The discovery of phthalazinone-based human H1 and H3 single-ligand antagonists suitable for intranasal administration for the treatment of allergic rhinitis

Article abstract of DOI:10.1021/jm1013874

A series of potent phthalazinone-based human H₁ and H ₃ bivalent histamine receptor antagonists, suitable for intranasal administration for the potential treatment of allergic rhinitis, were identified. Blockade of H₃ receptors is thought to improve efficacy on nasal congestion, a symptom of allergic rhinitis that is currently not treated by current antihistamines. Two analogues (56a and 56b) had slightly lower H₁ potency (pA₂ 9.1 and 8.9, respectively, vs 9.7 for the clinical gold-standard azelastine, and H₃ potency (pK _i 9.6 and 9.5, respectively, vs 6.8 for azelastine). Compound 56a had longer duration of action than azelastine, low brain penetration, and low oral bioavailability, which coupled with the predicted low clinical dose, should limit the potential of engaging CNS-related side-effects associated with H ₁ or H₃ antagonism.

Full text of DOI:10.1021/jm1013874

ARTICLE
pubs.acs.org/jmc
The Discovery of Phthalazinone-Based Human H1 and H3
Single-Ligand Antagonists Suitable for Intranasal Administration for
the Treatment of Allergic Rhinitis
Panayiotis A. Procopiou,*,† Christopher Browning,‡ Jennifer M. Buckley,† Kenneth L. Clark,‡ Lise Fechner,†
Paul M. Gore,† Ashley P. Hancock,† Simon T. Hodgson,† Duncan S. Holmes,† Michael Kranz,||
Brian E. Looker,† Karen M. L. Morriss,† Daniel L. Parton,† Linda J. Russell,‡ Robert J. Slack,‡ Steven L. Sollis,†
Sadie Vile,† and Clarissa J. Watts§
†Departments of Medicinal Chemistry, ‡Respiratory Biology, §Drug Metabolism and Pharmacokinetics, and Cheminformatics,
GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2NY, United Kingdom
S Supporting Information
b
ABSTRACT: A series of potent phthalazinone-based human
H1 and H3 bivalent histamine receptor antagonists, suitable for
intranasal administration for the potential treatment of allergic
rhinitis, were identified. Blockade of H3 receptors is thought to
improve efficacy on nasal congestion, a symptom of allergic
rhinitis that is currently not treated by current antihistamines.
Two analogues (56a and 56b) had slightly lower H1 potency
(pA2 9.1 and 8.9, respectively, vs 9.7 for the clinical gold-
standard azelastine, and H3 potency (pKi 9.6 and 9.5, respec-
tively, vs 6.8 for azelastine). Compound 56a had longer duration of action than azelastine, low brain penetration, and low oral
bioavailability, which coupled with the predicted low clinical dose, should limit the potential of engaging CNS-related side-effects
associated with H1 or H3 antagonism.
throughout the CNS and periphery and play a critical role in
regulating inflammatory responses and CNS activity such as
wakefulness. H1 antagonists, also known as H1 blockers or
antihistamines, are the most commonly used first-line medica-
tions for allergic rhinitis.5,6 H2 receptors regulate gastric acid
secretion, and H2 antagonists are used clinically to treat excess
acid production and gastric ulceration.7 The third histamine
receptor subtype (H3) is a presynaptic autoreceptor that controls
the synthesis and release of histamine as well as other neuro-
transmitters such as acetylcholine, dopamine, GABA, glutamate,
5-HT, and noradrenaline.8 Consequently, many applications
have been proposed for H3 receptor ligands, particularly in
the CNS, where centrally acting H3 antagonists may provide
novel therapies for neurological disorders such as epilepsy,
Parkinson’s disease, Alzheimer’s disease, attention-deficit hyper-
activity disorder, sleep disturbances, cognition, schizophrenia,
and obesity.9-11 The more recently identified fourth receptor
subtype (H4) appears to be restricted to cells of the immune and
inflammatory systems, and a physiological role for this receptor
remains to be identified.12
1. INTRODUCTION
Allergic rhinitis, also known as “hay fever,” affects at least 10-
25% of the world’s population and has shown a steady increase in
prevalence during the last 40 years.1 The prevalence of allergic
rhinitis may be significantly underestimated because of misdiag-
nosis, underdiagnosis, and failure of patients to seek medical
attention. There are two types of allergic rhinitis, seasonal and
perennial. The symptoms of seasonal allergic rhinitis include at
the early stage nasal itching, irritation, and sneezing and at the
late stage rhinorrhea and nasal congestion. The symptoms of
perennial allergic rhinitis are similar, however, nasal congestion
may be more pronounced. Either type of allergic rhinitis may
also cause other symptoms such as irritation of the throat and/
or eyes, epiphora, and edema around the eyes.2 In addition to
the classical symptoms, it is now recognized that allergic
rhinitis has a significant impact on quality of life, such as social
life, sleep disturbance as a result of nasal congestion, which in
turn leads to reduced performance at work and school.3
Allergic rhinitis and other allergic conditions are associated
with the release of histamine from various cell types, but
particularly mast cells.
First-generation H1 receptor antagonists, while effective,
caused sedation attributed to their ability to blockade H1 receptors
The physiological effects of histamine are mediated by four
major G-protein-coupled receptors, termed H1, H2, H3, and H4,
which differ in their expression, signal transduction, and hista-
mine-binding characteristics.4 H1 receptors are widely distributed
Received: October 25, 2010
Published: March 07, 2011
r
2011 American Chemical Society
2183
dx.doi.org/10.1021/jm1013874 J. Med. Chem. 2011, 54, 2183–2195
|

Journal of Medicinal Chemistry
ARTICLE
Chart 1. Representative H1 Receptor Antagonists Used
Clinically
Scheme 1a
a Reagents and conditions: (a) NaH, DMF, 69%; (b) hydrazine hydrate,
EtOH, reflux, 78%; (c) HCHO, HCO2H, 100 °C, 55%.
Scheme 2a
Chart 2. Phthalazinone Template
in the CNS. Second-generation H1 receptor antagonists were
developed with reduced CNS penetration and a corresponding
reduction in the sedative side effects. Oral second-generation H1
receptor antagonists such as cetirizine, desloratadine, fexofena-
dine, loratadine, and levocetirizine (Chart 1)13 are effective in
treating all the symptoms of allergic rhinitis apart from the nasal
congestion. Hence they are often used in combination with
R-adrenergic agonist decongestants such as pseudoephedrine.
However, the use of R-adrenergic agonists is limited due to their
potential to produce hypertension, agitation, and insomnia. In
addition to the oral antihistamines, there are topical treatments
available, such as azelastine nasal spray, which has been shown to
benefit patients who have not responded adequately to lorata-
dine and fexofenadine and is significantly more efficacious than
cetirizine and levocabastine in patients with seasonal allergic
rhinitis.14,15
Histamine H3 receptors are expressed widely on both CNS
and peripheral nerve endings and mediate the inhibition of
neurotransmitter release.8,16 Activation of H3 receptor by hista-
mine modulates outflow to resistance and capacitance vessels,
causing vasodilation in rats,17 guinea pigs,18 and cats.19 In vitro
electrical stimulation of peripheral sympathetic nerves in isolated
human saphenous vein20,21 and porcine nasal mucosa22 results in
an increase in noradrenaline release and smooth muscle contrac-
tion, which can be inhibited by histamine H3 receptor agonists
such as (R)-R-methylhistamine. In addition, activation of H3
receptors in isolated human nasal turbinate mucosa inhibits
sympathetic vasoconstriction.23 The same report also described
high distribution of H3 receptors in the human nasal mucosa. It is
a Reagents and conditions: (a) 1, NaH, DMF, 100%; (b) 4 M HCl-
dioxane, 93%; (c) HCHO, HCO2H, 100 °C, 90%.
thought that activation of the H3 receptor on the presynaptic
terminals of sympathetic neurones reduces noradrenaline release
and this may contribute, together with the activation of the
postsynaptic H1 receptors, to the nasal blockage caused by
histamine release. Consistent with this hypothesis, combination
treatment of H1 and H3 antagonists have been shown to reverse
the effects of mast cell activation on nasal airway resistance and
nasal cavity volume in a cat model of nasal congestion in vivo.24-26
Further evidence for the contribution of H3 receptors to hista-
mine-induced blockage of the nasal airway in normal healthy
human volunteers was provided recently by acoustic rhino-
metry.27 Scientists at the Schering-Plough Research Institute
have published the first dual H1H3 receptor antagonist based on
the first-generation H1 antagonist chlorpheniramine and a 4-sub-
stituted imidazole with H1 Ki = 7 nM and H3 Ki = 15 nM.28 In
principle, there are two ways of targeting dual H1H3 pharmacol-
ogy, either by using a combination of two individual selective
antagonists or identifying a molecule that exhibits antagonism at
both receptors. The former approach is easier to achieve but
more expensive, as it requires the development and progression
through expensive clinical trials of two chemical entities. We have
examined both strategies, and in a recent paper we have reported
our efforts in identifying a novel, nonbrain-penetrant histamine
2184
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Scheme 3a
a Reagents and conditions: (a) diisopropyl azodicarboxylate, triphenylphosphine, THF, 80%; (b) 4 M HCl in dioxane, 90%; (c) HCHO, HCO2H,
reflux, 85%.
