Acylguanidines as bioisosteres of guanidines: NG-acylated imidazolylpropylguanidines, a new class of histamine h2 receptor agonists

Article abstract of DOI:10.1021/jm800841w

N¹-Aryl(heteroaryl)alkyl-N²-[3-(1H-imidazol-4-yl) propyl]guanidines are potent histamine H₂-receptor (H₂R) agonists, but their applicability is compromised by the lack of oral bioavailability and CNS penetration. To improve pharmacokinetics, we introduced carbonyl instead of methylene adjacent to the guanidine moiety, decreasing the basicity of the novel H₂R agonists by 4-5 orders of magnitude. Some acylguanidines with one phenyl ring were even more potent than their diaryl analogues. As demonstrated by HPLC-MS, the acylguanidines (bioisosteres of the alkylguanidines) were absorbed from the gut of mice and detected in brain. In GTPase assays using recombinant receptors, acylguanidines were more potent at the guinea pig than at the human H₂R. At the hH₁R and hH₃R, the compounds were weak to moderate antagonists or partial agonists. Moreover, potent partial hH₄R agonists were identified. Receptor subtype selectivity depends on the imidazolylpropylguanidine moiety (privileged structure), opening an avenue to distinct pharmacological tools including potent H₄R agonists.

Full text of DOI:10.1021/jm800841w

J. Med. Chem. 2008, 51, 7193–7204
7193
Acylguanidines as Bioisosteres of Guanidines: NG-Acylated Imidazolylpropylguanidines, a New
Class of Histamine H2 Receptor Agonists
Prasanta Ghorai,† Anja Kraus,† Max Keller,† Carsten Go¨tte,† Patrick Igel,† Erich Schneider,‡ David Schnell,‡
Gu¨nther Bernhardt,† Stefan Dove,† Manfred Zabel,§ Sigurd Elz,† Roland Seifert,‡ and Armin Buschauer†,*
Departments of Pharmaceutical/Medicinal Chemistry, Pharmacology and Toxicology, Center for Chemical Analysis, Faculty of Chemistry and
Pharmacy, UniVersity of Regensburg, UniVersita¨tsstrasse 31, D-93053 Regensburg, Germany
ReceiVed July 9, 2008
N1-Aryl(heteroaryl)alkyl-N2-[3-(1H-imidazol-4-yl)propyl]guanidines are potent histamine H2-receptor (H2R)
agonists, but their applicability is compromised by the lack of oral bioavailability and CNS penetration. To
improve pharmacokinetics, we introduced carbonyl instead of methylene adjacent to the guanidine moiety,
decreasing the basicity of the novel H2R agonists by 4-5 orders of magnitude. Some acylguanidines with
one phenyl ring were even more potent than their diaryl analogues. As demonstrated by HPLC-MS, the
acylguanidines (bioisosteres of the alkylguanidines) were absorbed from the gut of mice and detected in
brain. In GTPase assays using recombinant receptors, acylguanidines were more potent at the guinea pig
than at the human H2R. At the hH1R and hH3R, the compounds were weak to moderate antagonists or
partial agonists. Moreover, potent partial hH4R agonists were identified. Receptor subtype selectivity depends
on the imidazolylpropylguanidine moiety (privileged structure), opening an avenue to distinct pharmacological
tools including potent H4R agonists.
Introduction
Histamine receptors (H1R, H2R, H3R, and H4R) are cell
surface receptors and belong to the class 1 or rhodopsin-like
family of G-protein coupled receptors, possessing seven trans-
membrane domains (TMsa), three extracellular loops, and three
intracellular loops.1-3 The histamine H2 receptor (H2R)4 is Gs-
coupled and therefore activates adenylyl cyclase.1 H2Rs are
located on gastric parietal cells and in several other tissues and
cells including leukocytes, heart, airways, uterus vascular smooth
muscles, and in the brain.1,5,6 Numerous H2R agonists as well
as antagonists have been identified. H2R antagonists were
introduced into the clinic for the treatment of gastroduodenal
ulcer and gastresophageal reflux disease. By contrast, H2R
agonists are not routinely used in therapy but are valuable
pharmacological tools. Nevertheless, such compounds are of
potential therapeutic value as positive inotropic vasodilators for
the treatment of acute heart failure,7 as differentiation-inducing
Figure 1. Structures of histamine and selected histamine H2 receptor
agonists.
agents in acute myelogenous leukemia,8 or as anti-inflammatory
agents.9 Compared to amine-type H2R agonists such as hista-
mine, dimaprit, and amthamine, guanidine-type compounds are
much more potent. Impromidine (Figure 1), which is about 50
times more potent than histamine at the guinea pig right atrium,
was investigated for the treatment of patients suffering from
severe catecholamine-insensitive congestive heart failure.10,11
Arpromidine (Figure 1) and related N-[3-(1H-imidazol-4-yl)-
propyl]guanidines were developed as positive inotropic vasodi-
lators (“cardiohistaminergics”12) and were superior to impro-
midine in terms of potency and hemodynamic profile when
tested in the guinea pig under physiological conditions and in
a pathophysiological model of severe congestive heart failure
(vasopressin-induced acute heart failure).7,13,14 Arpromidine and
related guanidine-type agonists represent the most potent H2R
agonists known so far, achieving up to 400 times the potency
of histamine on the spontaneously beating guinea pig right
atrium. The binding site of histamine in the H2R was identified
by in vitro mutagenesis studies and modeling approaches based
on the crystal structure of rhodopsin.15 The interaction of
arpromidine and guanidine-type agonists in general may be
interpreted by analogy with the model proposed for histamine.6
The guanidino group undergoes an ionic interaction with Asp-
98 in transmembrane domain 3 (TM3), and the imidazole ring
forms hydrogen bonds with Asp-186 and Tyr-182 in TM5.
Additionally, residues in TM6 like Phe-254 face the imida-
zolylpropylguanidine moiety of arpromidine.6,16,17 This binding
* To whom correspondence should be addressed. Phone: +49-941
9434827. Fax: +49-941 9434820. E-mail: armin.buschauer@chemie.uni-
regensburg.de.
†
Pharmaceutical/Medicinal Chemistry, Institute of Pharmacy.
Pharmacology and Toxicology, Institute of Pharmacy.
Center for Chemical Analysis, X-ray Laboratory.
‡
§
a Abbreviations: C1, C2, C3, intracellular loops of a GPCR; Cbz,
benzyloxycarbonyl; CDI, 1,1′-carbonyldiimidazole; DIPEA, diisopropyl-
ethylamine, DMAP, 4-dimethylaminopyridine; DMF, dimethylformamide;
DMSO, dimethylsulfoxide; E1, E2, E3, extracellular loops of a GPCR; GTP,
guanosine triphosphate; GPCR, G protein-coupled receptor; GsR, R-subunit
of the Gs protein that mediates adenylyl cyclase activation; Gꢀ1γ2, G protein
ꢀ1- and γ2-subunit; GRi, R-subunit of the Gi protein that mediates inhibition
of adenylyl cyclase; gpH2R, guinea pig histamine H2 receptor; gpH2R-GsRS,
fusion protein of gpH2R and GsR; H2R, histamine H2 receptor; H1R,
histamine H1 receptor; hH2R, human histamine H2 receptor; hH2R-GsR,
fusion protein of the human histamine H2 receptor and GsR; hH3R, human
histamine H3 receptor; hH4R, human histamine H4 receptor; hH4R-RGS19,
fusion protein of hH4R and RGS19; HR-MS, high resolution mass
spectrometry; Ms, mesyl, methanesulfonyl; NPY, neuropeptide Y; PE,
petroleum ether; RGS, regulator of G protein signaling proteins; SEM,
standard error of the mean; SIM, single ion monitoring; TM, transmembrane
domain of a GPCR; TFA, trifluoroacetic acid; Tf, triflyl, trifluoromethyl-
sulfonyl; Tr, trityl, triphenylmethyl.
10.1021/jm800841w CCC: $40.75
2008 American Chemical Society
Published on Web 10/25/2008

7194 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22
Ghorai et al.
Scheme 1. (I) Synthetic Route for the Preparation of the
Building Blocks 16, 17, and 23; (II) Synthesis of the
Guanidinylation Reagents 13 and 21a
Figure 2. Crystal structure (ORTEP diagram) of trityl-protected
acylguanidine 49a shown in two different views (A and B). CCDC
686506 contains the supplementary crystallographic data for 49a
(available free of charge at The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif).
site model is generally confirmed using the recent crystal
structure of the ꢀ2-adrenoceptor18,19 as a template (see below).
The strongly basic guanidino group, considered a mimic of the
primary amino group in histamine, is essential for the H2R
agonistic activity of arpromidine and analogues, but it is also
responsible for very low oral bioavailability and lack of
penetration across the blood-brain barrier.13,20,21 This is very
unfortunate because, to this end, very little is known about the
function of the H2R in the brain.22-25 The guanidine-type
agonists are nearly quantitatively protonated at physiological
pH and are virtually inactive after oral administration. In
principle, this problem can be solved by prodrug strategies as
demonstrated by the introduction of alkoxycarbonyl groups at
the guanidine group.12 However, such derivatives were not
active as H2R agonists until ester cleavage and decarboxylation.
Centrally active H2R agonists could not be obtained with this
approach. The present study describes the successful initial step
of developing orally active nonprodrug H2R agonists as
pharmacological tools to investigate the role of the H2R in the
brain. Starting from guanidine-type H2R agonists, structurally
related compounds with reduced basicity were prepared in order
to obtain agonists with an improved pharmacokinetic profile.
