Ring-substituted histaprodifen analogues as partial agonists for histamine H1 receptors: Synthesis and structure-activity relationships

Article abstract of DOI:10.1016/S0223-5234(00)00105-7

Thirteen racemic benzene ring-substituted analogues of histaprodifen (8a; 2-[2-(3,3-diphenylpropyl)-1H-imidazol-4-yl]ethanamine), a novel lead for potent and selective histamine H₁-receptor agonists, have been prepared from substituted 4,4-diphenylbutyronitriles 5 via cyclization of the corresponding methyl butyrimidates 6 with 2-oxo-4-phthalimido-1-butyl acetate in liquid ammonia, followed by deprotection. Nitriles 5 were accessible by alkylation of either substituted diphenylmethanes with 3-bromopropionitrile or diethyl malonate with substituted 1-chloro-diphenylmethanes and subsequent standard reactions. The title compounds 8 displayed partial agonism on conctractile H₁ receptors of the guinea-pig ileum (E(max) = 2-98% relative to histamine) and, compared with the endogenous agonist, were endowed with agonist potencies of 4-92%. The meta fluorinated (8c) and meta chlorinated (8f) analogues showed the highest relative potency in this series (95% confidence limits 85-99% and 78-102%), but did not exceed the value of the lead 8a (99- 124%). Compound 8c (2-[2-[3-(3-fluorophenyl)-3-phenylpropyl]-1H-imidazol-4- yl]ethanamine) was a partial agonist at contractile H₁ receptors of the guinea-pig aorta (relative potency 154% vs. 100% for histamine) and at relaxation-mediating endothelial H₁ receptors of the rat aorta (relative potency 556% vs. 100% for histamine) and matched with the functional behaviour of 8a. Agonism observed for each compound was sensitive to blockade by the selective H₁-receptor antagonist mepyramine (pA₂ ? 9 (guinea-pig) and pA₂ ? 8 (rat aorta)). All histaprodifen analogues 8 stimulated neither histaminergic H₂/H₃ nor cholinergic M₃ receptors. They displayed only low to moderate affinity for these sites (H₂: pD'₂ < 5; H₃/M3: pA₂ < 6). With regard to the substitution pattern on the benzene ring, there was no correlation between the histaprodifen series and the corresponding derivatives of another selective H₁-receptor agonist, viz. 2- phenylhistamine. (C) 2000 Editions scientifiques et medicales Elsevier SAS.

Full text of DOI:10.1016/S0223-5234(00)00105-7

Eur. J. Med. Chem. 35 (2000) 41−52
© 2000 Éditions scientifiques et médicales Elsevier SAS. All rights reserved
S0223523400001057/FLA
41
Original article
Ring-substituted histaprodifen analogues as partial agonists for histamine H1
receptors: synthesis and structure–activity relationships#
Sigurd Elz, Kai Kramer, Christian Leschke, Walter Schunack*
Institut für Pharmazie, Freie Universität Berlin, Königin-Luise-Strasse 2 + 4, D-14195 Berlin (Dahlem), Germany
Received 14 June 1999; accepted 2 July 1999
Abstract – Thirteen racemic benzene ring-substituted analogues of histaprodifen (8a; 2-[2-(3,3-diphenylpropyl)-1H-imidazol-4-
yl]ethanamine), a novel lead for potent and selective histamine H1-receptor agonists, have been prepared from substituted 4,4-
diphenylbutyronitriles 5 via cyclization of the corresponding methyl butyrimidates 6 with 2-oxo-4-phthalimido-1-butyl acetate in liquid
ammonia, followed by deprotection. Nitriles 5 were accessible by alkylation of either substituted diphenylmethanes with 3-bromopropionitrile
or diethyl malonate with substituted 1-chloro-diphenylmethanes and subsequent standard reactions. The title compounds 8 displayed partial
agonism on contractile H1 receptors of the guinea-pig ileum (Emax = 2–98% relative to histamine) and, compared with the endogenous agonist,
were endowed with agonist potencies of 4–92%. The meta fluorinated (8c) and meta chlorinated (8f) analogues showed the highest relative
potency in this series (95% confidence limits 85–99% and 78–102%), but did not exceed the value of the lead 8a (99–124%). Compound 8c
(2-[2-[3-(3-fluorophenyl)-3-phenylpropyl]-1H-imidazol-4-yl]ethanamine) was a partial agonist at contractile H1 receptors of the guinea-pig
aorta (relative potency 154% vs. 100% for histamine) and at relaxation-mediating endothelial H1 receptors of the rat aorta (relative potency
556% vs. 100% for histamine) and matched with the functional behaviour of 8a. Agonism observed for each compound was sensitive to
blockade by the selective H1-receptor antagonist mepyramine (pA2 ≈ 9 (guinea-pig) and pA2 ≈ 8 (rat aorta)). All histaprodifen analogues 8
stimulated neither histaminergic H2/H3 nor cholinergic M3 receptors. They displayed only low to moderate affinity for these sites (H2: pD’2
< 5; H3/M3: pA2 < 6). With regard to the substitution pattern on the benzene ring, there was no correlation between the histaprodifen series
and the corresponding derivatives of another selective H1-receptor agonist, viz. 2-phenylhistamine. © 2000 Éditions scientifiques et médicales
Elsevier SAS
guinea-pig aortic H1 receptor / H1-receptor agonists / histamine H1 receptor / histaprodifen / rat aortic H1 receptor
1. Introduction
time [3–5]. The first selective compounds, which drew
even with histamine, became available when the aromatic
portion of 2-phenylhistamine (figure 1), a moderately
active but selective H1-receptor agonist [6, 7], was sub-
stituted systematically in the meta position, preferentially
by halogen or halogen-containing groups [7, 8]. The
relative potency of the most prominent member of this
series of primary amines, 2-(3-trifluoromethylphenyl)-
histamine (figure 1) [7], was recently improved by Nα-
methylation of the side-chain nitrogen, leading to the
secondary amine Nα-methyl-2-(3-trifluoromethylphenyl)-
histamine (figure 1) which acts as a full H1-receptor
agonist on the guinea-pig ileum and displays a relative
potency of 174% compared with histamine [9].
The important role of the biogenic amine histamine
(figure 1) in causing a large number of physiological and
pathophysiological effects via interaction with histamine
H1, H2 and H3 receptors has been well established [2].
Over the years highly potent and selective antagonists for
the three histamine receptor subtypes have become avail-
able, as well as highly potent agonist ligands for H2 and
H3 receptors [2]. By contrast, efforts of several groups to
produce selective H1-receptor agonists endowed with
potency superior to histamine have failed for a long
#Presented in part: XIIIth International Congress of Pharmaco-
logy, Munich, July 26–31, 1998 [1]
*Correspondence and reprints:
A structurally divergent family of H1-receptor agonists
have been discovered in the late nineties after replacing
the 2-phenyl substituent of the imidazole nucleus by a
sieker@schunet.pharmazie.fu-berlin.de

