Dibenzo[b,f][1,4]oxazepines and dibenzo[b,e]oxepines: Influence of the chlorine substitution pattern on the pharmacology at the H1R, H4R, 5-HT2AR and other selected GPCRs

Article abstract of DOI:10.1016/j.phrs.2016.09.042

Inspired by VUF6884 (7-Chloro-11-(4-methylpiperazin-1-yl)dibenzo[b,f][1,4]oxazepine), reported as a dual H₁/H₄ receptor ligand (pK_i: 8.11 (human H₁R (hH₁R)), 7.55 (human H₄R (hH₄R))), four known and 28 new oxazepine and related oxepine derivatives were synthesised and pharmacologically characterized at histamine receptors and selected aminergic GPCRs. In contrast to the oxazepine series, within the oxepine series, the new compounds showed high affinity to the hH₁R (pK_i: 6.8–8.7), but no or moderate affinity to the hH₄R (pK_i: ≤ 5.3). For one oxepine derivative (1-(2-Chloro-6,11-dihydrodibenzo[b,e]oxepin-11-yl)-4-methylpiperazine), the enantiomers were separated and the R-enantiomer was identified as the eutomer at the hH₁R (pK_i: 8.83 (R), 7.63 (S)) and the guinea-pig H₁R (gpH₁R) (pK_i: 8.82 (R), 7.41 (S)). Molecular dynamic studies suggest that the tricyclic core of the compounds is bound in a similar mode into the binding pocket, as described for doxepine in the hH₁R crystal structure. Moreover, docking studies of all oxepine derivatives at the hH₁R indicate that the oxygen and the position of the chlorine in the tricyclic core determines, if the R- or the S-enantiomer is the eutomer. For some of the oxazepines and oxepines the affinity to other aminergic GPCRs is in the same range as to hH₁R or hH₄R, thus, those compounds have to be classified as dirty drugs. However, one oxazepine derivative (3,7-Dichloro-11-(4-methylpiperazin-1-yl)dibenzo[b,f][1,4]oxazepine was identified as dual hH₁/h5-HT_2A receptor ligand (pK_i: 9.23 (hH₁R), 8.74 (h5-HT_2AR), ≤7 at other analysed GPCRs), whereas one oxepine derivative (1-(3,8-Dichloro-6,11-dihydrodibenzo[b,e]oxepin-11-yl)-4-methylpiperazine) was identified as selective hH₁R antagonist (pK_i: 8.44 (hH₁R), ≤6.7 at other analyzed GPCRs). Thus, the pharmacological results suggest that the oxazepine/oxepine moiety and additionally the chlorine substitution pattern toggles receptor selectivity and specificity.

Full text of DOI:10.1016/j.phrs.2016.09.042

Pharmacological Research 113 (2016) 610–625
Contents lists available at ScienceDirect
Pharmacological Research
journal homepage: www.elsevier.com/locate/yphrs
Dibenzo[b,f][1,4]oxazepines and dibenzo[b,e]oxepines: Influence of
the chlorine substitution pattern on the pharmacology at the H1R,
H4R, 5-HT2AR and other selected GPCRs
Franziska Naporraa, Susanne Gobledera, Hans-Joachim Wittmannb, Julia Spindlera,
Michael Bodensteinerc, Günther Bernhardtb, Harald Hübnerd, Peter Gmeinerd,
Sigurd Elza, Andrea Strasserb,∗
a Department of Pharmaceutical/Medicinal Chemistry I, Institute of Pharmacy, University of Regensburg, D-93040 Regensburg, Germany
b Department of Pharmaceutical/Medicinal Chemistry II, Institute of Pharmacy, University of Regensburg, D-93040 Regensburg, Germany
c Institute of Inorganic Chemistry, University of Regensburg, D-93040 Regensburg, Germany
d Department of Chemistry and Pharmacy, Medicinal Chemistry, Friedrich-Alexander-Universität Erlangen-Nürnberg, D-91052 Erlangen, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Inspired by VUF6884 (7-Chloro-11-(4-methylpiperazin-1-yl)dibenzo[b,f][1,4]oxazepine), reported as a
dual H1/H4 receptor ligand (pKi: 8.11 (human H1R (hH1R)), 7.55 (human H4R (hH4R))), four known and 28
new oxazepine and related oxepine derivatives were synthesised and pharmacologically characterized at
histamine receptors and selected aminergic GPCRs. In contrast to the oxazepine series, within the oxepine
series, the new compounds showed high affinity to the hH1R (pKi: 6.8–8.7), but no or moderate affinity
to the hH4R (pKi: ≤ 5.3). For one oxepine derivative (1-(2-Chloro-6,11-dihydrodibenzo[b,e]oxepin-11-
yl)-4-methylpiperazine), the enantiomers were separated and the R-enantiomer was identified as the
eutomer at the hH1R (pKi: 8.83 (R), 7.63 (S)) and the guinea-pig H1R (gpH1R) (pKi: 8.82 (R), 7.41 (S)).
Molecular dynamic studies suggest that the tricyclic core of the compounds is bound in a similar mode
into the binding pocket, as described for doxepine in the hH1R crystal structure. Moreover, docking
studies of all oxepine derivatives at the hH1R indicate that the oxygen and the position of the chlorine
in the tricyclic core determines, if the R- or the S-enantiomer is the eutomer. For some of the oxazepines
and oxepines the affinity to other aminergic GPCRs is in the same range as to hH1R or hH4R, thus, those
compounds have to be classified as dirty drugs. However, one oxazepine derivative (3,7-Dichloro-11-
(4-methylpiperazin-1-yl)dibenzo[b,f][1,4]oxazepine was identified as dual hH1/h5-HT2A receptor ligand
(pKi: 9.23 (hH1R), 8.74 (h5-HT2AR), ≤7 at other analysed GPCRs), whereas one oxepine derivative (1-(3,8-
Dichloro-6,11-dihydrodibenzo[b,e]oxepin-11-yl)-4-methylpiperazine) was identified as selective hH1R
antagonist (pKi: 8.44 (hH1R), ≤6.7 at other analyzed GPCRs). Thus, the pharmacological results suggest
that the oxazepine/oxepine moiety and additionally the chlorine substitution pattern toggles receptor
selectivity and specificity.
Received 27 May 2016
Received in revised form
28 September 2016
Accepted 29 September 2016
Available online 30 September 2016
Keywords:
Histamine H1 receptor
Histamine H4 receptor
Dual/selective H1R/H4R ligands
Dual hH1R/h5-HT2AR ligands
Oxazepine derivatives
Oxepine derivatives
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
GPCRs: the ␣-adrenergic (␣1, ␣2), the ␤-adrenergic (␤1, ␤2, ␤3), the
dopamine (D1, D2, D3, D4, D5), the histamine (H1, H2, H3, H4), the
G-protein coupled receptors (GPCR) contain seven transmem-
brane domains, which span the lipid bilayer and interact by their
intracellular part with G-proteins [1,2]. The family A, i. e. the
rhodopsin-like GPCRs, represent the largest class of the GPCRs
[3] and can be divided in several subfamilies, e.g., the aminergic
serotonin (5-HT1A, 5-HT1B, 5-HT1D, 5-HT1E, 5-HT1F, 5-HT2A, 5-HT2B,
5-HT2C, 5-HT4, 5-HT5A, 5-HT6 and 5-HT7) and the muscarinic (M1,
M2, M3, M4, M5) receptors (http://www.guidetopharmacology.org/
). Within the present study, we focus on the histamine H1 receptor
(H1R) [4] and histamine H4 receptor (H4R) [5,6]. Currently, there is a
discussion on the synergistic involvement of the H1R and the H4R in
inflammatory processes [7–9]. In this context, dual H1/H4 receptor
ligands could not only serve as valuable molecular tools in proof of
concept studies, but harbour a potential in the treatment of allergic
∗
Corresponding author.
E-mail address: andrea.strasser@chemie.uni-regensburg.de (A. Strasser).
http://dx.doi.org/10.1016/j.phrs.2016.09.042
1043-6618/© 2016 Elsevier Ltd. All rights reserved.

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
611
red) of the seven-membered cycle were observed to act as partial
agonists at the hH4R (Fig. 2). The inverse agonists shown in Fig. 2
have only low affinities (<6) or potencies (<5) at the hH4R [21].
However, within the series of partial agonists, having high affinity
to the hH1R, but low affinity to the hH4R, repositioning of the chlo-
rine in loxapine (2) resulted in VUF6884 (3), with high affinity to
the hH1R and the hH4R (Fig. 1) [22].
Doxepine (4) is a highly active antagonist at the hH1R, but
described as an inverse agonist with low potency at the hH4R
[21] (Fig. 3). Inspired by the successful synthesis of the dual H1/H4
receptor ligand (3), starting from the H1R antagonist (2), the idea
of the present study was to exchange the basic side chain of
doxepine (4) into an N-methylpiperazine moiety, which was sug-
gested to be important for affinity to hH4R for several ligands
[17,25]. Additionally different chlorine substitution patterns at the
dibenzo[b,e]oxepine derivatives were introduced (Fig. 3) with the
purpose to obtain antagonists or inverse agonists at the hH1R and
the hH4R, by analogy with VUF6884 (3) based on loxapine (2). Fur-
thermore, according to results of Smits et al. [22], the mono- and
di-chloro substitution pattern of the oxazepine derivatives was
systematically changed in order to optimize the affinity to both
receptor subtypes. In addition to 4 known oxazepine derivatives
[22] 11 new oxazepine and 17 new oxepine derivatives were syn-
thesised and pharmacologically characterized at the hH1R and the
hH4R. As such tricyclic compounds, e.g. clozapine (1), are known
as dirty drugs, sharing high affinity to several aminergic GPCRs
[20], six oxazepines and four oxepines were consequently charac-
terized at selected aminergic GPCRs in order to explore the receptor
specificity.
Fig. 1. Structures of clozapine, loxapine and VUF6884 including pharmacological
data at the hH1R and the hH4R. Pharmacological data were taken from (a) Appl et al.
[21] and (b) Smits et al. [22].
2. Materials and methods
2.1. Capillary electrophoresis
diseases [7–12]. Furthermore, dual or selective H1 and/or H4 recep-
tor ligands represent an important tool to study these receptors on
a molecular level, in order to increase the understanding of selec-
tivity and functionality. Various ligands, binding with high affinity
either to the human H1R (hH1R) or to the human H4R (hH4R) are
known [13–16]. However, due to the low homology between the
orthosteric binding site of the hH1R and the hH4R [17], it is a chal-
lenge to develop dual H1/H4 receptor ligands. Nevertheless, some
ligand classes, showing affinity to the hH1R and hH4R, e.g. quina-
zoline [18], aminopyrimidine [19] and astemizole [17] derivatives
are described.
With the aim to identify a new antagonistic/inverse agonistic
H1/H4 receptor scaffold beside the quinazoline scaffold [18], we
focused on tricyclic compounds. Clozapine (1) (Fig. 1) is a so called
dirty drug, because it shows affinity to several GPCRs [20]. The
exchange of the NH group in clozapine by an oxygen resulted in
the dibenzo[b,f][1,4]oxazepine derivatives, e.g., 2–3 (Fig. 1). Com-
pared to loxapine (2), which has high affinity to the hH1R, but not
to the hH4R [21], VUF6884 (3) shows high affinity to both recep-
tor subtypes [22]. These data suggest that the affinity profile of the
dibenzo[b,f][1,4]oxazepine derivatives at the hH4R depends on the
chlorine substitutent pattern. However, VUF6884 (3) is described
as an antagonist at the hH1R and as an agonist at the hH4R [23].
A large number of H1R antagonists, antidepressants and antipsy-
chotics were pharmacologically characterized at the hH1R and the
hH4R [8,21,24]. All analyzed compounds were neutral antagonists
at the hH1R. An analysis of the pharmacological data showed a
switch between partial and inverse agonism at the hH4R (Fig. 2):
All compounds with the basic moiety connected to the “one-atom
spacer” (Fig. 2, blue) of the seven-membered cycle were observed
to act as inverse agonists at the hH4R. By contrast, all compounds
with the basic moiety connected to the “two-atom spacer” (Fig. 2,
Enantioseparation of 16e was essentially performed as
described previously with a Biofocus 3000 capillary electrophore-
sis system (Bio-Rad, Munich, Germany) [26]. Analyses were carried
out in an uncoated fused silica capillary (MicroQuartz, Munich,
Germany) of a total length of 75 cm (effective length: 70.4 cm) and
an inner diameter of 50 ␮m. The capillary was flushed with 0.1 M
NaOH for 5 min, then with water for 2 min, and finally with run-
ning buffer for 10 min. After hydrodynamic injection of the samples
at a constant pressure-time integral (20 psi s), analysis was per-
formed at a field strength of 266.7 V/cm (positive mode: + → −).
UV detection was performed at 210 nm. For the preparation of the
running buffer, 125 mM NaH2PO4·H2O and 5 M urea were dissolved
in water, and the pH was adjusted to 2.5 by adding appropriate
amounts of 85% H3PO4. In this buffer, ␥-cyclodextrin was dissolved
to a concentration of 100 mM. The compounds 16e, (R)-16e or
(S)-16e were dissolved in MeOH/10 mM HCl (80:20 (v/v)) at a con-
centration of 10 mM. The samples were prepared in a buffered (pH
2.5) solution, containing 15 mM NaH2PO4·H2O and 0.6 M urea. The
concentrations of the analytes were 500 ␮M ((R)-16e or (S)-16e) or
1 mM (16e).
2.2. Pharmacological methods: materials
[3H]Mepyramine
(30.0 Ci/mmol)
and
[3H]histamine
(14.2 Ci/mmol) were from PerkinElmer Life Sciences (Boston,
MA) or Hartmann Analytic (Braunschweig, Germany).
[␥32P]GTP was synthesized as described previously [27]. [32P]Pi
(8500–9100 Ci/mmol orthophosphoric acid) was from PerkinElmer
Life Sciences (Boston, MA). [35S]GTP␥S (1000 Ci/mmol) was from
Hartmann Analytic (Braunschweig, Germany) or PerkinElmer
Life Sciences (Boston, MA). Unlabeled GTP␥S was from Roche