Since structure-activity relationship studies in the arpromidine
series revealed that a third substituent on the guanidine group
is not tolerated, our attempts focused on modifications of the
connecting chains, more precisely, one methylene group adjacent
to the guanidine in arpromidine-like H2R agonists was replaced
with an electron-withdrawing carbonyl group. This approach
was stimulated by findings in the neuropeptide Y (NPY) receptor
field. Specifically, the affinities of NG-acylated argininamide-
type NPY Y1 antagonists were retained or even increased
compared to the respective nonacylated parent compounds.26-28
The basicity of the acylguanidines (pKa values around 8) is 4-5
orders of magnitude lower than that of the corresponding
guanidines. Hence, on one hand, acylguanidines are sufficiently
basic to undergo key interactions with acidic residues of the
receptor. On the other hand, a considerable portion remains
uncharged at physiological pH, thus facilitating diffusion across
a Reagents and conditions: (I) (i) anhydrous Na2SO4, H2SO4/conc, MeOH/
abs, 30 h, reflux; (ii) H2, Pd/C (10%) (cat.), MeOH, 5 bar, 24 h, rt; (iii)
TrCl (1.1 equiv), NEt3 (2.8 equiv), MeCN, 12 h, rt; (iv) LiAlH4 (2 equiv),
THF/abs, Et2O/abs, 2 h, reflux; (v) LiAlH4 (1.2 equiv), THF/abs, 2 h, rt;
(vi) 13 (1.8 equiv), PPh3 (1.5 equiv), DIAD (1.5 equiv), THF/abs, 24 h, rt;
(vii) TrCl, Et3N, 20 h, 0 °C-rt; (viii) 5% TFA/CH2Cl2, 10 min, 0 °C, 45
min, rt; (ix) 21, Et3N, CH2Cl2, 3 h, rt; (x) H2, Pd/C (10%), MeOH/THF
(1:1), 5 bar, 24 h, rt. (II) (xi) CbzCl, NaOH, CH2Cl2/H2O (2:1), 0 °C, 20 h;
(xii) Tf2O, NaH, THF, -45-0 °C, 16 h.
membranes. A route for the preparation of NG-acylated imida-
zolylpropylguanidines was established, and the synthesized
compounds were pharmacologically characterized on isolated
guinea pig organs (ileum: H1R; right atrium: H2R), on human
H1R-expressing U373-MG cells, and in GTPase assays using
membrane preparations of Sf9 insect cells expressing guinea
pig (gp) or human (h) histamine receptors. Moreover, absorption
after oral administration and brain penetration of representative
compounds was explored in nude mice.
Chemistry. Toward the synthesis of NG-acylated guanidines,
a guanidine building block is coupled to carboxylic acids to
make use of common coupling reagents. The protected alkylated
guanidines were prepared starting from alcohols (10, 12) or
amine 20 (Scheme 1).29 Guanidine hydrochloride was treated
with sodium hydroxide and benzyloxycarbonyl chloride to yield
a nucleophilic guanidinylation reagent (13) toward alcohols. The
N,N′-di-Cbz-protected guanidine 13 was subsequently depro-
tonated with sodium hydride, followed by the treatment with
triflic anhydride to provide an electrophilic guanidinylation
reagent (21) toward amines.

Acylguanidine-Type Histamine H2 Receptor Agonists
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22 7195
Scheme 2. General Procedures for the Synthesis of Pertinent
25a-d and 29, which were subsequently hydrogenated and
hydrolyzed to give the 3,3-disubstituted propanoic acids 27a-d
and 31. 1-(Pyridin-2-yl)ethanol (32a) was prepared from me-
thylmagnesium bromide and pyridine-2-carbaldehyde. 1-(Thia-
zol-2-yl)ethanol (32b) was synthesized from bromothiazole and
acetaldehyde by using known procedures. The alcohols 32a,b
were then converted to the mesylates 33a,b, followed by
nucleophilic displacement with dimethyl malonate, hydrolysis
under alkaline conditions, and decarboxylation to generate the
3-substituted propanoic acids 35a,b. Hydrogenation of the
phenylalkanoic acids 36a,b over Rh/Al2O3 catalyst resulted in
the corresponding cyclohexylalkanoic acids 37a,b. The 4-ha-
lobenzaldehydes 38a,b were converted to the 3-(4-halophenyl)-
propenoic acids 39a,b and then treated with AlCl3 and aryl
halide to provide the acids 40a,b. The 3,3-diarylpropanoic acids
43a-c were prepared via Horner-Wadsworth-Emmons reac-
tion of the corresponding ketones with triethyl phosphonoacetate,
followed by in situ hydrolysis and final hydrogenation over Pd/C
catalyst. 3-(1-Trityl-1H-imidazol-4-yl)propanoic acid (44) was
prepared by hydrolysis of the corresponding methyl ester 8. Acid
45 was synthesized from (5-methyl-1H-imidazol-4-yl)methanol
and sulfanylacetic acid.
Carboxylic Acidsa
According to Scheme 3, the trityl-protected guanidine building
blocks (16, 17, 23) were deprotonated with NaH and coupled
to the pertinent carboxylic acids, which were activated with
N,N′-carbonyldiimidazole (CDI) at room temperature. The
resulting trityl-protected acylguanidines (46-52a, 55-59a,
61-77a) were deprotected using trifluoroacetic acid (TFA) to
yield the acylguanidines as TFA salts (46-52, 55-59, 61-77).
The diphenylpropanoylguanidines 53 and 54 (structures see
Table 1) were synthesized as previously described.32 The
carboxylic acids 44 and 45 were converted to the acyl chorides,
allowed to react with S-methylisothiourea to give the corre-
sponding 1-(3,3-diarylpropionyl)-2-methylisothioureas, which
were treated with 3-(1H-imidazole-4-yl)propylamine to provide
NG-acylated guanidines 53 and 54. Compound 60 (Table 1) was
prepared by hydrogenolytic debenzylation (H2, Pd/C (10%))
of 59.
a Reagents and conditions: (i) n-BuLi, Et2O, -78 °C, 1 h, rt, overnight;
HCl, 0 °C, 20 min; (ii) triethyl phosphonoacetate, NaH, THF, reflux, 24 h;
(iii) H2, Pd/C (10%), MeOH, rt, 24 h; (iv) LiOH, DME/H2O, rt, 3-5 h; (v)
Mg, THF, rt; (vi) MsCl, DMAP, CH2Cl2; (vii) DMM, NaH, THF, rt, 12 h;
(viii) (a) NaOH, reflux; (b) HCl, reflux, 12 h; (ix) H2, Rh/Al2O3, AcOH, rt,
5-6 bar, 40 h; (x) Ac2O, NaOAc, reflux, 6 h; (xi) PhCl or PhBr, AlCl3, rt,
t
3 h; (xii) (a) triethyl phosphonoacetate, KOtBu, BuOH, reflux, 12-16 h;
(b) 1 N NaOH, MeOH, reflux, 24h; (xiii) H2, Pd/C (10%), MeOH, rt, 24 h;
(xiv) LiOH/aq (1 N) (1.2 equiv), THF, rt, 24 h; (xv) HSCH2COOH, Na2CO3,
AcOH, reflux, 24 h.
N-[3-(1-Trityl-1H-imidazol-4-yl)propyl]guanidine (16) was
synthesized starting from urocanic acid (6). After esterification,
hydrogenation of the double bond and trityl protection of the
imidazole NH the ester function was reduced with LiAlH4.30
The alcohol 10 was coupled to the di-Cbz-protected guanidine
13 under Mitsunobu conditions29,31 and subsequent cleavage
of the protecting groups generated the imidazolylpropylguani-
dine building block 16. Starting from 3-phenylbutanoic acid
(11), the arylalkylguanidine building block 17 was obtained by
the reduction of the acid with LiAlH4, yielding 12 followed by
subsequent coupling to guanidine 13 and removal of the Cbz-
groups by analogy with the aforementioned procedure for
compound 16. The imidazolylethylguanidine building block 23
was synthesized starting from histamine dihydrochloride (18).
After trityl protection of both imidazole NH and primary NH2,
the side chain of 19 was selectively detritylated by using a low
concentration of TFA and the guanidine group was introduced
by Goodman’s procedure29 using N,N′-di-Cbz-N′′-triflate pro-
tected guanidine (21) to provide 22. Again, hydrogenation over
Pd/C yielded the imidazolylethylguanidine building block (23).
The pertinent alkanoic acids were synthesized by applying
standard synthetic methods as summarized in Scheme 2. The
ketones 24a-d were prepared from nitriles via the addition of
aryl lithium intermediates, which were generated by lithium-
halogen exchange from the corresponding heteroaryl bromides,
followed by acid hydrolysis. The ketones 24a-d and 28 were
treated with triethyl phosphonoacetate to give the compounds
A small sample of intermediate 49a was isolated and
crystallized for X-ray analysis. The crystal structure (view A)
shows that a hydrogen bond is possible between the oxygen of
amide CO and the hydrogen of that guanidine nitrogen, which
is attached to the alkyl group. In view B, the coplanar orientation
of the guanidine and the CO group is obvious.
Pharmacological Results and Discussion. The synthesized
compounds were investigated for H2R agonistic activity on the
isolated spontaneously beating guinea pig right atrium4 (positive
chronotropic response). Most of the acylguanidines were also
investigated for H1R antagonism on the isolated guinea pig ileum
and on U-373 MG human cells. The pharmacological data are
summarized in Table 2. A selection of compounds was investigated
for H2R selectivity versus H1R, H3R, and H4R using human and
guinea pig histamine receptor models (Table 3) (results of
investigations in GTPase assays on hH2R, gpH2R, hH1R, and
gpH1R were already published in part elsewhere33-38).