Scheme 4a
H3 selective antagonist with optimized pharmacokinetic pro-
perties suitable for oral administration.29 In this paper, we are
reporting our efforts in identifying an H1H3 dual antagonist sui-
table for intranasal administration for the potential treatment of
allergic rhinitis. Our efforts to identify an H1H3 dual antagonist
suitable for oral administration will be reported in due course.30
2. OPTIMIZATION OF H1 ANTAGONISM
Azelastine (Chart 1), a second-generation antihistamine,
originally developed for oral administration but now also mar-
keted for topical administration, is available as a racemic mixture.
It is about 10 times more potent than the first-generation H1
antagonist chlorpheniramine31 and is oxidatively metabolized to
an active metabolite, desmethylazelastine. Adverse effects of
azelastine nasal spray include bitter taste (20%), somnolence
(11%), and nasal burning (4%).14,15 Azelastine was resolved, and
the potency of its enantiomers was reported to be the same as
a Reagent and conditions: (a) (()-(1-methyl-3-pyrrolidinyl)methanol
(17), diisopropyl azodicarboxylate, PPh3, THF.
that of the racemic mixture,32 however, there are no reports
associating the adverse effects of azelastine with one of its
enantiomers. We established that the H3 antagonism in dual
H1H3 antagonists is tolerant to a range of structural modifica-
tions, whereas the H1 antagonism is much more sensitive to
variations. For this reason, we were interested in finding more
potent homochiral phthalazinone analogues containing N2
substituents other than the azepine found in azelastine. Although
the replacement of azelastine’s azepine group with a 2-pyrroli-
dinomethyl group was reported to have led to increased in vivo
activity in guinea pigs,33 the effect of this group and that of its
individual enantiomers on human cell lines was not reported. It
was therefore decided to synthesize a variety of phthalazinone
analogues with different N2-substituents in order to optimize the
H1 potency and then further optimize the phthalazinone C4
substituent before attempting the investigation of dual H1H3
antagonists (Chart 2).
butoxycarbonyl)amino]propyl bromide (7), followed by depro-
tection of 8 to give 9 and Eschweiler-Clarke dimethylation
(Scheme 2).
The 2-pyrrolidinomethyl enantiomers 10a and 11a were
synthesized from 1 and the appropriate Boc-protected pro-
linol using Mitsunobu conditions as shown in Scheme 3. The
Mitsunobu condensation of 1 with (S)-(-)-2-(hydroxymethyl)-
1-methylpyrrolidine was reported to give a mixture of the
expected product and a rearrangement product due to formation
of an aziridinium intermediate, which then ring-expanded.36
Because this was due to the nucleophilicity of the pyrrolidine
nitrogen, we envisaged using the Boc-protected prolinols, where
the nucleophilicity of the nitrogen is reduced by the butoxycar-
bonyl group. The Mitsunobu couplings proceeded under these
modified conditions in high yields and without any rearrange-
ment to give 12 and 13, respectively. Cleavage of the Boc
protecting group with hydrogen chloride in dioxane gave the
secondary amines 14 and 15, and finally Eschweiler-Clarke
methylation gave tertiary amines 10a and 11a.
2.1. Chemistry. The phthalazinone core (1)34 was alkylated
with N-(2-bromoethyl)phthalimide (2) in the presence of so-
dium hydride in DMF to give 3, followed by deprotection
with hydrazine hydrate to give 4 and Eschweiler-Clarke alkyla-
tion to provide the required dimethylaminoethyl derivative 535
(Scheme 1). The homologous N,N-dimethylaminopropyl deri-
vative 6 was prepared in a similar way from 1 and 3-[(tert-
The regioisomeric 3-pyrrolidinomethyl analogue 16 was pre-
pared by Mitsunobu alkylation of 1 with (1-methyl-3-pyrrolidinyl)
2185
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Scheme 5a
a Reagents and conditions: (a) allylamine, 87 °C, 60%; (b) BOC2O, Et3N, DCM, 91%; (c) Grubb’s I catalyst, DCM, 40%; (d) H2, PtO2, EtOH, 100%;
(e) 1, di-tert-butyl azodicarboxylate, PPh3, THF, 99%; (f) TFA, DCM, 82%; (g) HCHO, HCO2H, 80 °C, 54%.
under Mitsunobu conditions to provide 35. Deprotection under
acidic conditions provided 36, which was then converted to the key
intermediate 37 by Eschweiler-Clarke methylation. The required
products 10 were obtained in one simple step by Negishi coupling,
following treatment with a variety of organozinc reagents.
Table 1. Phthalazinone C4-Substituents
R =
R =
a
b
c
d
e
f
4-Cl-C6H4CH2-
j
3-F-C6H4CH2-
4-MeO-C6H4CH2-
4-HO-C6H4CH2-
4-EtO-C6H4CH2-
4-F-C6H4CH2-
k
l
3-Cl-C6H4CH2-
3-MeO-C6H4CH2-
3-Me-C6H4CH2-
3,4-F-C6H4CH2-
3,4-MeO-C6H4CH2-
c-C6H11CH2-
2.2. Results and Discussion. The affinity of compounds was
evaluated in vitro at recombinant human histamine H1 receptor
in intact CHO cells by means of plate-based calcium imaging.
Inhibition of agonist-induced cellular calcium mobilization was
monitored using the calcium sensing dye Fluo-4AM in a FLIPR
instrument.38 The compounds were also evaluated in mem-
branes from CHO cells transfected with cloned human histamine
H3 receptors for their ability to reduce histamine stimulated
GTP-γ-S binding as determined by scintillation proximity detec-
tion. The adrenergic R1A and R1B receptor affinity of the test
compounds was assessed in intact Ratl fibroblast cells by means
of plate-based calcium imaging, and the data from all the above
screens are summarized in Table 2. Azelastine was used as a
standard, with affinities (pKi) for H1 and H3 8.9 and 6.8,
respectively. The effect of the phthalazinone N2-basic amino
group substituent (azepine, piperidine, pyrrolidine, or alkyl
chain) was first examined, and the groups that had affinities for
the H1 receptor higher than that of azelastine were the two
enantiomers of 2-pyrrolidinomethyl derivatives (10a and 11a)
and the simple N,N-dimethylaminoethyl derivative (5). It was
apparent that a two-carbon chain between the point of attach-
ment on the phthalazinone to the basic nitrogen (5, 10) were
preferred over the three-carbon analogues (6, 16, 19), whereas
the piperidino analogue (18), being of intermediate length,
showed similar affinity to azelastine. This is rationalized by
comparing azelastine’s binding in the H1 homology model. This
model was based on the bovine rhodopsin structure and was
constructed well before the recently published X-ray crystal
structures of a number of 7TM receptors. The basic nitrogen
of azelastine was tethered to Asp-107 of the TM3,39 allowing the
bulk of the ligand to extend into the large pocket formed by TM3,
TM4, TM5, and TM6 (Figure 1). Two Tyr side-chains from the
opposing TM3 and TM6 were observed to interact with each
other, thereby forming a channel between the ligand binding
pocket and Asp-107. Both Tyr hydroxyl groups can form
hydrogen bonds to the phthalazinone carbonyl in some of the
low-energy conformations. The chlorobenzyl group reached
between three aromatic rings of TM5 and TM6, thereby blocking
m
n
o
p
q
r
4-Me-C6H4CH2-
4-t-Bu-C6H4CH2-
4-MeO2C-C6H4CH2-
4-HO2C-C6H4CH2-
g
h
i
C6H5CH2CH2-
CH3CH2CH2CH2CH2-
methanol (17) without any rearrangements occurring, whereas the
piperidine 1833 and pyrrolidinoethyl derivative 1934 were synthe-
sized according to published procedures (Scheme 4).
The azelastine regioisomer 20 was prepared by the route
outlined in Scheme 5. Chlorohydrine 2137 was heated in excess
allyl amine to give 22, which was converted to the Boc-protected
amine 23. Ring-closing metathesis of 23 with Grubbs I catalyst
furnished the 5,6-olefin 24 as the major product, accompanied by
the regioisomeric 6,7-olefin. Hydrogenation of 24 (and its regio-
isomer) provided 25, which was then coupled to 1 using the
Mitsunobu reaction. The resulting phthalazinone 26 was depro-
tected to azepine 27 and then reductively methylated to azelas-
tine regioisomer 20.
The C4-phthalazinone substituents (Table 1) were intro-
duced in two ways. The first route was a linear route and is
outlined in Scheme 6. Reaction between phthalic anhydride
(28) and 4-methoxyphenylacetic acid (29) in the presence of
sodium acetate gave 30, which was treated with hydrazine to
provide the phthalazinone 31. Alkylation with N-Boc-D-proli-
nol using Mitsunobu conditions gave 32, which was depro-
tected with TFA to 33 and then converted to the N-methyl
derivative 10b using Eschweiler-Clarke conditions. Reaction
of 10b with boron tribromide gave phenol 10c, which was
converted to the ethyl ether 10d using the Mitsunobu reaction
(Scheme 6).
The second route (Scheme 7) was a much more versatile route,
suitable for array chemistry and allowing late-stage derivatization at
C4. It started from the commercially available 4-chloro-1(2H)-
phthalazinone (34), which was alkylated with N-Boc-D-prolinol
2186
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Scheme 6a
a Reagents and conditions: (a) NaOAc, 240 °C, 75%; (b) hydrazine sulfate, NaOH, EtOH, 95 °C, 91%; (c) N-Boc-D-prolinol, PPh3, di-tert-butyl
azodicarboxylate, THF, 91%; (d) TFA, DCM, 100%; (e) HCHO, HCO2H, 95 °C, 80%; (f) BBr3, DCM, 5 °C, 88%; (g) EtOH, PPh3, di-tert-butyl
azodicarboxylate, THF, 20 °C, 40%.