612
F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
Fig. 2. Maximal agonistic effects (Emax) of selected tricyclic compounds, comprising a “central” seven-membered ring, at the hH4R. Data were taken from Appl et al. [21],
unless otherwise indicated: (a) Ref. [8], (b) Ref. [24].
(Mannheim, Germany), GDP (guanosine diphosphate) and
polyethyleneimine were from Sigma-Aldrich Chemie GmbH
(Munich, Germany). Histamine was from Tocris (Avonmouth,
Bristol, UK).
coexpressing either the hH1R and the RGS4 or the hH4R, G␣i2 and
␤1␥2, respectively, were used.
G
2.5. Cell culture and membrane preparation
The Sf9 insect cells were from BD Biosciences (BD Biosciences,
Heidelberg, Germany). Analysis with regard to a contamina-
tion with mycoplasma was performed using the Venor GeM
Mycoplasma Detection Kit (Minerva Biolabs, Berlin, Germany) and
was negative for the Sf9 cells. Cell culture, infection of the cells with
the respective baculoviruses and membrane preparation was per-
formed as described previously [28–30]. Protein concentration was
determined by the method of Lowry using bovine serum albumin
as standard [31].
2.3. Preparation of compound stock solutions
All compounds were dissolved in 50% (v/v) DMSO and 50% (v/v)
Millipore water with appropriate equivalents of aqueous HCl. Lig-
and concentrations were used in the range of 0.01 nM up to 1 mM,
as appropriate.
2.4. Pharmacological assays – general remarks
For pharmacological characterization of the compounds at the
hH1R, guinea-pig H1R (gpH1R) or the hH4R, Sf9 cell membranes,
2.6. Radioligand competition binding assays
The radioligand competition binding assays at the hH1R, gpH1R
and hH4R were performed as described previously [19,32–35].
Briefly, the reaction mixtures, containing binding buffer (12.5 mM
MgCl2, 1 mM EDTA, 75 mM Tris-HCl, pH 7.4), ligand (different
concentrations within a concentration range from 1 nM to 1 mM
as appropriate), radioligand (hH1R: 5 nM [3H]mepyramine, hH4R:
10 nM [3H]histamine) and Sf9 membranes expressing the respec-
tive histamine receptor, were incubated with shaking at 250 rpm
for 90 min at room temperature. Non-specific binding was deter-
mined in the presence of 10 ␮M diphenhydramine (hH1R and
gpH1R) or thioperamide (hH4R). Bound radioligand was separated
from free radioligand by filtration through glass microfibre filters
(Whatman GF/C) (in case of hH4R pretreated with polyethylen-
imine 0.3% (w/v)) using a 96-well Brandel harvester. The filters
were washed three times with binding buffer (4 ◦C). Subsequently,
each well filter piece was punched, transferred into 96-well
sample Plates 1450-401 (Perkin Elmer, Rodgau, Germany), sup-
plemented with 20 ␮l of scintillation cocktail (Rotiscint Eco plus,
Roth, Karlsruhe, Germany) and incubated in the dark under shak-
Fig. 3. Approach to new oxepine and oxazepine derivatives as selective or dual
H1/H4 receptor ligands.

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
613
Scheme 1. Synthesis of the dibenzo[b,f][1,4]oxazepine derivatives 7a–7i and 8a–8f, according to Smits et al. [22], and 9a–9c according to Chao et al. [56].
tion of 0.20–0.30 nM. [3H]N-methylscopolamine (specific activity
80 Ci/mmol; Biotrend, Cologne, Germany) was used for binding
experiments with the muscarinic M1, M2, and M3 receptors at
final concentrations of 0.10–0.20 nM, 0.20–0.30 nM, and 0.10 nM,
respectively. Unspecific binding was determined in the presence
of haloperidol (10 ␮M for dopamine receptors), atropine (10 ␮M
for muscarinic receptors), WAY100635 (10 ␮M for 5-HT1A) or
ketanserin (10 ␮M for 5-HT2A).
ing (200 rpm). The radioactivity was measured with a Micro Beta2
1450 scintillation counter. The Ki values were calculated according
to Cheng and Prusoff [36] with KD values (hH1R: [3H]mepyramine,
4.49 0.35 nM; gpH1R: [3H]mepyramine, 2.53 0.23 nM; hH4R:
[3H]histamine, 9.7 1.7 nM) determined previously [28,34,37].
Binding studies at the dopamine receptor subtypes were car-
ried out in analogy to a previously described method [38].
Accordingly, competition binding experiments with the human
D2long
, D2short [39], D3 [40], and D4.4 [41] receptors were
The resulting competition curves were analyzed by nonlinear
regression using the algorithms of PRISM 5.01 or 6.0 (GraphPad
Software, San Diego, CA). The data were initially fitted using a
sigmoid model to provide an IC50 value, representing the con-
centration corresponding to 50% of maximal displacement. IC50
values were transformed to Ki values according to the equation
of Cheng and Prusoff [36]. The resulting Ki values are expressed
logarithmically as pKi. All data are the means of three independent
experiments SEM, each performed in triplicate (unless otherwise
indicated by a footnote in the respective table).
done using preparations of membranes from CHO cells stably
expressing the corresponding receptor and [3H]spiperone (specific
activity = 81 Ci/mmol, PerkinElmer, Rodgau, Germany) at a final
concentration of 0.20–0.30 nM as radioligand. The assays were
carried out in binding buffer (50 mM Tris pH 7.4, 5 mM MgCl2,
1 mM EDTA, 100 ␮g/mL bacitracin, 100 ␮g/mL soybean trypsin
inhibitor) at a final volume of 200 ␮l with protein concentrations
of 1–10 ␮g/assay tube, KD values of 0.064–0.12 nM, 0.067–0.12 nM,
0.10–0.18 nM, and 0.17–0.30 nM and corresponding Bmax values of
520–1400 fmol/mg, 2600–4800 fmol/mg, 2300–4200 fmol/mg, and
1300–4500 fmol/mg for the D2long, D2short, D3 and D4.4 receptors,
respectively. Competition binding experiments at the serotoner-
gic receptor subtype 5-HT1A were performed using membranes
prepared from porcine cerebral cortex [38] (protein concen-
tration of 60–80 ␮g/mL) in the presence of the radioligand
[3H]WAY100635 (specific activity = 80 Ci/mmol, Biotrend, Cologne,
Germany) (0.15–0.20 nM final concentration, KD = 0.060–0.090 nM,
Bmax = 75–85 fmol/mg). Radioligand displacement experiments
with the human receptors D1, 5-HT2A, M1, M2, and M3 were per-
formed using preparations of membranes from HEK 293T cells
which were transiently transfected with the respective plasmid
(all from Missouri S&T cDNA Resource Center (UMR), Rolla, MO)
using the Mirus TransIT-293 transfection reagent (purchased from
MoBiTec, Goettingen, Germany) [42]. In detail, the assays were
carried out with a protein concentration of 2–20 ␮g per assay
tube, KD values of 0.32–0.75 nM, 0.19–0.24 nM, 0.050–0.12 nM,
0.15–0.29 nM, and 0.082 nM and corresponding Bmax values
of 2200–8900 fmol/mg, 1000–2300 fmol/mg, 700–1100 fmol/mg,
500–1600 fmol/mg, and 1800–3100 fmol/mg for D1, 5-HT2A, M1,
M2, and M3, respectively. The radioligand [3H]SCH23390 (specific
activity 80 Ci/mmol; Biotrend, Cologne, Germany) was used for
D1 binding at a final concentration of 0.40–0.70 nM. For the 5-
HT2A receptor the radioligand [3H]ketanserin (specific activity 47
Ci/mmol; Biotrend, Cologne, Germany) was used at a concentra-
2.7. GTPꢀS binding studies
The [35S]GTP␥S binding assay at the human histamine H2
receptor (hH2R) (Sf9 membranes expressing the hH2R-Gs␣s fusion
protein), the human histamine H3 receptor (hH3R) (Sf9 insect
cell membranes, coexpressing the hH3R, mammalian G␣i2 and
G␤1␥2) and the hH4R (Sf9 membranes coexpressing the hH4R,
mammalian G␣i2 and G␤1␥2) was performed as described previ-
ously [37,43–47]. Briefly, the respective membranes were thawed,
sedimented at 4 ◦C by centrifugation at 13000g, 10 min, and resus-
pended in binding buffer (12.5 mM MgCl2, 1 mM EDTA, 75 mM
Tris/HCl (pH 7.4)). Each reaction mixture, containing the mem-
brane, expressing the respective receptor, 1 ␮M GDP, 100 mM NaCl
(only in case of the hH4R), 0.05% (w/v) bovine serum albumin,
0.2 nM [35S]GTP␥S and the investigated ligand (different concen-
trations within a concentration range from 0.1 nM to 100 ␮M as
appropriate) was incubated under shaking at 200 rpm at room
temperature for 120 min. Non-specific binding was determined in
the presence of 10 ␮M unlabelled GTP␥S. The bound [35S]GTP␥S
was separated from free [35S]GTP␥S by filtration through glass
microfibre filters using a 96-well Brandel harvester. The filters
were washed three times with binding buffer (4 ◦C). Subsequently,
each well filter piece was punched, transferred into 96-well
sample Plates 1450-401 (Perkin Elmer, Rodgau, Germany), sup-