42
solution of diphenylmethanes 2b, h, i and n with
3-bromopropionitrile (method A), or by alkylation of
diethyl
malonate
with
substituted
chloro-
diphenylmethanes 3c–g and j–m. The alkylated diethyl
malonates were decarboxylated, subsequently reduced
and chlorinated to give the 3,3-diphenyl-1-chloro-
propanes 4c–g and j–m, which afforded the desired
nitriles by Kolbe nitrile synthesis (method B). Compound
4o was prepared from commercially available 3,3-di-(4-
fluorophenyl)propanoic acid by reduction to the corre-
sponding propanol with lithium aluminum hydride, fol-
lowed by chlorination with thionylchloride. The addition
of dry methanol to the nitriles 5b–o led to the methyl
imidates 6b–o upon treatment with an excess of thionyl-
chloride. The synthesis of the substituted, side-chain
protected histaprodifen derivatives 7b–o was performed
by cyclocondensation of 6b–o in liquid ammonia with
2-oxo-4-phthalimido-1-butyl acetate which was available
in four steps from butin-1,4-diol [8]. After deprotection
by acidic hydrolysis or hydrazinolysis, the target com-
pounds 8b–o were purified by chromatography and crys-
tallized as dihydrogen maleates (table I). Histaprodifen
(8a) was available according to method A using commer-
cially available diphenylmethane (2a) [11].
Figure 1. Structure of histamine and prominent H1-receptor
agonists.
3. Pharmacology
3,3-diphenylpropyl moiety [10, 11]. Prototypic members
of this series are histaprodifen and its Nα-methylated
derivative (figure 1) which possess 101–528% and
343–2 825% relative potency, respectively, in several
functional H1-receptor assays [11]. The aim of the present
study was to evaluate whether structure–activity relation-
ships of substituted histaprodifen derivatives are congru-
ent with those of the 2-phenylhistamine familiy, as far as
the substitution pattern of the benzene rings is concerned.
Provided that both series were in agreement with each
other, this would allow histaprodifen analogues with
enhanced H1-receptor agonist potency to be obtained. For
the sake of a more convenient synthetic access to a larger
series of compounds, only racemic primary amines, i.e.,
analogues of histaprodifen, were synthesized and evalu-
ated in functional in vitro receptor assays.
Histaprodifen (8a) and the new histaprodifen ana-
logues 8b–o were routinely screened on the guinea-pig
ileum preparation to characterize functional H1-receptor
potency and the relative maximum effect (formerly
termed ‘intrinsic activity’ [13]), compared with the stan-
dard agonist histamine. Direct or indirect muscarinic
effects on the ileal muscle were excluded by the presence
of atropine (100 nM) during all experiments. For com-
pounds which produced contraction under these condi-
tions, the susceptibility of the contractile effect to block-
ade by the competitive H1-receptor antagonist
mepyramine was verified. Histaprodifen (8a) itself has
been shown to produce a concentration-dependent right-
ward shift of the histamine concentration–effect curve,
when the compound is incubated for a limited period, just
like a partial agonist [11]. From this experiment the
affinity of 8a was estimated as pKP which is defined as
the negative decadic logarithm of the dissociation con-
stant of a partial-agonist/receptor complex [14, 15]. Simi-
lar experiments with 8b–o allowed the estimation of the
affinity for the majority of the new compounds. Com-
pound 8c, which appeared to be the most interesting
analogue of the study, was additionally characterized by
two other assays predictive of H1-receptor agonism, viz.
the contraction of guinea-pig aorta [11, 16] and the
2. Chemistry
Ring-substituted congeners 8b–o of histaprodifen (8a)
were synthesized by cyclocondensation of a substituted
α-hydroxyketone with the appropriate methyl imidates
6b–o in liquid ammonia under pressure [12] (figure 2).
The nitriles 5b–o required for the acid-catalysed forma-
tion of the methyl imidates 6b–o were either prepared by
direct alkylation of a potassium amide/liquid ammonia-

43
NO-mediated relaxation of precontracted rat aorta [11,
17]. Finally, a functional selectivity profile vis-à-vis to
H2, H3 and M3 receptors was determined for some of the
compounds. In the case of 8c, this profiling was extended
to other monoaminergic receptor interactions.
4. Results and discussion
4.1. In vitro H1-receptor pharmacology
In the presence of the muscarinic receptor antagonist
atropine (100 nM), histaprodifen analogues 8b–o dis-
played partial agonism on the H1 receptor of the guinea-
pig ileum reaching maximum effects in the range of
2–98% relative to histamine (table II). Compared with
histamine (relative potency 100%) and histaprodifen (8a,
111% [11]), the potencies of the new agonists were lower
(4–52%) or equivalent in the case of the meta fluoro (8c,
92%, figure 3, upper panel) and the meta chloro (8f, 89%)
analogue. The concentration–effect curves of all partial
agonists, which reached more than 30% of the maximal
histamine contraction, were shifted to the right by nano-
molar concentrations of the competitive H1-receptor an-
tagonist mepyramine and allowed the calculation of
nanomolar affinities for this antagonist (pA2 range
8.85–9.37, table III), which is in good agreement with
literature data [2]. The contractions evoked by the hista-
prodifen analogues, either in the absence or particularly
in the presence of mepyramine, developed less rapidly
compared with histamine, which may be indicative of
slower binding kinetics. The contractile effect of partial
agonists which produced less than 30% contraction (8j,
m and o) was abolished by 100 nM mepyramine. There-
fore, the ileal contraction evoked by the title compounds
in the presence of 100 nM atropine was specifically
mediated by H1 receptors. Earlier studies on H1-receptor
agonists of the 2-phenylhistamine [9] and the histapro-
difen series [11] have presented evidence that a reason-
able affinity estimate, viz. the pKP value [14, 15], may be
deduced from functional experiments on the guinea-pig
ileum by construction of a second histamine curve in the
presence of the other agonist. This was even possible for
‘full’ agonists of both series (e.g., 8a [11]) after allowing
the contraction to fade time-dependently to a lower level.
Following the same experimental protocol, pKP values
could be calculated for eleven of the new histaprodifen
analogues (table II; figure 3, lower panel for 8c). These
data run parallel to the rank order of agonist potency,
although pKP values are approximately 0.6 logarithmic
units smaller than the corresponding pEC50 values (for
correlation of both parameters see figure 4).
Figure 2. Synthesis of histaprodifen (8a) and substituted ana-
logues 8b–o: (a) LiAlH4, AlCl3, Et2O, ∆; (b) BrC2H4CN,
THF/Et2O, KNH2, liq. NH3; (c) KCN, KI, DMSO, ∆; (d)
MeOH, SOCl2, –30 °C; (e) liq. NH3, 60 °C; (f) 6 N HCl, ∆, or
N2H4, MeOH. The 5 steps leading from 3c–g and j–m to 4c–g,
j–m and o are described in detail in the Experimental protocols
(Method B). For X and X≠, see tables I and II.

