2-aminonicotinic acid 1-
oxide derivatives as the first chemically stable class of 3-HAO
inhibitors, unrelated to the unstable nucleus of the substrate, 3-
HANA. While several compounds showed activity in human
and rat brain homogenates, the most potent one, 6, was also
tested in a rat model of a neurodegenerative condition, where it
was able to acutely shift the balance between neurotoxic
(QUIN) and neuroprotective (KYNA) KP metabolites toward
the latter.
9489
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
filtered, and concentrated under reduced pressure. The desired
products were purified by silica gel flash chromatography, eluting
with chloroform, then chloroform/methanol, from 99/1 to 95/5
(yield: 68−88%). Compounds 11, 25a−c, 30a−b, 33, 40a−d, 43, and
46 were obtained by this method.
General Procedure for the Hydrolysis of Methyl Esters. A fine
suspension of methyl ester in 10% (w/v) NaOH in methanol (10 mL/
mmol) was heated to reflux and stirred at this temperature for 1 h.
TLC after this time, eluting with chloroform/ethanol/formic acid, 8.7/
1.0/0.3, showed complete consumption of the starting material. The
reaction was stopped.
After cooling to room temperature, methanol was removed by
evaporation in vacuo. The crude, white solid was solved in minimum
amount of water. The clear, colorless solution so obtained was cooled
to 0 °C and carefully acidified to pH 4−5, with 3 N aqueous HCl. A
white precipitate immediately formed. The desired product was
collected by filtration in vacuo in quantitative yield. Compounds 4, 5,
and 8 were obtained by this method.
General Procedure for the Amination of 2-Chloropyridines. A
solution of the substrate in ammonia saturated ethanol (1 mL/mmol)
was placed in a sealed steel bomb, heated to 160 °C, and maintained at
this temperature for 24 h. After cooling to room temperature, solvent
was removed by evaporation in vacuo and the desired product was
purified by silica gel flash chromatography, eluting with petroleum
ether/ethyl acetate, from 9/1 to 6/4 (yield: 73−85%). Compounds
27, 29a−b, and 32 were obtained by this method.
General Procedure for the One-Pot Synthesis of 2-Amino-3-
cyanopyridines. A 10 mL microwave tube was charged with
malononitrile (36, 1.00 equiv), appropriate aldehyde (35, 1.00
equiv), appropriate ketone (37, 1.00 equiv), and ammonium acetate
(38, 1.50 equiv). The tube was placed in a microwave oven and
irradiated at 100 °C for 15 min (maximum power input, 150 W;
maximum pressure, 160 PSI; power max, ON; stirring, ON). The
resulting crude was purified by silica gel flash chromatography, eluting
with appropriate petroleum ether/ethyl acetate mixtures, affording the
target products in 37−65% yield. Compounds 39a−d were obtained
by this method.
General Procedure for the Hydrolysis of Nitriles and 5-Chloro-2-
aminonicotinamide 1-Oxide (46) to the Corresponding Carboxylic
Acids. A fine suspension of the substrate in 10% (w/v) aqueous KOH
(1 mL/mmol) was heated to reflux and stirred at this temperature for
3 h. TLC after this time, eluting with chloroform/ethanol/formic acid,
8.7/1.0/0.3, showed complete consumption of the starting material.
The reaction was stopped.
After cooling to room temperature, the clear, colorless solution so
obtained was cooled to 0 °C and carefully acidified to pH 4−5, with 3
N aqueous HCl. A white precipitate immediately formed. The desired
product was collected by filtration in vacuo in quantitative yield.
Compounds 6, 7, and 13−21 were obtained by this method.
Synthesis of 2-Amino-6-methylnicotinamide 1-Oxide (10).
Compound 27 was treated under the oxidative conditions described
above. After complete oxidation to compound 11 (TLC monitoring),
saturated aqueous NaHCO3 (10 mL/mmol) was added to the reaction
mixture and this was stirred for 30 min at room temperature. TLC
after this time, eluting with chloroform/methanol, 9/1, showed almost
complete consumption of compound 11. The reaction was stopped.
The reaction mixture was extracted with chloroform. The organic
phase was washed with brine, dried over anhydrous Na2SO4, filtered,
and concentrated under reduced pressure. The crude was purified by
silica gel flash chromatography, eluting with chloroform/methanol,
from 95/5 to 85/15, affording pure, target product in 90% yield, as a
light-brown powder.
Synthesis of 2-Chloro-6-ethylnicotinonitrile (31). Sodium hydride,
60% dispersion in mineral oil (1.50 equiv), was carefully added, under
a nitrogen atmosphere, to a stirred, cooled (0 °C) solution of 2-chloro-
6-methylnicotinonitrile (26, 1.00 equiv) in dry N,N-dimethylforma-
mide (2.5 mL/mmol). After stirring at this temperature for 15 min,
iodomethane (4.00 equiv) was added. The reaction mixture was
allowed to warm to room temperature and stirred for 4 h under
nitrogen. TLC after this time, eluting with petroleum ether/ethyl
Clearly, several issues must be carefully considered for the
further development of the compounds described here, and for
compound 6 in particular. Given the low molecular weight, the
presence of the N-oxide, and the high value of the polar surface
area, it can be expected that blood−brain barrier penetrability,
as well as metabolic liability, will need extensive rounds of
optimization. Nevertheless, it is important to point out that the
findings reported here provide a novel pharmacological tool for
the further characterization of the role of the KP in the onset
and progression of several neurodegenerative diseases as well as
for the design of innovative therapeutic strategies for the
treatment of these conditions in humans.
EXPERIMENTAL SECTION
■
Chemistry. General Remarks. Reactants were purchased from
both Sigma-Aldrich and Alfa Aesar at reagent purity and, unless
otherwise noted, were used without any further purification. Dry
solvents used in the reactions were obtained by distillation of technical
grade materials over appropriate dehydrating agents. MCRs were
performed using a CEM Microwave Synthesizer-Discover model.
Reactions were monitored by thin layer chromatography on silica gel-
coated aluminum foils (silica gel on Al foils, SUPELCO Analytical,
Sigma-Aldrich) at both 254 and 365 nm wavelengths. Where indicated,
intermediates and final products were purified by silica gel flash
chromatography (silica gel, 0.040−0.063 mm), using appropriate
solvent mixtures.
Intermediates and final products were characterized by HPLC/MS
analysis (HPLC: Agilent 1100 series, equipped with a Waters
Symmetry C18, 3.5 μm, 4.6 mm × 75 mm column; MS: Applied
1
Biosystem/MDS SCIEX, with API 150EX ion source) and H NMR
and 13C NMR analyses (BRUKER AVANCE 300 MHz and BRUKER
AVANCE 400 MHz). Chemical shifts (δ scale) are reported in parts
1
per million (ppm) relative to TMS. H NMR spectra are reported in
this order: multiplicity and number of protons, signals were
characterized as, s (singlet), d (doublet), dd (doublet of doublets), t
(triplet), q (quartet), hept (heptaplet), m (multiplet), bs (broad
signal).
All compounds were tested as 95% purity samples or higher (by
HPLC/MS), except compound 22, which was tested at 92% purity.
Synthetic Procedures. General Procedure for the Esterification of
Nicotinic Acids. (Trimethylsilyl)diazomethane, 2.0 M in diethyl ether
(2.00 equiv) was added dropwise, over a period of 15 min, to a stirred,
cooled (0 °C) suspension of substrate (1.00 equiv) in dry toluene/
methanol, 3/2 (v/v, 10 mL/mmol), under a nitrogen atmosphere. The
reaction mixture was kept under stirring at this temperature for 30
min. At the end of this time, the reaction mixture became a clear,
yellow solution. TLC, eluting with chloroform/methanol, 9/1, showed
complete consumption of the starting material. The reaction was
stopped.