614
F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
plemented with 20 ␮l of scintillation cocktail (Rotiscint Eco plus,
Roth, Karlsruhe, Germany) and incubated in the dark under shaking
(200 rpm). The counting was performed with a Micro Beta2 1450
scintillation counter. The data obtained by the [35S]GTP␥S bind-
ing assay were analysed by nonlinear regression and best-fitted
to sigmoidal concentration-response curves. All data are means of
at least three independent experiments SEM, each performed in
triplicate (unless otherwise indicated by a footnote in the respec-
tive table). For data analysis the software Prism 5.01 (GraphPad
Software Inc., San Diego, CA, USA) was used.
concentration-response curves for histamine in the range of 10 nM
up to the respective maximum (0.03–10 mM) were recorded. For
each compound, at least twenty concentration response-curves
were measured.
For analysis of the data, the Schild equation (r − 1) = [B]/KB was
used [51]. Therein, r represents the concentration ratio (concentra-
tion of histamine, required to elicit a given response in the presence
of an antagonist divided by the concentration of histamine required
to produce exactly the same response in the absence of an antag-
onist), [B] represents the concentration of the antagonist and KB
the equilibrium dissociation constant for the antagonist B with the
receptor. The affinities of all analyzed antagonists, except 7b, were
expressed as pA2 values, determined according to the method of
Arunlakshana and Schild [52], using at least three different antago-
nist concentrations in a range of at least 1 or 2 orders of magnitude.
The slope s and the pA2 value were calculated by fitting the data
points by linear regression using GraphPad Prism 5.01 (GraphPad
Software Inc., San Diego, CA, USA). For compound 7b, which showed
a depression already at antagonist concentrations ≥ 3.16 nM, the
Schild equation was used to estimate KB for single antagonist con-
centrations, where a depression was not observed. The statistical
error was denoted as SEM.
2.8. GTPase assay
The GTPase assay at the hH1R was performed, as described pre-
viously [28,34]. Briefly, Sf9 membranes, coexpressing the hH1R and
RGS4 were thawed and sedimented by centrifugation (13000g at
4 ◦C, 10 min). The resulting membrane pellet was resuspended in
10 mM Tris-HCl (pH 7.4). The reaction mixture in each vessel con-
tained Sf9 cell membranes, coexpressing hH1R and RGS4 (10–20 ␮g
protein/cup), 1.0 mM MgCl2, 100 ␮M adenylyl imidodiphosphate,
100 ␮M ATP, 100 nM GTP, 100 ␮M EDTA, 5 mM creatine phosphate,
40 ␮g creatine kinase, 0.2% (w/v) BSA in 50 mM Tris-HCl (pH 7.4),
and the investigated ligand at various concentrations (concentra-
tion range from 1 nM to 1 mM, as appropriate). All reaction mixtures
(80 ␮l) were incubated at 25 ◦C for 3 min. Afterwards, 20 ␮l of [␥-
32P]GTP (0.1 ␮Ci/cup) were added into each tube. After incubation
at 25 ◦C for 20 min, the reactions were terminated by addition of
900 ml of slurry (5% (w/v) activated charcoal and 50 mM NaH2PO4
(pH 2.0)). Thereafter, the mixtures were centrifuged at 15000g,
room temperature, for 15 min. 600 ␮l of the supernatant fluid were
collected, and 32Pi was determined by Cherenkov counting. Because
the analyzed compounds had been identified as neutral antago-
nists, the GTPase assay was performed in the antagonist mode in
the presence of 200 nM histamine. All data are means of at least
three independent experiments SEM, each performed in tripli-
cate (unless otherwise indicated by a footnote in the respective
table). For data analysis the software Prism 5.01 (GraphPad Soft-
ware Inc., San Diego, CA, USA) was used.
2.10. Molecular modelling
Models of the hH1R and the hH4R in its inactive state were gen-
erated by homology modelling using the crystal structure of the
hH1R with the antagonist doxepine bound (3RZE) [53], as tem-
plate, as described previously using the software SYBYL 7.0 (Tripos
Inc) [17,19]. Briefly, the cocrystallized lysozyme was deleted and
the N-terminus and the E2-loop were completed, using a pre-
viously described protocol [54]. Due to its length, the I3-loops
of both receptors were not modelled. However, to close the gap
between the intracellular parts of TM V and TM VI, the amino
acids of the I3-loop (hH1R: Leu231 to Val404, hH4R: Gly215 to His292
)
were replaced by eight alanines. Internal water molecules were
included into both models, as described [54]. The compounds 7e,
(R)-16e and (S)-16e were manually docked into the binding pocket
of the hH1R and the hH4R using SYBYL 7.0 (Tripos Inc.). The result-
ing ligand-receptor-complexes were minimized and embedded
in a POPC lipid bilayer. For the subsequent molecular dynamic
(MD) studies, the software package GROMACS 4.0.2 (http://www.
gromacs.org) was used. The force field parameters for the lig-
ands were obtained from the PRODRG server (http://davapc1.bioch.
dundee.ac.uk/prodrg/), and the partial charges were calculated
according to Gasteiger-Hückel using SYBYL 7.0. The force field
parameters for the POPC lipid were obtained from http://moose.
bio.ucalgary.ca/index.php?page=Structures and Topologies. Intra-
cellular and extracellular water, as well as an appropriate number of
sodium and chloride ions were included into the simulation boxes
to achieve electroneutrality using the commands genbox and genion
of the software package GROMACS 4.0.2. All simulation boxes con-
tained a number of 115 POPC molecules and about 11000 to 12000
water molecules. The ffG53a6 force field [55] was used for the
hH1R and the hH4R. After minimization of the simulation boxes, the
MD simulations (5 ns equilibration phase, at least 10 ns productive
phase) were performed with GROMACS 4.0.2. Each simulation was
performed four times, using different seed values.
2.9. Histamine H1 receptor assay on the isolated guinea-pig ileum
Selected compounds were pharmacologically characterized at
the gpH1R. For this purpose, assays on the isolated guinea-pig ileum
with cumulative concentration effect curves of histamine in the
presence of an antagonist were performed, as described in detail
previously [48]. Briefly, guinea-pigs of either sex (250 g–500 g)
were stunned by a blow on the neck and exsanguinated by cut-
ting the carotid artery. After removing the ileum, it was cut
into segments of 1.5 cm–2.5 cm length. The ileum segments were
mounted isotonically in 20 ml organ baths, containing aqueous
Tyrode’s solution (NaCl 137 mM, KCl 2.7 mM, CaCl2 1.8 mM, MgCl2
1.0 mM, NaH2PO4 0.4 mM, NaHCO3 11.9 mM, glucose 5.0 mM) [49]
and were connected to a transducer (Type TIT 1100, FMI GmbH,
Seeheim, preload 5 mN), was warmed to a constant temperature
of 37 ◦C and gassed with 95% O2 and 5% CO2. The contraction
of the ileum segments was measured by means of an ampli-
fier (Transducer coupler 4711, FMI GmbH, Seeheim, Germany)
and recorded by a 6-channel-x/t-writer (Kompensograph C 1015,
Siemens, Germany). To block cholinergic muscarinic receptors,
atropine was added to achieve a final concentration of 0.05 ␮M [50].
After 30 min of equilibration, the tissues were prestimulated three
times with histamine (first 1 ␮M, then 10 ␮M) and washed out
for 8 min, followed by a recovery period of 10 min. Subsequently,
cumulative concentration-response curves were registered for his-
tamine in the concentration range of 10 nM to 30 ␮M. After adding
the antagonist and an incubation period of 15 min, cumulative
3. Results and discussion
3.1. Synthesis
All loxapine derived compounds 7a–7i, 8a–f, 9a–c were
prepared according to a synthetic pathway already described

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
615
Scheme 2. Synthesis of the dibenzo[b,e][1,4]oxepine derivatives 16a–16q, according to a protocol described previously [57].
(Scheme 1, cf. Supporting information) [22]. Briefly, substituted
benzoic acids were converted into the corresponding acid chlo-
rides, which were added to a solution containing the corresponding
aminophenol and trimethylamine in THF, yielding the amide
derivatives 5a–5i. After adding pulverized NaOH and heating at
reflux in DMF, ring closure of the amides was obtained, yield-
ing the compounds 6a–6i. By adding POCl3, the compounds 6a–6i
were converted in situ into the corresponding reactive iminochlo-
rides, which were allowed to react with N-methylpiperazine or
piperazine in toluene to obtain the dibenzo[b,f][1,4]oxazepine
derivatives 7a–7i and 8a–8f. Compounds (7g–7i) bearing a NO2
substituent were converted into the corresponding amines 9a–9c
by reduction with SnCl2 according to Chao et al. [56]. The analyti-
cal data of the compounds 7a, 7c, 7d and 7f are in good accordance
with data in literature [22].
All doxepine derived compounds 16a–16q were synthesized as
described [57] (Scheme 2, cf. Supporting information). In the first
step, an esterification of the corresponding 2-methyl-benzoic acid
with iodoethane and potassium carbonate in DMF as deprotonat-
ing reagent was carried out to obtain the compounds 10a–10d.
After adding NBS and AIBN, acting as radical starter, the compounds
11a–11d were obtained and added to a solution, containing 21
% (w/w) NaOEt and the appropriate phenol derivative in ethanol,
yielding the compounds 12a–12p. Afterwards, the ester, which was
introduced in the first step as a protecting group, was cleaved
under basic conditions, yielding the compounds 13a–13p. The
compounds were converted with thionylchloride into the cor-
responding reactive acid chlorides. To obtain the ring closure
14a–14p, a Friedel-Crafts-acylation with aluminium chloride was
carried out. As starting material the derivatives 14a–14p and the
unsubstituted carba analogue 14q were reduced to the correspond-
ing alcohol (racemic in case of 15a–15p) with sodium borohydride.
By adding thionylchloride, the alcohols were converted in situ
into the corresponding chlorides, which immediately reacted with
N-methylpiperazine in chloroform to yield the final compounds
16a–16q.
acid and (+)-O,O’-di-p-toluoyl-d-tartaric acid, respectively (cf. Sup-
porting information). The separation of the racemate 16l into the
enantiomers by crystallization, as described for 16e, failed.
3.2. Optical purity and determination of the absolute
configuration of both enantiomers of 16e
By the determination of the crystal structures (cf. Supporting
information) the absolute configuration could be assigned to both
enantiomers (Fig. 4). The absolute structure of (R)-16e was deter-
mined by anomalous dispersion. The minor component ((S)-16e)
was located from the difference fourier map and the amount refined
to 0.040(2) in a restrained model of the second part. For the final
refinement the occupancy was fixed to allow a stable refinement
of the displacement parameters.
As shown by capillary electrophoresis, the enantiomers (R)-16e
and (S)-16e were obtained with a purity of 96.8% and 96.7%, respec-
tively. This corresponds to an enantiomeric excess of 93.6% for
(R)-16e and of 93.4% for (S)-16e (Fig. 5).
The racemate 16e was separated into the enantiomers (R)-16e
and (S)-16e by crystallization with (−)-O,O’-di-p-toluoyl-l-tartaric
Fig. 4. Crystal structures of (A) (R)-16e, crystallized with (−)-O,O’-di-p-toluoyl-l-
tartaric acid and (B) (S)-16e, crystallized with (+)-O,O’-di-p-toluoyl-d-tartaric acid.