44
Table I. Preparative and analytical data of histaprodifens 8b–o.a
Compound
X
X≠
Formula
Molecular
mass
M.p. (°C)
Method
+FAB spectra, m/z
(rel. intensity)
8b
8c
8d
8e
8f
8g
8h
8i
8j
8k
8l
8m
8n
8o
2-F
3-F
4-F
H
H
H
H
H
H
H
H
H
H
H
H
H
4-F
C20H22FN3 2C4H4O4
C20H22FN3 2C4H4O4 0.25H2O
C20H22FN3 2C4H4O4
C20H22ClN3 2C4H4O4
C20H22ClN3 2C4H4O4
C20H22ClN3 2C4H4O4
C20H22BrN3 2C4H4O4
C20H22BrN3 2C4H4O4
C20H22BrN3 2C4H4O4
C21H25N3 2C4H4O4
555.6
559.1
555.6
527.1
527.1
527.1
616.5
616.5
616.5
551.6
551.6
551.6
614.6
573.6
142–143
147–149
147–149
137–139
143–144
134–136
132–134
138–139
144–145
137–139
143–145
141–142
133–134
128–129
A
B
B
B
B
B
A
A
B
B
B
B
A
B
324 (100), 307 (10)
324 (100), 307 (19)
324 (100), 307 (13)
340 (100), 323 (12)
340 (100), 323 (10)
340 (100), 323 (8)
386 (93), 384 (100)
386 (94), 384 (100)
386 (98), 384 (100)
320 (100), 303 (6)
320 (100), 303 (11)
320 (100), 303 (9)
374 (100), 345 (8)
342 (100), 325 (11)
2-Cl
3-Cl
4-Cl
2-Br
3-Br
4-Br
2-CH3
3-CH3
4-CH3
3-CF3
4-F
C21H25N3 2C4H4O4
C21H25N3 2C4H4O4
C21H22F3N3 2C4H4O4 0.5H2O
C20H21F2N3 2C4H4O4
a For structure see figure 2. All compounds gave satisfactory analyses (C, H, N) and spectra.
Compound 8c (‘meta fluorohistaprodifen’) was se-
lected for further characterization due to its equipotency
with 8a in terms of relative potency, maximum effect, and
affinity (table II). Compound 8c contracted guinea-pig
aortic segments in the absence of endothelium in a
mepyramine-sensitive fashion (figure 5, upper panel).
The affinity of mepyramine was also found to be in the
lower nanomolar range. The guinea-pig aorta was less
sensitive to H1-receptor agonists, especially to the full
agonist histamine, which confirms earlier observa-
tions [7, 9, 11]. Relative to histamine, the potency of 8c
was increased by a factor of 1.7 and matched with the
potency of 8a [11]. The attenuated relative maximum
contraction evoked by 8c typifies the partial agonist
Table II. Contraction of guinea-pig ileal whole segments by histamine and histaprodifens 8a–o in the presence of 0.1 µM atropine.
H1-receptor agonism antagonism vs. histamine
Compounda
X
X≠
nb
pEC50 ± SEM
relative 95% confidence Emax ± SEM nb
potency limits
pKPc ± SEM cd [µM]
(t [min])
8a [11]
8b
8c
8d
8e
8f
8g
8h
8i
8j
8k
8l
8m
8n
H
2-F
3-F
4-F
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F
34
10
13
6
6
10
8
11
8
9
8
6
6.74 ± 0.02
5.66 ± 0.06
6.66 ± 0.015
6.33 ± 0.10
6.18 ± 0.06
6.65 ± 0.03
5.72 ± 0.05
6.31 ± 0.05
6.42 ± 0.03
5.61 ± 0.09
6.25 ± 0.04
5.99 ± 0.06
n.d.
111
9
99–124
7–12
85–99
23–79
22–42
78–102
8–14
32–52
43–62
5–13
28–44
14–27
–
100
12
10
10
6
–
–
4
5
7
9
22
6
10
5
–
–
6.04 ± 0.05
4.83 ± 0.05
6.22 ± 0.08
5.58 ± 0.11
n.d.
3–30 (e)
10–20 (3)
3–10 (10)
10 (5)
–
86 ± 2
98 ± 1f
85 ± 3
89 ± 2
93 ± 1h
58 ± 4
86 ± 2
87 ± 2
12 ± 2
82 ± 5
88 ± 3
2 ± 1
92f
43
30
89g
10
41
52
8
2-Cl
3-Cl
4-Cl
2-Br
3-Br
4-Br
2-CH3
3-CH3
4-CH3
3-CF3
F
n.d.
–
5.28 ± 0.04
5.37 ± 0.10
5.52 ± 0.07
5.43 ± 0.02
5.73 ± 0.04i
5.33 ± 0.07
5.67 ± 0.05
4.88 ± 0.12
n.d.
5 (0.5)
10 (10)
30 (3)
20–30 (0.5)
1–30 (5)
10 (10)
10–30 (0.5)
20 (0.5)
–
35
20
n.d.
4
7
100
18
6
3
5.27 ± 0.05
5.52 ± 0.09
3–5
3–17
–
62 ± 4
23 ± 9
100
8o
histamine
–
–
> 100 6.70 ± 0.02
–
–
a For structure see figure 2. b Number of experiments. c Negative logarithm of the partial-agonist/receptor dissociation constant KP [14, 15].
d Concentration of partial agonist. Incubation time given in parentheses. e Incubation time 10 min (3 and 10 µM) or 3 min (30 µM).
f Significantly different from 100 (P < 0.05). g Not significantly different from 100 (P > 0.05). h Significantly different from 100 (P < 0.001).
i Calculated from a Kaumann-Marano plot [15], with slope constrained to unity (0.99 ± 0.07, P > 0.50). n.d. not determined.

45
Table III. Mepyramine antagonism of ileal contractions evoked
by histamine and histaprodifens 8a–o.
Agonista nb
pA2 ± SEM Emax (agonist) ± SEM cc [µM]
8a [11] 17
9.11 ± 0.04d see [11]
9.11 ± 0.05 50 ± 16
9.15 ± 0.03 91 ± 4
9.37 ± 0.10 73 ± 3
31 ± 17
9.24 ± 0.04 72 ± 6
8.93 ± 0.14 78 ± 6
51 ± 6
1–30
8b
8c
8d
3
3
4
1
1
1
3
1
1
3
1
1
1
8e
8f
3
4
8g
8h
8i
8j
8k
8l
3
3
4
3
4
4
9.13 ± 0.10 34 ± 2
8.88 ± 0.04 69 ± 8
9.20 ± 0.07 79 ± 5
n.d.
0
100
1
3
8.95 ± 0.14 63 ± 8
8.85 ± 0.12 47 ± 4
8m
8n
8o
–
n.d.
9.18 ± 0.12 14 ± 5
n.d.
9.07 ± 0.03e see [11]
–
–
3
3
1
0
100
0.3–100
histamine 29
a For structure see figure 2 and table I. b Number of experiments.
c Concentration of mepyramine. d Schild plot slope 1 (0.90 ± 0.06,
P > 0.05). e Schild plot slope 1 (0.97 ± 0.04, P > 0.20). n.d. not
determined.
character of all histaprodifens studied so far in this
preparation [11]. Interestingly, the pEC50 of 8c and re-
ported histaprodifens [11] was close to the pKP value
determined on the guinea-pig ileum (e.g., 5.80 ± 0.05 vs.
6.22 ± 0.08 for 8c), which indicates that fractional recep-
tor occupation and biological response run parallel in the
aorta assay.
Figure 3. Upper panel: contraction of guinea-pig ileum indu-
ced by histamine (▼, n > 10), meta fluorohistaprodifen (8c) in
the absence (●, n = 13, Emax = 98 ± 1%) and presence (·, n =
3, 91 ± 4%) of the competitive H1-receptor antagonist mepyra-
mine (1 nM, 15 min, pA2 = 9.15 ± 0.03), 8d (▲, n = 6,
85 ± 3%), and 8b (■, n = 10, 86 ± 2%) in the absence of
mepyramine. Lower panel: contraction of guinea-pig ileal
segments induced by histamine in the absence (▼, n = 10) and
presence of 8c (●, 3 µM, 95 ± 2%, n = 5; and ■, 10 µM,
84 ± 3%, n = 5). Open symbols represent the maximum
contraction elicited by 8c (88 ± 2% (,) and 94 ± 1% ([))
which faded to 50 ± 4% (▼) and 44 ± 7% (■) during an
equilibration period of 10 min. For 8c the affinity parameter
pKP [14, 15] was calculated as 6.22 ± 0.08 (n = 10).
Figure 4. Correlation of agonist potency (abscissa: pEC50) and
receptor affinity (ordinate: pKP value) of 15 histaprodifen
analogues on the guinea-pig ileum. Data are from this study (●)
or taken from [11] (·). r2 was 0.7782 (P < 0.001).