Solvents were evaporated under reduced pressure. The crude,
solved in chloroform, was washed with saturated aqueous NaHCO3
and brine. The organic phase was dried over anhydrous Na2SO4,
filtered, and concentrated under reduced pressure. The desired
product was purified by silica gel flash chromatography, eluting with
chloroform, then chloroform/methanol, 95/5 (yield: 75−95%).
Compounds 9 and 24a−c were obtained by this method.
General Procedure for the Oxidation of the Pyridine Nitrogen. A
1/2 (v/v) solution of 35% (w/w) aqueous H2O2 in absolute ethanol
was stirred over anhydrous Na2SO4 (5 g/30 mL) at room temperature
for 3 h. After filtration, methyltrioxorhenium(VII) (MTO, 0.10 equiv)
was added to this oxidant solution (4 mL/mmol), followed by
substrate (1.00 equiv). The reaction mixture was stirred at room
temperature for the appropriate time (24 h for halogen-substituted
substrates, 3−4 h for all the others), after which TLC, eluting with
chloroform/methanol, 9/1, showed almost complete consumption of
the starting material. The reaction was stopped.
Brine was added to the reaction mixture, and it was extracted with
chloroform. The organic phase was dried over anhydrous Na2SO4,
9490
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
acetate, 7/3, showed almost complete consumption of the starting
material. The reaction was stopped.
(bs + dd, 2 + 1 H), 8.10−8.12 (dd, 1 H), 8.20−8.22 (dd, 1 H). 13C
NMR (CDCl3, 75.5 MHz) δ: 51.41 (−OCH3), 105.62 (3-C), 112.09
(5-CH), 139.59 (4-CH), 153.15 (6-CH), 159.03 (2-C), 166.92
(−COO−). HPLC/MS: found m/z [M + H]+ 152.9.
The reaction mixture was cooled to 0 °C, then water was carefully
added. The mixture was extracted with diethyl ether. The organic
phase was washed with brine, dried over anhydrous Na2SO4, filtered,
and concentrated under reduced pressure. The crude was purified by
silica gel flash chromatography, eluting with petroleum ether/ethyl
acetate, from 9/1 to 8/2, affording the target product in 69% yield as a
colorless oil.
Synthesis of Pyrido[2,3-d]pyrimidin-4(3H)-one (34). A 10 mL
microwave tube was charged with a mixture of 2-aminonicotinic acid
(23a, 1.00 equiv) and formamide (5.00 equiv). The tube was placed in
a microwave oven and irradiated at 125 °C for 20 min (maximum
power input, 60 W; maximum pressure, 160 PSI; power max, ON;
stirring, ON). The resulting crude was purified by silica gel flash
chromatography, eluting with chloroform/methanol/acetic acid, 95.0/
4.5/0.5, affording the target product in 47% yield as a white powder.
Synthesis of Pyrido[2,3-d]pyrimidin-4(3H)-one 8-Oxide (12).
Compound 34 was treated under the oxidative conditions described
above. After 4 h stirring at room temperature, TLC confirmed
complete consumption of the starting material. The reaction mixture
was heated to reflux, then allowed to slowly cool to room temperature.
The target product crystallized and was collected by filtration in vacuo
in 71% yield.
Synthesis of 2-Amino-5-methylnicotinonitrile (42). A mixture of
3-bromo-2-amino-5-methylpyridine (41, 1.00 equiv), potassium
hexacyanoferrate(II) trihydrate {K4[Fe(CN)6]·3H2O, 0.40 equiv},
tetrakis(triphenylphosphine)palladium(0) [Pd(PPh3)4, 0.05 equiv],
and 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU, 0.25 equiv) in tert-
butanol/water, 1/1 (v/v, 3 mL/mmol) was heated to 85 °C and
stirred at this temperature for 18 h. TLC after this time, eluting with
chloroform/methanol, 95/5, showed incomplete consumption of the
starting material. The reaction was stopped.
The reaction mixture was extracted with chloroform. The organic
phase was washed with brine, dried over anhydrous Na2SO4, filtered,
and concentrated under reduced pressure. The crude was purified by
silica gel flash chromatography, eluting with petroleum ether/ethyl
acetate, from 9/1 to 7/3, affording pure, target product in 12% yield as
a white powder. Starting material was recovered in 78% yield.
Synthesis of 5-Chloro-2-aminonicotinamide (45). A 10 mL
microwave tube was charged with a mixture of 3-bromo-5-chloro-2-
aminopyridine (44, 1.00 equiv), potassium hexacyanoferrate(II)
trihydrate {K4[Fe(CN)6]·3H2O, 0.40 equiv}, tetrakis-
(triphenylphosphine)palladium(0) [Pd(PPh3)4, 0.05 equiv], and 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU, 0.25 equiv) in tert-butanol/
water, 1/1 (v/v, 3 mL/mmol). The tube was placed in a microwave
oven and irradiated at 120 °C for 10 min (maximum power input, 250
W; maximum pressure, 160 PSI; power max, OFF; stirring, ON). TLC
after this time, eluting with chloroform/methanol, 95/5, showed
complete consumption of the starting material.
Methyl 2-Aminonicotinate 1-Oxide (25a). Obtained in 88% yield
from 24a. 1H NMR (DMSO-d6, 300 MHz) δ: 3.87 (s, 3 H), 6.66−6.71
(dd, 1 H), 7.69 (bs, 2 H), 7.71−7.75 (dd, 1 H), 8.37−8.39 (dd. 1 H).
13C NMR (DMSO-d6, 75.5 MHz) δ: 52.45 (−OCH3), 107.31 (3-C),
110.61 (5-CH), 128.36 (4-CH), 140.25 (6-CH), 151.38 (2-C), 165.71
(−COO−). HPLC/MS: found m/z [M + H]+ 169.2.
2-Aminonicotinic Acid 1-Oxide (4). Obtained in quantitative yield
1
from 25a. H NMR (DMSO-d6, 400 MHz) δ: 6.66−6.70 (dd, 1 H),
7.71 (bs, 2 H), 7.74−7.76 (dd, 1 H), 8.33−8.35 (dd, 1 H). 13C NMR
(DMSO-d6, 100.6 MHz) δ: 109.19 (3-C), 110.94 (5-C), 129.68 (4-
CH), 140.52 (6-CH), 152.11 (2-C), 167.79 (−COO−). HPLC/MS:
found m/z [M + H]+ 155.1.
Methyl 6-Chloro-2-aminonicotinate (24b). Obtained in 93% yield
from 6-chloro-2-aminonicotinic acid (23b). 1H NMR (DMSO-d6, 400
MHz) δ: 3.82 (s, 3 H), 6.65−6.67 (d, 1 H), 7.54 (bs, 2 H), 8.04−8.06
(d, 1 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 52.51 (−OCH3),
104.16 (3-C), 111.61 (5-CH), 143.33 (4-CH), 154.06 (6-C), 159.75
(2-C), 166.73 (−COO−). HPLC/MS: found m/z [M + H]+ 187.3.
Methyl 6-Chloro-2-aminonicotinate 1-Oxide (25b). Obtained in
70% yield from 24b. 1H NMR (DMSO-d6, 400 MHz) δ: 3.88 (s, 3 H),
7.03−7.05 (d, 1 H), 7.69−7.71 (d, 1 H), 7.95 (bs, 2 H). 13C NMR
(DMSO-d6, 100.6 MHz) δ: 53.07 (−OCH3), 105.71 (3-C), 111.92 (5-
CH), 128.18 (4-CH), 142.93 (6-C), 153.10 (2-C), 166.00 (−COO−).