Most of the synthesized acylguanidines were full or nearly
full H2R agonists on the spontaneously beating guinea pig right
atrium (Table 2). The suitability of a carbonyl group as a
bioisosteric replacement for a methylene group is strongly
dependent on the substitution pattern of the H2R agonist
molecule, i.e., the substituents R2, R3, and the length of the
connecting chain X (see structures in Table 1). As arpromidine
and related phenyl(heteroaryl)alkyl substituted imidazolylpro-
pylguanidines are the most potent H2R agonists known so far,

7196 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22
Ghorai et al.
Scheme 3. General Procedure for Coupling of Acids with Guanidine Building Blocksa
a Reagents and conditions: (i) CDI (1.2 equiv), NaH (60% dispersion in mineral oil) (2 equiv), THF/abs, 5 h, rt; (ii) 20% TFA, CH2Cl2, 6-10 h, rt.
Substitution patterns, see Table 1.
Table 1. Substitution Patterns of Synthesized Acylguanidines 46-77
no.
R2
R3
X
no.
R2
CH3
CH3
CH3
CH3
CH3
Ph
H
H
H
H
CH3
CH3
c-hexyl
H
H
H
R3
X
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
5-Me-4-imidazolyl
Ph
H
CH2S
CH2
CH2
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
Ph
Ph
Ph
Ph
Ph
CH2
CH2
CH2
CH2
Ph
Ph
4-F-C6H4
Ph
Ph
2-thienyl
2-pyridyl
2-thiazolyl
4-F-C6H4
4-F-C6H4
3,4-diF-C6H3
4-Cl-C6H4
4-Br-C6H4
Ph
4-F-C6H4
Ph
3,4-diF-C6H3
3,4-diF-C6H3
3,4-diF-C6H3
H
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
Ph
Ph
Ph
Ph
4-Cl-C6H4
4-Br-C6H4
2-pyridyl
2-pyridyl
2-thiazolyl
2-thiazolyl
1-benzyl-2-imidazolyl
2-imidazolyl
4-imidazolyl
CH2
(CH2)2
(CH2)3
c-hexyl
c-hexyl
CH2
c-hexyl
c-hexyl
c-hexyl
CH2
(CH2)2
highest potency was expected to reside in the corresponding
acylated analogues, too. “Oxo-arpromidine” (56) was 3.5 times
less potent than arpromidine (5) or half as potent as impromidine
(4); the intrinsic activity remained unaffected. The oxo-
derivative of impromidine (46) was likewise nearly as potent
as the parent compound (4), still acting as a full agonist. The
exchange of the p-fluoro substituted phenyl ring in “oxo-
arpromidine” (56) with an unsubstituted phenyl residue (55) and
further exchange of the 2-pyridyl ring, giving the 3,3-diphe-
nylpropanoyl analogue 48, resulted in about the same agonistic
potency. The introduction of a thiazole ring (57) was tolerated
as well. In the 3,3-diphenylpropanoyl series, p-fluorination and
m,p-difluorination of one ring had a negligible effect on potency
(51 and 52 vs 48). However, the 3,3-bis(4-fluorophenyl)-
derivative (50) was about 7 times less potent than the nonha-
logenated parent compound 48. This is different from the
structure-activity relationships in nonacylated diphenylpropyl-
substituted imidazolylpropylguanidines, where highest potency
was found in fluorinated compounds.20 Fluoro- and chloro- p,p′-
disubstituted diphenylpropanoylguanidines have similar poten-
cies (50 vs 53) and are superior to the bromo analogue (54).
The introduction of different heterocycles instead of an unsub-
stituted phenyl ring in compound 52 resulted in a similar (58,
60) or lower (59) agonistic potency on the guinea pig atrium.
In general, unsubstituted or benzyl-substituted imidazole moi-
eties in the side chain of the acylguanidines were less tolerated
than a thiazole ring. As in the impromidine or arpromidine
series, the trimethylene spacer between imidazole ring and
guanidine group represents the optimum. Compound 47, the
shorter homologue of 48, was significantly less potent and
efficacious. Replacement of methylene by carbonyl in this part
of the molecule was not tolerated; compound 61, the constitu-
tional isomer of the 3-phenylbutanoyl analogue 63, was inactive
as an agonist on the guinea pig right atrium.
The tendency toward lower pEC50 values of the acylguanidines
compared to the alkylguanidines13,20 was confirmed by inves-
tigation of related compounds such as 55, 57, 58, and 60 and
was most obvious for the mono- or difluorinated derivatives
50-52 and 56. Nevertheless, it is conceivable that under in vivo
conditions reduced potency is compensated by improved phar-
macokinetic properties so that a decrease in potency can be
accepted to a certain degree. However, a loss of potency is not
inevitable. Compounds 48, 53 and in particular the 3-(het-
ero)arylbutanoyl substituted guanidines 63-66 were as potent
as the corresponding 3-(hetero)arylbutylguanidines. The poten-
cies are similar, comparing the alkyl- and the acylguanidine
series when one of the aryl rings is replaced with a methyl group.
As in the arpromidine series, a three-membered carbon chain
between guanidine and aromatic ring is optimal. In the series
of NG-acylated guanidines, the 3-phenylbutanoyl derivative (63)
was the most potent compound being 64 times more potent than
histamine on the guinea pig right atrium. The thienyl (64),
pyridyl (65), and thiazolyl (66) analogues are about 26 times
more potent than the reference compound still acting as full
agonists. In contrast to the alkylguanidine series where the
cyclohexyl analogues were about equipotent with the corre-

Acylguanidine-Type Histamine H2 Receptor Agonists
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22 7197
Table 2. Histamine H2 Receptor Agonism on the Guinea Pig Right Atrium, H1 Receptor Antagonism on the Isolated Guinea Pig Ileum, and on U-373
MG Human Cells (Ca2+-Assay)
histamine H2 receptor agonism
histamine H1 receptor antagonism
guinea pig ileum
U-373 MG cells (Ca2+-assay)
isolated guinea pig right atrium
a
no.
pEC50 ( SEM
relative potencyb
Emax (%)c
Nd
pA2 ( SEM (or pD′2 ( SEM)e
Nd
KB (µM)f
1
4
5
6.00 ( 0.02
7.70 ( 0.10
8.01 ( 0.10
7.51 ( 0.09
6.19 ( 0.17
7.55 ( 0.09
5.98 ( 0.18
6.69 ( 0.07
7.26 ( 0.09
7.27 ( 0.05
6.51 ( 0.10
6.11 ( 0.10
7.29 ( 0.03
7.47 ( 0.12
7.42 ( 0.03
7.30 ( 0.06
6.40 ( 0.08
7.01 ( 0.16
5.19 ( 0.19 g
5.44 ( 0.20
7.80 ( 0.07
7.42 ( 0.10
7.42 ( 0.06
7.41 ( 0.11
5.93 ( 0.12
6.74 ( 0.01
6.78 ( 0.10
7.25 ( 0.10
6.33 ( 0.03
6.86 ( 0.11
7.17 ( 0.07
7.32 ( 0.07
7.00 ( 0.10
7.21 ( 0.11
6.43 ( 0.06
100
5000
10200
3200
154
3530
95
490
100
100
100
>50
4
5
4
3
5
3
4
4
4
3
3
4
3
3
3
3
3
3
3
4
3
3
4
3
3
3
4
4
3
3
3
3
3
3
5.47 ( 0.01
7.65 ( 0.01
7
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
101 ( 2
64 ( 3
85 ( 3
69 ( 5
76 ( 2
95 ( 3
81 ( 3
60 ( 4
60 ( 4
97 ( 1
100 ( 1
100 ( 1
90 ( 4
54 ( 5
100 ( 4
7 ( 1h
88 ( 5
99 ( 2
99 ( 2
98 ( 2
91 ( 2
89 ( 4
97 ( 2
97 ( 5
94 ( 2
86 ( 5
86 ( 3
101 ( 3
102 ( 1
95 ( 1
98 ( 3
89 ( 7
5.37 ( 0.05
5.01 ( 0.10 [4.55 ( 0.06]
6.13 ( 0.05 [5.20 ( 0.09]
12
11 16
10 18
8.5
8.7
4.0
1.4
[5.95 ( 0.15]
[5.39 ( 0.03]
[6.06 ( 0.14]
4
4
4
1830
1850
320
1.2
130
1960
2930
2630
2000
251
4.95 ( 0.04
5.43 ( 0.03
5.33 ( 0.05
5.33 ( 0.07
10
12
18
16
4.9
5.2
6.8
4.6
2.9
6.2
9.3
8.8
3.0
2.1
7.4
11.1
15.8
2.8
9.2
7.1
1030
28
6350
2630
2610
2590
86
546
607
1780
211
729
1470
2090
1000
1610
269
<5.0
5.87 ( 0.14
2
4
>5.0
2
2
5
2
4
4
5
2
4
2
4
6.24 ( 0.02
5.67 ( 0.14
5.74 ( 0.20
6.25 ( 0.03
5.63 ( 0.01
5.66 ( 0.09
5.29 ( 0.07
5.61 ( 0.07
5.21 ( 0.26
6.13 ( 0.02
8.8
0.14
6.0
4.3
1.6
1.1
a pEC50 was calculated from the mean shift ∆pEC50 of the agonist curve relative to the histamine reference curve by equation: pEC50 ) 6.00 + 0.13 +
∆pEC50. Summand 0.13 represents the mean desensitization observed for control organs when two successive curves for histamine were performed (0.13 (
0.02, N ) 16). The SEM given for pEC50 is the SEM calculated for ∆pEC50. b Relative potency to histamine ) 100%. c Efficacy, maximal response (%),
relative to the maximal increase in heart rate induced by the reference compound histamine. d Number of experiments. e pD′2 values given in brackets for
compounds producing a significant, concentration-dependent reduction of histamine’s maximal response. f KB values for the inhibition of the histamine (30
µM) induced increase in cellular calcium, mean of two experiments, SEM < 10%. g pA2 (antagonist). h Emax at 58 µM 61, Emax of histamine in the presence
of 58 µM 61 was 45 ( 5%.