46
Precontracted rat aortic rings with intact endothelium
were relaxed to a submaximal extent by increasing
concentrations of 8c in a mepyramine-sensitive manner.
While 8c was five times as potent as histamine, the
affinity of mepyramine was reduced by one order of
magnitude, which confirms literature values for me-
pyramine in rat tissues [2, 11] (figure 5, lower panel).
4.2. Structure–activity relationships
with regard to H1 receptors
Concerning the substituent position, the potency pat-
tern of 8b–m was heterogeneous. Relative agonist poten-
cies followed the rank order meta > para > ortho (X = F;
8b–d), meta > ortho > para (X = Cl; 8e–g), or ortho >
meta > para (X = Br, CH3; 8h–m). Obviously, smaller
electronegative atoms are favoured in the meta position
(8c and 8f), while in the ortho position less electronega-
tive substituents are beneficial for optimum potency.
Except for 8d, a para substitution was accompanied by a
prominent loss of contractile potency (Emax = 2–58%).
The relative maximum effect decreased rather homoge-
neously in the order meta > ortho > para. The introduc-
tion of a second para fluoro substituent was detrimental
for both agonist potency and maximum effect (8o). The
structure–activity relationships of histaprodifens differ
from those of the 2-phenylhistamine series (figure 6).
Meta fluorohistaprodifen (8c) and its chloro analogue 8f
were equipotent with the unsubstituted lead 8a, while
meta halogenated 2-phenylhistamines were significantly
more potent than 2-phenylhistamine [7]. Another key
compound which supported this dissimilarity, was meta
trifluoromethylhistaprodifen (8n) which was 32 times
less potent than 2-(3-trifluoromethylphenyl)histamine
[7]. With regard to the relative maximum effect, the
2-phenylhistamine analogues of para substituted histap-
rodifens 8d, 8g and 8m also displayed significantly
reduced values compared with the respective meta sub-
stituted congener.
Based on molecular dynamics simulations, binding
models for histamine and unsubstituted histaprodifens
have been proposed recently [11]. The cited study has
revealed essential differences in the putative binding
mode of histamine and 8a. Both phenyl rings of 8a fill out
a lipophilic receptor pocket so that no space is available
for substituents much larger than hydrogen. In the light of
the fact that in the 2-phenylhistamine series, meta sub-
stituents such as iodine or bromine result in a three-fold
increase of potency [7], it would appear that the orienta-
tion of 2-phenylhistamines relative to the H1 receptor is
substantially different from the position adopted by hi-
staprodifens.
Figure 5. Upper panel: contraction of endothelium-denuded
guinea-pig aortic rings by histamine (▼, n = 11, Emax
=
110 ± 1%) and 8c (●, n = 12, 65 ± 4%) in the absence of
antagonist, and by 8c in the presence (·, n = 5, 52 ± 4%) of
mepyramine (5 nM, 30 min, pA2 = 8.76 ± 0.07). Only second
curves are shown. The maximum relative to the second hista-
mine curve (▼) was 59 ± 3% (●) and 47 ± 4% (·). Relative
potency of 8c was 154% (126–188) compared with 100% for
histamine (P < 0.001). From a Schild plot analysis [19] for
mepyramine (1–1 000 nM), a pA2 of 9.11 ± 0.02 (n = 34, slope
constrained to unity) was determined using histamine as the
agonist [11] (data not shown). Lower panel: Relaxation of rat
aortic rings with intact endothelium relative to the relaxation
induced by carbachol (0.3–1 mM). Histamine (▼, n = 6) and 8c
(●, n = 10) relaxed rings precontracted with 15.8 nM U46619
(a thromboxane A2-receptor agonist) by 94 ± 3 and 59 ± 4%,
respectively. Incubation of mepyramine (100 nM, 75 min)
produced a rightward shift of the concentration–relaxation
curves of histamine (n = 8, not shown) and 8c (·, n = 4). pA2
values for mepyramine were 8.00 ± 0.07 (vs. histamine) and
8.02 ± 0.17 (vs. 8c). In accordance with literature [2] the
affinity of mepyramine for the rat H1 receptor is one order of
magnitude lower than for H1 receptors of other species. The
relative maximum effect of 8c was 68 ± 3%, and relative
potency was 556% (449–689) compared with 100% for hista-
mine. Only one curve was established with each preparation.

47
affinity for H2 and H3 receptors, respectively (table IV).
All compounds studied on the spontaneously beating
guinea-pig right atrium (H2 receptors) depressed concen-
tration–frequency curves of histamine with low potency
(pD2’ = 4.3–4.9). When studied as potential antagonists of
(R)-α-methylhistamine on the field-stimulated longitudi-
nal muscle of the guinea-pig ileum, histaprodifens with
small or medium-sized ortho substituents (8b, 8e and 8k)
elicited a significant rightward displacement of the con-
centration–relaxation curve (pA2 = 5.6–5.8), while the
others were inactive, at least at the concentrations stud-
ied. It was not possible to characterize the potential
H3-receptor antagonism in full detail since at higher
concentrations the histaprodifen analogues blocked the
muscarinic M3 receptors of the preparation and disturbed
the electrically evoked, acetylcholine-mediated contrac-
tion of the tissue by themselves. Moreover, meta fluoro-
histaprodifen (8c), as the most interesting congener of 8a,
was subjected to a broader spectrum of functional recep-
tor assays (see table IV, footnote e) and, all in all,
displayed the same affinity profile as 8a [11].
Figure 6. Plot of relative agonist potencies of 2-phenyl-
histamine and nine phenyl-ring mono-substituted analogues
(ordinate, data from [4, 7, 8]) versus relative agonist potencies
of histaprodifen (8a) and the nine correspondingly mono-
substituted histaprodifens (abscissa, data from [11] and this
study). Substitution included 2-F, 3-F, 4-F, 2-Cl, 3-Cl, 4-Cl,
3-Br, 3-CH3 and 3-CF3 groups. Data were measured on the
guinea-pig ileum. No significant correlation was observed (r2 =
0.095, P > 0.39). Therefore, it is unlikely that structure–activity
relationships of 2-phenylhistamines and histaprodifenes run
parallel.
5. Conclusions
Histaprodifen (8a, 2-(3,3-diphenylpropyl)histamine,
figure 1) is the prototype of a new familiy of potent and
selective histamine H1-receptor agonists which offers a
wide array of structural modifications to gain more
insight into structure–activity relationships of H1-
histaminergic agonists and into the underlying molecular
principles of H1-receptor activation [11]. The presented
investigation conclusively demonstrates that an optimi-
zation of agonist potency in this series is not attainable by
4.3. Functional receptor selectivity of
histaprodifen analogues
As far as other members of the family of histaminergic
receptors are concerned, the new histaprodifen analogues
lacked agonist potency and possessed low or moderate
Table IV. Functionally determined antagonist affinity of histaprodifens 8a–i, k and l for histamine H2, H3 and muscarinic M3 receptors.
Compounda
H2 (guinea-pig right atrium) H3 (guinea-pig ileum) M3 (guinea-pig ileum)
No.
X
X≠
nb
pD’2 ± range
cd
nb
pA2 (95 % conf. lim.)c cd
nb
pA2 ± SEMc
cd
8a [11]
8b
8ce
8d
8e
8f
8g
8h
8i
8k
8l
H
2-F
3-F
4-F
2-Cl
3-Cl
4-Cl
2-Br
3-Br
2-CH3
3-CH3
H
H
H
H
H
H
H
H
H
H
H
2
2
2
2
2
3
2
2
2
2
3
4.46 ± 0.08
4.68 ± 0.10
4.47 ± 0.16
4.37 ± 0.01
4.77 ± 0.01
4.74 ± 0.15
4.50 ± 0.07
4.94 ± 0.03
4.31 ± 0.02
4.47 ± 0.18
4.76 ± 0.34
20
20
20
30
20
10–30
20
16–20
20
20
5
5
4
3
5
5
5
4
–
4
3
5.51 (5.00–5.80)
5.64 (5.50–5.75)
5.52 (5.10–5.78)
< 5.7
5.66 (5.53–5.77)
< 6.0
< 6.0
< 6.0
n.d.
2
2
2
2
1
1
1
1
11
10
12
8
12
8
6
11
–
5.55 ± 0.04
5.54 ± 0.04
5.63 ± 0.10
5.42 ± 0.07
5.74 ± 0.05
5.82 ± 0.11
5.62 ± 0.08
5.75 ± 0.05
n.d.
3–30
3–30
3–10
3–10
1–10
3–10
3–10
1–10
–
5.84 (5.59–6.03)
< 5.7
1
2
8
8
5.74 ± 0.07
5.43 ± 0.07
1–10
3–10
10–20
a For structure see figure 2. b Number of experiments. c Apparent pA2 (see Experimental protocols). d Concentration(s) of 8 [µM]. e Further
affinities of 8c: â1: pD’ < 4.5 (n = 2); α1D: pA2 = 5.77 ± 0.12 (n = 5); 5-HT1B: pA2 = 5.09 ± 0.15 (n = 4); 5-HT2A: pA2 = 5.33 ± 0.04 (n =
2
8); 5-HT3: pA2 < 5.5 (n = 6); 5-HT4: pA2 < 5.5 (n = 3). For assays see Experimental protocols. n.d. not determined.