HPLC/MS: found m/z [M + H]+ 203.2.
6-Chloro-2-aminonicotinic Acid 1-Oxide (5). Obtained in
quantitative yield from 25b. 1H NMR (DMSO-d6, 300 MHz) δ:
7.00−7.03 (d, 1 H), 7.67−7.70 (d, 1 H), 7.95 (bs, 2 H). 13C NMR
(DMSO-d6, 75.5 MHz) δ: 102.26 (3-C), 110.28 (5-CH), 139.27 (4-
CH), 153.59 (6-C), 158.98 (2-C), 168.43 (−COO−). HPLC/MS:
found m/z [M + H]+ 189.5.
2-Amino-6-methylnicotinonitrile (27). Obtained in 85% yield from
2-chloro-6-methylnicotinonitrile (26). 1H NMR (DMSO-d6, 300
MHz) δ: 2.30 (s, 3 H), 6.50−6.53 (d, 1 H), 6.76 (bs, 2 H), 7.70−
7.73 (d, 1 H). 13C NMR (DMSO-d6, 75.5 MHz) δ: 24.31 (−CH3),
86.10 (3-C), 111.58 (5-CH), 117.36 (−CN), 142.09 (4-CH), 159.55
(2-C), 162.57 (6-C). HPLC/MS: found m/z [M + H]+ 134.2.
2-Amino-6-methylnicotinonitrile 1-Oxide (11). Obtained in 78%
yield from 27. 1H NMR (DMSO-d6, 400 MHz) δ: 2.41 (s, 3 H), 6.77−
6.79 (d, 1 H), 7.45−7.47 (d, 1 H), 7.71 (bs, 2 H). 13C NMR (DMSO-
d6, 100.6 MHz) δ: 18.65 (−CH3), 89.03 (3-C), 113.15 (5-CH), 115.95
(−CN), 129.37 (4-CH), 151.62 (2-C), 152.61 (6-C). HPLC/MS:
found m/z [M + H]+ 150.0.
2-Amino-6-methylnicotinic Acid 1-Oxide (6). Obtained in
1
quantitative yield from 11. H NMR (DMSO-d6, 400 MHz) δ: 2.42
(s, 3 H), 6.72−6.74 (d, 1 H), 7.62−7.64 (d, 1 H), 7.73 (bs, 2 H),
13.48 (bs, 1 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 18.78 (−CH3),
106.20 (3-C), 111.52 (5-CH), 127.92 (4-CH), 150.63 (2-C), 152.07
(6-C), 168.10 (−COO−). HPLC/MS: found m/z [M + H]+ 169.4.
2-Amino-6-methylnicotinic Acid (7). Obtained in quantitative yield
from 27. 1H NMR (DMSO-d6, 300 MHz) δ: 2.30 (s, 3 H), 6.46−6.49
(d, 1 H), 7.16 (bs, 2 H), 7.91−7.93 (d, 1 H). 13C NMR (DMSO-d6,
75.5 MHz) δ: 24.10 (−CH3), 102.71 (3-C), 111.37 (5-CH), 140.34
(4-CH), 159.31 (2-C), 162.57 (6-C), 168.59 (−COO−). HPLC/MS:
found m/z [M + H]+ 153.3.
Brine was added to the reaction mixture and it was extracted with
chloroform. The organic phase was washed with brine, dried over
anhydrous Na2SO4, filtered, and concentrated under reduced pressure.
The crude was purified by filtration over a pad of silica gel, washing
with chloroform/methanol, from 95/5 to 85/15. The desired product
was isolated in 79% yield as a yellow powder.
Synthesis of 5-Bromo-2-aminonicotinic Acid 1-Oxide (22). 5-
Bromo-2-aminonicotinic acid (47) was treated under the oxidative
conditions described above. After overnight stirring at room
temperature, TLC showed almost complete consumption of the
starting material. The reaction was stopped by addition of 1 N aqueous
NaOH. The reaction mixture became an orange solution, and it was
cooled to 0 °C over an ice bath. Then 3 N aqueous HCl was added
dropwise (pH 4−5). A white precipitate immediately formed, and it
was collected by filtration in vacuo in quantitative yield (purity: 92%
by HPLC/MS analysis) and with unreacted starting material (47) as
the sole impurity.
Methyl 2-Amino-6-methylnicotinate 1-Oxide (9). Obtained in
1
75% yield from 6. H NMR (DMSO-d6, 400 MHz) δ: 2.42 (s, 3 H),
3.86 (s, 3 H), 6.75−6.77 (d, 1 H), 7.63−7.65 (d, 1 H), 7.72 (bs, 2 H).
13C NMR (DMSO-d6, 100.6 MHz) δ: 18.78 (−CH3), 52.78
(−OCH3), 105.22 (3-C), 111.76 (5-CH), 127.37 (4-CH), 150.99
(2-C), 151.81 (6-C), 166.46 (−COO−). HPLC/MS: found m/z [M +
H]+ 183.3.
2-Amino-6-methylnicotinamide 1-Oxide (10). Obtained in 90%
yield from 27. 1H NMR (DMSO-d6, 300 MHz) δ: 2.39 (s, 3 H), 6.69−
6.71 (d, 1 H), 7.55 (bs, 1 H), 7.59−7.62 (d, 1 H), 7.85 (bs, 2 H), 8.09
(bs, 1 H). 13C NMR (DMSO-d6, 75.5 MHz) δ: 18.05 (−CH3), 108.22
Characterization of Intermediates and Final Products. Methyl 2-
Aminonicotinate (24a). Obtained in 90% yield from 2-aminonicotinic
1
acid (23a). H NMR (CDCl3, 300 MHz) δ: 3.88 (s, 3 H), 6.56−6.63
9491
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
(3-C), 110.39 (5-CH), 124.89 (4-CH), 148.57 (2-C), 151.12 (6-C),
168.45 (−CON−). HPLC/MS: found m/z [M + H]+ 168.3.
Methyl 6-Chloronicotinate (24c). Obtained in 95% yield from 6-
chloronicotinic acid (23c). 1H NMR (CDCl3, 300 MHz) δ: 3.98 (s, 3
H), 7.42−7.45 (d, 1 H), 8.24−8.27 (dd, 1 H), 9.00−9.01 (d, 1 H). 13C
NMR (CDCl3, 75.5 MHz) δ: 52.61 (−OCH3), 124.19 (5-CH), 125.00
(3-C), 139.59 (4-CH), 151.18 (2-CH), 155.68 (6-C), 164.88
(−COO−). HPLC/MS: found m/z [M + H]+ 172.2.
Methyl 6-Chloronicotinate 1-Oxide (25c). Obtained in 68% yield
from 24c. 1H NMR (DMSO-d6, 300 MHz) δ: 3.88 (s, 3 H), 7.71−7.73
(d, 1 H), 7.91−7.93 (d, 1 H), 8.68 (s, 1 H). 13C NMR (CDCl3, 75.5
MHz) δ: 53.49 (−OCH3), 125.87 (5-CH), 127.92 (4-CH), 128.01 (3-
C), 140.78 (2-CH), 144.90 (6-C), 163.29 (−COO−). HPLC/MS:
found m/z [M + H]+ 188.3.
115.11 (−CN), 139.90 (4-C), 150.47 (2-C), 152.16 (6-C). HPLC/
MS: found m/z [M + H]+ 164.3.