Scheme 7a
a Reagents and conditions: (a) N-Boc-D-prolinol, PPh3, di-tert-butyl azodicarboxylate, THF, 100%; (b) 4 M HCl in dioxane, 66%; (c) HCHO, HCO2H,
80 °C, 82%; (d) organozinc chloride or bromide, Pd[(PPh3)]4, THF, 80 °C.
the Trp residue in TM6 implied in the 7TM agonist state.40 The
edge of the phthalazinone core phenyl ring coordinated to
the backbone of TM5. The effect of larger groups between
the phthalazinone N2 group and the basic group was to push the
phthalazinone nucleus onto the backbone of TM5, however,
the combination of a flexible linker and a smaller heterocycle
in the pyrrolidinomethyl analogues, relative to the azepine in
azelastine, allowed for the phthalazinone moiety to adopt an
orientation almost identical to that of azelastine (Figure 1).
The regioisomer 20 of azelastine had lower H1 affinity than
azelastine, and this is presumed to be due to its rigid structure,
which after binding of the basic nitrogen to Asp-107 forces the
phthalazinone core in an unfavorable orientation. Because the R
enantiomer of the 2-pyrrolidinomethyl analogue (10a) had the
highest H1 affinity (9.8), this analogue was chosen for the
optimization of the phthalazinone C4 substituent. Preferred
substituents at C4 were found to be the 4-chloro- (10a),
4-methoxy- (10b), 4-hydroxy- (10c), 4-methyl- (10f), and
4-fluoro-benzyl (10e) analogues. Bulky substituents at the
para-position of the C4-benzyl substituent, such as tert-butyl-
(10g), methyl ester (10h), ethyl ether (10d), and ionic groups,
such as carboxylic acid (10i), were not well tolerated. Alkyl
substituents, such as cyclohexylmethyl (10p), phenethyl
(10q), and pentyl (10r) had affinities much lower than that
of a substituted benzyl group. The meta-substituted benzyl
analogues of the highest affinity antagonists, such as the chloro-
(10k), methyl- (10m), and methoxy- (10l), had lower affinity
than the corresponding para analogues, whereas the meta-
fluoro (10j) analogue had only slightly lower affinity. Two 3,4-
disubstituted analogues were prepared, of which the dimethoxy
analogue (10o) showed much reduced affinity, whereas the
3,4-difluoro analogue (10n) had affinity similar to that of the
4-fluoro analogue (10e), indicating that there is only limited
space available in the pocket in which the benzyl substituent
2187
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Table 2. Antagonist Affinitya of Target Compounds at the
Human H1 Receptor (Determined by Fluorescence Imaging
Plate Reader), Human H3 Receptor (Determined by a Func-
tional GTPγ[S]-Assay), and Affinity at the Human r1A, and
r
1B Receptors
compd
H1 pKi (n)
H3 pKi (n)
R
1A pKi (n)
R1B pKi (n)
azelastine 8.9 ( 0.0 (364) 6.8 ( 0.0 (56) 7.3 ( 0.0 (145) 7.3 ( 0.0 (97)
5
6
9.6 ( 0.1 (8)
8.0 ( 0.1 (6)
<6.3 (7)
6.9 ( 0.1 (4) 6.5 ( 0.1 (3)
6.8 ( 0.2 (3)
6.7 ( 0.0 (3)
6.5 ( 0.1 (3)
10a
10b
10c
10d
10e
10f
10g
10h
10i
10j
10k
10l
10m
10n
10o
10p
10q
10r
11
9.8 ( 0.1 (10) 6.4 ( 0.0 (2) 7.6 ( 0.1 (12) 7.9 ( 0.2 (9)
10.0 ( 0.0 (8) 7.1 ( 0.0 (2) 6.9 ( 0.1 (4)
6.8 ( 0.2 (4)
6.4 ( 0.1 (3)
<5.7 (3)
9.8 ( 0.1 (8)
8.3 ( 0.2 (6)
6.6 ( 0.1 (4) 6.6 ( 0.2 (3)
6.5 ( 0.2 (2) 6.1 (1)
9.5 ( 0.1 (10) <6.3 (8)
7.2 ( 0.2 (4)
8.0 ( 0.3 (3)
<5.7 (3)
<5.7 (2)
<5.7 (3)
7.1 ( 0.2 (4)
7.8 ( 0.1 (3)
<5.7 (3)
<5.7 (3)
<5.7 (3)
Figure 1. (A) One of the low energy conformations of azelastine
docked into the homology model of H1. (B) Overlay of the docked
poses of azelastine (green) and 11a.
9.7 ( 0.1 (4)
6.4 ( 0.0 (6)
6.2 ( 0.1 (6)
<5.8 (6)
<6.2 (5)
<6.2 (6)
<6.2 (4)
<6.2 (6)
9.3 ( 0.0 (6)
8.8 ( 0.1 (6)
8.7 ( 0.1 (7)
9.1 ( 0.0 (6)
9.6 ( 0.1 (6)
5.8 ( 0.0 (3)
7.1 ( 0.0 (6)
7.5 ( 0.0 (2)
6.7 ( 0.0 (6) 7.3 ( 0.3 (4)
6.4 ( 0.1 (4)
6.2 ( 0.3 (2)
<5.7 (4)
6.3 ( 0.1 (3)
6.5 ( 0.1 (3)
<5.7 (5)
<6.2 (5)
6.2 (1)
6.4 ( 0.2 (5)
5.8 ( 0.1 (2)
Chart 3. Strategy for Dual H1 H3 Antagonism on a Single
Ligand
6.6 ( 0.1 (5) 6.3 ( 0.1 (3)
<6.2 (6)
<6.2 (10)
6.7 (1)
6.7 ( 0.1 (3)
<5.7 (5)
<5.7 (3)
<5.7 (3)
6.7 (1)
<5.5 (2)
6.7 (1)
8.8 ( 0.1 (20) <6.2 (6)
9.5 ( 0.1 (10) <6.4 (6)
6.3 ( 0.1 (3)
7.4 ( 0.2 (10) 7.7 ( 0.2 (9)
6.0 ( 0.1 (2)
16
18
19
8.2 ( 0.1 (8)
9.0 ( 0.2 (10) 6.9 ( 0.1 (7) 7.8 ( 0.1 (6)
7.2 ( 0.1 (18) 7.1 ( 0.2 (5) 6.5 ( 0.1 (5)
<6.4 (8)
7.0 ( 0.1 (6)
7.1 ( 0.1 (6)
8.2 ( 0.3 (6)
6.5 (1)
20
38
39
40
8.6 ( 0.0 (4)
7.3 ( 0.1 (8)
7.8 ( 0.1 (6)
7.8 ( 0.1 (4)
<6.2 (4)
6.8 ( 0.2 (2)
6.8 ( 0.1 (2)
9.3 ( 0.1 (8) 7.0 ( 0.1 (10) 6.9 ( 0.1 (7)
8.4 ( 0.1 (2) 7.5 ( 0.1 (3)
9.6 ( 0.1 (4) 6.9 ( 0.2 (2)
7.5 ( 0.2 (3)
7.1 ( 0.1 (2)
41
54
55
56a
56b
56c
57
7.7 ( 0.1 (13) 8.7 ( 0.2 (8) 6.8 ( 0.1 (10) 7.0 ( 0.1 (8)
of attachment of the H3 fragment. Furthermore, our early
investigations with a nonphthalazinone H1 fragment indicated
that the human H1 antagonism of dual H1H3 antagonists was
found to be very sensitive to the amino group of the H3 fragment.
For this reason, the homopiperidinopropyloxyphenyl group,
which was found to be one of the higher affinity H3 fragments
in our ketopiperazine selective H3 antagonists,29 was chosen for
attachment onto the basic nitrogen of the optimized H1 phtha-
lazinones 10a-c and 5 identified above.
7.4 ( 0.0 (6)
7.4 ( 0.1 (4)
9.4 ( 0.1 (6) 6.5 ( 0.1 (4)
8.1 ( 0.1 (4) 5.8 (1)
6.5 ( 0.2 (3)
<5.7 (2)
8.0 ( 0.1 (36) 9.6 ( 0.0 (33) 7.4 ( 0.0 (17) 7.5 ( 0.1 (14)
8.1 ( 0.1 (8)
7.9 ( 0.1 (8)
7.9 ( 0.1 (6)
7.6 ( 0.1 (6)
9.5 ( 0.0 (8) 7.2 ( 0.1 (5)
9.4 ( 0.1 (6) 7.0 ( 0.0 (3)
9.4 ( 0.0 (6) 7.6 ( 0.2 (3)
9.6 ( 0.1 (6) 7.5 ( 0.2 (3)
7.3 ( 0.1 (5)
6.9 ( 0.0 (3)
7.3 ( 0.1 (4)
7.4 ( 0.0 (3)
58
a Mean ( SEM (where applicable) of estimated functional pKi. For n < 3
SEM is the SD. n = number of experiments.
3.1. Chemistry. A series of azelastine derivatives 38-4141
incorporating a linker group consisting of one to four methylenes
to the H3 fragment was prepared by alkylation of des-methyl
azelastine (42) and an appropriate electrophile, such as 43
(Chart 4). The mesylate 43 was prepared by the method outlined
in Scheme 8. Thus, reaction of 4-iodophenol (45) with 1-bromo-
3-chloropropane gave chloride 46, which was treated with
hexamethyleneimine to provide amine 47. Sonogashira coupling
of 47 with 3-butyn-1-ol yielded 48, which was converted to the
alcohol 49 by catalytic hydrogenation and then treated with
methanesulfonyl chloride to give the mesylate 43.