48
substitution of the aromatic rings of 8a. At best a meta
fluoro substituent, as present in 8c (‘meta fluorohista-
prodifen’), is tolerated without loss of H1-receptor activ-
ity compared with 8a. Partial H1-receptor agonism is
exhibited when 8c is studied as a constrictor of guinea-
pig aorta and as a relaxant agent on the precontracted rat
aorta, respectively. With regard to receptor selectivity,
compound 8c also perfectly matches with 8a. Contrary to
results in the 2-phenylhistamine series [3, 7, 8], the
desired increase of agonist potency after introduction of a
meta halogen or trifluoromethyl substituent in 8a is not
observed. Hence it follows that structure–activity rela-
tionships in both series are different. Therefore, it will be
necessary to focus on the modification of other parts of
the histaprodifen molecule to increase H1-receptor po-
tency.
Et2O (100 mL). This suspension was stirred for 30 min
before a solution of the respective diphenylmethanol (1b,
h–i or n, 31 mmol) dissolved in Et2O (50 mL) was added
dropwise. After heating to reflux for 12 h, a mixture of
Et2O/MeOH (10 mL each) was added carefully under
cooling in an ice-bath. After hydrolysis with 1 N HCl
(100 mL) the product was separated by extraction with
Et2O (3 × 50 mL). After removal of the solvent, the pure
diphenylmethane was obtained as a colourless oil after
chromatographic purification using petrolether/CH2Cl2
(9:1) as eluent.
(3-Bromophenyl)phenylmethane (2i). Yield: 6.6 g
1
(86%). H-NMR (CDCl3) δ 7.39–7.13 (m, 9H, 9 Ph-H),
3.98 (s, 2H, CH2). EI-MS m/z 248 (M+•, 26), 246 (M+•,
31), 168 (36), 167 (100), 165 (37), 152 (22). IR (KBr,
cm–1) 2 910 m, 1 567 st, 1 474 st, 1 071 st, 850 m. Anal.
C13H11Br (C, H, N).
6. Experimental protocols
6.1.2.2. Diphenylbutyronitriles 5a, b, h–i and n
General procedure:
6.1. Chemistry
A solution of the respective diphenylmethane (2a, b,
h–i or n) (68 mmol) in a mixture of dry Et2O/THF
(25 mL each) was added dropwise over 15 min to a
stirred suspension of potassium (3.2 g, 82 mmol) and
catalytic amounts of Fe(NO3)3 in liquid NH3 (100 mL,
–70 °C). The deep red suspension was stirred for 30 min
and a solution of 3-bromopropionitrile (9.4 g, 70 mmol)
in dry Et2O/THF (20 mL) was added. After the deep red
colour had disappeared, NH3 was allowed to evaporate.
The organic layer was separated and the residue extracted
twice with Et2O (40 mL). The combined extracts were
washed with 0.1 N HCl (2 × 50 mL), water (100 mL),
brine (50 mL) and dried over Na2SO4. The solvent was
removed under reduced pressure and the resulting colour-
less oil purified by column chromatography using
CH2Cl2/petrolether (40/60) as eluent.
6.1.1. General procedures
Melting points were determined on an Electrothermal
IA 9000 digital apparatus and are uncorrected. IR spectra
were recorded on a Perkin Elmer 1420 spectrometer.
1H-NMR spectra were recorded on a Bruker Avance-TM-
DPX 400 (400 MHz) spectrometer. Chemical shifts are
given in ppm downfield from TMS as internal reference.
1H-NMR data are reported in the order: multiplicity (s,
singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br,
broad; *, exchangeable by D2O), approximate coupling
constant J, number of protons, and location of protons
(Im, imidazole; Ph, phenyl; Mal, maleic acid). 13C-NMR
spectra are presented accordingly. Mass spectra were
recorded using a Finnigan MAT CH7A (70 eV, EI
spectra) or a Finnigan MAT CH5DF (+FAB spectra).
Elemental analyses (C, H, N, Vario EL) were within ±
0.4% of the theoretical values unless otherwise indicated.
Yields were not optimized. Chromatographic separation
was achieved by column chromatography using silica gel
60 (Merck No. 9285, 230–400 mesh). Preparative rota-
tory layer chromatography was performed using a Chro-
matotron 7924T (Harrison Research, CA, USA) and glass
rotors coated with 4 mm layers of silica gel 60 PF254
containing gypsum (Merck).
4-(3-Bromophenyl)-4-phenylbutyronitrile (5i). Yield:
1
7.9 g (39%). H-NMR (CDCl3) δ 7.41–7.15 (m, 9H, 9
Ph-H), 4.06 (t, J = 7.9 Hz, 1H, CHCH2), 2.43–2.37 (m,
4H, CH2CH2CN). EI-MS m/z 301 (M+•, 15), 299 (M+•,
15), 247 (59), 245 (63), 168 (89), 167 (100), 166 (78),
165 (75). IR (KBr, cm–1) 2 246 m (CN), 1 592 m,
1 567 m, 1 495 m, 1 473 m, 1 452 m, 1 424 m, 1 217 st,
1 075 st. Anal. C16H14BrN (C, H, N).
6.1.3. Synthesis of diphenylbutyronitriles 5
from diethylmalonates (method B)
6.1.2. Synthesis of diphenylbutyronitriles
5
from
diphenylmethanols 1 via diphenylmethanes 2 (method A)
6.1.3.1. 1-Chloro-3,3-diphenylpropanes 4c–g and j–m
General procedure:
A solution of diethyl malonate (8.0 g, 50 mmol) in dry
DMF (30 mL) was added to a suspension of NaH (2.0 g,
50 mmol) and catalytic amounts of NaI in dry DMF
6.1.2.1. Diphenylmethanes 2b, h–i and n
General procedure:
A solution of anhydrous AlCl3 in Et2O (50 mL) was
added to a suspension of LiAlH4 (1.5 g, 40 mmol) in dry