2-Amino-4,6-dimethylnicotinic Acid 1-Oxide (14). Obtained in
1
quantitative yield from 30a. H NMR (DMSO-d6, 300 MHz) δ: 2.44
(s, 2× 3 H), 6.73 (s, 1 H), 8.04 (bs, 2 H). 13C NMR (DMSO-d6, 75.5
MHz) δ: 18.14 (−CH3), 22.60 (−CH3), 109.64 (3-C), 115.27 (5-
CH), 146.47 (4-C), 149.40 (2-C), 151.81 (6-C), 167.73 (−COO−).
HPLC/MS: found m/z [M + H]+ 183.3.
2-Amino-4-ethyl-6-methylnicotinonitrile (39a). Obtained in 37%
yield from propionaldehyde, malononitrile, acetone, and ammonium
acetate. 1H NMR (CDCl3, 300 MHz) δ: 1.24−1.29 (t, 3 H), 2.39 (s, 3
H), 2.67−2.75 (q, 2 H), 5.33 (bs, 2 H), 6.47 (s, 1 H). 13C NMR
(CDCl3, 75.5 MHz) δ: 13.88 (−CH3), 24.64 (−CH3), 27.45
(−CH2−), 88.36 (3-C), 113.12 (5-CH), 116.36 (−CN), 158.69 (4-
C), 159.59 (2-C), 162.12 (6-C). HPLC/MS: found m/z [M + H]+
162.4.
2-Amino-4-ethyl-6-methylnicotinonitrile 1-Oxide (40a). Obtained
in 76% yield from 39a. 1H NMR (DMSO-d6, 400 MHz) δ: 1.18−1.21
(t, 3 H), 2.40 (s, 3 H), 2.60−2.66 (q, 2 H), 6.75 (s, 1 H), 7.66 (bs, 2
H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 14.25 (−CH3), 18.61
(−CH3), 26.84 (−CH2−), 89.17 (3-C), 112.72 (5-CH), 114.88
(−CN), 145.71 (4-C), 150.87 (2-C), 152.32 (6-C). HPLC/MS: found
m/z [M + H]+ 178.3.
6-Chloronicotinic Acid 1-Oxide (8). Obtained in quantitative yield
1
from 25c. H NMR (DMSO-d6, 400 MHz) δ: 7.72−7.74 (d, 1 H),
7.90−7.93 (d, 1 H), 8.68 (s, 1 H), 14.05 (bs, 1 H). 13C NMR (DMSO-
d6, 100.6 MHz) δ: 126.26 (5-CH), 127.80 (4-CH), 129.66 (3-C),
140.95 (2-CH), 144.33 (6-C), 164.34 (−COO−). HPLC/MS: found
m/z [M + H]+ 174.0.
Pyrido[2,3-d]pyrimidin-4(3H)-one (34). Obtained in 47% yield
from 2-aminonicotinic acid (23a). 1H NMR (DMSO-d6, 400 MHz) δ:
7.54−7.57 (dd, 1 H), 8.33 (s, 1 H), 8.50−8.52 (d, 1 H), 8.94−8.96 (d,
1 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 118.29 (4a-C), 123.00 (2-
CH), 136.09 (6-CH), 149.27 (5-CH), 156.20 (7-CH), 159.08 (8a-C),
162.16 (4-CON−). HPLC/MS: found m/z [M + H]+ 148.0.
Pyrido[2,3-d]pyrimidin-4(3H)-one 8-Oxide (12). Obtained in 71%
2-Amino-4-ethyl-6-methylnicotinic Acid 1-Oxide (15). Obtained
1
in quantitative yield from 40a. H NMR (DMSO-d6, 400 MHz) δ:
1.13−1.17 (t, 3 H), 2.46 (s, 3 H), 2.75−2.81 (q, 2 H), 6.76 (s, 1 H),
7.90 (bs, 2 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 15.26 (−CH3),
18.23 (−CH3), 27.74 (−CH2−) 110.13 (3-C), 113.77 (5-CH), 149.42
(4-C), 151.01 (2-C), 151.34 (6-C), 167.42 (−COO−). HPLC/MS:
found m/z [M + H]+ 194.4.
1
yield from 34. H NMR (CD3CO2D, 400 MHz) δ: 7.66−7.70 (dd, 1
H), 8.49−8.51 (d, 1 H), 8.70 (s, 1 H), 9.07−9.08 (d, 1 H). 13C NMR
(CD3CO2D, 100.6 MHz) δ: 121.28 (4a-C), 122.74 (2-CH), 129.75
(5-CH), 145.14 (6-CH), 150.74 (7-CH), 151.83 (8a-C), 160.91 (4-
CON−). HPLC/MS: found m/z [M + H]+ 164.4.
2-Amino-4-isopropyl-6-methylnicotinonitrile (39b). Obtained in
40% yield from isobutyraldehyde, malononitrile, acetone, and
1
ammonium acetate. H NMR (DMSO-d6, 400 MHz) δ: 1.19−1.20
2-Chloro-6-ethylnicotinonitrile (31). Obtained in 69% yield from
(d, 6 H), 2.29 (s, 3 H), 2.94−3.05 (hept, 1 H), 6.52 (s, 1 H), 6.65 (bs,
2 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 22.45 (2× −CH3), 24.89
(−CH3), 32.40 (−CH−), 86.23 (3-C), 109.14 (5-CH), 116.89
(−CN), 160.71 (4-C), 162.52 (2-C), 162.80 (6-C). HPLC/MS:
found m/z [M + H]+ 176.3.
1
2-chloro-6-methylnicotinonitrile (26). H NMR (CDCl3, 400 MHz)
δ: 1.29−1.33 (t, 3 H), 2.85−2.90 (q, 2 H), 7.24−7.26 (d, 1 H), 7.90−
7.92 (d, 1 H). 13C NMR (CDCl3, 100.6 MHz) δ: 13.12 (−CH3), 31.49
(−CH2−), 107.65 (3-C), 115.03 (−CN), 120.77 (5-CH), 142.58 (4-
CH), 151.97 (6-C), 169.01 (2-C). HPLC/MS: found m/z [M + H]+
167.3.
2-Amino-6-ethylnicotinonitrile (32). Obtained in 78% yield from
31. 1H NMR (CDCl3, 400 MHz) δ: 1.25−1.28 (t, 3 H), 2.65−2.71 (q,
2 H), 5.30 (bs, 2 H), 6.58−6.60 (d, 1 H), 7.60−7.62 (d, 1 H). 13C
NMR (CDCl3, 100.6 MHz) δ: 12.19 (−CH3), 30.61 (−CH2−), 86.92
(3-C), 104.21 (5-CH), 119.85 (−CN), 147.58 (4-CH), 167.48 (2-C),
172.13 (6-C). HPLC/MS: found m/z [M + H]+ 148.3.
2-Amino-4-isopropyl-6-methylnicotinonitrile 1-oxide (40b). Ob-
1
tained in 74% yield from 39b. H NMR (DMSO-d6, 400 MHz) δ:
1.21−1.23 (d, 6 H), 2.41 (s, 3 H), 2.98−3.08 (hept, 1 H), 6.80 (s, 1
H), 7.62 (bs, 2 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 18.76
(−CH3), 22.55 (2× −CH3), 32.10 (−CH−), 88.70 (3-C), 110.21 (5-
CH), 114.91 (−CN), 149.52 (4-C), 151.04 (2-C), 152.28 (6-C).
HPLC/MS: found m/z [M + H]+ 192.0.