616
F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
Table 1
Binding affinities of the dibenzo[b,f][1,4]oxazepine derivatives 7a–7i, 8a–8f and 9a–9c at hH1R and hH4R, determined by radioligand-competition binding assays.
cpd.
pKi
cpd.
pKi
cpd.
pKi
cpd.
pKi
hH1R
hH4R
hH1R
hH4R
hH1R
hH4R
hH1R
hH4R
7a
7b
7c
7d
7e
7f
8.73 0.19
9.29 0.03
7.76 0.11b
8.56 0.03
9.23 0.28
8.82 0.05
6.25 0.12a
6.00 0.13
6.99 0.01b
6.47 0.15c
6.90 0.10
6.34 0.08d
8a
8b
8c
8d
8e
8f
8.30 0.16
9.06 0.13
8.26 0.15
8.91 0.20
8.59 0.22
8.76 0.08
5.34 0.09
5.35 0.02
6.58 0.10
6.08 0.05
5.95 0.06
5.58 0.05
–
7g
–
–
7h
7i
–
–
–
9a
–
–
9b
9c
–
–
9.00 0.08
–
5.94 0.04
–
8.21 0.07
–
–
7.99 0.13
8.50 0.09
5.06 0.07
–
–
6.44 0.03
6.02 0.04
–
–
8.59 0.04e
8.69 0.04f
5.87 0.06e
6.71 0.03f
a
pKi (hH4R) = 6.88 (Smits et al. [22]).
b
c
d
e
f
pKi (hH1R) = 8.11, pKi (hH4R) = 7.55 (Smits et al. [22]).
pKi (hH4R) = 7.37 (Smits et al. [22]).
pKi (hH4R) = 6.51 (Smits et al. [22]).
Slight impurities are visible in the [1H]NMR and MS (cf. Supporting information).
Slight impurities are visible in the [1H]NMR (cf. Supporting information).
Fig. 6. Displacement curves with selected compounds 7b, 7c and 7e at the hH1R
(red) and the hH4R (blue), determined by radioligand competition binding.
et al. [22]. The affinities of the methylated derivatives 7a–7f are in
a range from 7.8 to 9.3 at hH1R and in a range from 6.0 to 7.0 at
hH4R. In general, the data show that the chloro substitution pat-
tern has an influence regarding the affinity and selectivity at the
hH1R and the hH4R (Table 1): For the mono-Cl derivative 7c, the
smallest selectivity (6) and for the mono-Cl derivative 7b the high-
est selectivity (1950) towards the hH1R was observed. Compound
7b showed highest affinity to the hH1R, but lowest affinity to the
hH4R. In contrast, compound 7c showed lowest affinity to the hH1R
but highest affinity to the hH4R. The mono-Cl substitution pattern
of 7b and 7c was combined in the di-Cl substituted compound 7e,
showing the same affinity to the hH1R as 7b and at the hH4R as 7c
(Fig. 6). Thus, by combination of the Cl-substitution pattern of 7b
and 7c, the high affinity of 7b to the hH1R and of 7c to the hH4R
resides in one compound, 7e.
The affinities of the demethylated analogues 8a–8f to the
hH1R differed within 0.5 orders of magnitude compared to their
methylated counterparts 7a–7f. In contrast, demethylation at the
piperazine moiety resulted in a decrease in affinity to the hH4R by
up to one order of magnitude (Table 1). Thus, it can be concluded
that for all analyzed compounds, methylation at the piperazine
moiety increased affinity to the hH4R and for some of the analyzed
compounds at hH1R. The exchange of the chloro group by the nitro
group (7g–7i) resulted in no or only slight differences in affinity of
7g and 7i at hH1R and hH4R, compared to the respective chlorine
compounds 7b and 7f. In contrast, for 7h, a decrease in affinity up
to one order of magnitude compared to 7e was observed. Beside
a possible influence of the NO2-group in 7h onto the affinities, it
has to be taken into account, that some impurities are present in
7h (Table 1, cf. Supporting information). An exchange of the chloro
group by an amino substituent resulted in a decrease in affinity
Fig. 5. Electropherograms using ␥-CD as a chiral selector: (A) racemate 16e, (B)
(R)-16e, (C) (S)-16e; both enantiomers were isolated by crystallization.
3.3. Pharmacological characterization of
dibenzo[b,f][1,4]oxazepine derivatives at the hH1R and the hH4R
The affinities of the dibenzo[b,f][1,4]oxazepine derivatives
7a–7i, 8a–8f and 9a–9c to the hH1R and the hH4R are shown in
Table 1. The affinities of 7a, 7c, 7d and 7f at hH1R or hH4R are
in general in good accordance with the data published by Smits

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
617
Table 2
Potencies and efficacies of the dibenzo[b,f][1,4]oxazepine derivatives 7a–7f, 8a–8f and 9b–9c at the human histamine receptor subtypes hH1R, hH2R, hH3R and hH4R. At
hH1R, the data were determined by the GTPase assay in the presence of 200 nM histamine. At hH2R, hH3R and hH4R, the data were determined by the [35S]GTP␥S-binding
assay.
cpd.
hH1R
hH2R
hH3R
hH4R
pKB
pEC50
Emax
pEC50
Emax
pEC50
Emax
7a
7b
7c
7d
7e
7f
8a
8b
8c
8d
8e
8f
8.59 0.23
8.58 0.06
7.89 0.11
8.85 0.30
8.30 0.10
8.47 0.22
8.02 0.09
8.39 0.12
7.94 0.14
8.19 0.13
7.77 0.04
8.19 0.06
7.71 0.26
8.24 0.20
<5
<5
−0.26 0.02
−0.22 0.06
n.d.
<5
<5
n.d.
n.d.
<5
−0.32 0.03
−0.28 0.01
n.d.
6.16 0.04
6.08 0.05
6.58 0.19
6.20 0.05
7.12 0.22
6.20 0.24
5.21 0.15
5.35 0.19
6.11 0.07
5.76 0.10
5.87 0.12
5.39 0.09
6.40 0.13
6.11 0.16
0.62 0.14
0.68 0.20
0.45 0.19
0.55 0.19
0.57 0.20
0.57 0.16
0.37 0.09
0.40 0.07
0.34 0.11
0.38 0.12
0.50 0.18
0.48 0.10
0.35 0.14
0.35 0.08
n.d.
n.d.
5.45 0.22
<5
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
<5
n.d.
n.d.
−0.16 0.02
−0.21 0.04
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
−0.26 0.01
−0.28 0.01
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
<5
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
<4
9b
9c
−0.25 0.02
−0.32 0.03
<4
−0.31 0.03
<4
−0.39 0.01
Fig. 7. Concentration-response curves of histamine and compound 7e at the hH1R,
determined in the steady-state GTPase assay.
Fig. 8. Concentration-response curves of compound 7e at the hH2R, the hH3R, and
the hH4R determined in a [35S]GTP␥S binding assay.
by up to one order of magnitude (7b and 9a, 7e and 9b) at both
receptors, whereas there was no difference in case of 7f and 9c.
All dibenzo[b,f][1,4]oxazepines were studied in functional
assays (hH1R: GTPase-assay, hH4R: [35S]GTP␥S-assay). The deriva-
tives 7a–7f and 8a–8f acted as neutral antagonists at the hH1R,
but as partial agonists at the hH4R, as exemplarily shown for 7e
(Figs. 7 and 8, Table 2). In order to obtain information on the func-
tional selectivity at the four human histamine receptor subtypes,
the selected compounds 7a, 7b, 7e, and 7f were analyzed with
respect to [35S]GTP␥S binding at hH2R and hH3R (Table 2). The
potencies of the compounds were low at the hH2R and the hH3R
(pEC50 ≤ 5) and the compounds showed inverse agonism, as exem-
plarily shown for 7e (Fig. 8). Thus, the compounds show >10-fold
selectivity to the hH1R and the hH4R over the hH2R and the hH3R.
Fig. 9. Contraction of guinea-pig ileum (whole segments) by histamine in the
absence and presence of the selected inhibitors 7b, 7c and 7e (representative curves
are shown).
At the hH1R, the pKB values of 7a–7f ranged from 7.9 to 8.9,
and those of the demethylated counterparts 8a–8f from 7.7 to 8.4
(Table 2). By analogy to the affinities, the pKB values of 8a–8f dif-
fered up to 0.6 orders of magnitude compared to the methylated
derivatives 7a–7f. The pEC50 values of the methylated compounds
7a–7f at the hH4R were in a range from 6.1 to 7.1, whereas the pEC50
values of the demethylated counterparts 8a–8f were in a range
from 5.2 to 6.1. Thus, the potencies of the demethylated compounds

618
F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
Table 3
Contraction of guinea-pig ileum (whole segments) by histamine in absence and presence of the inhibitors 7a–7f, 8a–8f.
cpd.
depression
concentration range for
Schild analysis
pA2
slope sa
7a
7b
7c
7d
7e
7f
8a
8b
8c
8d
8e
8f
≥316 nM
≥3.16 nM
≥3.16 ␮M
≥31.6 nM
≥31.6 nM
≥31.6 nM
≥10 ␮M
3.16 nM–100 nM
–
9.03 0.05
–
0.93 0.05
–
b
31.6 nM–1 ␮M
1 nM–10 nM
0.32 nM–10 nM
0.32 nM –10 nM
1 nM–3.16 ␮M
1 nM–100 nM
1 nM–1 ␮M
1 nM–100 nM
10 nM–100 nM
3.16 nM–31.6 nM
7.57 0.14
8.38 0.19
8.65 0.12
9.04 0.10
7.85 0.08
8.91 0.08
8.35 0.06
8.64 0.06
7.96 0.04
8.36 0.08
1.11 0.14
1.76 0.48**,c
0.71 0.22
0.71 0.18
0.91 0.08
1.06 0.16
0.61 0.06***,c
1.11 0.10
1.06 0.09
1.50 0.09***,c
≥316 nM
≥3.16 ␮M
≥316 nM
≥316 nM
≥100 nM
a
Slope according to data analysis using the Schild equation.
Estimation 7b: 0.316 nM: 9.34 0.22 (n = 5); 1 nM: 9.70 0.19 (n = 4); n: number of experiments.
Slope significantly different from 1.
p < 0.01.
b
c
**
***
p < 0.001.
Table 4
were considerably lower than those of the methylated compounds
at the hH4R. The maximal agonistic effects (Emax) relative to his-
tamine (Emax = 1) of the methylated derivatives 7a–7f were in the
range of 0.4 to 0.7, and those of the demethylated counterparts
8a–8f in the range of 0.3 to 0.5. Thus, the maximal agonistic effect
at the hH4R tends to be smaller for the demethylated compounds
compared to the methylated ones.
Binding affinities of the dibenzo[b,e]oxepine derivatives 16a–16q at hH1R, gpH1R
and hH4R, determined by radioligand-competition binding assays.
cpd.a
pKi (hH1R)
pKi (hH4R)
16a
16b
16c
16d
16e
(R)-16e
(S)-16e
16f
16g
16h
16i
16j
16k
16l
8.12 0.03b
7.13 0.04
8.71 0.13
8.48 0.09
≤5
≤4
≤4
≤4
8.42 0.04 (gpH1R: 8.24 0.09c)
8.83 0.09 (gpH1R: 8.82 0.13b)
7.63 0.04b (gpH1R: 7.41 0.03b)d
8.47 0.08e
≤5
3.4. Pharmacological characterization of
dibenzo[b,f][1,4]oxazepine derivatives at the isolated guinea-pig
ileum
≤5
≤5
≤5
7.70 0.11b
7.06 0.10
7.42 0.07
≤4
≤5
The concentration-response curves of histamine in the presence
of dibenzo[b,f][1,4]oxazpepine derivatives 7a–7f and 8a–8f were
additionally analyzed at the isolated guinea-pig ileum (Table 3).
The concentration response curves are exemplarily shown for 7b,
7c and 7e (Fig. 9).
≤5
6.83 0.09
≤5
8.50 0.16f
≤5
8.44 0.10
7.40 0.06
8.65 0.16
8.56 0.03
7.10 0.07
8.06 0.04
5.27 0.02
≤5
16m
16n
16o
16p
16q
≤4
All compounds acted as silent antagonists at the gpH1R, with
pA2 values in the range of 7.6 to 9.0. As already observed in the
binding studies at hH1R, demethylation at the piperazine moiety
(cpds. 8a–8f) led, compared to the methylated counterparts 7a–7f,
to a difference in pA2 values up to ∼one order of magnitude.
For compounds 7c, 8a and 8c up to an antagonist concentration
of at least 1 ␮M, no depression of the histamine concentration-
response curves were observed. In contrast, for 7b, a depression
was already observed at an antagonist concentration of 3.16 nM.
For most of the compounds, except 7d, 8c and 8f, the slope, accord-
ing to the Schild-equation, was not significantly different from 1.
Taking into account the depression of the concentration-response
curves of histamine in the presence of antagonist and slopes sig-
nificantly different from 1, for some compounds it can be assumed
that they do not act as “pure”, competitive antagonists at the gpH1R
of the guinea-pig ileum.
≤5
≤5
≤4
a
Histamine: pKi (hH1R): 5.62 0.03 [77]; pKi (hH4R): 8.07 0.14 [78].
Number of experiments: 5.
Number of experiments: 7.
b
c
d
Traces of (R)-16e, which were shown by CE to be present in (S)-16e, can pretend
a slightly higher pKi for (S)-16e.
e
Number of experiments: 4.
Number of experiments: 6.
f
in a decrease (16b: ∼10-fold, 16g: ∼3-fold), and a chlorine at R2
(16c) led to an increase (16c: ∼4-fold) in affinity compared to
16a. A chlorine at R3, R4 or R6 led to an increase in affinity com-
pared to the unsubstituted compound 16a. Within the series of the
mono-substituted compounds, the affinity decreased in the series
R2 (16c) > R3 (16d) ≈ R5 (16f) ≈ R4 (16e) > R6 (16g) > R1 (16b). The
combination of the chloro substitution at R1, R2, R3 and R4, R5, R6
resulted in the di-substituted compounds 16h–16p. Compared to
the mono-substituted compounds 16c, 16d, 16e, 16f, the corre-
sponding di-substituted compounds 16k, 16l, 16n and 16o showed
an affinity in a similar range (Table 4).
The competition binding studies showed that the (R)-16e is
the eutomer at hH1R and gpH1R (Table 4, Fig. 10). Compared to
the distomer (S)-16e, the eutomer (R)-16e was about one order of
magnitude more affine to the hH1R and the gpH1R, respectively.
The oxepine derivatives 16a–16q had weak affinity (pKi < 5) at
the hH4R, only 16l exhibited a pKi > 5 (Table 4), and thus, represent
selective hH1R ligands, compared to the hH4R. These data indi-
3.5. Pharmacological characterization of the dibenzo[b,e]oxepine
derivatives at the hH1R and the hH4R
The seventeen dibenzo[b,e]oxepine derived compounds
16a–16q were characterized at the hH1R and the hH4R by
radioligand competition binding assays (Table 4).
The pKi values of the compounds were ranged in dependence of
the chloro substitution pattern from 6.8 to 8.7 at the hH1R (Table 4).
The affinity of the dibenzo[b,e]oxepine 16a was in the same range
as that of the carba analogue 16q. This indicates that the exchange
of a CH2 into an O at position X has no influence on the affin-
ity to the hH1R. A chlorine atom at R1 (16b) or R6 (16g) resulted