The homologue of 43, mesylate 44, was prepared by the
method outlined in Scheme 9. Thus, Mitsunobu alkylation of
phenol 5042 with alcohol 5143 provided 52, which was reduced
with lithium aluminum hydride to give alcohol 53 and, in turn,
converted to mesylate 44. Alkylating agents 43 and 44 were
found to deteriorate with time at room temperature, so they were
stored at -20 °C or made freshly before use.
Alkylation of 4 with mesylate 43 gave 54, which was then
convertedtothetertiaryamine55 by Eschweiler-Clarke alkylation
(Scheme 10).
binds. The highest affinity analogues 10a-c, 10e, and 10f were
also found to be selective for H1 against H3, R1A and R1B,
however, 10f had reduced selectivity against R1A and R1B.
3. DUAL H1H3 ANTAGONISM
Having established the optimal C4 and N2 phthalazinone
substituents for maximal antagonism at the human H1 receptor,
attention was focused on identifying dual human H1H3 antagon-
ism in a single ligand. Examination of the H1 homology model
indicated the existence of a narrow channel between TM2, TM3,
TM6, and TM7, extending parallel to the helix axes deeper into
the 7TM bundle, which may be reached from the basic center of
azelastine (Figure 1). From our preliminary investigations with
azelastine analogues, it was found that high H1 antagonism was
retained with a large number of substituents on its basic nitrogen.
It was envisioned that introduction of fragments associated with
H3 antagonism, such as a phenoxypropylamino group, would
bring about dual H1H3 antagonism (Chart 3).
It was necessary to investigate first the optimal chain-length of
the linker group between azelastine’s basic nitrogen and the point
2188
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Chart 4. Variation of the Linking Group between the H1 and H3 Fragments a
aReagents and conditions: (a) Alkyl mesylate, NaHCO3, MeCN, reflux.
Scheme 8a
a Reagents and conditions: (a) 1-bromo-3-chloropropane, K2CO3, 2-butanone, reflux, 84%; (b) hexamethyleneimine, NaI, 2-butanone, 80 °C, 83%; (c)
3-butyn-1-ol, Et3N, (PPh3)2PdCl2, CuI, THF, 20 °C, 77%; (d) H2, 10% Pd/C, HCl, EtOH, MeOH, 88%; (e) MsCl, iPr2NEt, DCM, 20 °C, 94%.
Scheme 9a
a Reagents and conditions: (a) PPh3, di-tert-butyl azodicaroxylate, THF, 69%; (b) LiAlH4, THF, 68%; (c) MsCl, iPr2NEt, DCM, 20 °C, 100%.
The chloro- and methoxy- analogues (56a and 56b, res-
pectively) were prepared by alkylation of the appropriate amine
(14 and 33, respectively) with the mesylate 43, whereas the
hydroxy- analogue (56c) was prepared from 56b by treatment
with boron tribromide. The homologue of 56a (57) was
prepared by reacting 14 with mesylate 44 (Scheme 11).
A second, nonconvergent route was used for the synthesis of 56a
and its enantiomer 58. This route, outlined in Scheme 12, was not
suitable for array chemistry, however, it was more suitable for larger
scale work. All the intermediates in this route (59-63) were either
neutral or monobasic compounds and therefore more readily
purified by chromatography than dibasic compounds.
group linked to the H3 fragment with an alkyl-chain linker of one
to four methylenes (compounds 38-41) in order to determine
the chain-length requirement for the linker. In the H1 homology
model, the H3 fragment of the bivalent ligands could be threaded
into a narrow channel between TM2, TM3, TM6, and TM7.
While the longer-chain analogues fit well into the available space,
molecule 38 additionally formed an H-bond with the Asp-73 side
chain in TM2 (Figure 2). The compounds were not docked into
the H3 homology model where this narrow channel also existed,
albeit with a different shape. The qualitative modeling results
were confirmed by the low nanomolar affinity of 38-41 at the
H3 receptor and high nanomolar affinity at H1. However, the
shorter-chain analogue (38) was less active than the longer
analogues, which means that the additional H bond to Asp-73
3.2. Results and Discussion. The first dibasic, bivalent ligands
to be tested were the analogues bearing the azelastine H1 azepine
2189
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Scheme 10a
a Reagents and conditions: (a) 43, NaHCO3, MeCN, reflux, 21%; (b) HCHO, HCO2H, 100 °C, 66%.
Scheme 11a
a Reagents and conditions: (a) 43, NaHCO3, MeCN, reflux, 35%; (b) BBr3, DCM, 5 °C, 14%; (c) 44, NaHCO3, MeCN, reflux, 21%.
was either not formed or the H1 homology model did not reflect
the observed SAR.
progressed. The analogues 56a-c and 57 had high H3-receptor
affinity (pKi values between 9.4 and 9.6), which is almost 3 orders
of magnitude greater than azelastine (H3 pKi 6.8). However, their
H1-receptor functional pKi values were about 8, which repre-
sented a drop of 2 orders of magnitude by comparison with their
monovalent analogues. Despite the reduction in H1-receptor
affinity, compounds 56a-c and 57 are to our knowledge among the
most potent bivalent H1H3 antagonists known.28 The R-enantiomer
56a had greater H1 affinity than the S-isomer 58, as was the case
with the monovalent analogues (cf 10a with 11).
Compounds 38 and 40 were more potent at H3 (about 100-
fold) and were not progressed; the data is shown in Table 2.
Compounds 39 and 41 had a more acceptable difference of about
10-fold, however, 41 was more selective than 39 for H1 over R1A
and R1B, hence the four methyl linker was chosen for optimizing
the H1 fragment. Analogues 54 and 55, bearing flexible chain
linkers, had greatly reduced H1 affinity by comparison to the
monovalent ligand 5 (Table 2) and were therefore not
2190
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Scheme 12a
a Reagents and conditions: (a) MsCl, Et3N, Et2O, 0 °C, 100%; (b) 14, K2CO3, 2-butanone, reflux, 24 h, 60%; (c) BBr3, DCM, -60 °C, 93%; (d)
1-bromo-3-chloropropane, K2CO3, 2-butanone, reflux, 18 h, 85%; (e) hexamethyleneimine, KI, K2CO3, 2-butanone, reflux, 41 h, 48%.
Table 3. Antagonist pA2 Affinity of 56a-c and 57 at the
Human H1 Receptor, Determined by Fluorescence Imaging
Plate Reader and in Vitro Duration
compd
pA2 ( SEMa
n
wash-out
56a
9.1 ( 0.1
8.9 ( 0.1
8.4 ( 0.1
8.9 ( 0.1
9.7 ( 0.1
11
10
8
slower
56b
slower
56c
slower
57
6
no difference
reference
azelastine
19
a All pA2 values taken from curve shifts generated at 30 min incubation
time and with 100 nM antagonist.
Figure 2. One of the low-energy conformations of the extended
azelastine ligand 40 docked into the homology model of H1.
versus time plot. Results were statistically analyzed and related to
azelastine in the same assay and expressed as slower, no-difference,
or faster wash-out than azelastine, with slower wash-out equating to
longer duration of action. Duration of action for bivalent antago-
nists 56a-c and 57 is summarized Table 3. Antagonists 56a-c
were all found to be longer-acting than azelastine, however, 57 had
the same duration as azelastine.
The specificity profile of compounds 56a and 56b was
evaluated by comparison to that of the clinical gold-standard,
azelastine, and all three compounds were found to have a similar
profile against R1A and R1B (Table 2). In addition, 56a had low
affinity for the human H2 and H4 receptors (pIC50 5.0). How-
ever, 56a showed significant affinity (pIC50 7.3) for the hERG
channel, having a very similar affinity to azelastine (pIC50 7.0) in
the hERG dofetilide binding assay. Further assessment of the
hERG effects of 56a in an electrophysiology patch clamp
confirmed the dofetilide displacement data. However, subse-
quent evaluation in rabbit hearts in vitro (SCREENIT model)
suggested that the risk of 56a-induced QT prolongation is low at
therapeutically relevant concentrations.