49
(60 mL). When the formation of hydrogen had ceased, a
solution of the respective chloro-diphenylmethane (3c–g
or j–m) (40 mmol) in dry DMF (30 mL) was slowly
added. After stirring for 2 h at 60 °C, the mixture was
heated to 90 °C for an additional 36 h, cooled, and poured
into water (300 mL). The mixture was extracted with
Et2O (3 × 70 mL), the organic solvents were removed by
evaporation, and the residue was dissolved in a solution
of KOH (10.0 g) in water (30 mL) and EtOH (20 mL),
and refluxed for 12 h. After cooling to ambient tempera-
ture, by-products were separated by extraction with Et2O
(50 mL). The aqueous layer was diluted with 3 N NaOH
(100 mL) and extracted with Et2O (2 × 50 mL). The
organic layer was discarded and the aqueous phase
acidified with 6 N H2SO4. After extraction with Et2O (3
× 70 mL) and removal of the solvent under reduced
pressure the 2-(diphenylmethyl)malonic acid was ob-
tained as a pale yellow solid. The malonic acid derivative
(34 mmol) was converted into the corresponding 3,3-
diphenylpropanoic acid by decarboxylation at 180 °C in
vacuo (10 min). After cooling, the residue was dissolved
in a mixture of 3 N NaOH (100 mL) and MeOH (50 mL)
and extracted twice with a mixture of petrolether/Et2O
(1:1). The aqueous layer was acidified (6 N H2SO4) and
the product extracted with Et2O (3 × 70 mL). After
removal of the solvent under reduced pressure, the
substituted propionic acid was obtained as a pale brown
solid which was pure enough for the next step.
3-(3-Fluorophenyl)-3-phenylpropionic acid. Yield:
8.2 g (77%). M.p. 107 °C. H-NMR (DMSO-d6) 12.14
1
(br*, 1H, COOH), 7.36–7.16 (m, 8H, 5 Ph-H, 3 F-Ph-H),
3
3
4
6.98 (ddd, JH–F = JH–H = 8.5 Hz, JH–H = 2.1 Hz, 1H,
F-Ph-4-H), 4.44 (t, J = 8.0 Hz, 1H, CHCH2), 3.04 (m, 2H,
CHCH2). EI-MS m/z 244 (M+•, 42), 198 (22), 185 (100),
165 (19). IR (KBr, cm–1) 3 423 br, 1 712 st, 1 589 m,
1 487 st, 1 290 st. Anal. C15H13FO2 (C, H, N).
3-(3-Fluorophenyl)-3-phenyl-1-propanol [18]. Yield:
5.9 g (77%). Colourless oil. 1H-NMR (CDCl3) 7.31–7.18
(m, 6H, 5 Ph-H, F-Ph-5-H), 7.03 (d, J = 7.8 Hz, 1H,
F-Ph-6-H), 6.99–6.94 (m, 1H, F-Ph-2-H), 6.87 (ddd,
3JH–H = 3JH–F = 8.2 Hz, 4JH–H = 2.5 Hz, 1H, F-Ph-4-H),
4.15 (t, J = 7.9 Hz, 1H, CHCH2), 3.60 (t, J = 6.4 Hz, 2H,
CH2OH), 2.29 (dt, J = 6.6 Hz, 2H, CHCH2CH2), 1.31
(br*, 1H, OH). EI-MS m/z 230 (M+•, 27), 212 (51), 185
(100), 165 (25). IR (KBr, cm–1) 3 339 br, 1 612 m,
1 589 m, 1 487 m, 1 447 m, 1 246 m, 1 033 m. Anal.
C15H15FO (C, H, N).
1-Chloro-3-(3-fluorophenyl)-3-phenylpropane
(4c).
1
Yield: 5.1 g (86%). Colourless oil. H-NMR (CDCl3) δ
7.32–7.19 (m, 6H, 5 Ph-H, F-Ph-5-H), 7.03 (d, J = 7.6
3
Hz, 1H, F-Ph-6-H), 6.94 (d, JH–F = 10.1 Hz, 1H,
3
3
4
F-Ph-2-H), 6.89 (ddd, JH–H = JH–F = 8.4 Hz, JH–H
=
2.3 Hz, 1H, F-Ph-4-H), 4.22 (t, J = 7.8 Hz, 1H, CHCH2),
3.44 (t, J = 6.5 Hz, 2H, CH2Cl), 2.47 (dt, J = 7.0 Hz, 2H,
CHCH2). EI-MS m/z 248 (M+•, 13), 185 (100), 171 (6),
166 (28). IR (KBr, cm–1) 1 612 m, 1 589 st, 1 488 st,
1 447 st. Anal. C15H14ClF (C, H, N).
The above described propionic acid derivative or
commercially available 3,3-di-(4-fluorophenyl)propionic
acid (33 mmol) was subsequently converted into the
corresponding alcohol upon treatment with LiAlH4 in dry
Et2O according to standard methods, and purified by
column chromatography (CH2Cl2/MeOH (9/1)). The sub-
stituted 3,3-diphenyl-1-propanol (24 mmol) was then
converted into the corresponding 1-chloro-3,3-
diphenylpropane (4c–g and j–m) by reaction with thio-
nylchloride (10 mL) and a catalytic amount of 4-DMAP
in CH2Cl2 (20 mL), followed by column chromatography
using petrolether/CH2Cl2 (8:2) as eluent.
6.1.3.2. Diphenylbutyronitriles 5c–g, j–m and o
General procedure:
The above described 1-chloropropanes 4c–g, j–m and
o (21 mmol), potassium cyanide (1.4 g, 22 mmol) and a
catalytic amount of KI were dissolved in DMSO (50 mL)
and stirred for 1 h at ambient temperature, 1 h at 65 °C,
and 3 h at 95 °C. After cooling, the mixture was diluted
with water (500 mL), and the butyronitrile was isolated as
a colourless liquid by extraction with Et2O (3 × 50 mL),
followed by column chromatography using CH2Cl2/
petrolether (1:1) as eluent.
1-Chloro-3-(3-fluorophenyl)-3-phenylpropane (4c):
4-(3-Fluorophenyl)-4-phenylbutyronitrile (5c):
2-((3-Fluorophenyl)phenylmethyl)malonic acid. Yield:
5.2 g (45%). M.p. 164 °C (dec.). H-NMR (CDCl3) 7.80
(br*, 2H, 2 COOH), 7.35–7.16 (m, 6H, 5 Ph-H, F-Ph-5-
Colourless liquid. Yield: 4.5 g (92%). 1H-NMR
(CDCl3) 7.34–7.20 (m, 6H, 5 Ph-H, F-Ph-5-H), 7.03 (d, J
= 7.8 Hz, 1H, F-Ph-6-H), 6.93–6.87 (m, 2H, F-Ph-2-H,
F-Ph-4-H), 4.06 (t, J = 7.9 Hz, 1H, CHCH2), 2.37 (dt, J
= 7.2 Hz, 2H, CH2CH2CN), 2.27 (t, J = 7.6 Hz, 2H,
CH2CN). 13C-NMR (CDCl3) 163.1 (d, 1JC–F = 246.8 Hz,
1
3
H), 7.02 (d, J = 7.8 Hz, 1H, F-Ph-6-H), 6.93 (d, JH–F
=
=
3
3
9.8 Hz, 1H, F-Ph-2-H), 6.98–6.86 (ddd, JH–H = JH–F
4
8.4 Hz, JH–H = 2.3 Hz, 1H, F-Ph-4-H), 4.64 and 4.28
(2d, J = 12.2 Hz, 2H, CHCH). EI-MS m/z 288 (M+•, 8),
270 (12), 242 (34), 224 (18), 185 (100), 166 (45). IR
(KBr, cm–1) 2 692 w, 1 723 st, 1 593 st, 1 490 st, 1 408 st,
1 108 m. Anal. C16H13FO4 (not available).
3
F-Ph-C(3)), 145.4 (d, JC–F = 6.7 Hz, F-Ph-C(1)), 142.0
3
(Ph-C(1)), 130.3 (d, JC–F = 8.0 Hz, F-Ph-C(5)), 129.0,
4
127.7, 127.1 (5 Ph-C(2–6)), 123.4 (d, JC–F = 2.7 Hz,
2
F-Ph-C(6)), 119.1 (CN), 114.5 (d, JC–F = 21.4 Hz,