2-Amino-4-isopropyl-6-methylnicotinic Acid 1-Oxide (16). Ob-
tained in quantitative yield from 40b. 1H NMR (DMSO-d6, 300 MHz)
δ: 1.17−1.19 (d, 6 H), 2.54 (s, 3 H), 3.25−3.38 (hept, 1 H), 6.96 (s, 1
H), 8.02 (bs, 2 H). 13C NMR (DMSO-d6, 75.5 MHz) δ: 18.32
(−CH3), 22.80 (2× CH3), 30.95 (−CH−), 110.37 (5-CH), 113.32 (3-
C), 149.92 (4-C), 156.42 (2-C), 157.58 (6-C), 166.34 (−COO−).
HPLC/MS: found m/z [M + H]+ 211.4.
2-Amino-4-phenyl-6-methylnicotinonitrile (39c). Obtained in 65%
yield from benzaldehyde, malononitrile, acetone, and ammonium
acetate. 1H NMR (CDCl3, 300 MHz) δ: 2.48 (s, 3 H), 5.33 (bs, 2 H),
6.66 (s, 1 H), 7.49−7.52 (m, 3 H), 7.58−7.59 (m, 2 H). 13C NMR
(CDCl3, 75.5 MHz) δ: 24.81 (−CH3), 87.25 (3-C), 114.00 (5-CH),
117.12 (−CN), 128.12 (2× Ph−CH), 128.86 (2× Ph−CH), 129.73
(Ph−CH), 136.74 (Ph−C), 154.57 (4-C), 160.10 (2-C), 162.33 (6-
C). HPLC/MS: found m/z [M + H]+ 210.4.
2-Amino-6-ethylnicotinonitrile 1-Oxide (33). Obtained in 71%
1
yield from 32. H NMR (CDCl3, 300 MHz) δ: 1.32−1.37 (t, 3 H),
2.98−3.06 (q, 2 H), 6.45 (bs, 2 H), 6.67−6.70 (d, 1 H), 7.32−7.34 (d,
1 H). 13C NMR (CDCl3, 75.5 MHz) δ: 10.33 (−CH3), 24.52
(−CH2−), 90.43 (3-C), 111.61 (5-CH), 114.95 (−CN), 129.13 (4-
CH), 152.12 (2-C), 157.10 (6-C). HPLC/MS: found m/z [M + H]+
164.4.
2-Amino-6-ethylnicotinic Acid 1-Oxide (13). Obtained in
quantitative yield from 33. 1H NMR (DMSO-d6, 400 MHz) δ:
1.19−1.23 (t, 3 H), 2.82−2.85 (q, 2 H), 6.66−6.69 (d, 1 H), 7.66−
7.68 (d, 1 H), 7.75 (bs, 2 H), 13.49 (bs, 1 H). 13C NMR (DMSO-d6,
100.6 MHz) δ: 10.87 (−CH3), 24.63 (−CH2−), 106.75 (3-C), 109.56
(5-CH), 128.19 (4-CH), 152.12 (2-C), 154.74 (6-C), 168.22
(−COO−). HPLC/MS: found m/z [M + H]+ 183.4.
2-Amino-4,6-dimethylnicotinonitrile (29a). Obtained in 83% yield
from 2-chloro-4,6-dimethylnicotinonitrile (28a). 1H NMR (DMSO-d6,
400 MHz) δ: 2.26 (s, 3 H), 2.27 (s, 3 H), 6.46 (s, 1 H), 6.64 (bs, 2 H).
13C NMR (DMSO-d6, 100.6 MHz) δ: 20.19 (−CH3), 24.60 (−CH3),
87.52 (3-C), 113.46 (5-CH), 117.01 (−CN), 152.89 (4-C), 160.54 (2-
C), 161.98 (6-C). HPLC/MS: found m/z [M + H]+ 148.4.
2-Amino-4,6-dimethylnicotinonitrile 1-Oxide (30a). Obtained in
75% from 29a. 1H NMR (DMSO-d6, 300 MHz) δ: 2.32 (s, 3 H), 2.37
(s, 3 H), 6.72 (s, 1 H), 7.64 (bs, 2 H). 13C NMR (DMSO-d6, 75.5
MHz) δ: 18.51 (−CH3), 19.84 (−CH3), 89.93 (3-C), 114.23 (5-CH),
2-Amino-4-phenyl-6-methylnicotinonitrile 1-Oxide (40c). Ob-
tained in 70% yield from 39c. H NMR (DMSO-d6, 400 MHz) δ:
1
2.47 (s, 3 H), 6.93 (s, 1 H), 7.51−7.56 (m, 3 H), 7.60−7.61 (m, 2 H),
7.75 (bs, 2 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 18.63 (−CH3),
87.74 (3-C), 113.57 (5-CH), 115.75 (−CN), 128.75 (2× Ph−CH),
129.23 (2× Ph−CH), 129.83 (Ph−CH), 136.47, (Ph−C) 141.15 (4-
C), 150.87 (2-C), 153.05 (6-C). HPLC/MS: found m/z [M + H]+
226.5.
2-Amino-4-phenyl-6-methylnicotinic Acid 1-Oxide (17). Obtained
in quantitative yield from 40c. 1H NMR DMSO-d6, 400 MHz) δ: 2.43
9492
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
(s, 3 H), 6.71 (s, 1 H), 7.29 (bs, 2 H), 7.32−7.45 (m, 5 H). 13C NMR
(DMSO-d6, 100.6 MHz) δ: 18.38 (−CH3), 108.15 (3-C), 114.16 (5-
CH), 128.20 (Ph−C), 128.29 (2× Ph−CH), 128.58 (2× Ph−CH),
139.51 (Ph−C), 140.02 (4-C), 147.84 (2-C), 150.40 (6-C), 168.32
(−COO−). HPLC/MS: found m/z [M + H]+ 245.4.
5-Chloro-2-aminonicotinamide 1-Oxide (46). Obtained in 71%
yield from 45. 1H NMR (DMSO-d6, 400 MHz) δ: 7.75 (bs, 1 H), 7.85
(bs, 3 H), 8.22 (bs, 1 H), 8.54 (d, 1 H). 13C NMR (DMSO-d6, 100.6
MHz) δ: 111.02 (3-C), 115.22 (5-C), 126.41 (4-CH), 137.91 (6-CH),
150.98 (2-C), 167.61 (−CON−). HPLC/MS: found m/z [M + H]+
188.4.
2-Amino-4,6-diisopropylnicotinonitrile (39d). Obtained in 50%
yield from isobutyraldehyde, malononitrile, 3-methyl-2-butanone, and
ammonium acetate. 1H NMR (DMSO-d6, 400 MHz) δ: 1.15−1.17 (d,
6 H), 1.20−1.21 (d, 6 H), 2.77−2.87 (hept, 1 H), 2.96−3.07 (hept, 1
H), 6.52 (s, 1 H), 6.59 (bs, 2 H). 13C NMR (DMSO-d6, 100.6 MHz)
δ: 22.36 (2× −CH3), 22.47 (2× −CH3), 32.56 (−CH−), 36.40
(−CH−), 86.73 (3-C), 106.70 (5-CH), 116.87 (−CN), 160.79 (4-C),
163.06 (2-C), 171.08 (6-C). HPLC/MS: found m/z [M + H]+ 204.4.
2-Amino-4,6-diisopropylnicotinonitrile 1-Oxide (40d). Obtained
in 73% yield from 39d. 1H NMR (DMSO-d6, 400 MHz) δ: 1.21−1.23
(d, 6 H), 1.24−1.26 (d, 6 H), 3.01−3.12 (hept, 1 H), 3.55−3.65 (hept,
1 H), 6.62 (s, 1 H), 7.65 (bs, 2 H). 13C NMR (DMSO-d6, 100.6 MHz)
δ: 20.06 (2× −CH3), 22.49 (2× −CH3), 28.75 (−CH−), 32.38
(−CH−), 88.60 (3-C), 106.13 (5-CH), 114.89 (−CN), 149.93 (4-C),
152.37 (2-C), 159.02 (6-C). HPLC/MS: found m/z [M + H]+ 220.4.