Table 3. Agonist/Antagonist Activities on Recombinant Histamine Receptors in GTPase Assaysa
hH1R + RGS4
no. efficacy EC50[nM] (KB[nM]) relative potency efficacy EC50(nM) (KB(nM)) relative potency efficacy
hH2R-GsRb
gpH2R-GsRb
EC50(nM) relative potency
1
1.00
190 ( 8.6
100
6.4
1.0
1.3
1.8
1.00
1200 ( 300
78 ( 42
120 ( 45
67 ( 2
109 ( 31
100
1500
1000
1800
1156
1.00
1,200 ( 200
6 ( 1
100
19000
8570
10000
5710
48 0.19 ( 0.02
57
63 0.35 ( 0.05
64
(2980 ( 719)
(19000 ( 898)
(14300 ( 5303)
(10700 ( 1479)
0.69 ( 0.09
0.80 ( 0.04
0.87 ( 0.01
0.97 ( 0.01
0.93 ( 0.32
0.94 ( 0.05
1.03 ( 0.06
1.05 ( 0.13
14 ( 4
12 ( 1
21 ( 19
hH3R + GRi2+ Gꢀ1γ2 + RGS4
EC50(nM) (KB(nM))
hH4R-RGS19 + GRi2+ Gꢀ1γ2
no.
efficacy
relative potency
efficacy
1.00
0.76 ( 0.03
0.47 ( 0.03
0.84 ( 0.06
0.76 ( 0.03
EC50(nM)
relative potency
1
1.00
25.1 ( 3.1
(17.1 ( 1.5)
(112.7 ( 4.5)
2.45 ( 0.65
1.64 ( 0.96
100
11.6 ( 2.5
8.6 ( 0.9
66.3 ( 10.1
15.3 ( 0.3
6.3 ( 1.2
100
135
17
76
184
48
57
63
64
0.24 ( 0.02
0.29 ( 0.01
1024
1530
a Membrane preparations of Sf9 insect cells expressing hH1R (coexpressed with RGS4), hH2R-GsR, or gpH2R-GsR fusion protein, hH3R (coexpressed with
GRi, Gꢀ1γ2 and RGS4) or hH4R-RGS19 fusion protein coexpressed with GRi2 Gꢀ1γ2 were used. Regulator of G-protein signaling proteins (RGS-proteins)
enhance agonist-steady-state GTP hydrolysis and, thereby, enhance the signal-to noise ratio of the assays.59 For experimental protocols, cf. Supporting
Information. b Data from refs 37, 38 for comparison.
sponding phenyl derivatives,20 the acylguanidines exhibited
different agonistic potencies after introduction of a cyclohexyl
ring. Compounds 63 and 73 showed the most pronounced
differences: while 63 was 64 times more potent than histamine,
the corresponding cyclohexyl analogue 73 was only 15 times
more potent than the reference compound.

7198 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22
Ghorai et al.
Figure 3. Concentration-response curves on the guinea pig right
atrium. Histamine (pEC50 ) 6.06 ( 0.09, N ) 4), compound 63 alone
(pEC50 ) 7.73 ( 0.07, relative potency 63.5 (37.7-107.0), Emax ) 99
( 2%, N ) 4), and 63 in the presence of the H2R antagonist cimetidine
(10 µM, preincubation for 30 min, pA2 ) 6.31 ( 0.06, N ) 3). Data
( SEM.
Representative concentration-response curves for histamine
and compound 63 are shown in Figure 3. The acylguanidine
63 is 64 times more potent than the reference agonist, and the
positive chronotropic response is inhibited by the H2R antagonist
cimetidine, resulting in a rightward shift of the concentration-
response curve. The pA2 determined for cimetidine (6.31) versus
63 as the agonist was not significantly different from the values
obtained versus other agonists like histamine, impromidine, or
arpromidine.
H2R Binding Model and Species Selectivity. The recently
described crystal structures of the ꢀ2-adrenoceptor18 are suited
as a template for homology models of H2Rs because of the
obvious evolutionary relatedness of both receptor subtypes
(sequence identity of about 37%). Figure 4 presents a model of
the gpH2R with a focus on the binding site of imidazolylpro-
pylguanidines. Compound 56, the N1-acylated analogue of
arpromidine, is docked in a favorable extended conformation.
Compared to the crystal structure of 49a (see Figure 2), the
spatial arrangement of the N1-acyl and the N2-imidazolylpropyl
moieties with respect to the guanidino group is changed from
z(N1), e(N2) into e(N1), z(N2). Consequently, the intramolecular
H bond of the CO oxygen is formed with the NH2 moiety of
the protonated species. Attempts to dock compound 56 in the
arrangement of 49a were not successful. The docking mode is
not essentially different from the mode suggested for arpromi-
dine in a model based on bovine rhodopsin.17 Compared to
carazolol complexed with the ꢀ2-adrenoceptor,18,19 the position
of the heterocycles (imidazole vs. carbazole), and of the basic
groups (guanidino carbon vs amino nitrogen) as well as the
projection of the side chains (acyl vs isopropyl) principally
correspond.
Figure 4A shows that the acylated arpromidine analogue may
fit like arpromidine itself into a gpH2R pocket between TMs 2,
3, 5, 6, and 7. The imidazolylpropyl moiety predominantly
contacts amino acids in TMs 5 and 6. The 2-pyridyl group fits
to a relatively narrow site mainly formed by TMs 2 and 7. The
4-fluorophenyl moiety may project outward along residues in
TM7.
Figure 4B represents the suggested binding mode in more
detail. Presumably, the imidazolylpropylguanidine moiety binds
to H2R like histamine (1). Studies with H2R mutants proved an
ionic interaction of the protonated amino group of histamine
with Asp-98 (3.32).39 (Numbers in parentheses indicate the
generic numbering scheme of amino acids in TMs 1-7 proposed
by Ballesteros and Weinstein.40) The second and third site of
the widely accepted three-point model for biogenic amine/GPCR
Figure 4. Model of the gpH2R binding site for compound 56. Both
panels show the side chains and CR atoms of all amino acids within 3
Å around the ligand and, additionally, the putative toggle switch Trp-
247. The backbone and the C atoms of the amino acids are individually
drawn in spectral colors: TM1, red; TM2, orange; TM3, yellow; TM4,
green; E2, cyan; TM5, greenblue; TM6, blue; TM7, purple. All
nitrogens, blue; oxygens, red, C atoms of the ligand, gray. (A) Volume
(space fill) representation of the binding site within a tube model of
the backbone. Additionally, the gpH2R specific residue Tyr-17 in TM1
is shown. (B) Detailed model of the gpH2R-compound 56 interactions.
Drawn are the CR trace (lines), binding site CR atoms and side chains
(sticks), and the ligand (balls and sticks).
interaction could principally be formed by the couples Asp-
186/Thr-190 (5.42/5.46)39 or Tyr-182/Asp-186 (5.38/5.42).16
Although docking in the first mode is generally possible, too,
the pose in Figure 4 reflects the second mode of imidazole
binding. This assumption is also in agreement with a pH-
dependent model of H2R activation that suggests subsequent
tautomerization of the imidazole ring into the Nπ-H form caused
by neutralization of the imidazole upon binding and ac-
companied by proton transfers from Tyr-182 to Nπ and from
Nτ to Asp-186, respectively.41 Interactions of nontautomeric
agonists with H2Rs are compatible with this model, too. In
conclusion, the suggested binding of the imidazolylpropylguani-
dine moiety is governed by two H bonds of the imidazole
nitrogens with Tyr-182 (5.38) and Asp-186 (5.42), respectively,
and by a strong salt bridge or a geometrically favorable charge-
assisted H-bond of the guanidino group with the carboxylate
function of Asp-98 (3.32). The imidazolylpropyl side chain
additionally fits into a pocket consisting of Val-176 (E2), Tyr-
250 (6.51), Phe-251 (6.52), and Phe-254 (6.55).
The model predicts that N1-propanoyl and N1-propyl chains
may fill the same pocket. Because of a possible H bond with
the hydroxy group of Tyr-278 (7.43), the CO oxygen should
actually lead to higher affinity of the acylated derivatives.

Acylguanidine-Type Histamine H2 Receptor Agonists
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22 7199
However, in the ꢀ2-adrenoceptor structure, the corresponding
tyrosine OH is occupied by a strong intramolecular H bond with
the aspartate at position 3.32. On refining the gpH2R model in
complex with compound 56, Tyr-278 approaches a conformation
that rather enables an additional H bond with the guanidine
moiety than interaction with the CO oxygen. Another group
that may contribute to binding of the acyl oxygen is the hydroxy
function of Tyr-94 (3.28).
The complex of gpH2R with compound 56 was derived from
the inactive state of the ꢀ2-adrenoceptor. A reliable model of
the fully active state cannot be obtained without appropriate
crystal structures. However, the docking pose in Figure 4
contains interactions probably inducing correlated rotamer
changes especially of the aromatic residues Tyr-250 (6.51), Phe-
251 (6.52), and Trp-247 (6.48). The tryptophan may serve as
toggle switch, leading to subsequent modification of the proline
kink in TM6.18,43 Tyr-250 and Trp-247 are in contact with Leu-
274 (7.39) and, via water, the backbone oxygen of Gly-277
(7.42), respectively, suggesting that flexibility of TM7 could
also affect the toggle switch. Thus, gpH2R-selective potency
and efficacy may indeed be due to specific interactions of
imidazolylpropylguanidines with TM7.