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
619
Table 5
Pharmacological analysis of histamine and selected compounds 16l, 16e, (R)-16e, and (S)-16e at the human histamine receptor subtypes hH2R, hH3R and hH4R.
hH2R
hH3R
hH4R
pEC50
Emax
pEC50
Emax
pEC50
Emax
histamine
16l
16e
(R)-16e
(S)-16e
6.43 0.23
5.19 0.14
n.d.
n.d.
n.d.
1.00
8.20 0.04
5.08 0.24
n.d.
n.d.
n.d.
1.00
8.08 0.15
1.00
−0.19 0.04
−2.6 1.1a
<5
<5
<5
<5
inv. ag.b
inv. ag.b
inv. ag.b
inv. ag.b
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
a
Relative to histamine (Emax = 1.00).
Invserse agonism was observed at ligand concentrations ≥ 10 ␮M.
b
Table 6
Contraction of guinea-pig ileum (whole segments) by histamine in absence and presence of the inhibitors 16a, 16e, (R)-16e, (S)-16e and 16l.
cpd.
depression
concentration range for
Schild analysis
pA2
slope sa
16a
16e
(R)-16e
(S)-16e
16l
≥3.16 ␮M
≥1 ␮M
3.16 nM–1 ␮M
3.16 nM–1 ␮M
1 nM–1 ␮M
10 nM–3.16 ␮M
1 nM–100 nM
8.77 0.29
8.55 0.30
8.68 0.21
8.03 0.23
9.39 0.36
0.89 0.04*,b
0.94 0.04c
≥316 nM
≥316 nM
≥316 nM
0.78 0.03***,b
0.78 0.03***,b
0.98 0.07c
a
Slope according to data analysis using the Schild equation.
Slope significantly different from 1.
Slope not significantly different from 1.
p < 0.1.
b
c
*
***
p < 0.001.
of 16l, which acts as inverse agonist, at hH2R, hH3R and hH4R is very
moderate (pEC50 ≤ 5). Thus, within the human histamine receptors,
16l shows high subtype selectivity towards the H1R.
3.6. Pharmacological characterization of selected
dibenzo[b,e]oxepine derivatives at the isolated guinea-pig ileum
The concentration-response curves of histamine in the presence
of dibenzo[b,e]oxepine derivatives 16a, 16e, (R)-16e, (S)-16e and
16l were additionally analyzed at the isolated guinea-pig ileum
(Table 6). The concentration response curves and respective Schild
plots are exemplarily shown for 16e, (R)-16e and (S)-16e (Fig. 11).
The compounds acted as silent antagonists at the gpH1R, with
pA2 values in the range of 8.0 to 9.4. For the enantiomers (R)-16e
and (S)-16e the slope is quite identical, but significantly different
from 1 (Table 6). Thus, it can be assumed that they do not act as
“pure”, competitive antagonists at the gpH1R of the guinea-pig
ileum. Strikingly, the slope for the racemate 16e is considerably
different from the slope of both enantiomers, but there is no obvi-
ous explanation for this observation. However, in analogy to the
results of the competition binding assay at gpH1R (Table 4), the
(R)-enantiomer is the eutomer.
3.7. Binding mode of 7e, (R)-16e and (S)-16e at the hH1R and the
hH4R
The MD studies of the oxazepine derivative 7e at the hH1R
and the hH4R and of the oxepine derivatives (R)-16e and (S)-16e
at the hH1R resulted in stable binding modes in the orthosteric
binding pocket of both receptors. The binding modes are quite sim-
ilar to that found in the crystal structure of the hH1R in complex
with the H1R antagonist doxepine [53]. In all cases, the positively
charged N-methylpiperazine, in chair conformation, formed a sta-
ble electrostatic interaction with the highly conserved Asp3.32 (cf.
Supporting information). The tricyclic oxazepine or oxepine moi-
ety is embedded in a pocket between TM III, TM V and TM VI (cf.
Fig. 10. Displacement curves of 16e, (R)-16e and (S)-16e at the hH1R and the gpH1R
determined by radioligand competition binding. Representative curves are shown.
cate that the tricyclic core of the oxepines may not fit well into
the orthosteric binding pocket of the hH4R.
To obtain information regarding selectivity, the oxepine deriva-
tive 16l was studied at the hH2R, hH3R and hH4R in the
Supporting information), mainly formed by the amino acids Tyr3.33
,
Thr3.37, Thr5.42, Phe5.47, Phe6.44 and Trp6.48, which are identical at
the hH1R and the hH4R, and Ile3.40, Trp4.56, Lys5.39, Ala5.43, Asn5.46
[
35S]GTP␥S-binding assay (Table 5). The data show that the potency

620
F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
Fig. 11. Contraction of guinea-pig ileum (whole segments) by histamine in the absence and presence of the selected inhibitors 16e, (R)-16e and (S)-16e (representative
curves and respective Schild plots are shown).
and Phe6.52 in case of hH1R and Val3.40, Val4.56, Leu5.39, Ser5.43
Glu5.46 and Ser6.52 in case of the hH4R.
,
accordance to the experimental data (Table 4). In contrast, for 16d,
the (S)-enantiomer is predicted as the more affine enantiomer. The
energetically disfavored position of the oxygen of (S)-16d is out-
balanced by the energetically favored position of the chlorine at
R3 (Fig. 12C). Considering the crystal structure and the energetic
data of the modeling studies, it can be speculated that the (R)- and
(S)-enantiomer of the oxepine derivatives differ in their orienta-
tion in the binding pocket (Fig. 12C). In case of the (R)-enantiomer,
Based on the results of the MD simulation of (R)-16e and
(S)-16e, the enantiomers (R)-16a-(R)-16p and (S)-16a-(S)-16p
were docked into the binding pocket of the hH1R and the
resulting ligand-receptor complexes were minimized with SYBYL
7.0 (Tripos Inc) (Fig. 12A). For each ligand, three slightly dif-
ferent docking poses were generated and the mean docking
energy between the tricyclic core of each ligand and the hH1R
was calculated in order to study the influence of the chlo-
rine substitution pattern. The docking energy for each racemic
oxepine derivative 16a–16p was calculated according to the
equation: ꢁE = −RT ln (1/2 exp(−ꢁEcalc, (R)−enantiomer/(RT)) + 1/2
exp(−ꢁEcalc, (S)-enantiomer/(RT))) and correlated with the experi-
mentally determined pKi values (Fig. 12B). For the unsubstituted
16a, the docking study suggested that the (R)-enantiomer is more
affine to the hH1R, than the (S)-enantiomer. A structural anal-
ysis of the ligand-receptor model indicated that in case of the
(S)-enantiomer, the oxygen of the ligand is closer to oxygens of
the receptor, resulting in a repulsive electrostatic interaction. For
all compounds with one or two chlorine in the tricyclic core,
the R-enantiomers are predicted to bind preferably to the hH1R,
except for 16d, 16g, 16o and 16p. Furthermore, the docking stud-
ies, predicting (R)-16e as the more affine enantiomer, are in good
the chlorine at the positions R4, R5 or R6 is neighbored to Thr5.42
,
Trp4.56, Lys5.39 and to water molecules. In contrast, the positions
R4, R5 and R6 are neighbored to Thr3.37, Ile3.40, Phe6.44 and Trp6.48
for the (S)-enantiomer. The matter of fact is reversed for R1, R2 and
R3 (Fig. 12C). Based on the modeling studies, it can be suggested
that the more voluminous chlorine at R4 or R5 ((R)-enantiomer) or
R3 and R2 ((S)-enantiomer) fills the space within the binding site
(Lys5.39, Phe6.52) of the hH1R (Fig. 12C), resulting in an increase
in affinity. A chlorine at position R6 ((R)-enantiomer) or R1 ((S)-
enantiomer) led to a steric clash with Thr5.42. In contrast, a chlorine
at position R1 ((R)-enantiomer) or R6 ((S)-enantiomer) led to a steric
clash with Thr3.37, resulting in a shift of ∼0.7 to ∼0.9 Å of the aro-
matic moiety towards TM VI. Furthermore, a chlorine at R2 or R3
((R)-enantiomer) or R5 or R4 ((S)-enantiomer) led to a steric clash
with Ile3.40 and Trp6.48 (Fig. 12C). Additionally, the modelling data
indicate that in case of the (S)-enantiomer, the oxygen of the ligand