In addition to the assays reported in Section 2.2, a more precise,
lower throughput, modified version of the human H1 FLIPR assay
was run, which provided apparent pA2 values. Antagonist pA2
values were determined by generating histamine concentration-
response curves either in the absence or presence of a single
concentration of antagonist (100 nM) at 30 min incubation. The
histamine concentration-response curves were analyzed using a
four-parameter logistic equation to determine the midpoint (EC50)
of the curve. Antagonist pA2 values were calculated using the
equation pA2 = log(DR - 1) - log[B] where DR, the dose ratio, is
the EC50 in the presence of antagonist divided by the EC50 for the
control curve, and [B] is the molar concentration of the antagonist
tested. The data for dual antagonists 56a-c and 57 are summarized
in Table 3 and compared to azelastine. Compound 56a was the
most potent H1 antagonist (pA2 9.1). Duration of action in vitro
was determined in the FLIPR assay by incubation of adherent
CHO cells with antagonist for 30 min, followed by washing, and
then by repeat histamine challenges at intervals of 90 and 270 min
at 37 °C. Agonist dose ratios were converted to receptor occu-
pancies, which were plotted against time. A measure of duration
was obtained from the gradient of the percent receptor occupancy
The in vitro rate of metabolism of 56a in human liver micro-
somes was moderate; the rate in human hepatocytes, however,
was low. The rate of in vitro hepatic metabolism was generally
2191
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
Table 4. In Vivo Pharmacokinetic Profile of 56a and Azelastine in Male CD Rats and Male Beagle Dogs
compound dose (mg/kg) (route)
species
Cl (mL/min/kg) [range] % LBF
Vss (L/kg) [range]
t1/2 (h) [range]
F (%) [range]
56a 3.0 (po)/1.0 (iv)
rat (n = 7)
dog (n = 3)
rat (n = 3)
dog (n = 1)
30 [16-49] 35%
27 [23-34] 87%
35 [28-40] 41%
22
3.9 [2.0-4.6]
22 [5.9-51]
1.8 [1.6-2.1]
4.9
2.2 [1.6-3.5]
9.4 [3.4-20]
1.0 [1.0-1.1]
3.2
4 [1-10]
19 [17-23]
7 [7-7]
46
56a 3.0 (po)/1.0 (iv)
azelastine 3.0 (po)/1.0 (iv)
azelastine 2.0 (po)/1.0 (iv)
moderate for rat and dog and was in agreement with its clearance
obtained in vivo. The pharmacokinetic profile of compound 56a
was evaluated in rat and dog and summarized in Table 4. In the
male CD rat, 56a had moderate in vivo clearance, and in the male
Beagle dog, the clearance was high (35% and 87% of liver blood
flow, respectively). Over a 24 h period, excretion of unchanged
56a in dog urine was <1% following both oral and intravenous
administration, suggesting that elimination via renal clearance
was low. The steady state volume of distribution (Vss) for 56a
was moderate for the rat and high for the dog (3.9 and 22 L/kg,
respectively). However, there was significant variability in the
estimatation of Vss for the dog. After intravenous administration,
56a had a moderate blood half-life in the rat and a long blood
half-life in the dog (2.2 and 9.4 h, respectively). In the dog, the
long half-life is volume driven and significant variability was
observed in line with the variability observed with Vss. Systemic
exposure following oral administration of 56a was very low in the
rat (4%) and higher in the dog (19%). Following oral adminis-
tration of 56a, low concentrations of the compound were
detected in blood sampled from the hepatic portal vein (1-6%
of the dose) due to either low absorption and/or high first-pass
metabolism. The bioavailability of azelastine in humans following
intranasal administration was reported to be 40%.13 The rat and
dog pharmacokinetic parameters for azelastine are included in
Table 4 and clearly show that 56a had twice as long as azelastine’s
half-life and that the oral bioavailability of 56a was lower than
that of azelastine’s in both species.
To avoid sedation, it is necessary to have a low brain con-
centration of H1 receptor antagonist following an intranasal
therapeutic dose. Azelastine gave a brain-blood ratio of 7.5
(range 6.2-8.6) following intravenous infusion to the male CD rat.
However, 56a gave a brain-blood ratio of 0.6 (range 0.3-1.1).
Therefore, the risk of engaging CNS related side-effects associated
with either H1 or H3 would be very low considering the clinical
dose and the low bioavailability of 56a.
Several second-generation H1 antagonists are substrates and
modulators of hepatic cytochrome P450, in particular the sub-
type CYP3A4. Azelastine inhibits mainly CYP2D6 (IC50 1 μM).
For compounds designed for oral dosing, it is essential to avoid
any drug-drug interactions that might lead to CYP450 activa-
tion and reduced safety in the clinic. 56a showed an acceptable in
vitro inhibition profile against CYP1A2, CYP2C9, and CYP2C19
(IC50 >3 μM), however, it was found to be a potent inhibitor of
CYP2D6 (IC50 0.04 μM) and CYP3A4 (IC50 0.3 μM). As its oral
bioavailability is low, and the intended clinical intranasal dose is
predicted to be low, coadministration with 56a is unlikely to
result in significant clinical drug-drug interactions. Further-
more, there was no evidence of time-dependent P450 inhibition
for the CYP3A4 and CYP2D6 isoforms.
response with a duration of up to 72 h, whereas azelastine anta-
gonized the response with a duration of <24 h at 1 mg/mL. These
data are consistent with potential clinical application of a once a
day intranasal treatment for allergic rhinitis. Further pharmaco-
logical findings, such as the time-dependent increase in the
binding affinity of 56a and the in vivo inhibition of histamine-
induced nasal airway resistance in guinea pigs, will be published
elsewhere.44
Various salts of 56a were prepared, and two of these are
noteworthy: the dihydrochloride and the 1,5-naphthalene dis-
ulfonate salts. The former was a highly soluble salt (>11 mg/mL
in PBS at pH 5) and log D 2.86, whereas the latter was a crystal-
line salt with low solubility (20 μg/mL) and log D 3.0. For
comparison, the solubility of azelastine hydrochloride salt was
144 μg/mL in PBS (pH 7.4) and its log D 2.25.
4. CONCLUSION
A series of potent phthalazinone-based human H1H3 bivalent
human histamine receptor antagonists were synthesized. Com-
pounds 56a and 56b had H1 potency slightly lower than that of
the clinical gold-standard, azelastine, and H3 potency nearly 3
orders of magnitude greater than that of azelastine. In addition,
56a and 56b had longer duration of action in vitro on histamine-
induced stimulation of human recombinant H1 receptors than
azelastine. Furthermore, following intranasal dosing, 56a is a
potent inhibitor of histamine-induced nasal airway resistance in
guinea pigs in vivo, with significantly longer duration of action
than that of azelastine and efficacy at a 10 times lower concen-
tration. The in vivo pharmacokinetic characteristics of 56a were
indicative of limited systemic exposure from the swallowed
portion of an intranasal dose, which should limit the potential
for unwanted side effects. Brain penetration from iv dosing in the
rat was lower than azelastine’s, which coupled with the low
clinical dose and low bioavailability, should limit the potential for
engaging CNS related side effects associated with either H1 or H3
antagonism. CYP2D6 and CYP3A4 were inhibited by 56a, how-
ever, as both the intended clinical dose and oral bioavailability
are predicted to be low, coadministration with 56a is unlikely to
result in significant clinical drug-drug interactions. There was
also no evidence of time-dependent P450 inhibition for either
CYP3A4 or CYP2D6. On the basis of the above data, 56a was
chosen for further progression as an intranasal treatment for
allergic rhinitis.
5. EXPERIMENTAL SECTION
Organic solutions were dried over anhydrous Na2SO4 or MgSO4.
TLC was performed on Merck 0.25 mm Kieselgel 60 F254 plates.
Products were visualized under UV light and/or by staining with
aqueous KMnO4 solution. LCMS analysis was conducted on a Supelco-
sil LCABZþPLUS column (3.3 cm ꢀ 4.6 mm i.d.) eluting with 0.1%
formic acid and 0.01 M ammonium acetate in water (solvent A) and
0.05% formic acid and 5% water in acetonitrile (solvent B), using the
following elution gradient 0.0-0.7 min 0% B, 0.7-4.2 min 100% B,
In vivo pharmacology of 56a and azelastine was studied in
conscious guinea pigs in which nasal congestion was induced by
intranasal histamine instillation and measured indirectly by
plethysmography. Compound 56a dosed by intranasal instilla-
tion of 0.1 and 1 mg/mL antagonized the histamine induced
2192
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
4.2-5.3 min 0% B, 5.3-5.5 min 0% B at a flow rate of 3 mL min-1. The
mass spectra were recorded on a Fisons VG Platform spectrometer using
electrospray positive and negative mode (ESþve and ES-ve). Column
chromatography was performed on Flashmaster II. The Flashmaster II is
an automated multiuser flash chromatography system, available from
Argonaut Technologies Ltd., which utilizes disposable, normal phase,
SPE cartridges (2-100 g). Mass-directed autopreparative HPLC
(MDAP) was conducted on a Waters FractionLynx system comprising
a Waters 600 pump with extended pump heads, Waters 2700 auto-
sampler, Waters 996 diode array, and Gilson 202 fraction collector on a
10 cm ꢀ 2.54 cm i.d. ABZþ column, eluting with 0.1% formic acid in
water (solvent A) and 0.1% formic acid in acetonitrile (solvent B), using
an appropriate elution gradient over 15 min at a flow rate of 20 mL min-1
and detecting at 200-320 nm at room temperature. Mass spectra were
recorded on Micromass ZMD mass spectrometer using electrospray
positive and negative mode, alternate scans. The software used was
MeCN and applied to a SCX cartridge (20 g), preconditioned with
MeOH then MeCN, and eluting with MeCN, followed by a solution of
10% aqueous 0.88 ammonia in MeCN. The appropriate fractions were
concentrated under reduced pressure to give 56a (651 mg, 35%): LCMS
RT = 2.52 min, ESþve m/z 641 [M þ H]þ and 321/322 [M/2 þ H]þ.
1H NMR (DMSO-d6) δ 8.38 (1H, dd, J = 7.7, 1.6 Hz), 7.93 (1H, m),
7.86 (1H, m), 7.82 (1H, m), 7.30 (4H, m), 7.03 (2H, d, J = 8.5 Hz), 6.80
(2H, d, J = 8.5 Hz), 4.36 (1H, m), 4.33 (2H, s), 4.14 (1H, dd, J = 13.1, 8.0
Hz), 3.98 (2H, t, J = 6.1 Hz), 3.14 (1H, m), 3.03 (1H, dd, J = 7.8, 4.5 Hz),
2.84 (1H, m), 2.75 (6H, m), 2.50 (2H, t, J = 6.9 Hz), 2.31 (2H, m), 1.97
(2H, m), 1.82 (4H, m), 1.68 (8H, m), 1.55 (4H, m).