The suggested pocket for the 2-pyridyl group consists of Gln-
79 (2.65), Asp-271 (7.36), Leu-274 (7.39), Trp-275 (7.40), Tyr-
278 (7.43) and, possibly through interaction via a water
molecule, Tyr-94 (3.28). Trp-275 shapes a boundary of the
binding site, disallowing bulky substituents especially in the
4-position. Considering the p,p′-difluorinated diphenylpro-
panoylguanidine 50 and its dichlorinated analogue 53, the
potency of the former is much lower when compared with the
corresponding bis(4-fluorophenyl)propylguanidine (pEC50 of
6.69 vs 7.7520), whereas 53 is as potent as its propyl analogue.
Thus, the tolerance for a certain bulk is greater in the case of
the alkylguanidines compared to the alkanoylguanidines. This
finding and, in general, the different structure-activity relation-
ships in both series as well as the mostly lower potency of the
acylguanidines must be due to the acyl moiety itself. Because
there is no evidence of principally different binding modes and
interactions (see above), the most probable reason is the
restricted flexibility of the acylguanidines (loss of one freely
rotatable bond compared to the alkylguanidine analogues), i.e.,
the contribution of conformational energy (strain) is simply
higher in the case of the present series. This hypothesis is
supported by the fact that the 3-phenylbutanoyl analogue 63,
whose methyl group may favorably fit into the “2-pyridyl
pocket”, has the highest potency within the series.
Recently, the NG-acylated imidazolylpropylguanidines were
also investigated in a membrane steady-state GTPase activity
assay using fusion proteins of human and guinea pig H2R and
the short splice variant GsR, GsRS (selected data in Table 3, for
additional data cf. refs 37, 38). Similarly to the results on the
guinea pig right atrium, acylguanidines with only one phenyl
or cyclohexyl residue show similar or even higher agonistic
potencies compared to the diaryl analogues at gpH2R-GsRS as
well as hH2R-GsRS.
37,38 Furthermore, as it was shown for N-[3-
(1H-imidazol-4-yl)propyl]guanidines,17 the acylated analogues
generally activate gpH2R with higher potency and efficacy than
hH2R.37,38
Histamine Receptor Subtype Selectivity. All investigated
acylguanidines (46-77, Table 2) were devoid of histamine H1R
agonistic activity on the guinea pig ileum as well as on human
U-373 MG cells. In both test systems, where only weak H1R
antagonism was found, for instance, on the guinea pig ileum,
the compounds were 1-2.5 orders of magnitude less potent than
the moderate H1R antagonist arpromidine (5).
As shown for a set of representative compounds in Table 3,
very weak hH1R antagonistic (57, 64) or partial agonistic activity
(48, 63) was found for most of the acylated imidazolylpropy-
lguanidines studied so far38 in a steady state GTPase assay on
hH1Rs. By contrast, potent partial agonists at the human hH4R
as well as partial agonists and moderate to potent antagonists
at the hH3R were found among the investigated acylguanidines.
This is in agreement with reports on H3R antagonism,44,45 H4R
affinity,46 and H4R (partial) agonism47 of the parent guanidine-
type H2R agonist, impromidine. Surprisingly, the imidazolyl-
propylguanidine portion also confered weak to moderate binding
affinity to nonpeptidic NPY Y1 and Y4 receptor antagonists.48,49
Because acylated and nonacylated imidazolylpropylguanidines
are both capable of interacting with several targets, these
structural moieties may be considered privileged structures.
The p-fluorophenyl moiety of compound 56 projects upward
into the extracellular region of the gpH2R and interacts with
Tyr-78 (2.64) and Asp-271 (7.36). The slight advantage of the
p-F substituent (56 vs 55) may be due to interactions with Lys-
175 (E2) and/or (via water) Glu-267 (7.32).
Although most amino acids interacting with guanidine-type
agonists are identical in the human (hH2R) and guinea pig
(gpH2R) H2 receptor, the guanidine-type H2R agonists are less
potent and less efficient at the H2R of human neutrophils than
at the H2R of the guinea pig atrium.9,13,42 They were also less
potent and efficient at hH2R-GsRS fusion proteins compared to
gpH2R-GsRS fusion proteins in a membrane steady-state GTPase
assay, whereas for the small H2R agonists, such differences were
not detected.17 The model in Figure 4 indicates that Asp-271
in TM7 of the gpH2R is part of the binding site of the dia-
rylpropanoyl moiety. Asp-271 faces both aryl rings of compound
56 (distance 3 to 4 Å) so that weak anion-π interactions are
possible just in the case of electron-withdrawing substituents
like fluorine. In the hH2R, this residue is Ala-271, which may
in part explain the lower potency at this subtype. Site-directed
mutagenesis and previous molecular modeling studies of the
gpH2R suggested that an interaction between Tyr-17 in TM1
and Asp-271 in TM7 accounts for the high efficacy of
guanidines. In the light of the ꢀ2-adrenoceptor based model,
however, the proposed hydrogen bond6,17 is unlikely due to the
large distance of both residues (see Figure 4A). Possibly this
proposed hydrogen bonding can be mediated by water molecules
and include Gln-79 (2.65). Such interactions could not occur
in the hH2R possessing Cys-17 and Ala-271 at these positions.6,17
Hence, the guanidines may stabilize an active conformation in
gpH2R more efficiently and potently than in hH2R.17
Brain Penetration of Selected Compounds. The pKa value
of acylguanidinium cations is by 4-5 orders of magnitude lower
than that of guanidinium ions (guanidinium: pKa ≈ 12.5;
acetylguanidinium: pKa ) 7.650). On one hand, the acylguanidines
are still sufficiently basic to form a cation, which is supposed
to interact with Asp-98 in transmembrane domain 3 of the H2R
by analogy with the binding mode suggested for arpromidine-
like agonists.6,17 On the other hand, the reduced basicity of
acylguanidines results in penetration across the blood-brain
barrier, as a considerable portion of the substance remains
uncharged under physiological conditions. This has been shown
for the representative compounds 58 and 63 in mice.
As shown in Figure 5 as an example at 15 min after
intravenous administration, both compounds (63, 58) were
detected in the brain at approximately equimolar concentrations
compared to plasma (estimation of teminal half-life in plasma

7200 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22
Ghorai et al.
group, the new compounds are superior with respect to
pharmacokinetic properties. As confirmed by HPLC-MS analysis
studies, these compounds are absorbed after oral administration
and are capable of penetrating through the blood-brain barrier.
Thus, this concept is promising with respect to the development
of orally active and centrally available acylguanidines, for
instance, as pharmacological tools to study the role of H2R in
the brain. Relatively low receptor subtype selectivity due to the
privileged structure, imidazolylpropylguanidine, is characteristic
of the compounds presented in this proof-of-concept study.
However, the H2R selectivity of the compounds reported in this
study has been overcome by replacing the imidazole ring with
other heterocycles (to be reported elsewhere). Moreover, the
unexpected high potency of several acylguanidines at HRs other
than H2 provides a sound basis for the further development of
ligands with a different receptor profile, for instance, selective
tools for the H4R.
Figure 5. Detection of acylguanidines 58 and 63 by HPLC-MS analysis
in the brain of nude mice 15 min after iv administration (i.s. ) internal
standard; N1-[3-(1-benzyl-1H-imidazol-2-yl)-3-(3,4-difluorophenyl)pro-
panoyl]-N2-[3-(1H-imidazol-4-yl)propyl]guanidine (59)). The concen-
trations in the brain amounted approximately to those in the plasma
(data not shown).
Experimental Section
Where indicated, reactions were carried out under a dry, oxygen-
free atmosphere of N2 using Schlenk technique or under argon
atmosphere. Commercially available reagents were used as received.
DMF, MeCN, and CH2Cl2 were distilled over P4O10 and stored
under N2 over 3 Å molecular sieves. EtOH and MeOH were dried
over Mg and stored under N2. THF, 1,4-dioxane, and Et2O were
dried with Na/benzophenone and stored over Na wire under N2.
EtOAc, PE, CHCl3, CH2Cl2, MeOH, and hexane for chromato-
graphic separations were distilled before use. For column chroma-
tography silica gel Geduran 60 (Merck, 0.063-0.200 mm) was
used. TLC analysis was done on silica gel 60 F254 (Merck) coated
on aluminum sheets. NMR spectra were recorded on Bruker Avance
300 (1H: 300 MHz, 13C: 75.5 MHz) and Bruker Avance 600 (1H:
600 MHz; 13C: 150.29 MHz) with TMS as internal standard. X-ray
analysis was performed by the Crystallography Laboratory. El-
emental analysis (Heraeus Elementar Vario EL III) and mass
spectrometry (Finnigan ThermoQuest TSQ 7000) were done by the
Central Analytical Laboratory (Universita¨t Regensburg). Preparative
chromatography was performed with a Knauer K-1800 pump, using
a Eurosphere-100 (250 mm × 32 mm) column, was attached to a
Knauer-Detector K-2000 UV detector. The parameters of the
preparative chromatography were as follows: UV detection was
done at 254 or 210 nm, respectively, temperature was 25 °C, and
the flow rate was 40-37 mL/min. The mobile phase was 0.1%
TFA in Millipore water and MeOH or MeCN. Analytical chroma-
tography was performed on a system from Thermo Separation
Products equipped with a SN 400 controller, P4000 pump, an
AS3000 autosampler, and a Spectra Focus UV-vis detector. The
columns were either column A: Nucleodur 100-5 C18 (250 mm
× 4.0 mm, 5 µm), column B: Luna C18 (150 mm × 4.6 mm, 3
µm), column C: Eurosphere-100 C-18 (250 mm × 4.0 mm, 5 µm)
or column D: Purospher 100 C18 (250 mm × 4.0 mm, 5 µm). The
temperature was 30 °C, and the UV detection was set to 254 and
210 nm. The mobile phase was 0.05% TFA in Millipore water and
MeCN.