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
621
Fig. 12. Binding mode of oxepine derivatives at the hH1R. (A) 16a–16p, docked into the hH1R (overlay of all R-enantiomers is shown); (B) Correlation of the docking energy
(ꢁEcalc) between the tricyclic core of (R)-16a-(R)-16p (green), (S)-16a-(S)-16p (red) and calculated for the racemate (black) versus the experimentally determined pKi values;
the linear regression with the 95% CI (dotted line) was calculated for the data of the racemate; (C) Structure-activity-relationships of the oxepines 16a–16p at the hH1R,
derived from the experimental and modelling results.
is closer to an area of the receptor with a partially negative charge,
resulting in a repulsive electrostatic interaction, compared to the
(R)-enantiomer.
In contrast to the oxazepine derivatives the oxepine deriva-
tives showed no or only moderate affinity to the hH4R. However,
as expected, based on experimental data, previously determined
(Fig. 2) [21], the oxazepine derivatives showed partial agonism at
the hH4R, but the oxepine derivatives showed inverse agonism at
ligand concentrations ≥10 ␮M. In literature, it is suggested that the
highly conserved Trp6.48 changes its conformation during receptor
activation [58]. Thus, based on the experimental data, it might be
suggested that the different positions of one aromatic moiety in the
oxazepines and oxepines influence the conformation of the Trp6.48
and determine partial or inverse agonism.
3.8. Selectivity between hH1R and hH4R on a molecular level
Numerous studies were performed to identify amino acids of
the TM domains and the extracellular loops at the H1R and H4R,
being involved in ligand binding [4,23,47,59–68]. Additionally, at
the hH3R and rH3R the amino acids 3.37 and 3.40 were identified
to be responsible for species differences [69–71]. The amino acids
at 3.40 (hH1R: Ile, hH4R: Val), 4.56 (hH1R: Trp, hH4R: Val), 5.38
(hH1R: Phe, hH4R: Ile), 5.39 (hH1R: Lys, hH4R: Leu), 5.43 (hH1R:
Ala, hH4R: Ser), 5.46 (hH1R: Asn, hH4R: Glu), 6.52 (hH1R: Phe,
hH4R: Ser) and 7.42 (hH1R: Gly, hH4R: Gln) form the orthosteric
binding pocket and are shown or suggested to be involved in lig-
and binding at the histamine receptors [17]. Furthermore, it was
shown by mutagenesis studies that the E2-loop or distinct amino
acids of the E2-loop have influence on ligand affinity and recep-
Fig. 13. Polar plot of pKi values of clozapine 1, 7a–7f, 16b, 16c, 16f and 16l at selected
non-histaminergic GPCRs.

622
F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
Table 7
Pharmacological analysis (pKi values) of clozapine 1 and selected compounds 7a–7f, 16b, 16c, 16f and 16l at the selected aminergic GPCRs.
hD1R
hD2Rlong
hD2Rshort
hD3R
hD4.4
R
p5-HT1A
R
h5-HT2A
R
hM1R
hM2R
hM3R
1
7a
7b
7c
7d
7e
7.45 0.33
7.24 0.05
6.78 0.06
7.69 0.15
8.17 0.16
6.57 0.20
7.21 0.14
6.06 0.21
6.04 0.32
6.02 0.17
5.75 0.27
7.48 0.13
6.21 0.13
6.68 0.09
6.85 0.01
7.86 0.12
6.68 0.04
7.65 0.01
5.75 0.15
5.94 0.11
6.29 0.02
5.75 0.13
7.64 0.14
6.23 0.02
6.72 0.03
7.05 0.08
7.89 0.12
6.94 0.10
7.71 0.16
5.82 0.10
6.61 0.11
6.50 0.03
6.28 0.19
6.62 0.14
6.18 0.07
6.58 0.16
6.30 0.33
7.24 0.10
6.47 0.19
7.29 0.12
6.70 0.03
6.33 0.05
6.68 0.03
6.14 0.04
7.98 0.13
6.85 0.04
6.81 0.06
7.04 0.46
8.69 0.22
6.33 0.05
7.36 0.07
<5
6.74 0.28
7.07 0.21
5.71 0.08
6.03 0.13
7.34 0.28
5.30 0.13
5.44 0.09
<5
7.81 0.27
7.81 0.07
8.72 0.15
8.24 0.16
8.63 0.06
8.74 0.18
9.35 0.10
7.32 0.02
7.00 0.08
7.22 0.07
6.71 0.06
8.97 0.15
7.59 0.31
7.48 0.26
7.21 0.08
8.65 0.06
7.06 0.14
7.82 0.23
7.23 0.07
6.98 0.02
7.81 0.21
6.46 0.24
7.41 0.22
6.70 0.24
6.65 0.01
6.79 0.29
7.20 0.04
6.34 0.38
6.54 0.34
7.07 0.03
6.71 0.06
7.29 0.20
6.59 0.09
8.10 0.16
7.31 0.34
6.74 0.03
6.87 0.21
8.06 0.23
6.40 0.16
7.02 0.20
7.28 0.04
6.80 0.04
7.77 0.15
6.27 0.16
7f
16b
16c
16f
16l
5.55 0.04
5.65 0.20
5.27 0.11
<5
<5
<5
Fig. 14. Affinity profile of 7e, 7f and cyproheptadine at selected human aminergic GPCRs. Affinity data for cyproheptadine were taken from: hH1R [72], hH4R [8], hD1R, hD2R,
hD3R, h5-HT1AR [74], D4R [75], h5-HT2AR [73], hM1R, hM2R, hM3R [76]. pKi values of 7e and 7f refer to the p5-HT1AR and of cyproheptadine to the h5-HT1AR.
tor activation [35,47,61–63,68]. Thus, it may be suggested that the
mentioned amino acids (at positions 3.40, 4.56, 5.38, 5.39, 5.43,
5.46, 6.52 and 7.42) may be responsible for the different pharma-
cological profiles of the oxazepines 7a–7f, 8a–8f and 9a–9c and
oxepines 16a–16p at the hH1R and the hH4R. The amino acid at
2.61, shown to be involved in species differences between gpH1R
and hH1R [59], is too far away from the oxazepines and oxepines
in the binding pocket, thus an influence of this amino acid is not
expected. However, this was experimentally proven by mutagene-
sis studies and pharmacological characterization of the respective
mutants.
pounds 7a–7f, 16b, 16c, 16f and 16l were determined at selected
GPCRs (Fig. 13, Table 7). The affinities of most of the compounds to
the analyzed receptors were in a pKi range from 6 to 8.
The data show that the dibenzo[b,f][1,4]oxazepine deriva-
tives 7a–7f have a high affinity to the human 5-HT2A receptor
(h5-HT2AR) in
a pKi range of 7.8 to 9.3. Compared to the
dibenzo[b,f][1,4]oxazepine derivatives 7a–7c, 7e, 7f the compound
7d had a high affinity in a pKi range of 7.2 to 8.7 at all of the analyzed
non-histamine receptors. Compound 7e, which was identified as
the one with the highest affinities to the hH1R and the H4R, had
an affinity in a comparable range at the hH1R and the h5-HT2A
R
It is a challenge to identify compounds with high affinity to both
hH1R and hH4R, and no or moderate affinity to the hH2R and hH3R.
A reason for this is the high homology between the amino acids
forming the orthosteric binding site of hH4R and hH3R and the
low homology between the amino acids forming the binding site
hH1R and hH4R. However, some ligand classes, showing affinity
to the hH1R and hH4R, e.g. quinazoline 17 [18] (pKi: hH1R: 7.70
[18], 6.26 [17]; hH4R: 8.12 [18], 7.37 [17]), aminopyrimidine 18
(pKi: hH1R: 5.95; hH4R: 6.02) [19] and astemizole derivative 19
(pKi: hH1R: 7.07; hH4R: 5.61) [17] were described (cf. Supporting
information). At the hH4R, the quinazoline 17 and the aminopy-
rimidine 18 are described as inverse agonists at the hH4R [18,19],
whereas the astemizole derivative 19 and the oxazepine VUF6884
3 are described as partial agonists [17,22].
(Fig. 14). Additionally, the affinity of 7e was in a comparable range at
the hH4R and all investigated non-histamine-receptors except the
porcine 5-HT1AR (p5-HT1AR) and h5-HT2AR (Fig. 14). The oxazepine
derivatives 7e and 7f showed a similar pharmacological profile at
the analyzed aminergic GPCRs as cyproheptadine (Fig. 14). Cypro-
heptadine, 7e and 7f showed high affinity (pKi > 8) to the hH1R
[72] and the h5-HT2AR [73], compared to other aminergic GPCRs
(Fig. 14). However, the selectivity of 7e and 7f towards the h5-
HT2AR compared to the p5-HT1AR (∼2700 (7e), ∼8100 (7f)) was
considerably higher than for cyproheptadine (∼35). Furthermore,
the data showed, compared to cyproheptadine, a decrease in affin-
ity of 7e and 7f to the D3R, M2R and M3R, and an increase to the
hH4R. However, within the oxazepine series, compound 7e was the
only one with high affinity to the hH1R and the h5-HT2AR (pKi ∼ 9),
but moderate affinity to the other analyzed GPCRs (pKi ≤ 7).
The compounds 16b, 16c, 16f and 16l showed good affinity to
the h5-HT2AR, and they were more than 100 fold selective towards
the h5-HT2AR, compared to the p5-HT1AR. Whereas for 16b no
selectivity between the hH1R and h5-HT2AR was observed, the com-
pounds 16c, 16f and 16l showed a preference of the hH1R (∼50 (16c,
3.9. Specificity of selected compounds at selected aminergic
GPCRs
In order to study the specificity with regard to other aminergic
GPCRs, the binding affinity of clozapine (1) and the selected com-