4-[(4-Chlorophenyl)methyl]-2-({(2R)-1-[4-(4-{[3-(hexahydro-
1H-azepin-1-yl)propyl]oxy}phenyl)butyl]-2-pyrrolidinyl}
methyl)-1(2H)-phthalazinone, (56a)-1,5-Naphthalene Disulfo-
nate Monohydrate Salt. Free base 56a (400 mg, 0.62 mmol) was
dissolved in methanol (4.44 mL). A solution of 1,5-naphthalene disul-
fonic acid (232 mg, 0.80 mmol) in methanol (1 mL) was added, and the
resulting gummy solution was heated. Small amounts of solid began to
form, and on cooling a solid precipitated. The slurry was stirred at room
temperature for approximately 1 h, and methanol (2 mL) was added to
mobilize the slurry, which was heated and cooled again and left to stir for
a further hour. The solid was isolated by filtration and dried in vacuo at
40 °C to give 56a-1,5-naphthalenedisulfonate salt (465 mg, 73%):
1
MassLynx 3.5 with OpenLynx and FractionLynx options. H NMR
spectra were recorded at 400 MHz. The chemical shifts are expressed in
ppm relative to tetramethylsilane. High resolution positive ion mass
spectra were acquired on a Micromass Q-Tof 2 hybrid quadrupole time-
of-flight mass spectrometer. Optical rotations were measured with an
Optical Activity AA100 digital polarimeter. Analytical chiral HPLC was
conducted on Chiralpak column (250 mm ꢀ 4.6 mm) eluting with 15%
EtOH-heptane for 30 min at room temperature, flow rate 1 mL min-1
,
20
1
mp (DSC) 234-240 °C; [R]D þ8.0 (c 1.05 in DMSO). H NMR
(DMSO-d6) δ 9.10 (2H, br), 8.86 (2H, d, J = 9 Hz), 8.31 (1H, br d,
J = 7 Hz), 8.00-7.84 (5H, m), 7.43-7.33 (6H, m), 7.07 (2H, d,
J = 9 Hz), 6.82 (2H, m), 4.55 (2H, d, J = 5 Hz), 4.31 (2H, m), 3.96
(2H, t, J = 6 Hz), 3.85 (1H, m), 3.62 (1H, m), 3.53-3.27 (9H, m),
3.26-3.19 (2H, m), 3.19-3.03 (4H, m), 2.22-1.43 (18H, m). Found:
injection volume 15 μL, detecting at 215 nm. The purity of all
compounds screened in the biological assays was examined by LCMS
analysis and was found to be g95%, unless otherwise specified.
The purity of crystalline salts was additionally assessed by elemental
microanalysis.
C, 61.8; H, 6.2; N, 5.8; S, 6.8; Cl, 3.6. C39H49ClN4O2 C10H8O6S2
requires C, 62.1; H, 6.3; N, 5.9; S, 6.8; Cl, 3.7%
4-[(4-Chlorophenyl)methyl]-2-{2-[[4-(4-{[3-(hexahydro-1H-
azepin-1-yl)propyl]oxy}phenyl)butyl](methyl)amino]ethyl}-1
(2H)-phthalazinone (55). A suspension of 4-[(4-chlorophenyl)
methyl]-2-(2-{[4-(4-{[3-(hexahydro-1H-azepin-1-yl)propyl]oxy}phenyl)
butyl]amino}ethyl)-1(2H)-phthalazinone (54) (16 mg, 0.027 mmol) in
formaldehyde (37 wt % in water, 2 mL) and formic acid (0.20 mL) was
heated at 100 °C with stirring for 40 min. After cooling, the mixture was
concentrated in vacuo. The residue was then heated on a steam bath,
under high vacuum for 2 h, to give 55 (11 mg, 66%) without further
purification: LCMS RT = 2.47 min, 95%, ESþve m/z 615 [M þ H]þ
and 308/309 [M/2 þ H]þ. 1H NMR δ (CD3OD) 8.37-8.33 (1H, m),
7.90-7.85 (1H, m), 7.84-7.76 (2H, m), 7.28 (2H, d, J = 8 Hz), 7.23
(2H, d, J = 8 Hz), 6.91 (2H, d, J = 8.5 Hz), 6.72 (2H, d, J = 8.5 Hz), 4.36
(2H, t, J = 6.5 Hz), 4.29 (2H, s), 3.94 (2H, t, J = 6 Hz), 2.85 (2H, t,
J = 6.5 Hz), 2.83-2.73 (6H, m), 2.45-2.38 (4H, m), 2.30 (3H, s),
2.01-1.92 (2H, m), 1.76-1.60 (8H, m), 1.48-1.38 (4H, m). HRMS
ESþve m/z: calcd for C37H48ClN4O2, 615.3466; found, 615.3489.
4-[(4-Chlorophenyl)methyl]-2-({(2R)-1-[4-(4-{[3-(hexahydro-
1H-azepin-1-yl)propyl]oxy}phenyl)butyl]-2-pyrrolidinyl}
methyl)-1(2H)-phthalazinone (56a) Free Base. A mixture of 14
(1.017 g, 2.87 mmol), 43 (1.115 g, 2.91 mmol), and sodium bicarbonate
(474 mg, 5.64 mmol) in MeCN (50 mL) was heated at 80 °C with
stirring for 5 days under a nitrogen atmosphere. The cooled reaction
mixture was partitioned between water and EtOAc. The aqueous layer
was washed with further EtOAc (ꢀ2). The combined organic extracts
were dried (MgSO4) and concentrated in vacuo. The residue (1.35 g)
was dissolved in DMF-TFA (2:1, 15 mL) and purified by preparative
HPLC, using a Kromasil C8 column (28 cm ꢀ 5 cm), eluting with a
gradient of 5-45% (MeCN containing 0.25% TFA)-(0.25% TFA in
water) over 40 min, followed by holding the final solvent ratio for a
further 15 min. The relevant fractions were combined and concen-
trated in vacuo to leave an aqueous solution. This was applied to an
Amberchrom CG-161 M column (25 cm ꢀ 2.5 cm), and the column was
washed with water to remove excess TFA and eluted with MeCN to
afford the product as the trifluoroacetate salt. This was redissolved in
3
4-[(4-Chlorophenyl)methyl]-2-({(2R)-1-[4-(4-{[3-(hexahydro-
1H-azepin-1-yl)propyl] oxy}phenyl)butyl]-2-pyrrolidinyl}
methyl)-1(2H)-phthalazinone, (56a)-Dihydrochloride. Free
base 56a (3.85 g, 6.0 mmol) was dissolved in MeOH (100 mL), and
2 M hydrochloric acid (12 mL, 24 mmol) was added. The solution was
then evaporated under reduced pressure, and the residue was dissolved
in MeOH (50 mL) and re-evaporated. The addition of MeOH and
evaporation was repeated three times, and the residue was dried in vacuo
to give 56a 2HCl (4.3 g, 100%). 1H NMR (DMSO-d6) δ 10.60 (1H,
3
br s), 10.49 (1H, br s), 8.30 (1H, dd, J = 7.5, 1.5 Hz), 7.96 (1H, d,
J = 7.5 Hz), 7.93-7.88 (1H, m), 7.89-7.84 (1H, m), 7.38 (2H, d, J = 8.5
Hz), 7.34 (2H, d, J = 8.5 Hz), 7.09 (2H, d, J = 8.5 Hz), 6.84 (2H, d, J = 8.5
Hz), 4.62 (1H, dd, J = 14, 4.5 Hz), 4.55 (1H, dd, J = 14, 7 Hz), 4.37
(1H, d, J = 16.5 Hz), 4.33 (1H, d, J = 16.5 Hz), 4.00 (2H, t, J = 6 Hz),
3.85-3.77 (1H, m), 3.64-3.55 (1H, m), 3.46-3.31 (3H, m), 3.22-3.15
(2H, m), 3.14-3.02 (4H, m), 2.53-2.47 (2H, m), 2.23-2.07 (4H, m),
1.99-1.49 (14H, m).
2-({(2R)-1-[4-(4-{[3-(Hexahydro-1H-azepin-1-yl)propyl]oxy}
phenyl)butyl]-2-pyrrolidinyl}methyl)-4-{[4-(methyloxy)phenyl]
methyl}-1(2H)-phthalazinone (56b) diformate. Was prepared by a
procedure similar to that described for 56a substituting 14 with 33 LCMS
RT = 2.37 min, 95%, ESþve m/z 637 (M þ H)þ, 319 (M/2 þ H)þ.
1H NMR δ (CD3OD) 8.48 (2H, br s), 8.40-8.35 (1H, m), 7.98 (1H, m),
7.90-7.80 (2H, m), 7.22 (2H, d, J= 9 Hz), 7.06 (2H, d, J= 9 Hz), 6.86-6.80
(4H, m), 4.58 (2H, m), 4.28 (2H, s), 4.03 (2H, t, J = 6 Hz), 3.88-3.80 (1H,
m), 3.72 (3H, s), 3.66-3.58 (1H, m), 3.40-3.32 (5H, m), 3.11-2.95 (2H,
m), 2.53 (2H, t, J = 7 Hz), 2.30-2.16 (4H, m), 2.11-1.88 (8H, m), 1.78-
1.57 (8H, m). HRMS ESþve m/z: calcd for C40H53N4O3, 637.4118; found,
637.4125
2-({(2R)-1-[4-(4-{[3-(Hexahydro-1H-azepin-1-yl)propyl]oxy}
phenyl)butyl]-2-pyrrolidinyl}methyl)-4-[(4-hydroxyphenyl)
methyl]-1(2H)-phthalazinone (56c). A solution of 56b diformate
(100 mg, 0.13 mmol) in DCM (10 mL) was cooled in an ice-bath under
2193
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
nitrogen and then treated with boron tribromide solution in hexanes
(1M, 0.3 mL), followed by another portion (0.3 mL) after 2 h. The
mixture stood at room temperature for a total of 2 days and 4 h, and then
the solvents were removed under reduced pressure. The residue was
dissolved in MeOH-DMSO (1:1, 2 mL) and purified by MDAP to give
18 mg, which was repurified by MDAP to give 56c (13 mg, 14%) LCMS
RT = 2.37 min, 92%, ESþve m/z 623 (M þ H)þ, 312 (M/2 þ H)þ. 1H
NMR δ (CD3OD) 8.42 (2H, br s), 8.40-8.37 (1H, m), 8.00 (1H, m),
7.90-7.81 (2H, m), 7.12 (2H, d, J = 8 Hz), 7.07 (2H, d, J = 9 Hz), 6.82
(2H, d, J = 9 Hz), 6.70 (2H, d, J = 8 Hz), 4.60 (2H, m), 4.24 (2H, s), 4.02
(2H, t, J = 6 Hz), 3.45-3.30 (5H, m), 3.20-3.00 (2H, m), 3.11-2.95
(2H, m), 2.54 (2H, t, J = 7 Hz), 2.33-2.16 (4H, m), 2.13-1.88 (8H, m),
1.78-1.58 (8H, m).