50
2
F-Ph-C(2)), 113.8 (d, JC–F = 20.9 Hz, F-Ph-C(4)), 49.5
(Ph-CH-Ph), 30.1 (CH2CH2CN), 15.7 (CH2CH2CN).
EI-MS m/z 239 (M+•, 20), 185 (100), 170 (7), 165 (36).
IR (KBr, cm–1) 3 027 m, 2 936 m, 2 246 m (CN), 1 590
st, 1 490 st, 1 449 st, 1 425 w. Anal. C16H14FN (C, H, N).
F-Ph-6-H), 7.05–7.01 (m, 1H, F-Ph-2-H), 6.90 (ddd,
3JH–H = 3JH–F = 8.4 Hz, 4JH–H = 2.2 Hz, 1H, F-Ph-4-H),
6.25 (s, 4H, 4 Mal-H), 4.04 (t, J = 7.8 Hz, 1H, CHCH2),
3.27 (t, J = 7.5 Hz, 2H, CH2CH2N), 3.03 (t, J = 7.5 Hz,
2H, Im(C4)-CH2), 2.90 (t, J = 8.0 Hz, 2H, Im(C2)-CH2),
2.59–2.53 (m, 2H, CHCH2). 13C-NMR (CD3OD) δ 170.9
(4 COOH), 164.4 (d, 1JC–F = 244.6 Hz, F-Ph-C(3)) 149.2
6.1.4. Synthesis of methyl butyrimidates 6b–o
3
(Im-C(2)), 148.1 (d, JC–F = 6.8 Hz, F-Ph-C(1)), 144.3
(Ph-C(1)), 136.6 (4 CHCOOH), 131.4 (d, 3JC–F = 8.2 Hz,
6.1.4.1. General procedure
A solution of the respective butyronitriles 5b–o
(17 mmol) in dry MeOH (30 mL) was chilled to –30 °C.
After stirring for 5 min, thionylchloride (3.0 mL) was
added to the solution. After 7 days in a freezer the solvent
was removed in vacuo. The bulk was pure enough for the
subsequent cyclization and was therefore used without
further purification.
F-Ph-C(5)), 130.0 (Im-C(4)), 129.9 and 128.8 (4 Ph-
4
C
(2,3,5,6)), 127.9 (Ph-C(4)), 124.7 (d, JC–F = 2.7 Hz,
2
F-Ph-C(6)), 117.6 (Im-C(5)), 115.4 (d, JC–F = 21.6 Hz,
F-Ph-C(2)), 114.3 (d, JC–F = 21.1 Hz, F-Ph-C(4)), 51.8
2
(Ph-CH-Ph), 39.2 (CH2NH2), 33.6 (CHCH2CH2), 25.8
(Im-C(2)-CH2), 23.9 (Im-C(4)-CH2). +FAB-MS (Xe/
glycerol) m/z 324 ([M + H]+, 100), 307 (19), 295 (17). IR
(KBr, cm–1) 3 419 m, 1 619 st, 1 583 st, 1 486 st,
1 358 m, 867 m. Anal. C20H22FN3 2 C4H4O4 0.25 H2O
(C, H, N).
6.1.4.2. Methyl 4-(3-fluorophenyl)-4-
phenylbutyrimidate hydrochloride (6c)
Colourless semisolid material. Yield: 4.8 g (83%).
+FAB-MS (Xe/3-NO2-benzyl-OH) m/z 272 ([M + H]+,
100), 258 (3), 241 (3), 185 (3), 185 (53). Anal.
C17H18FNO HCl (not available).
6.2. Pharmacology
6.2.1. Data handling and pharmacological parameters
Data are presented as mean ± standard error (SEM or
SE) or with 95% confidence limits. Graphical data
(figures 3 and 5) are given as mean ± SEM (SEM does
not appear when smaller than symbol). Significant differ-
ences of means or the difference between a mean and a
constant value (P < 0.05) were discerned by Student’s t
test. Agonists were characterized by relative potency
compared with histamine, calculated as the antilog of
(pEC50(agonist) – pEC50(histamine)). The first concen-
tration–effect curve (CEC) for histamine served as an
internal reference when two CECs were performed on
each preparation. The daily mean of untreated histamine
control organs served as reference when only one CEC
per organ was established. Emax [%] of the new agonists
(formerly termed ‘intrinsic activity’ [13]) was calculated
relative to histamine. The negative logarithm of the
dissociation constant of the partial-agonist/receptor com-
plex, pKP, was calculated by a weighted linear regression
procedure from individual sets of equi-effective concen-
trations of histamine in the absence and presence of a
maximal or supramaximal concentration of the partial
agonist [14, 15]. The affinity of the H1-receptor antago-
nist mepyramine was calculated as pA2 from experiments
with histamine or the new agonists in the absence or
presence of suitable concentrations of mepyramine. Ap-
parent pA2 values were obtained from experiments with a
single mepyramine concentration according to the equa-
tion pA2 = –log10 c(mepyramine) + log10 (r – 1), where
r is the ratio of agonist concentrations in the absence and
6.1.5. Synthesis of histaprodifens 8b–o
6.1.5.1. General procedure
Equimolar amounts of 2-oxo-4-phthalimido-1-butyl
acetate (4.2 g, 15 mmol) and the respective methyl butyr-
imidate 6b–o (15 mmol) were dissolved in liquid NH3
(150 mL) in an autoclave (1 000 mL, Kotter, Germany).
After stirring for 12 h at room temperature the mixture
was heated to 60 °C (24–26 bar) for 6 h. After the
evaporation of NH3, the crude amines 8b–o were ob-
tained after deprotection by acidic hydrolysis with 6 N
HCl under reflux for 6 h. After alkalization with excess 3
N NaOH, the substituted histaprodifens were extracted
with 3 portions of CH2Cl2/MeOH (9:1). Alternatively, the
protecting phthaloyl group was cleaved by refluxing the
crude mixture containing 7b–o in a mixture of MeOH
(50 mL) and hydrazine hydrate (1.0 g, 20 mmol) for 3 h.
After removal of the solvent, the free bases were ex-
tracted with CH2Cl2/MeOH (3 × 50 mL). For all com-
pounds, rotatory chromatography (CH2Cl2/MeOH (10/1),
NH3-saturated) afforded the purified histaprodifen bases
8b–o which were crystallized as dihydrogen maleates
(table I).
6.1.5.2. 2-[2-[3-(3-Fluorophenyl)-3-phenylpropyl]-
1H-imidazol-4-yl]ethanamine (8c)
Yield: 0.68 g (8%). M.p. 147–149 °C. 1H-NMR
(CD3OD) δ 7.33–7.27 (m, 5H, 5 Ph-H), 7.21–7.17 (m,
3
2H, F-Ph-5-H, Im-5-H), 7.11 (d, JH–H = 7.9 Hz, 1H,