2-Amino-4,6-diisopropylnicotinic Acid 1-Oxide (18). Obtained in
quantitative yield from 40d. 1H NMR (DMSO-d6, 300 MHz) δ: 1.18−
1.21 (d, 6 H), 1.24−1.26 (d, 6 H), 3.32−3.37 (hept, 1 H), 3.55−3.60
(hept, 1 H), 6.69 (s, 1 H), 7.71 (bs, 2 H). 13C NMR (DMSO-d6, 75.5
MHz) δ: 20.86 (2× −CH3), 22.97 (2× −CH3), 28.63 (−CH−), 30.73
(−CH−), 106.26 (5-CH), 117.90 (3-C), 135.69 (4-C), 150.25 (2-C),
157.36 (6-C), 166.90 (−COO−). HPLC/MS: found m/z [M + H]+
239.1.
5-Chloro-2-aminonicotinic Acid 1-Oxide (21). Obtained in
1
quantitative yield from 46. H NMR (DMSO-d6, 400 MHz) δ: 7.72
(s, 1 H), 7.76 (bs, 2 H), 8.64 (s, 1 H), 13.89 (bs, 1 H). 13C NMR
(DMSO-d6, 100.6 MHz) δ: 108.51 (3-C), 115.33 (5-C), 128.51 (4-
CH), 139.23 (6-CH), 151.49 (2-C), 166.90 (−COO−). HPLC/MS:
found m/z [M + H]+ 189.4.
5-Bromo-2-aminonicotinic Acid 1-Oxide (22). Obtained in
quantitative yield from 5-bromo-2-aminonicotinic acid (47). Purity:
92%. 1H NMR (DMSO-d6, 300 MHz) δ: 7.77 (bs, 2 H), 7.80−7.81 (d,
1 H), 8.66−8.67 (d, 1 H), 13.77 (bs, 1 H). 13C NMR (DMSO-d6, 75.5
MHz) δ: 101.40 (3-C), 109.10 (5-C), 131.17 (4-CH), 140.93 (6-CH),
151.63 (2-C), 166.82 (−COO−). HPLC/MS: found m/z [M − H]−
230.2.
3-HAO Assay in Vitro. Human brain samples (cortex) were
obtained from the Maryland Brain Collection, a repository of post-
mortem tissue maintained at the Maryland Psychiatric Research
Center. This tissues, obtained from five donors (age, 44 12.5; post-
mortem interval, 10
4.3; M/F, 4/1; 3 Caucasian, 2 African-
American) who had died without a history of neurological or
psychiatric disease, were used for the in vitro testing of newly
synthesized compounds. Male Sprague−Dawley rats (220−300 g)
were used to provide tissue for both in vitro and in vivo studies.
Human tissue was homogenized (1:25, w/v) in 60 mM MES [2-(N-
morpholino)ethane sulfonic acid] buffer, pH 6.5, and 80 μL of the
tissue preparation were incubated for 1 h at 37 °C in a solution
containing 150 μM Fe(NH4)2SO4, 0.01% ascorbic acid and 5 nCi (3
μM) [1-14C]-3-HANA (6 mCi/mmol; NEN Dupont, Wilmington,
DE, USA) in a total volume of 200 μL. Solutions of the test
compounds (or vehicle) were added in a volume of 20 μL, while the
tissues were still on ice. Blanks were obtained using 20 μL of the
inhibitor 4-Cl-3-HANA (100 μM in assay). The reaction was
terminated by the addition of 50 μL of 6% perchloric acid, and the
resulting precipitate was removed by centrifugation (16000g, 10 min).
The supernatant was applied to a Dowex 50W (H+-form) ion
exchange column, and newly formed 14C-QUIN was eluted and then
quantitated by liquid scintillation spectrometry. The same protocol
was followed to measure 3-HAO activity in rat brain (cortex), using 80
μL of the original tissue homogenate (1:10, w/v).
2-Amino-5,6-dimethylnicotinonitrile (29b). Obtained in 73% yield
1
from 2-chloro-5,6-dimethylnicotinonitrile (28b). H NMR (DMSO-
d6, 400 MHz) δ: 2.07 (s, 3 H), 2.28 (s, 3 H), 6.51 (bs, 2 H), 7.57 (s, 1
H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 17.46 (−CH3), 23.02
(−CH3), 86.61 (3-C), 117.93 (−CN), 119.70 (5-C), 142.26 (4-CH),
158.53 (2-C), 161.86 (6-C). HPLC/MS: found m/z [M + H]+ 148.4.
2-Amino-5,6-dimethylnicotinonitrile 1-Oxide (30b). Obtained in
77% yield from 29b. 1H NMR (DMSO-d6, 400 MHz) δ: 2.18 (s, 3 H),
2.40 (s, 3 H), 7.40 (s, 1 H), 7.50 (bs, 2 H). 13C NMR (DMSO-d6,
100.6 MHz) δ: 15.10 (−CH3), 18.27 (−CH3), 87.86 (3-C), 116.09
(−CN), 120.79 (5-C), 129.77 (4-CH), 150.87 (2-C), 151.10 (6-C).
HPLC/MS: found m/z [M + H]+ 164.3.
2-Amino-5,6-dimethylnicotinic Acid 1-Oxide (19). Obtained in
1
quantitative yield from 30b. H NMR (DMSO-d6, 400 MHz) δ: 2.21
(s, 3 H), 2.42 (s, 3 H), 7.52 (bs, 2 H), 7.55 (s, 1 H), 13.25 (bs, 1 H).
13C NMR (DMSO-d6, 100.6 MHz) δ: 15.17 (−CH3), 18.48 (−CH3),
105.42 (3-C), 118.68 (5-C), 128.74 (4-CH), 149.83 (2-C), 150.33 (6-
C), 168.09 (−COO−). HPLC/MS: found m/z [M + H]+ 183.3.
2-Amino-5-methylnicotinonitrile (42). Obtained in 12% yield from
QUIN-Induced Striatal Lesions in Rats. Striatal lesions were
induced by a focal intrastriatal infusion of 300 nmoles of the
excitotoxin QUIN (Sigma Chemical Co., St. Louis, MO, USA),
administered over 10 min in 2.5 μL of phosphate buffered saline
(PBS), pH 7.4.29 The injection coordinates were: AP, 0.8 mm anterior
to bregma; L, 2.7 mm from the midline; V, 4.7 mm below the dura.
1
3-bromo-2-amino-5-methylpyridine (41). H NMR (DMSO-d6, 400
MHz) δ: 2.12 (s, 3 H), 6.61 (bs, 2 H), 7.69−7.70 (d, 1 H), 8.06−8.07
(d, 1 H). 13C NMR (DMSO-d6, 100.6 MHz) δ: 16.81 (−CH3), 89.28
(3-C), 117.60 (−CN), 121.12 (5-C), 142.30 (4-CH), 153.79 (6-CH),
158.78 (2-C). HPLC/MS: found m/z [M + H]+ 134.3.
2-Amino-5-methylnicotinonitrile 1-Oxide (43). Obtained in 77%
yield from 42. 1H NMR (DMSO-d6, 400 MHz) δ: 2.13 (s, 3 H), 7.46
(s, 1 H), 7.50 (bs, 2 H), 8.30 (s, 1 H). 13C NMR (DMSO-d6, 100.6
MHz) δ: 16.88 (−CH3), 91.37 (3-C), 115.71 (−CN), 122.15 (5-C),
131.43 (4-CH), 141.18 (6-CH), 150.83 (2-C). HPLC/MS: found m/z
[M + H]+ 150.0.