Figure 6. Representative chromatograms of acylguanidines 58 and
63 detected in the plasma of nude mice by HPLC-MS analysis 30 min
(dashed line) and 180 min (solid line) after oral administration (i.s. )
internal standard; N1-[3-(1-benzyl-1H-imidazol-2-yl)-3-(3,4-difluo-
rophenyl)propanoyl]-N2-[3-(1H-imidazol-4-yl)propyl]guanidine (59)).
from preliminary pharmacokinetic experiments: 67 min for 63
and 28 min for 58).
Resorption of Compounds 58 and 63 after Peroral
Administration. Alkylguanidine-type H2R agonists such as
arpromidine had to be converted to prodrugs by introduction
of hydrolyzable electron-withdrawing substituents at the third
guanidine nitrogen to obtain orally active “cardiohistaminer-
gics”.12 As shown in Figure 6 the conversion of a methylene
group adjacent to the guanidine group into a carbonyl group as
in 58 and 63 led to absorption from the gastrointestinal tract.
Substantial amounts were detected in the plasma of mice after
oral administration. As the prototype acylguanidine-type H2R
agonists described in this article are not ideal with respect to
receptor selectivity, detailed pharmacokinetic studies will be the
subject of future work with optimized compounds.
Synthetic protocols for the preparation of starting material and
analytical data of the following compounds are available as
Supporting Information: building blocks 10, 12, 13, 19-21, 24a-d,
25a-d, 29, 26a-d, 30, 27a-d, 31, 32a,b, 33a,b, 34a,b, 35a,b,
37a,b, 39a,b, 40a,b, 42a-c, 43a-c, 44, and 45, trityl-protected
intermediates 47a-52a, 55a-59a, 61a-77a, and acylguanidines
47-52, 55, 57-62, 65-77.
General Procedure for the Synthesis of N,N′-Bis(benzyloxy-
carbonyl)alkylguanidines (14, 15). A solution of the alcohol (11.9
mmol), di-Cbz protected guanidine 1329(21.1 mmol), and PPh3 (18
mmol) in 100 mL of THF/abs was cooled to -5 °C under an argon
atmosphere. DIAD (18 mmol) 30 mL THF/abs was added dropwise
at such a rate that the reaction mixture was completely colorless
before addition of the next drop. After the addition was complete,
the reaction mixture was stirred at room temperature for 24 h. The
Conclusion
Alkanoylguanidines and alkylguanidines are bioisosters for
H2Rs, with the acylguanidines exhibiting a basicity 4-5 orders
of magnitude lower than that of the parent guanidines. Although
a decrease in potency was found when the methylene group in
arpromidine-like H2R agonists was exchanged by a carbonyl

Acylguanidine-Type Histamine H2 Receptor Agonists
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22 7201
solvent was removed in vacuo, and the crude product subjected to
flash-chromatography.
sodium hydride (2 mmol, 80 mg of a 60% dispersion in oil) was
added to a solution of an imidazolylalkylguanidine (1 mmol) in
DMF or THF (5 mL), and the mixture was heated at 50 °C for 30
min and then allowed to cool to rt. The two mixtures were combined
and stirred at rt for 5 h, then the solution was poured on water (10
mL) and extracted with EtOAc. The organic phase was washed
with saturated sodium bicarbonate and brine, dried over anhydrous
Na2SO4, and evaporated in vacuo. The residue was purified by a
Chromatotron under a NH3 atmosphere with CHCl3/MeOH as
solvent with yields between 70% and 30%. Subsequently, the
obtained trityl protected acylguanidines (46a not characterized, data
of 47a-52a, 55a-59a, 61a-77a cf. Supporting Information) were
dissolved in dry CH2Cl2 (8 mL), 2 mL of TFA were added drop
by drop with constant stirring for 15 min at rt, and stirring was
continued till the starting material had completely disappeared
(6-10 h). After removing the solvent in vacuo, the residue was
purified by preparative HPLC, using mobile phase 0.1% TFA in
water and MeOH or MeCN (experimental setup is given at general
conditions). All compounds were obtained with yields between 20%
and 40% based on the trityl protected products, and their purity
was checked by analytical HPLC (experimental setup is given at
general conditions, HPLC data available as Supporting Information).
Four representative compounds (46, 56, 63, 64) are described in
the following, the data of all other acylguanidines prepared
according to the general procedure are available as Supporting
Information.
N,N′-Bis(benzyloxycarbonyl)-[3-(1-trityl-1H-imidazol-4-yl-
)propyl]guanidine (14). Synthesized from 10 and 13; flash-
chromatography (PE/EtOAc 60/40); yield 70%; colorless foam-
1
like solid. H NMR (CDCl3) δ (ppm): 7.20-7.48 (m, 20H, Im-2-
H, Ar-H), 7.03-7.14 (m, 6H, Ar-H), 6.57 (s, 1H, Im-5-H), 5.19
(s, 2H, CH2Ph), 5.06 (s, 2H, CH2Ph), 4.02 (t, 2H, J ) 7.3 Hz,
CH2N), 2.59 (t, 2H, J ) 7.7 Hz, ImCH2), 1.95 (m, 2H,
CH2CH2CH2). ES-MS (CH2Cl2/MeOH + 10 mM NH4OAc) m/z
(%): 678 (MH+, 100); C42H39N5O4 (677.8).
N,N′-Bis(benzyloxycarbonyl)-(3-phenylbutyl)guanidine (15).
Synthesized from 12 and 13; flash-chromatography (PE/EtOAc 90/
10); yield 70%; colorless oil. 1H NMR (CDCl3) δ (ppm): 7.36-7.12
(m, 15H, Ar-H), 5.16 (s, 2H, CH2-Ar), 5.15 (s, 2H, CH2Ar),
3.99-3.77 (m, 2H, CH2N), 2.74 (m, 1H, CH3CH), 1.89 (m, 2H,
CH3CHCH2), 1.19 (d, 3H, J ) 6.9 Hz, CH3). ES-MS (CH2Cl2/
MeOH + 10 mM NH4OAc) m/z (%): 460 (MH+, 100); C27H29N3O4
(459.5).
N,N′-Bis(benzyloxycarbonyl)-[2-(1-trityl-1H-imidazol-4-yl-
)ethyl]guanidine (22). The amine 20 (1.0 mmol) was added to a
solution of 2129 (0.9 mmol) and triethylamine (1.0 mmol) in 5 mL
of CH2Cl2, and the mixture was allowed to stir at room temperature
until 21 was consumed (3 h) as evidenced by TLC. After the
reaction was complete, the mixture was diluted with CH2Cl2 (6
mL) and washed with 2 M sodium bisulfate, saturated sodium
bicarbonate, and brine. The organic extract was then dried over
sodium sulfate and filtered, and the solvent was removed under
reduced pressure. The crude product was further purified by flash
column chromatography with CH2Cl2. Yield 92%; colorless crystal-
line solid; mp ) 98 °C. 1H NMR (CDCl3) δ (ppm): 11.70 (s, 1H,
NH), 8.64 (t, J ) 5.4 Hz, 1H, NH), 7.43 (s, 1H, Im-2-H), 7.27-7.40
(m, 19H, Ar-H), 7.10-7.16 (m, 6H, Ar-H), 6.64 (s, 1H, Im-5-H),
5.15 (s, 2H, PhCH2O), 5.10 (s, 2H, PhCH2O), 3.74 (dd, J ) 6.3
Hz, J ) 12.0 Hz, 2H, CH2NH), 2.81 (t, J ) 6.4 Hz, Im-4-CH2).
ES-MS (CH2Cl2/MeOH + 10 mM NH4OAc), m/z: 664 (MH+), 243
([Ph3C]+); C41H37N5O4 (663.8).
General Procedure for the Synthesis of the Deprotected
Guanidines 16, 17, 23. To a solution of 14, 15, or 22 in THF/
MeOH (1:1) was added Pd/C (10%), and the mixture was
hydrogenated at 5 bar overnight. The mixture was filtered through
a small pad of celite, washed with MeOH, and the solvent was
removed in vacuo.
N-[3-(1-Trityl-1H-imidazol-4-yl)propyl]guanidine (16). Syn-
thesized from 14 (9.7 mmol) using Pd/C (10%) (cat.) in 165 mL
THF/MeOH (1:1); yield 98%; pale-yellow foam-like solid. 1H NMR
(CD3OD) δ (ppm): 7.38-7.14 (m, 16H, CPh3, Im-2-H), 6.70 (s,
1H, Im-5-H), 3.17 (t, 2H, J ) 6.9 Hz, CH2NH), 2.58 (t, 2H, J )
7.4 Hz, Im-4-CH2), 1.86 (m, 2H, Im-4-CH2CH2). ES-MS (CH2Cl2/
MeOH + 10 mM NH4OAc) m/z (%): 410 (MH+, 100); C26H27N5
(409.5).