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
623
16l), ∼20 (16f)). However, within the oxepine series, compound
References
16l showed high affinity to the hH1R, but moderate affinities to the
other analyzed GPCRs (pKi < 7).
[1] J.P. Overington, B. Al-Lazikani, A.L. Hopkins, How many drug targets are
there? Nat. Rev. Drug Discov. 5 (12) (2006) 993–996.
Thus, within the oxepine series, 16l was identified as selec-
tive H1 receptor antagonist, whereas within the oxazepine series
7e was identified as a dual H1/5-HT2A receptor ligand. These
results suggest that on one hand, the oxazepine/oxepine moi-
ety and on the other hand the chlorine substitution pattern acts
as a toggle switch between selective H1R antagonists and dual
H1/5-HT2A receptor ligands. This may be of importance for the
development of dual H1/5-HT2A antagonists, because structurally
related oxepine derivatives, e.g. 1-(11 H-spiro[cyclopentane-
1,10 -dibenzo[b,f]oxepin]-3-yl)azetidine-3-carboxylic acid, were
described as dual H1/5-HT2A receptor antagonists for the potential
treatment of sleep disorders [75].
[2] E. Jacoby, R. Bouhelal, M. Gerspacher, K. Seuwen, The 7 TM G-protein-coupled
receptor target family, ChemMedChem 1 (8) (2006) 761–782.
[3] B. Jassal, S. Jupe, M. Caudy, E. Birney, L. Stein, H. Hermjakob, P. D’Eustachio,
The systematic annotation of the three main GPCR families in Reactome,
Database (Oxford) 2010 (2010) baq018.
[4] N. Moguilevsky, F. Varsalona, J.-P. Guillaume, M. Noyer, M. Gillard, J. Daliers,
J.-P. Henichart, A. Bollen, Pharmacological and functional characterisation of
the wild-type and site-directed mutants of the human H1 histamine receptor
stably expressed in CHO cells, J. Recept. Sign. Transd. Res. 15 (1995) 91–102.
[5] T. Oda, S. Matsumoto, Identification and characterization of histamine H4
receptor, Yakugaku Zasshi 118 (1) (2001) 36–42.
[6] T. Nakamura, H. Itadani, Y. Hidaka, M. Ohta, K. Tanaka, Molecular cloning and
characterization of a new human histamine receptor, hH4R, Biochem,
Biophys. Res. Commun. 279 (2) (2000) 615–620.
[7] R.L. Thurmond, E.W. Gelfand, P.J. Dunford, The role of histamine H1 and H4
receptors in allergic inflammation: the search for new antihistamines, Nat.
Rev. Drug Discov. 7 (1) (2008) 41–53.
[8] K.F. Deml, S. Beermann, D. Neumann, A. Strasser, R. Seifert, Interactions of
histamine H1-receptor agonists and antagonists with the human histamine
H4-receptor, Mol. Pharmacol. 76 (5) (2009) 1019–1030.
4. Conclusion
[9] Y. Ohsawa, N. Hirasawa, The role of histamine H1 and H4 receptors in atopic
dermatitis: from basic research to clinical study, Allergol. Int. 63 (4) (2014)
533–542.
[10] A. Deiteren, J.G. De Man, N.E. Ruyssers, T.G. Moreels, P.A. Pelckmans, B.Y. De
Winter, Histamine H4 and H1 receptors contribute to postinflammatory
visceral hypersensitivity, Gut 63 (12) (2014) 1873–1882.
[11] A. Kollmeier, K. Francke, B. Chen, P.J. Dunford, A.J. Greenspan, Y. Xia, X.L. Xu, B.
Zhou, R.L. Thurmond, The histamine H4 receptor antagonist, JNJ 39758979, is
effective in reducing histamine-induced pruritus in a randomized clinical
study in healthy subjects, J. Pharmacol. Exp. Ther. 350 (1) (2014) 181–187.
[12] R.L. Thurmond, The histamine H4 receptor: from orphan to the clinic, Front.
Pharmacol. 6 (2015) 65.
[13] P. Panula, P.L. Chazot, M. Cowart, R. Gutzmer, R. Leurs, W.L. Liu, H. Stark, R.L.
Thurmond, H.L. Haas, International Union of Basic and Clinical Pharmacology.
XCVIII. Histamine receptors, Pharmacol. Rev. 67 (3) (2015) 601–655.
[14] C.M. Marson, Targeting the histamine H4 receptor, Chem. Rev. 111 (11) (2011)
7121–7156.
[15] R. Seifert, A. Strasser, E.H. Schneider, D. Neumann, S. Dove, A. Buschauer,
Molecular and cellular analysis of human histamine receptor subtypes,
Trends Pharmacol. Sci. 34 (1) (2013) 33–58.
[16] A. Strasser, H.-J. Wittmann, A. Buschauer, E.H. Schneider, R. Seifert,
Species-dependent activities of GPCR ligands: lessons from histamine
receptor orthologs, Trends Pharmacol. Sci. 34 (1) (2013) 13–32.
[17] E. Wagner, H.-J. Wittmann, S. Elz, A. Strasser, Pharmacological profile of
astemizole-derived compounds at the histamine H1 and H4 receptor – H1/H4
receptor selectivity, Naunyn Schmiedebergs Arch. Pharmacol. 387 (2014)
235–250.
[18] R.A. Smits, I.J. de Esch, O.P. Zuiderveld, J. Broeker, K. Sansuk, E. Guaita, G.
Coruzzi, M. Adami, E. Haaksma, R. Leurs, Discovery of quinazolines as
histamine H4 receptor inverse agonists using a scaffold hopping approach, J.
Med. Chem. 51 (24) (2008) 7855–7865.
[19] S.G. Hammer, S. Gobleder, F. Naporra, H.-J. Wittmann, S. Elz, M.R. Heinrich, A.
Strasser, 2,4-Diaminopyrimidines as dual ligands at the histamine H1 and H4
receptor-H1/H4-receptor selectivity, Bioorg. Med. Chem. Lett. 26 (2) (2016)
292–300.
[20] B.L. Roth, D.J. Sheffler, W.K. Kroeze, Magic shotguns versus magic bullets:
selectively non-selective drugs for mood disorders and schizophrenia, Nat.
Rev. Drug Discov. 3 (4) (2004) 353–359.
[21] H. Appl, T. Holzammer, S. Dove, E. Haen, A. Strasser, R. Seifert, Interactions of
recombinant human histamine H1R, H2R, H3R, and H4R receptors with 34
antidepressants and antipsychotics, Naunyn Schmiedebergs Arch. Pharmacol.
385 (2) (2012) 145–170.
[22] R.A. Smits, H.D. Lim, B. Stegink, R.A. Bakker, I.J. de Esch, R. Leurs,
Characterization of the histamine H4 receptor binding site. Part 1. Synthesis
and pharmacological evaluation of dibenzodiazepine derivatives, J. Med.
Chem. 49 (15) (2006) 4512–4516.
[23] A. Jongejan, H.D. Lim, R.A. Smits, I.J. de Esch, E. Haaksma, R. Leurs, Delineation
of agonist binding to the human histamine H4 receptor using mutational
analysis, homology modeling, and ab initio calculations, J. Chem. Inf. Model.
48 (7) (2008) 1455–1463.
Within the present work, 32 oxazepines and oxepines were
pharmacologically characterized at the hH1R and the hH4R. The
oxazepine derivatives showed (high) affinity to both receptors.
However, depending on the position of one or two chlorine atoms
in the tricyclic core and methylation of the piperazine moiety, they
showed up to ∼5000 fold selectivity towards the hH1R. All the
oxazepine derivatives were antagonists at the hH1R, but partial
agonists at the hH4R. In contrast, the oxepine derivatives showed
high affinity to the hH1R, but no or only moderate affinity to the
hH4R. As expected, considering the lead structure doxepine, in gen-
eral the oxepine derivatives showed inverse agonism at the hH4R,
but due to the low potency, this is only the case at ligand con-
centrations ≥10 ␮M. For one oxepine derivative, the racemate was
separated into its enantiomers, and the (R)-enantiomer was iden-
tified as the eutomer at the hH1R and the gpH1R. Furthermore,
all analyzed compounds showed affinity in the three to one digit
nM range to the h5-HT2AR and the hM1R depending on the chlo-
rine substitution pattern. Similar to clozapine, some compounds
showed also affinity to other aminergic GPCRs, demonstrating the
importance to characterize new compounds not only at the GPCR
of interest, but also at other GPCRs. Summing up, the oxazepine and
oxepine derivatives turned out to be valuable compounds to study
aminergic GPCRs on a molecular level and to understand specificity
and selectivity between those receptors.
Conflict of interest
All authors declare that there is no conflict of interest.
Acknowledgments
We are grateful to Christine Gebhardt-Braun, Kerstin Röhrl,
Maria Beer-Krön, Susanne Bollwein, Sieglinde Dechant for expert
technical assistance and Andreas Gruber (electronic workshop) for
repair of the CE apparatus. Furthermore, we would like to thank
Prof. Buschauer for his support. This work was supported by the
DFG (Deutsche Forschungsgemeinschaft) research training group
(GRK 1910) “Medicinal Chemistry of selective GPCR ligands”.
[24] H.D. Lim, R.M. van Rijn, P. Ling, R.A. Bakker, R.L. Thurmond, R. Leurs,
Evaluation of histamine H1-, H2-, and H3-receptor ligands at the human
histamine H4 receptor: identification of 4-methylhistamine as the first potent
and selective H4 receptor agonist, J. Pharmacol. Exp. Ther. 314 (3) (2005)
1310–1321.
[25] N. Terzioglu, R.M. van Rijn, R.A. Bakker, I.J. De Esch, R. Leurs, Synthesis and
structure-activity relationships of indole and benzimidazole piperazines as
histamine H4 receptor antagonists, Bioorg. Med. Chem. Lett. 14 (21) (2004)
5251–5256.
[26] A. Schuster, G. Bernhardt, E. Eibler, A. Buschauer, W. Hesselink, Chiral
separation of 3-phenyl-3-(2-pyridyl)propylamines, and analogous guanidines
and guanidine-N-carboxylic acid esters with high-performance liquid
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.phrs.2016.09.
042.