55%). LCMS and 1H NMR same as for 51a. Anal. Chiral HPLC
RT = 7.9 min, 2.1% (R-isomer) and 10.38 min, 97.9% (S-isomer)
’ ASSOCIATED CONTENT
S
Supporting Information. Preparative details and spec-
b
troscopic data for compounds 3-5, 8, 9, 6, 14, 10a, 11a, 16, 18,
22-27, 20, 30-33, 10b-10d, 35-37, 10e-r, 38-41, 52, 53,
44, 46, 47, 48, 49, 43, 54, 60-63, (S)-enantiomers of 61-63,
biological screens, table of LCMS purity and retention times, and
microanalytical data on 56a-1,5-naphthalenedisulfonate salt,
molecular modeling. This material is available free of charge via
the Internet at http://pubs.acs.org.
4-[(4-Chlorophenyl)methyl]-2-({(2R)-1-[5-(4-{[3-(hexahydro-
1H-azepin-1-yl)propyl]oxy}phenyl)pentyl]-2-pyrrolidinyl}
methyl)-1(2H)-phthalazinone (57). 5-(4-{[3-(Hexahydro-1H-
azepin-1-yl)propyl]oxy}phenyl)pentyl methanesulfonate (44) (142 mg,
0.36 mmol) was stirred with 14 (126 mg, 0.36 mmol) in MeCN (10 mL)
at 80 °C under nitrogen containing sodium bicarbonate (60 mg, 0.72 mmol)
for six days when reaction appeared almost complete. The mixture was
evaporated to dryness, and the residue in DCM was loaded onto a 20 g
silica cartridge which had been preconditioned with DCM. The cartridge
was eluted with DCM-EtOH-0.88 aq ammonia solution (200:8:1)
and then (100:8:1) to give impure product in three fractions (52, 74, and
25 mg). The 74 and 25 mg portions were combined and loaded onto two
20 cm ꢀ 20 cm silica preparative plates (1 mm thick layer), which were
developed twice in DCM-EtOH-0.88 aq ammonia solution (100:8:1).
The main band was taken off and eluted to give 57 (50 mg, 21%). LCMS
RT = 2.58 min, 100%, ESþve m/z 655 [M þ H]þ, ESþve m/z 329
[1/2M þ H]þ. 1H NMR (CDCl3) 8.48-8.43 (1H, m), 7.72-7.62 (3H,
m), 7.26 (2H, d, J = 8 Hz), 7.20 (2H, d, J = 8 Hz), 7.07 (2H, d, J = 8 Hz),
6.81 (2H, d, J = 8 Hz), 4.43 (1H, dd, J = 4, 13 Hz), 4.25 (2H, s), 4.07 (1H,
dd, J = 9, 13 Hz), 4.00 (2H, t, J = 6 Hz), 3.22-3.16 (3H, m), 3.02-2.93
(2H, m), 2.93-2.73 (4H, m), 2.52 (2H, t, J = 7.5 Hz), 2.40-2.29 (1H,
m), 2.27-2.18 (1H, m), 2.09-1.98 (2H, m), 1.90-1.68 (8H, m),
1.67-1.53(8H, m), 1.42-1.25 (2H, m). HRMS ESþve m/z: calcd for
C40H52ClN4O2, 655.3779; found, 655.3792.
4-[(4-Chlorophenyl)methyl]-2-({(2R)-1-[4-(4-{[3-(hexahydro-
1H-azepin-1-yl)propyl]oxy}phenyl)butyl]-2-pyrrolidinyl}
methyl)-1(2H)-phthalazinone (56a) from 63. (Scheme 12) To a
solution of 4-[(4-chlorophenyl)methyl]-2-{[(2R)-1-(4-{4-[(3-chloropropyl)
oxy]phenyl}butyl)-2-pyrrolidinyl]methyl}-1(2H)-phthalazinone (63)
(20 g, 34.6 mmol) in 2-butanone (200 mL) under nitrogen was added
potassium iodide (11.5 g, 69.2 mmol), potassium carbonate (9.6 g,
69.2 mmol), and hexamethylene imine (7.8 mL, 69.2 mmol). The
reaction mixture was heated at reflux for 41 h. The solid was removed by
filtration and washed with 2-butanone (2 ꢀ 100 mL). The combined
filtrate and washings were evaporated in vacuo, and the residue was
dissolved in MeOH-DMSO (1:1; 30 mL). The solution was applied to
two C18 reverse phase cartridges (330 g) and eluted using a gradient of
0-50% (MeCN containing 0.05% TFA)-(water containing 0.05%
TFA) over 12 CV. The required fractions were evaporated in vacuo,
and the residue was dissolved in MeOH. The solution was applied to
four amino propyl cartridges (70 g) and eluted with MeOH. The
required fractions were evaporated in vacuo to afford 56a (10.74 g,
48%): LCMS RT = 2.67 min, ESþve m/z 641/643 [M þ H]þ. Anal.
Chiral HPLC RT = 7.9 min, 100%.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: (þ44)1438 762883. Fax: (þ44)1438 768302. E-mail:
pan.a.procopiou@gsk.com.
’ ABBREVIATIONS USED
Fluo-4AM, 4-(6-acetoxymethoxy-2,7-difluoro-3-oxo-9-xanthenyl)-
40-methyl-2,20-(ethylenedioxy)dianiline-N,N,N0,N0-tetraacetic acid
tetrakis(acetoxymethyl) ester; GTP-γ-S, guanosine-50-[γ-thio]-
triphosphate; TBTU, (O-benzotriazol-1-yl)-N,N,N0,N0-tetramethyl-
uronium tetrafluoroborate
’ REFERENCES
(1) Bousquet, J.; Van Cauwnberge, P.; Khaltaev, N. ARIA Workshop
Group, World Health Organization. Allergic rhinitis and impact on
asthma. J. Allergy Clin. Immunol. 2001, 108(Suppl 5) S147-S336.
(2) Berger, W. E. Overview of allergic rhinitis. Ann. Allergy Asthma
Immunol. 2003, 90, 7–12.
(3) Leynaert, B.; Neukirch, C.; Liard, R.; Bousquet, J.; Neukirch, F.
Quality of life in allergic rhinitis and asthma. A population-based study of
young adults. Am. J. Respir. Crit. Care Med. 2000, 162, 1391–1396.
(4) B€aumer, W.; Rossbach, K. Histamine as an immunomodulator.
JDDG, J. Germ. Soc. Derm. 2010, 8, 495–504.
(5) Simons, F. E. R. H1-antihistamines: more relevant than ever in
the treatment of allergic disorders. J. Allergy Clin. Immunol. 2003, 112
(Suppl 4), S42–S52.
(6) Gushchin, I. S. Inverse agonists of H1 receptors as promising
antiallergy agents. Pharm. Chem. J. 2010, 44, 1–6.
(7) Ganellin, C. R. In Pharmacology of Histamine Receptors; Ganellin,
C. R., Parsons, M. E., Eds.; Wright: Bristol, UK, 1982, pp 11-102.
(8) Sander, K.; Kottke, T.; Stark, H. Histamine H3 receptor antago-
nists go to clinics. Biol. Pharm. Bull. 2008, 31, 2163–2181.
(9) Leurs, R.; Blandina, P.; Tedford, C.; Timmerman, H. In The
Histamine H3 Receptor; A Target for New Drugs, 1st ed.; Leurs, R.,
Timmerman, H., Eds.; Elsevier: Amsterdam, 1998.
(10) Leurs, R.; Blandina, P.; Tedford, C.; Timmerman, H. Ther-
apeutic potential of histamine H3 receptor agonists and antagonists.
Trends Pharmacol. Sci. 1998, 19, 177–183.
(11) yaz_ewska, D.; Kieꢀc-Kononowicz, K. Recent advances in hista-
mine H3 receptor antagonists/inverse agonists. Expert Opin. Ther. Pat.
2010, 20, 1147–1169.
(12) Bhatt, H. G.; Agrawal, Y. K.; Raval, H. G.; Manna, K.; Desai,
P. R. Histamine H4 receptor: a novel therapeutic target for immune and
allergic responses. Mini-Rev. Med. Chem. 2010, 10, 1293–1308.
(13) Lehman, J. M.; Blaiss, M. S. Selecting the optimal oral anti-
histamine for patients with allergic rhinitis. Drugs 2006, 66, 2309–2319.