N1-[2-(5-Methyl-1H-imidazol-4-ylmethylsulfanyl)acetyl]-N2-
[3-(1H-imidazol-4-yl)propyl]guanidine (46). Synthesized from (5-
methyl-1H-imidazol-4-ylmethylsulfanyl)acetic acid (45) and 16 via
1
detritylation of 46a; yield 54%; sticky oil. H NMR (CD3OD) δ
(ppm): 8.81 (s, 1H, Im-2-H), 8.75 (s, 1H, Im-2-H), 7.37 (s, 1H,
Im-5-H), 3.95 (s, 2H, SCH2CO), 3.41-3.33 (m, 2H, NHCH2), 3.31
(s, 2H, Im-4-CH2S), 2.90-2.80 (m, 2H, Im-4-CH2), 2.36 (s, 3H,
CH3), 2.04 (m, 2H, Im-4-CH2CH2). 13C NMR (CD3OD) δ (ppm):
173.4 (quart, CO), 155.4 (quart, C ) NH), 134.9 (+, Im-C-2), 134.3
(quart, Im-C-4), 134.2 (+, Im-C-2), 128.8 (quart, Im-C-4), 126.5
(quart, Im-C-5), 117.1 (+, Im-C-5), 41.6 (-, CH2NH), 35.7 (-,
SCH2CO), 27.9 (-, Im-4-CH2CH2), 25.0 (-, Im-4-CH2S), 22.5 (-,
Im-4-CH2), 8.9 (+, CH3). MS (ESI, CH2Cl2/MeOH + 10 mM
NH4OAc): 336 (MH+). HRMS [FAB(glycerol)]: m/z, calculated
for [C14H21N7OS + H]+ 336.1601, found, 336.1607; C14H21N7OS·
3TFA (677.5).
N1-[3-(4-Fluorophenyl)-3-(pyridin-2-yl)propanoyl]-N2-[3-(1H-
imidazol-4-yl)-propyl]guanidine (56). Synthesized from 3-(4-
fluorophenyl)-3-(pyridin-2-yl)propanoic acid (27b) and 16 via
detritylation of 56a; yield 39%; colorless sticky oil. 1H NMR
(CD3OD) δ (ppm): 8.78 (s, 1H, Im-2-H), 8.71 (d, J ) 5.2 Hz,1H,
Pyr-4-H), 8.57 (m, 1H, Pyr-H), 8.38 (t, J ) 7.8 Hz, 1H, Pyr-H),
8.00 (d, J ) 8.2 Hz, 1H, Pyr-H), 7.78 (t, J ) 6.7 Hz, 1H, Ph-H),
7.50-7.40 (m, 2H, Ph-H and Im-H), 7.09 (t, J ) 7.8 Hz, 2H, Ph-
H), 4.84 (t overlap with H2O, 1H, Ar2CH), 3.68 (dd, J ) 9.0 Hz,
J ) 17.3 Hz, 1H, CHHCO), 3.43 (dd, J ) 6.6 Hz, J ) 17.3 Hz,
1H, CHHCO), 3.35-3.30 (m, 2H, NHCH2), 2.82 (t, J ) 7.5 Hz,
2H, Im-4-CH2), 2.00 (m, 2H, Im-4-CH2CH2). 13C NMR (CD3OD)
δ (ppm): 174.4 (quart, CO), 163.7 (quart, d, J ) 246.1 Hz, CF),
159.8 (quart, Py-C-2), 155.1 (quart, C ) NH), 146.2 (+, Py-C-6),
144.5 (+, Im-C-2), 136.3 (quart, Ar-C-1), 134.9 (+, Pyr-C-4), 134.2
(quart, Im-C-4), 131.1 (+, d, J ) 8.2 Hz, 2C, Ar-C-2/-6), 126.6
(+, Pyr-C-3), 125.9 (+, Pyr-C-5), 117.1 (+, d, J ) 21.1 Hz, 2C,
Ar-C-3/-5), 117.0 (+, Im-C-5), 45.9 (+, COCH2CH), 41.8 (-,
COCH2), 41.6 (-, NHCH2), 28.0 (-, Im-4-CH2CH2), 22.4 (-, Im-
4-CH2). ES-MS (CH2Cl2/MeOH + 10 mM NH4OAc): 395 (MH+).
HRMS [EI-MS]: m/z, calculated for [C21H23FN6O] 394.1917, found,
394.1917; C21H23FN6O·3TFA (736.5).
3-Phenylbutylguanidine (17). Synthesized from 15 (2.46 mmol)
in 40 mL THF/MeOH (1:1) was added Pd/C (10%) (cat.); yield )
1
100%; yellow amorphous solid. H NMR (CDCl3) δ (ppm): 7.26
(m, 5H, Ar-H), 3.03 (t, 2H, J ) 7.0 Hz, NHCH2), 2.81 (m, 1H,
CH3CH), 1.88 (m, 2H, CH3CHCH2), 1.29 (d, 3H, J ) 6.9 Hz, CH3).
CI-MS (NH3) m/z (%): 192 (MH+, 100); C11H17N3 (191.3).
[2-(1-Trityl-1H-imidazol-4-yl)ethyl]guanidine (23). Synthe-
sized from 22 (5 mmol) in 15 mL THF: MeOH (1:1) was added
Pd/C (10%) (cat.); yield 96%; hygroscopic semisolid. 1H NMR
(CDCl3) δ (ppm): 8.51 (s, 1H, NH), 7.56 (s, 1H, Im-2-H),
7.43-7.01 (m, 15H, Ph-H), 6.86 (s, 1H, Im-5-H), 3.43 (t, J ) 7.1
Hz, 2H, CH2NH), 3.31 (m, 1H, NH), 2.80 (t, J ) 7.1 Hz, Im-4-
CH2). ES-MS (CH2Cl2/MeOH + 10 mM NH4OAc), m/z: 791 ([2
M + H]+), 396 (MH+), 243 ([Ph3C]+); C25H25N5 (395.5).
General Procedure for the Preparation of the N1-acyl-N2-
[1H-imidazol-4-yl)alkyl]guanidines (46-52, 55-59, 61-77).
Under an argon atmosphere, the pertinent carboxyclic acid (1.0
mmol) was added to a solution of CDI (0.178 g, 1.1 mmol) in DMF
or THF (3 mL), and the mixture was stirred at rt for 1 h. At the
same time, in a separate flask under argon atmosphere with stirring,
N1-[3-(1H-Imidazol-4-yl)propyl]-N2-(3-phenylbutanoyl)guani-
dine (63). Synthesized from 3-phenylbutanoic acid and 16 via
detritylation of 63a; yield: 20%; colorless sticky oil. 1H NMR
(CD3OD) δ (ppm): 8.77 (d, 1H, J ) 1.3 Hz, Im-2-H), 7.34 (s, 1H,
Im-5-H), 7.18-7.27 (m, 5H, Ph), 3.33 (m, 3H, Im-4-CH2CH2CH2,
COCH2CH), 2.78 (m, 4H, Im-4-CH2CH2, COCH2), 2.00 (m, 2H,
Im-4-CH2CH2), 1.31 (d, 3H, J ) 7.0 Hz, CH3). 13C NMR (CD3OD)

7202 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22
Ghorai et al.
δ (ppm): 176.02 (quart, CO), 155.31 (quart, CdNH), 146.41 (quart.
Ph-C-1), 134.93 (+, Im-C-2), 134.30 (quart, Im-C-4), 127.67,
127.95, 129.63 (+, arom. CH), 116.96 (+, Im-C-5), 46.06 (-,
COCH2), 41.49 (-, Im-4-CH2CH2CH2), 37.63 (+, COCH2CH),
27.91 (-, Im-4-CH2CH2), 23.49 (-, Im-4-CH2CH2), 22.20 (+,
CH3). HRMS: EI-MS: m/z for [C17H23N5O] calcd 313.1903, found
313.1902; C17H23N5O·2TFA (541.4).
addition of 100 µL of acetonitrile, mixing, and centrifugation (13000
rpm, 5 min, Biofuge 13, Heraeus, Germany). After removal of the
supernatants, residues were resuspended in 1 mL of 1% aqueous
TFA and 100 µL of acetonitrile, mixed, and centrifuged. Solid phase
extraction cartridges (Strata-X-C, 3 mL, 60 mg, Phenomenex,
Aschaffenburg, Germany), conditioned with 2 mL of methanol and
2 mL of water, were loaded with the combined supernatants.
Washing was performed with 2 mL of hydrochloric acid (0.1 N)
and 2 mL of a mixture of water and methanol (95/5). Cartridges
were dried prior to elution with 3 mL of an ammonia enriched
(3.5 N) mixture of acetonitrile and methanol (50/50). The eluates
were evaporated in a vacuum concentrator (Speed-Vac Plus,
Egelsbach, Germany), residues were dissolved in 300 µL of a
mixture of acetonitrile and 0.1% TFA in water (20/80), filtered (0.45
µm), and injected into the LC-MS system (LC: Agilent 1100, Palo
Alto, CA; MS: TSQ 7000, Thermo FINNIGAN, USA). HPLC
conditions: column: Synergi Hydro RP (250 mm × 4.6 mm, 4 µm,
Phenomenex, Aschaffenburg, Germany), mobile phase: mixtures
of acetonitrile (A) and 0.1% TFA in water (B), gradient: 0-6 min:
A/B: 20/80, 6-20 min: A/B: 20/80 f 32/68, 20-20.5 min: A/B:
32/68 f 80/20, 20.5-25 min: A/B: 80/20, flow rate: 0.8 mL/min,
oven temperature: 45 °C, UV-detection: 214 nm, injection: 50 µL.
MS-conditions: single ion monitoring (SIM), source type: ESI
(capillary temperature: 350 °C, spray voltage: 4.0 kV, sheath and
auxiliary gas: on).
Absorption after Oral Administration. A mixture of 63 and
58 was administered to male nude mice (nu/nu, weight: 30-35 g)
by gavage at a dose of 5 mg/kg each in isotonic sodium chloride
solution ((final DMSO concentration: 5%). Blood was collected
by cardiac puncture during sevofluran anesthesia and centrifuged.