51
presence of mepyramine that elicit 50% of the respective
maximum effect [13]. When a set of at least three
different mepyramine concentrations was studied, full
pA2 values were calculated from Arunlakshana-Schild
plots after constraining the Schild plot slope to unity [19].
Antagonist potencies of the new compounds at selected
neurotransmitter receptor subtypes were calculated as
apparent pA2 values or as pD’2 values [13], when only
depression of CECs instead of rightward displacement
was observed.
was corrected using the sensitivity change monitored by
untreated histamine control preparations.
6.2.4. Histamine H1-receptor assay on the
isolated rat aorta
Male Wistar rats were stunned (CO2) and decapitated.
The thoracic portion of the aorta was rapidly removed,
rinsed, and cleared of connective tissue [11, 17]. Rings of
2–4 mm length were cut and set up isometrically (initial
tension 10 mN) by means of two L-shaped stainless steel
hooks in a modified Krebs-Henseleit solution (37 °C,
gassed with 95% O2/5% CO2, 1.25 mM Ca2+, 1.20 mM
Mg2+). After an equilibration period of ca. 100 min,
vascular rings were contracted with a submaximal con-
centration of U46619, a TP (thromboxane-A2)-receptor
agonist (15.8 nM) in the presence of prazosin (100 nM).
When the effect had plateaued (usually after 45 min), a
cumulative concentration–relaxation curve was estab-
lished for histamine (0.1–1 000 µM) or a new agonist in
the absence or presence of mepyramine (100 nM, total
incubation time approximately 75 min). When the agonist
had elicited its maximum effect, a final relaxation was
induced by addition of the M3-receptor agonist carbachol
(300–1 000 µM).
6.2.2. Histamine H1-receptor assay on the
isolated guinea-pig ileum
Guinea-pigs of either sex were stunned by a blow on
the head and exsanguinated. The ileum was removed and
whole segments (2.0–2.5 cm) were mounted isotonically
(preload 0.5 g) at 37 °C in Tyrode’s solution, aerated with
95% O2/5% CO2, in the continuous presence of 0.1 µM
atropine, a concentration not affecting H1 receptors [7,
20]. During an equilibration period of ca. 80 min the
organs were stimulated three times with histamine (1 and
10 µM) followed by washout. Each preparation was used
to establish a cumulative concentration–effect curve for
histamine (0.01–30 µM) followed by a second curve for
a new agonist in the absence or presence of mepyramine
(1–100 nM, incubation time 10–15 min). The pEC50
difference was not corrected since two successive curves
for histamine were superimposable (n = 10). For the
determination of pKP, the partial agonist was not washed
out and incubated for 0.5–10 min (table II). A final
cumulative curve for histamine was then constructed.
6.2.5. Antagonist activity of title compounds at histamine
H2, H3 and other neurotransmitter receptors
In vitro experiments were performed according to
published protocols for the following functional receptor
assays: histaminergic H2 receptors of the spontaneously
beating guinea-pig right atrium (in the presence of 0.3
µM propranolol and 1 µM mepyramine) [20], histamin-
ergic H3 receptors of field-stimulated guinea-pig ileal
longitudinal muscle with adhering plexus myentericus (in
the presence of 1 µM mepyramine) [21], muscarinic M3
receptors of guinea-pig whole ileal segments (in the
presence of 1 µM mepyramine) [20], adrenergic α1D
receptors of rat thoracic aorta and â1 receptors of spon-
taneously beating guinea-pig right atrium (in the presence
of 1 µM mepyramine) [20], serotonergic 5-HT1B recep-
tors of guinea-pig arteria iliaca (in the presence of 1 µM
mepyramine) [20], 5-HT2A receptors of rat tail ar-
tery [20], 5-HT3 receptors of quiescent guinea-pig ileal
longitudinal muscle with adhering plexus myentericus (in
the presence of 1 µM mepyramine) [22], and 5-HT4
receptors of rat oesophagal tunica muscularis muco-
sae [22], respectively. For the new compounds, agonist
effects were not observed in all non-H1-receptor assays.
6.2.3. Histamine H1-receptor assay on the
isolated guinea-pig aorta
Thoracic aortae of sacrificed guinea-pigs were quickly
removed and cleared of connective tissue [7, 11, 16].
Rings of 2–4 mm length were cut and rolled with a pair
of tweezers to damage the endothelium. Organs were
mounted isometrically (initial tension 10 mN) by means
of two L-shaped stainless steel hooks in a modified
Krebs-Henseleit solution (37 °C, gassed with 95% O2/5%
CO2, 1.80 mM Ca2+, 1.20 mM Mg2+). During an equili-
bration period of ca. 130 min, organs were stimulated
three times with histamine (10 µM) followed by washout.
Cumulative concentration–effect curves for histamine
(0.1–300 µM) followed by a second curve for histamine
or a new agonist in the absence or presence of me-
pyramine (1–1 000 nM, incubation time 30 min) were
established in the presence of cimetidine, corticosterone,
cocaine (30 µM each), prazosin, yohimbine (0.3 µM
each), and propranolol (0.1 µM). The pEC50 difference

52
[8] Zingel V., Elz S., Schunack W., Eur. J. Med. Chem. 25 (1990)
673–680.
Acknowledgements
[9] Kramer K., Elz S., Pertz H.H., Schunack W., Bioorg. Med. Chem.
Lett. 8 (1998) 2583–2588.
This project was supported by Verband der Chemi-
schen Industrie, Fonds der Chemie (Frankfurt/Main, Ger-
many) (S.E., W.S.) and by the FNK of the Freie Univer-
sität Berlin (S.E., C.L., K.K.). The authors acknowledge
the contribution of Heinz H. Pertz to some experiments
with compound 8c, and of Inge Walther who performed
part of the functional selectivity assays.
[10] Elz S., Kramer K., Pertz H., Schunack W., Methylhistaprodifen and
congeners: highly potent and selective histamine H1-receptor ago-
nists. XIIIth International Congress of Pharmacology, Munich,
Germany, July 26–31, (1998). Naunyn-Schmiedeberg’s Arch. Phar-
macol. 358, Suppl. 2 (1998) R762.
[11] Elz S., Kramer K., Pertz H.H., Detert H., ter Laak A.M., Kühne R.,
Schunack W., J. Med. Chem. (2000), in press.
[12] Dziuron P., Schunack W., Arch. Pharm. (Weinheim, Ger.) 306
(1973) 347–350.
References
[13] Jenkinson D.H., Barnard E.A., Hoyer D., Humphrey P.P.A., Leff P.,
Shankley N.P., Pharmacol. Rev. 47 (1995) 255–266.
[1] Kramer K., Elz S., Schunack W., Analogues of histaprodifen: potent
partial agonists for histamine H1 receptors. XIIIth International
Congress of Pharmacology, Munich, Germany, July 26–31, (1998).
Naunyn-Schmiedeberg’s Arch. Pharmacol. 358, Suppl. 1 (1998)
R116.
[14] Marano M., Kaumann A.J., J. Pharmacol. Exp. Ther. 198 (1976)
518–525.
[15] Kaumann A.J., Marano M., Naunyn-Schmiedeberg’s Arch. Pharma-
col. 318 (1982) 192–201.
[2] Hill S.J., Ganellin C.R., Timmerman H., Schwartz J.-C., Shankley
N.P., Young J.M., Schunack W., Levi R., Haas H.L., Pharmacol.
Rev. 49 (1997) 253–278.
[16] Dodel R., Borchard U., Agents Actions 37 (1992) C410–C413.
[17] Carrier G.O., White R.E., Kirby M.L., Blood Vessels 21 (1984)
180–183.
[3] Zingel V., Leschke C., Schunack W., Prog. Drug Res. 44 (1995)
49–85.
[18] Bergmann F., Weizmann M., Diamant E., Patai S., Sznuszkovicz J.,
J. Am. Chem. Soc. 70 (1948) 1612–1617.
[4] Koper J.G., van der Vliet A., van der Goot H., Timmerman H.,
Pharm. Weekbl. Sci. Ed. 12 (1990) 236–239.
[19] Arunlakshana O., Schild H.O., Br. J. Pharmacol. Chemother. 14
(1959) 48–58.
[5] Young R.C., Ganellin C.R., Griffiths R., Mitchell R.C., Parsons
M.E., Saunders D., Sore N.E., Eur. J. Med. Chem. 28 (1993)
201–211.
[20] Pertz H., Elz S., J. Pharm. Pharmacol. 47 (1995) 310–316.
[21] Sasse A., Kiec-Kononowicz K., Stark H., Motyl M., Reidemeister
S., Ganellin C.R., Ligneau X., Schwartz J.-C., Schunack W., J. Med.
Chem. 42 (1999) 593–600.
[6] Dziuron P., Schunack W., Eur. J. Med. Chem. -Chim. Ther. 10
(1975) 129–133.
[22] Elz S., Keller A., Arch. Pharm. Pharm. Med. Chem. 328 (1995)
585–594.
[7] Leschke C., Elz S., Garbarg M., Schunack W., J. Med. Chem. 38
(1995) 1287–1294.