(14) Lee, C.; Corren, J. Review of azelastine nasal spray in the
treatment of allergic and non-allergic rhinitis. Expert Opin. Pharmacother.
2007, 8, 701–709.
4-[(4-Chlorophenyl)methyl]-2-({(2S)-1-[4-(4-{[3-(hexahydro-
1H-azepin-1-yl)propyl]oxy}phenyl)butyl]-2-pyrrolidinyl}
methyl)-1(2H)-phthalazinone (58). Compound 58 Was prepared
from 4-[(4-chlorophenyl)methyl]-2-{[(2S)-1-(4-{4-[(3-chloropropyl)
oxy]phenyl}butyl)-2-pyrrolidinyl]methyl}-1(2H)-phthalazinone (S-en-
antiomer of 63) by the procedure described above for 56a, yield (92 mg,
2194
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195

Journal of Medicinal Chemistry
ARTICLE
(15) Bernstein, J. A. Azelastine hydrochloride: a review of pharma-
cology, pharmacokinetics, clinical efficacy and tolerability. Curr. Med.
Res. Opin. 2007, 23, 2441–2452.
(34) Scheffler, G.; Engel, J.; Kutscher, B.; Sheldrick, W. S.; Bell, P.
Synthesis and X-ray structure analysis of azelastine. Arch. Pharm. 1988,
321, 205–208.
(16) Ishikawa, S.; Sperelakis, N. Histamine H3 receptor-mediated
inhibition of noradrenaline release in pig retina discs. Arch. Pharmacol.
1987, 324, 497–501.
(35) Engelbrecht, H. J.; Lenke, D.; Muller, H. Aminocarboxylic acid
derivatives having antiallergic/antiasthmatic effect and a process for their
preparation. German Patent DE1046625, 1958.
(17) Malinowska, B.; Schlicker, E. H3 receptor-mediated inhibition
of neurogenic vasopressor responses in the pithed rats. Eur. J. Pharmacol.
1991, 205, 307–310.
(36) Knaak, M.; Fleischhauer, I.; Charpentier, P.; Emig, P.; Kutscher,
B.; M€uller, A. Ring-rearrangement during the Mitsunobu alkylation of
phthalazinones and indazolols. Liebigs Ann. 1996, 1477–1482.
(37) De Camp Schuda, A.; Mazzocchi, P. H.; Fritz, G.; Morgan, T. A
Short and Efficient Synthesis of 4,5-Disubstituted-1-pentenes. Synthesis
1986, 309–312.
(18) Hey, J. A.; Del Prado, M.; Egan, R. W.; Kreuter, W.; Chapman,
R. W. Inhibition of sympathetic hypertensive responses in the guinea pig
by prejunctional H3-receptors. Br. J. Pharmacol. 1992, 107, 347–350.
(19) Hey, J. A.; Aslanian, R.; Bolser, D. C.; Chapman, R. W.; Egan,
R. W.; Rizzo, C. A.; Shih, N.-Y.; Fernandez, X.; McLeod, R. L.; West, R.;
Kreutner, W. Studies on the pharmacology of the novel histamine H3
receptor agonist Sch 50971. Arzneim. Forsch. Drug Res. 1998, 48, 881–888.
(20) Molderings, G. J.; Weissenborn, G.; Schlicker, E.; Likungu, J.;
G€othert, M. Inhibition of noradrenaline release from the sympathetic
nerves of the human saphenous vein by presynaptic histamine H3
receptors. Naunyn-Schmiedeberg's Arch. Pharmacol. 1992, 346, 46–50.
(21) Valentine, A. F.; Rizzo, C. A.; Rivelli, M. A.; Hey, J. A. Pharma-
cological characterization of histamine H3 receptors in human saphenous
vein and guinea pig ileum. Eur. J. Pharmacol. 1999, 366, 73–78.
(22) Varty, L. M.; Hey, J. A. Histamine H3 receptor activation
inhibits neurogenic sympathetic vasoconstriction in porcine nasal
mucosa. Eur. J. Pharmacol. 2002, 452, 339–345.
(38) Sullivan, E.; Tucker, E. M.; Dale, I. L. Measurement of [Ca2þ
]
using the Fluorometric Imaging Plate Reader (FLIPR). Methods Mol.
Biol. 1999, 114, 125–133.
(39) Shi, L.; Javitch, J. A. The binding site of aminergic G protein-
coupled receptors: the transmembrane segments and second extracel-
lular loop. Annu. Rev. Pharmacol. Toxicol. 2002, 42, 437–467.
(40) Lin, S. W.; Sakmar, T. P. Specific tryptophan UV-absorbance
changes are probes of the transition of rhodopsin to its active state.
Biochemistry 1996, 35, 11149–11159.
(41) Gore, P. M.; Hancock, A. P.; Hodgson, S. T.; Kindon, L. J.;
Procopiou, P. A. 2-Substituted 4-benzylphthalazinone derivatives as
histamine H1 and H3 antagonists. Patent WO 2007122156 A1 1 Nov
2007.
(42) Yi, C.-S.; Martinelli, L. C.; Blanton, C.; DeWitt, Jr. Synthesis of
N-methyl-1-oxa-5-aza[10]paracyclophane: a conformationally restricted
analog of phenoxypropylamines. J. Org. Chem. 1978, 43, 405–409.
(43) Strogryn, E. L. New synthesis of antimalarials related to
2-bromo-4,5-dimethoxy-N,N-bis(diethylaminoethyl)aniline. Terminal
nitrogen modifications. J. Med. Chem. 1970, 13, 864–866.
(44) Slack, R. J.; Russell, L. J.; Hall, D. A.; Luttmann, M. A.; Ford,
A. J.; Saunders, K. A.; Hodgson, S. T.; Connor, H. E.; Browning C.;
Clark, K. L. Pharmacological characterisation of GSK1004723, a novel,
long acting antagonist at histamine H1 and H3 receptors. Br. J. Pharmacol.,
in press.
(23) Varty, L. M.; Gustafson, E.; Laverty, M.; Hey, J. A. Activation of
Histamine H3 receptors in human nasal mucosa inhibits sympathetic
vasoconstriction. Eur. J. Pharmacol. 2004, 484, 83–89.
(24) McLeod, R. L.; Mingo, G. G.; Herczku, C.; DeGennaro-Culver,
F.; Kreutner, W.; Egan, R. W.; Hey, J. A. Combined histamine H1 and H3
receptor blockade produces nasal decongestion in an experimental
model of nasal congestion. Am. J. Rhinol. 1999, 13, 391–399.
(25) McLeod, R. L.; Rizzo, C. A.; West, R. E., Jr.; Aslanian, R.;
McCormick, K.; Bryant, M.; Hsieh, Y.; Korfmacher, W.; Mingo, G. G.;
Varty, L.; Williams, S. M.; Shih, N.-Y.; Egan, R. W.; Hey, J. A. Pharmaco-
logical characterization of the novel histamine H3-receptor antagonist
N-(3,5-dichlorophenyl)-N0-[[4-(1H-imidazol-4-ylmethyl)phenyl]-
methyl]-urea (Sch 79687). J. Pharmacol. Exp. Ther. 2003, 305, 1037–1044.
(26) Beaton, G.; Moree, W. J. The expanding role of H1 antihista-
mines: a patent survey of selective and dual activity compounds 2005-
2010. Expert Opin. Ther. Pat. 2010, 20, 1197–1218.
(27) Taylor-Clark, T.; Sodha, R.; Warner, B.; Foreman, J. Histamine
receptors that influence blockage of the normal human nasal airway. Br.
J. Pharmacol. 2005, 144, 867–874.
(28) Aslanian, R.; wa Mutahi, M.; Shih, N.-Y.; Piwinski, J. J.; West,
R.; Williams, S. M.; She, S.; Wu, R.-L; Hey, J. A. Identification of a dual
histamine H1/H3 receptor ligand based on the H1 antagonist chlorphe-
niramine. Bioorg. Med. Chem. Lett. 2003, 13, 1959–1961.
(29) Procopiou, P. A.; Ancliff, R. A.; Bamford, M. J.; Browning, C.;
Connor, H.; Davies, S.; Fogden, Y. C.; Hodgson, S. T.; Holmes, D. S.;
Looker, B. E.; Morriss, K. M. L.; Parr, C. A.; Pickup, E. A.; Sehmi, S. S.;
White, G. V.; Watts, C. J.; Wilson, D. M.; Woodrow, M. D. 4-Acyl-1-(4-
aminoalkoxyphenyl)-2-ketopiperazines as a novel class of non-brain-
penetrant histamine H3 receptor antagonists. J. Med. Chem. 2007,
50, 6706–6717.
(30) Hodgson, S. T.; unpublished results.
(31) Casale, T. B. The interaction of azelastine with human lung
histamine H1, β, and muscarinic receptor-binding sites. J. Allergy Clin.
Immunol. 1989, 83, 771–776.
(32) Fleischhauer, I.; Kutscher, B.; Engel, J.; Achterrath-Tucker-
mann, U.; Borbe, H. O.; Schmidt, J.; Szelenyi, I.; Camuglia, G. Chirality
of azelastin and flezelastine: investigation of their enantiomers. Chirality
1993, 5, 366–369.
(33) Vogelsang, D.; Scheffler, G.; Brock, N.; Lenke, D.; Vogelsang
Schafer, U. Basically substituted benzyl phthalazone derivatives, acid
salts thereof and process for the production thereof. U.S. Patent
US3813384, 1974.
2195
dx.doi.org/10.1021/jm1013874 |J. Med. Chem. 2011, 54, 2183–2195