3
Intrastriatal Injection of H-Kynurenine in Rats. The radioactive
tracer method introduced by Guidetti et al.26 was used to study the
effect of 3-HAO inhibition in the brain in vivo. The excitotoxically
lesioned rat striatum, which shows substantially activated KP
metabolism,26 was selected as the experimental paradigm. Thus, rats
that had received an intrastriatal injection of QUIN one week earlier
were anesthetized with chloral hydrate (360 mg/kg, ip), and 2.5 μCi
(35 μM) 3H-kynurenine (28 Ci/mmol; Amersham, Arlington Heights,
IL, USA) in PBS (pH 7.4) were infused into the striatum in a volume
of 6 μL over 10 min. The radioactive compound was applied either
alone or in combination with 6, using the same injection coordinates
listed above. Prior to use, 6 was dissolved in 1 M NaOH and the
solution was adjusted to physiological pH with diluted acetic acid and
PBS.
2-Amino-5-methylnicotinic Acid 1-Oxide (20). Obtained in
1
quantitative yield from 43. H NMR (DMSO-d6, 400 MHz) δ: 2.16
(s, 3 H), 7.47 (bs, 2 H), 7.59 (s, 1 H), 8.24 (s, 1 H). 13C NMR
(DMSO-d6, 100.6 MHz) δ: 17.11 (−CH3), 108.71 (3-C), 120.18 (5-
C), 130.06 (4-CH), 140.30 (6-CH), 150.20 (2-C), 167.78 (−COO−).
HPLC/MS: found m/z [M + H]+ 169.1.
3
The awake animals were euthanized 2 h after the start of the H-
5-Chloro-2-aminonicotinamide (45). Obtained in 79% yield from
3-bromo-5-chloro-2-aminopyridine (44). 1H NMR (DMSO-d6, 400
MHz) δ: 7.36 (bs, 2 H), 7.46 (bs, 1 H), 8.05−8.06 (m, 3 H). 13C
NMR (DMSO-d6, 100.6 MHz) δ: 109.94 (3-C), 116.84 (5-C), 136.83
(4-CH), 149.92 (6-CH), 158.15 (2-C), 169.09 (−CON−). HPLC/
MS: found m/z [M + H]+ 172.2.
kynurenine infusion, and their brain was quickly removed and placed
on ice. The lesioned striatum was rapidly dissected out, frozen on dry
ice, and stored at −80 °C until processed. After thawing, newly formed
3
3H-QUIN and H-KYNA were processed, detected, and analyzed as
described.26 The production of the two tritiated metabolites in the
9493
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
absence or presence of 6 was expressed as a percentage of the total
radioactivity recovered per striatum.
(2) Costantino, G. New promises for manipulation of kynurenine
pathway in cancer and neurological diseases. Expert Opin. Ther. Targets
2009, 13, 247−258.
(3) Schwarcz, R.; Bruno, J.; Muchowski, P.; Wu, H.-Q. Kynurenines
in the mammalian brain: when physiology meets pathology. Nature
Rev. Neurosci. 2012, 13, 465−477.
(4) Giorgini, F.; Guidetti, P.; Nguyen, Q.; Bennett, S.; Muchowski, P.
A genomic screen in yeast implicates kynurenine 3-monooxygenase as
a therapeutic target for Huntington disease. Nature Genet. 2005, 37,
526−531.
(5) Guillemin, G. J.; Brew, B. J.; Noonan, C. E.; Takikawa, O.;
Cullen, K. M. Indoleamine 2,3 dioxygenase and quinolinic acid
immunoreactivity in Alzheimer’s disease hippocampus. Neuropathol.
Appl. Neurobiol. 2005, 31, 395−404.
(6) Chugani, D. C. α-Methyl-L-tryptophan: mechanisms for tracer
localization of epileptogenic brain regions. Biomarkers Med. 2011, 5,
567−575.
Molecular Modeling Studies. The crystal structure of bovine 3-
HAO, PDB code 3FE5, was used for generating a homology model of
the human enzyme by point-mutating the nonconserved residues (see
Figure S1, Supporting Information). Protein Preparation Workflow
and docking were performed with the Schrodinger Maestro package.30
̈
The Protein Preparation Workflow31 was used to add hydrogen atoms,
assign bond orders, create zero-order bonds to the iron ion, optimize
the hydrogen bonding network and, finally, to refine the protein
structure with a maximum RMSD of tolerance of 0.3 Å. Partial charges
were assigned with the single-point energy calculation of Qsite32 to
ensure a more accurate charge assignment to the iron atom and the
surrounding binding site residues. We used DFT-B3LYP, lacvp++** as
basis set and continuum salvation model for the QM region (Fe2+
atom and residues Arg43, His47, Glu53, His91, Arg95, and Glu105).
OPLS2005 was used for residues in the MM region. Molecular models
of compounds 4−22 were built using Maestro, and tautomerization
and protonation state at pH 7.0 1 were assigned with LigPrep.33
Ligands geometry optimization and partial charges were calculated
with Jaguar, using DFT-B3LYP, lacvp++** as basis set, a maximum
number of 100 steps, and using the Poisson−Boltzmann solver as
solvent model for water. Docking simulations were performed using
Glide5.6. The binding site grid box was centered on the centroid of the
following residues: Fe2+, Arg43, His47, Glu53, Phe55, His91, Arg95,
and Glu105. During grid generation, the previously calculated protein
partial charges were retained. Standard precision (SP) mode was
applied. Two constraints were defined to avoid the generation of poses
not in agreement with experimental data, the iron−ligand interaction
and the hydrogen bond between ligand and Arg95 residue, and at least
one of them should be satisfied during docking. Ten poses for each
ligand were required.
(7) Chen, Y.; Stankovic, R.; Cullen, K. M.; Meininger, V.; Garner, B.;
Coggan, S.; Grant, R.; Brew, B. J.; Guillemin, G. J. The kynurenine
pathway and inflammation in amyotrophic lateral sclerosis. Neurotoxic.
Res. 2010, 18, 132−142.
(8) Lim, C.; Brew, B.; Sundaram, G.; Guillemin, G. Understanding
the roles of the kynurenine pathway in multiple sclerosis progression.
Int. J. Tryptophan Res. 2010, 3, 157−167.
(9) Erhardt, S.; Blennow, K.; Nordin, C.; Skogh, E.; Lindstrom, L. H.;
̈
Engberg, G. Kynurenic acid levels are elevated in the cerebrospinal
fluid of patients with schizophrenia. Neurosci. Lett. 2001, 313, 96−98.
(10) Maes, M.; Verkerk, R.; Bonaccorso, S.; Ombelet, W.; Bosmans,
́
E.; Scharpe, S. Depressive and anxiety symptoms in the early
puerperium are related to increased degradation of tryptophan into
kynurenine, a phenomenon which is related to immune activation. Life
Sci. 2002, 71, 1837−1848.
(11) Heyes, M. P.; Saito, K.; Major, E. O.; Milstien, S.; Markey, S. P.;
Vickers, J. H. A mechanism of quinolinic acid formation by brain in
inflammatory neurological disease. Attenuation of synthesis from L-
tryptophan by 6-chlorotryptophan and 4-chloro-3-hydroxyanthranilate.
Brain 1993, 116, 1425−1450.
ASSOCIATED CONTENT
■
S
* Supporting Information
Sequence alignment of 3-HAO from Ralstonia metallidurans
and Bos taurus. Glide Score per residue calculated for
compound 6, 3-HANA, and 4-Cl-3-HANA. This material is
available free of charge via the Internet at http://pubs.acs.org.