624
F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
chromatography and capillary zone electrophoresis, J. Chromatogr. A 793 2-heteroarylhistamines, and analogues, J. Med. Chem. 38 (8) (1995)
(1998) 77–90.
1287–1294.
[27] T.F. Walseth, R.A. Johnson, The enzymatic preparation of [␣-32P]nucleoside
triphosphates, cyclic [32P]AMP, and cyclic [32P]GMP, Biochim. Biophys. Acta
562 (1) (1979) 11–31.
[51] R.R. Neubig, M. Spedding, T. Kenakin, International Union of Pharmacology
Committee on Receptor Nomenclature and Drug Classification, International
Union of Pharmacology Committee on Receptor Nomenclature and Drug
Classification. XXXVIII. Update on terms and symbols in quantitative
pharmacology, Pharmacol. Rev. 55 (4) (2003) 597–606.
[28] R. Seifert, K. Wenzel-Seifert, T. Bürckstümmer, H.H. Pertz, W. Schunack, S.
Dove, A. Buschauer, S. Elz, Multiple differences in agonist and antagonist
pharmacology between human and guinea-pig histamine H1-receptor, J.
Pharmacol. Exp. Ther. 305 (2003) 1104–1115.
[52] O. Arunlakshana, H.O. Schild, Some quantitative uses of drug antagonists, Br. J.
Pharmacol. Chemother. 14 (1) (1959) 48–58.
[29] E.H. Schneider, A. Strasser, R.L. Thurmond, R. Seifert, Structural requirements
for inverse agonism and neutral antagonism of indole-, benzimidazole-, and
thienopyrrole-derived histamine H4 receptor ligands, J. Pharmacol. Exp. Ther.
334 (2) (2010) 513–521.
[53] T. Shimamura, M. Shiroishi, S. Weyand, H. Tsujimoto, G. Winter, V. Katritch, R.
Abagyan, V. Cherezov, W. Liu, G.W. Han, T. Kobayashi, R.C. Stevens, S. Iwata,
Structure of the human histamine H1 receptor complex with doxepin, Nature
475 (7354) (2011) 65–U82.
[30] N. Pop, P. Igel, A. Brennauer, C. Cabrele, G.N. Bernhardt, R. Seifert, A.
Buschauer, Functional reconstitution of human neuropeptide Y (NPY) Y2 and
Y4 receptors in Sf9 insect cells, J. Recept. Signal Trans. 31 (4) (2011) 271–285.
[31] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with
the Folin phenol reagent, J. Biol. Chem. 193 (1) (1951) 265–275.
[32] N. Kagermeier, K. Werner, M. Keller, P. Baumeister, G. Bernhardt, R. Seifert, A.
Buschauer, Dimeric carbamoylguanidine-type histamine H2 receptor ligands:
a new class of potent and selective agonists, Bioorg. Med. Chem. 23 (14)
(2015) 3957–3969.
[33] S. Cuboni, C. Devigny, B. Hoogeland, A. Strasser, S. Pomplun, B. Hauger, G.
Hofner, K.T. Wanner, M. Eder, A. Buschauer, F. Holsboer, F. Hausch, Loratadine
and analogues: discovery and preliminary structure-activity relationship of
inhibitors of the amino acid transporter B(0)AT2, J. Med. Chem. 57 (22) (2014)
9473–9479.
[54] H.-J. Wittmann, R. Seifert, A. Strasser, Sodium binding to hH3R and hH4R—a
molecular modeling study, J. Mol. Model. 20 (8) (2014) 2394–2404.
[55] C. Oostenbrink, A. Villa, A.E. Mark, W.F. van Gunsteren, A biomolecular force
field based on the free enthalpy of hydration and solvation: the GROMOS
force-field parameter sets 53A5 and 53A6, J. Comput. Chem. 25 (13) (2004)
1656–1676.
[56] Q. Chao, K.G. Sprankle, R.M. Grotzfeld, A.G. Lai, T.A. Carter, A.M. Velasco, R.N.
Gunawardane, M.D. Cramer, M.F. Gardner, J. James, P.P. Zarrinkar, H.K. Patel,
S.S. Bhagwat, Identification of N-(5-tert-butyl-isoxazol-3-yl)-N’-{4-[7-(2-
morpholin-4-yl-ethoxy)imidazo[2,1-b][1,3]benzothiazol-2-yl]phenyl}urea
dihydrochloride (AC220), a uniquely potent, selective, and efficacious
FMS-like tyrosine kinase-3 (FLT3) inhibitor, J.Med. Chem. 52 (23) (2009)
7808–7816.
[57] M. Kurokawa, Y. Masuda, S. Naruto, F. Sato, H. Uno, Dibenz[b, e]oxepin
derivatives and pharmaceutical composition containing the same, European
patent application PCT/JP87/00392, DE3788195D1 (1993).
[58] S. Ahuja, S.O. Smith, Multiple switches in G protein-coupled receptor
activation, Trends Pharmacol. Sci. 30 (9) (2009) 494–502.
[34] A. Strasser, B. Striegl, H.-J. Wittmann, R. Seifert, Pharmacological profile of
histaprodifens at four recombinant H1-receptor species isoforms, J.
Pharmacol. Exp. Ther. 324 (2008) 60–71.
[35] D. Wifling, K. Löffel, U. Nordemann, A. Strasser, G. Bernhardt, S. Dove, R.
Seifert, A. Buschauer, Molecular determinants for the high constitutive
activity of the human histamine H4 receptor: functional studies on
orthologues and mutants, Br. J. Pharmacol. 172 (2015) 785–798.
[36] Y. Cheng, W.H. Prusoff, Relationship between the inhibition constant (KI ) and
the concentration of inhibitor which causes 50 percent inhibition (I50) of an
enzymatic reaction, Biochem. Pharmacol. 22 (23) (1973) 3099–3108.
[37] E.H. Schneider, D. Schnell, D. Papa, R. Seifert, High constitutive activity and a
G-protein-independent high-affinity state of the human histamine
H4-receptor, Biochemistry 48 (6) (2009) 1424–1438.
[38] H. Hübner, C. Haubmann, W. Utz, P. Gmeiner, Conjugated enynes as
nonaromatic catechol bioisosteres: synthesis, binding experiments, and
computational studies of novel dopamine receptor agonists recognizing
preferentially the D3 subtype, J. Med. Chem. 43 (4) (2000) 756–762.
[39] G. Hayes, T.J. Biden, L.A. Selbie, J. Shine, Structural subtypes of the dopamine
D2 receptor are functionally distinct: expression of the cloned D2A and D2B
subtypes in a heterologous cell line, Mol. Endocrinol. 6 (6) (1992) 920–926.
[40] P. Sokoloff, B. Giros, M.P. Martres, M.L. Bouthenet, J.C. Schwartz, Molecular
cloning and characterization of a novel dopamine receptor (D3) as a target for
neuroleptics, Nature 347 (6289) (1990) 146–151.
[59] M. Bruysters, A. Jongejan, M. Gillard, F. van de Manakker, R.A. Bakker, P.
Chatelain, R. Leurs, Pharmacological differences between human and guinea
pig histamine H1 receptors: Asn84 (2.61) as key residue within an additional
binding pocket in the H1 receptor, Mol. Pharmacol. 67 (2005) 1045–1052.
[60] K. Wieland, A.M. Ter Laak, M.J. Smit, R. Kühne, H. Timmerman, R. Leurs,
Mutational analysis of the antagonist-binding site of the histamine H1
receptor, J. Biol. Chem. 274 (1999) 29994–30000.
[61] I. Brunskole, A. Strasser, R. Seifert, A. Buschauer, Role of the second and third
extracellular loops of the histamine H4 receptor in receptor activation,
Naunyn Schmiedebergs Arch. Pharmacol. 384 (3) (2011) 301–317.
[62] H.D. Lim, A. Jongejan, R.A. Bakker, E. Haaksma, I.J. de Esch, R. Leurs,
Phenylalanine 169 in the second extracellular loop of the human histamine H4
receptor is responsible for the difference in agonist binding between human
and mouse H4 receptors, J. Pharmacol. Exp. Ther. 327 (1) (2008) 88–96.
[63] A. Strasser, H.-J. Wittmann, R. Seifert, Ligand-specific contribution of the
N-terminus and E2-loop to pharmacological properties of the histamine
H1-receptor, J. Pharmacol. Exp. Ther. 326 (2008) 783–791.
[64] R. Leurs, M.J. Smit, C.P. Tensen, A.M. Ter Laak, H. Timmerman, Site-directed
mutagenesis of the histamine H1-receptor reveals a selective interaction of
asparagine207 with subclasses of H1-receptor agonists, Biochem. Biophys.
Res. Commun. 201 (1994) 295–301.
[65] K. Ohta, H. Hayashi, H. Mizuguchi, H. Kagamiyama, K. Fujimoto, H. Fukui,
Site-directed mutagenesis of the histamine H1 receptor: roles of aspartic
acid107 asparagine198 and threonine194, Biochem. Biophys. Res. Commun.
203 (1994) 1096–1101.
[66] M. Gillard, C. Van Der Perren, N. Moguilevsky, R. Massingham, P. Chatelain,
Binding characteristics of cetirizine and levocetirizine to human H1 histamine
receptors: contribution of Lys191 and Thr194, Mol. Pharmacol. 61 (2) (2002)
391–399.
[67] A. Jongejan, R. Leurs, Delineation of receptor-ligand interactions at the human
histamine H1 receptor by a combined approach of site-directed mutagenesis
and computational techniques – or – how to bind the H1 receptor, Arch.
Pharm. (Weinheim) 338 (5–6) (2005) 248–259.
[68] H.-J. Wittmann, R. Seifert, A. Strasser, Influence of the N-terminus and the
E2-loop onto the binding kinetics of the antagonist mepyramine and the
partial agonist phenoprodifen to H1R, Biochem. Pharmacol. 82 (12) (2011)
1910–1918.
[69] D. Schnell, A. Strasser, R. Seifert, Comparison of the pharmacological
properties of human and rat histamine H3-receptors, Biochem. Pharmacol. 80
(9) (2010) 1437–1449.
[41] V. Asghari, S. Sanyal, S. Buchwaldt, A. Paterson, V. Jovanovic, H.H. Van Tol,
Modulation of intracellular cyclic AMP levels by different human dopamine
D4 receptor variants, J. Neurochem. 65 (3) (1995) 1157–1165.
[42] N. Tschammer, J. Elsner, A. Goetz, K. Ehrlich, S. Schuster, M. Ruberg, J.
Kuhhorn, D. Thompson, J. Whistler, H. Hübner, P. Gmeiner, Highly potent
5-aminotetrahydropyrazolopyridines: enantioselective dopamine D3
receptor binding, functional selectivity, and analysis of receptor-ligand
interactions, J. Med. Chem. 54 (7) (2011) 2477–2491.
[43] C. Houston, K. Wenzel-Seifert, T. Bürckstummer, R. Seifert, The human
histamine H2-receptor couples more efficiently to Sf9 insect cell Gs-proteins
than to insect cell Gq-proteins: limitations of Sf9 cells for the analysis of
receptor/Gq-protein coupling, J. Neurochem. 80 (4) (2002) 678–696.
[44] K. Wenzel-Seifert, M.T. Kelley, A. Buschauer, R. Seifert, Similar apparent
constitutive activity of human histamine H2-receptor fused to long and short
splice variants of Gs␣, J. Pharmacol. Exp. Ther. 299 (3) (2001) 1013–1020.
[45] A. Rouleau, X. Ligneau, J. Tardivel-Lacombe, S. Morisset, F. Gbahou, J.C.
Schwartz, J.M. Arrang, Histamine H3-receptor-mediated [35S]GTP([S] binding:
evidence for constitutive activity of the recombinant and native rat and
human H3 receptors, Br. J. Pharmacol. 135 (2) (2002) 383–392.
[46] D. Schnell, K. Burleigh, J. Trick, R. Seifert, No evidence for functional selectivity
of proxyfan at the human histamine H3 receptor coupled to defined Gi/Go
protein heterotrimers, J. Pharmacol. Exp. Ther. 332 (3) (2010) 996–1005.
[47] D. Wifling, G. Bernhardt, S. Dove, A. Buschauer, The extracellular loop 2 (ECL2)
of the human histamine H4 receptor substantially contributes to ligand
binding and constitutive activity, PLoS One 10 (1) (2015) e0117185.
[48] S. Elz, K. Kramer, H.H. Pertz, H. Detert, A.M. ter Laak, R. Kühne, W. Schunack,
Histaprodifens: synthesis, pharmacological in vitro evaluation, and molecular
modeling of a new class of highly active and selective histamine H1-receptor
agonists, J. Med. Chem. 43 (6) (2000) 1071–1084.
[70] B.B. Yao, C.W. Hutchins, T.L. Carr, S. Cassar, J.N. Masters, Y.L. Bennani, T.A.
Esbenshade, A.A. Hancock, Molecular modelling and pharmacological analysis
of species-related histamine H3 receptor heterogeneity, Neuropharmacology
44 (2003) 773–786.
[71] T.W. Lovenberg, J. Pyati, H. Chang, S.J. Wilson, M.G. Erlander, Cloning of rat
histamine H3 receptor reveals distinct species pharmacological profiles, J.
Pharmacol. Exp. Ther. 293 (3) (2000) 771–778.
[72] N. Moguilevsky, F. Varsalona, M. Noyer, M. Gillard, J.P. Guillaume, L. Garcia, C.
Szpirer, J. Szpirer, A. Bollen, Stable expression of human
[49] H. Pertz, S. Elz, In-vitro pharmacology of sarpogrelate and the enantiomers of
its major metabolite: 5-HT2A receptor specificity, stereoselectivity and
modulation of ritanserin-induced depression of 5-HT contractions in rat tail
artery, J. Pharm. Pharmacol. 47 (4) (1995) 310–316.
[50] C. Leschke, S. Elz, M. Garbarg, W. Schunack, Synthesis and histamine H1
receptor agonist activity of a series of 2-phenylhistamines,
H1-histamine-receptor cDNA in Chinese hamster ovary cells. Pharmacological
characterisation of the protein, tissue distribution of messenger RNA and
chromosomal localisation of the gene, Eur. J. Biochem. 224 (2) (1994)
489–495.
[73] P.K. Gillman, Monoamine oxidase inhibitors, opioid analgesics and serotonin
toxicity, Br. J. Anaesth. 95 (4) (2005) 434–441.

F. Naporra et al. / Pharmacological Research 113 (2016) 610–625
625
[74] L. Toll, I.P. Berzetei-Gurske, W.E. Polgar, S.R. Brandt, I.D. Adapa, L. Rodriguez,
R.W. Schwartz, D. Haggart, A. O’Brien, A. White, J.M. Kennedy, K. Craymer, L.
Farrington, J.S. Auh, Standard binding and functional assays related to
medications development division testing for potential cocaine and opiate
narcotic treatment medications, NIDA Res. Monogr. 178 (1998) 440–466.
[75] M. Gianotti, M. Botta, S. Brough, R. Carletti, E. Castiglioni, C. Corti, M. Dal-Cin,
S. Delle Fratte, D. Korajac, M. Lovric, G. Merlo, M. Mesic, F. Pavone, L. Piccoli, S.
Rast, M. Roscic, A. Sava, M. Smehil, L. Stasi, A. Togninelli, M.J. Wigglesworth,
Novel spirotetracyclic zwitterionic dual H1/5-HT2A receptor antagonists for
the treatment of sleep disorders, J. Med. Chem. 53 (21) (2010) 7778–7795.
[76] T. Stanton, C. Bolden-Watson, B. Cusack, E. Richelson, Antagonism of the five
cloned human muscarinic cholinergic receptors expressed in CHO-K1 cells by
antidepressants and antihistaminics, Biochem. Pharmacol. 45 (11) (1993)
2352–2354.
[77] H.H. Pertz, T. Görnemann, B. Schurad, R. Seifert, A. Strasser, Striking
differences of action of lisuride stereoisomers at histamine H1 receptors,
Naunyn Schmiedebergs Arch. Pharmacol. 374 (3) (2006) 215–222.
[78] H.-J. Wittmann, S. Elz, R. Seifert, A. Strasser, N␣-methylated phenylhistamines
exhibit affinity to the hH4R—a pharmacological and molecular modelling
study, Naunyn Schmiedebergs Arch. Pharmacol. 384 (3) (2011) 287–299.