(12) Amori, L.; Guidetti, P.; Pellicciari, R.; Kajii, Y.; Schwarcz, R. On
the relationship between the two branches of the kynurenine pathway
in the rat brain in vivo. J. Neurochem. 2009, 109, 316−325.
(13) (a) Long, C. L.; Hill, H. N.; Weinstock, I. M.; Henderson, L. M.
Studies of the enzymatic transformation of 3-hydroxyanthranilate to
quinolinate. J. Biol. Chem. 1954, 211, 405−417. (b) Foster, A. C.;
White, R. J.; Schwarcz, R. Synthesis of quinolinic acid by 3-
hydroxyanthranilic acid oxygenase in rat brain tissue in vitro. J.
Neurochem. 1986, 47, 23−30.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +39 0521905054. E-mail: gabriele.costantino@unipr.
it.
Present Address
(14) Foster, A. C.; Zinkand, W. C.; Schwarcz, R. Quinolinic acid
phosphoribosyltransferase in rat brain. J. Neurochem. 1985, 44, 446−
454.
§X.-D.W.: Department of Neurology, Tianjin Huanhu Hospital,
Tianjin 300060, China.
(15) Linderberg, M.; Hellberg, S.; Bjork, S.; Gotthammar, B.;
̈
Author Contributions
Hogberg, T.; Persson, K.; Schwarcz, R.; Luthman, J.; Johansson, R.
̈
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Synthesis and QSAR of substituted 3-hydroxyanthranilic acid
derivatives as inhibitors of 3-hydroxyanthranilic acid dioxygenase (3-
HAO). Eur. J. Med. Chem. 1999, 34, 729−744.
(16) Schwarcz, R.; Okuno, E.; White, R. J.; Bird, E. D.; Whetsell, W.
O., Jr. 3-Hydroxyanthranilate oxygenase activity is increased in the
brains of Huntington disease victims. Proc. Natl. Acad. Sci. U. S. A.
1988, 85, 4079−4081.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
(17) (a) Zhang, Y.; Colabroy, K.; Begley, T.; Ealick, S. Structural
studies on 3-hydroxyanthranilate-3,4-dioxygenase: the catalytic mech-
anism of a complex oxidation involved in NAD biosynthesis.
Biochemistry 2005, 44, 7632−7643. (b) Colabroy, K.; Zhai, H.; Li,
T.; Ge, Y.; Zhang, Y.; Liu, A.; Ealick, S.; McLafferty, F.; Begley, T. The
mechanism of inactivation of 3-hydroxyanthranilate-3,4-dioxygenase
by 4-chloro-3-hydroxyanthranilate. Biochemistry 2005, 44, 7623−7631.
(18) Parli, C. J.; Krieter, P.; Schmidt, B. Metabolism of 6-
chlorotryptophan to 4-chloro-3-hydroxyanthranilic acid: a potent
inhibitor of 3-hydroxyanthranilic acid oxidase. Arch. Biochem. Biophys.
1980, 203, 161−166.
■
We are grateful to Dr. Francesca Notarangelo for discussions
and to Erin Stachowski for expert technical assistance. G.C.
gratefully acknowledges support from MIUR, PRIN 2010-11,
project no. 20103W4779_005.
REFERENCES
■
(1) Leklem, J. E. Quantitative aspects of tryptophan metabolism in
humans and other species: a review. Am. J. Clin. Nutr. 1971, 24, 659−
672.
9494
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495

Journal of Medicinal Chemistry
Article
(19) Đilovic,
́
I.; Gliubich, F.; Malpeli, G.; Zanotti, G.; Matkovic-
́
̌
Calogovic,
́
D. Crystal structure of bovine 3-hydroxyanthranilate 3,4-
dioxygenase. Biopolymers 2009, 91, 1189−1195.
(20) Presser, A.; Hufner, A. Trimethylsilyldiazomethanea mild and
̈
efficient reagent for the methylation of carboxylic acids and alcohols in
natural products. Monatsh. Chem. 2004, 135, 1015−1022.
(21) Murray, R. W.; Iyanar, K.; Chen, J.; Wearing, J. T. Oxidation of
organonitrogen compounds by the methyltrioxorhenium-hydrogen
peroxide system. Tetrahedron Lett. 1996, 37, 805−808.
(22) Thomas, A. A.; Le Huerou, Y.; De Meese, J.; Gunawardana, I.;
Kaplan, T.; Romoff, T. T.; Gonzales, S. S.; Condroski, K.; Boyd, S. A.;
Ballard, J.; Bernat, B.; DeWolf, W.; Han, M.; Lee, P.; Lemieux, C.;
Pedersen, R.; Pheneger, J.; Poch, G.; Smith, D.; Sullivan, F.; Weiler, S.;
Wright, S. K.; Lin, J.; Brandhuber, B.; Vigers, G. Synthesis, in vitro and
in vivo activity of thiamine antagonist transketolase inhibitors. Bioorg.
Med. Chem. Lett. 2008, 18, 2206−2210.
(23) Alexandre, F.-R.; Berecibar, A.; Besson, T. Microwave-assisted
Niementowski reaction. Back to the roots. Tetrahedron Lett. 2002, 43,
3911−3913.
(24) Shi, F.; Tu, S.; Fang, F.; Li, T. One-pot synthesis of 2-
aminopyridine derivatives under microwave irradiation without
solvent. ARKIVOC 2005, i, 137−142.
(25) Zhang, D.; Sun, H.; Zhang, L.; Zhou, Y.; Li, C.; Jiang, H.; Chen,
K.; Liu, H. An expedient Pd/DBU mediated cyanation of aryl/
heteroaryl bromides with K4[Fe(CN)6]. Chem. Commun. 2012, 48,
2909−2911.
(26) Guidetti, P.; Eastman, C. L.; Schwarcz, R. Metabolism of
[5-3H]kynurenine in the rat brain in vivo: evidence for the existence of
a functional kynurenine pathway. J. Neurochem. 1995, 65, 2621−2632.
(27) Li, X.; Guo, M.; Fan, J.; Tang, W.; Wang, D.; Ge, H.; Rong, H.;
Teng, M.; Niu, L.; Liu, Q.; Hao, Q. Crystal structure of 3-
hydroxyanthranilic acid 3,4-dioxygenase from Saccharomyces cerevisiae:
a special subgroup of the type III extradiol dioxygenases. Protein Sci.
2006, 15, 761−773.
(28) DeLano, W. L. PyMol Molecular Graphic System, version 1.2r1;
DeLano Scientific: South San Francisco, CA, 2009.
(29) Schwarcz, R.; Whetsell, W. O., Jr.; Mangano, R. M. Quinolinic
acid: an endogenous metabolite that produces axon-sparing lesions in
rat brain. Science 1983, 219, 316−318.
(30) Maestro 9.1; Schrodinger, LLC: New York, 2010.
̈
(31) Protein Preparation Wizard, Epik version 2.0, Impact version 5.5,
Prime version 2.1; Schrodinger, LLC, New York, 2009.
̈
(32) QSite 5.6; Schrodinger, LLC: New York, 2010.
̈
(33) LigPrep, version 2.3; Schrodinger, LLC: New York, 2009.
̈
(34) See as a general example: Katritzky, A. R.; Pilarski, B.; Urogdi, L.
Efficient conversion of nitriles to amides with basic hydrogen peroxide
in dimethyl sulfoxide. Synthesis 1989, 12, 949−950 and references
therein..
9495
dx.doi.org/10.1021/jm401249c | J. Med. Chem. 2013, 56, 9482−9495