Heparin plasma (200 µL) was used for sample preparation. After
addition of the internal standard (59), proteins were precipitated
by addition of 400 µL of ice-cold acetonitrile, strong mixing,
incubation at -20 °C for 30 min, and additional vigorous mixing.
After centrifugation at 5 °C (4000 rpm, 5 min, MinifugeT, Heraeus,
Hanau, Germany), supernatants were transferred to 1.5 mL reaction
vessels (Eppendorf, Hamburg, Germany) and residues were washed
with 300 µL of ice-cold acetonitrile followed by centrifugation.
Combined supernatants were evaporated to dryness in a vacuum
concentrator (Speed-Vac Plus, Egelsbach, Germany), residues were
dissolved in 300 µL of a mixture of acetonitrile and 0.1% TFA in
water (20/80), filtered (0.45 µm), and analyzed by LC-MS as
described above.
N1-[3-(1H-Imidazol-4-yl)propyl]-N2-[3-(thiophen-2-yl)bu-
tanoyl]guanidine (64). Synthesized from (3-thiophen-2-yl)butanoic
acid (31) and 16 via detritylation of 64a; yield 43%; colorless sticky
1
oil. H NMR (CD3OD) δ (ppm): 8.84 (s, 1H, Im-2-H), 7.48 (s,
1H, Im-5-H), 7.30 (m, 1H, Thio-H), 6.93 (m, 2H, Thio-H), 3.60
(m, 1H, COCH2CH), 3.32 (m, 2H, NHCH2), 2.92 (m, 4H, Im-4-
CH2 and one of COCH2), 2.00 (m, 2H, Im-4-CH2CH2), 1.42 (d, J
) 7.0 Hz, 3H, CH3). 13C NMR (CD3OD) δ (ppm): 175.6 (quart,
CO), 155.2 (quart, CdNH), 150.0 (+, Thio-C-2), 134.9 (+, Im-
C-2), 134.3 (quart, Im-C-4), 127.7 (+, Thio-C-5), 126.3, 124.4 (+,
Thio-C-3,4), 117.1 (+, Im-C-5), 47.0 (-, CHCH2CO), 41.6 (-,
CH2NH), 32.9 (+, CHCH2CO), 27.9 (-, Im-4-CH2CH2), 23.2 (+,
CH3), 22.5 (-, Im-4-CH2). MS (ESI, CH2Cl2/MeOH + 10 mM
NH4OAc): m/z 320 (MH+). HRMS: EI-MS: m/z for [C15H21N5OS]
calcd 319.1467, found 319.1469; C15H21N5OS·2TFA (547.5).
General Procedure for the Preparation of the N1-(3,3-diphe-
nylpropanoyl)-N2-[1H-imidazol-4-yl)propyl]guanidines 53, 54.32
The corresponding p,p′-dihalogenated 1-(3,3-diphenylpropanoyl)-
2-methylisothiourea (1 mmol), prepared from the respective car-
boxylic acid chloride and S-methylisothiourea hydroiodide in the
presence of triethylamine (2 equiv) in acetone at rt, was dissolved
in 10 mL of anhydrous MeCN and added to a solution of 3-(1H-
imidazol-4-yl)propylamine in 10 mL MeCN. The mixture was
stirred for 14 h at 60 °C, the solvent subsequently removed in vacuo,
redissolved in very little 0.1 N HCl/aq, and the solvent immediately
removed in vacuo. The remaining foam was repeatedly dissolved
in EtOH and the solvent removed in vacuo. The residue was purified
by a Chromatotron using CHCl3/MeOH (7:1).
N1-[3,3-Bis(4-chlorophenyl)propanoyl]-N2-[3-(1H-imidazol-4-
yl)propyl]guanidine (53). Synthesized from 1-[3,3-bis(4-chlo-
rophenyl)propanoyl]-2-methylisothiourea and 3-(1H-imidazol-4-
yl)propylamine; yield 33%; colorless hygroscopic foam. 1H NMR
(DMSO-d6) δ (ppm): 14.40 (br, 2H, NH2), 12.40 (s, 1H, NH), 9.06
(br, 1H, NH), 9.00 (s, 1H, Im-2-H), 7.45 (s, 1H, Im-5-H), 7.26-7.40
(m, 8H, Ar-H), 4.60-4.69 (m, 1H, Ar2CH), 3.11-3.37 (m, 4H,
Ar2CHCH2, NHCH2), 2.49 (d, J ) 7.5 Hz, 2H, Im-4-CH2),
1.81-1.90 (m, 2H, Im-4-CH2CH2). HRMS (FAB+, CH2Cl2,
MeOH) m/z, calculated for [C22H23Cl2N5O] 444.1357, found,
444.1351; C22H23Cl2N5O·2 TFA (672.4).
Pharmacological Methods. The experimental protocols for the
determination of pharmacological parameters on the guinea pig right
atrium (H2R), the guinea pig ileum (H1R), U373-MG cells (hH1R,
calcium assay), the GTPase assays on membrane preparations of
Sf9 cells expressing the human H1, H2, H3, or H4 receptor, or the
guinea pig H2R, follow the standard procedures provided as
Supporting Information.
N1-[3,3-Bis(4-bromophenyl)propanoyl]-N2-[3-(1H-imidazol-
4-yl)propyl]guanidine (54). Synthesized from 1-[3,3-bis(4-bro-
mophenyl)propanoyl]-2-methylisothiourea and 3-(1H-imidazol-4-
yl)propylamine; yield 48%; colorless hygroscopic foam. 1H NMR
(DMSO-d6) δ (ppm): 14.40 (br, 2H, NH2), 12.40 (s, 1H, NH), 9.10
(br, 1H, NH), 9.02 (s, 1H, Im-2-H), 7.42-7.56 (m, 4H, Ar-H, Im-
5-H), 7.25-7.35 (m, 4H, Ar-H), 4.62 (app-t, J ) 7.9 Hz, 1H,
Ar2CH), 3.20-3.30 (m, 4H, Ar2CHCH2, NHCH2), 2.62-2.71 (m,
2H, Im-4-CH2), 1.80-1.91 (m, 2H, Im-4-CH2CH2). HRMS (FAB+,
CH2Cl2, MeOH) m/z, calculated for [C22H23Br2N5O] 532.0347,
found, 532.0347; C22H23Br2N5O·2TFA (761.3).
Analysis of Compounds 63 and 58 in Brain and Blood.
Detection in Brain after Systemic Administration. To investigate
the penetration of 63 and 58 across the blood-brain barrier, a
mixture of these compounds was administered to male nude mice
(nu/nu, weight: 30-35 g) by retrobulbar injection in an isotonic
solution of sodium chloride (final DMSO concentration: 10%) at a
dosage of 10 mg/kg each (concentration: 2 + 2 mg/mL, injected
volume: 50 µL/10 g mouse). Mice were killed 15 min after injection
by cardiac puncture during anesthesia with sevofluran, followed
by removal of the brains. The tissues were homogenized (1000 rpm,
on ice) in 1% aqueous TFA (150 mg tissue/1 mL) with a
Potter-Elvehjem homogenizer (type 8, 5 mL-vessel, Teflon pestle,
Braun, Melsungen, Germany). One mL of the homogenate was
mixed vigorously with the internal standard (59), followed by the
Data Handling and Pharmacological Parameters. Data pre-
sented as mean ( SEM or SE or with 95% confidence limits (cl)
unless otherwise indicated. Agonist potencies are given in percent
or are expressed as pEC50 values (negative decadic logarithm of
the molar concentration of the agonist producing 50% of the
maximal response) and were corrected according to the long-term
mean value of the reference agonist histamine in our laboratory
(guinea pig atrium (H2): pEC50 ) 6.00 for histamine). Maximal
responses are expressed as Emax values (percentage of the maximal
response to histamine as reference). Antagonist affinities are
expressed apparent pA2 values and were calculated from the
following equation: pA2 ) -log c(B) + log(r - 1), where c(B) is
the molar concentration of antagonist and r the ratio of agonist
EC50 measured in the presence and absence of antagonist.51,52
Noncompetitive antagonists are characterized by estimation of a
pD′2 value according to the equation: pD′2 ) -log c(B) + log(100/
Emax - 1).53 Where appropriate, differences between means were
determined by Student’s t test, after checking the homogeneity of
the variances; P values <0.05 were considered to indicate a
significant difference between the mean values being compared.

Acylguanidine-Type Histamine H2 Receptor Agonists
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22 7203
(4) Black, J. W.; Duncan, W. A.; Durant, C. J.; Ganellin, C. R.; Parsons,
E. M. Definition and antagonism of histamine H2-receptors. Nature
(London) 1972, 236, 385–390.
(5) Traiffort, E.; Pollard, H.; Moreau, J.; Ruat, M.; Schwartz, J.-C.;
Martinez-Mir, M. I.; Palacios, J. M. Pharmacological Characterization
and Autoradiographic Localization of Histamine H2 Receptors in
Human Brain Identified with [125I]Iodoaminopotentidine. J. Neuro-
chem. 1992, 59, 290–299.
KB values for H1R antagonism in the calcium assay (U373-MG
cells)werecalculatedfromIC50 valuesaccordingtotheCheng-Prusoff
equation.54
Molecular Modeling. First, a rough model of gpH2R was directly
constructed from the PDB file 2RH1 of the ꢀ2-adrenoceptor crystal
structure.18 The backbone coordinates of TMs 1-7, helix 8, the
loops C1, E1, and C2, corresponding parts of E2, intrahelical water
molecules, as well as identical side chains were retained. All other
cocrystallized substructures and the lysozyme domain were deleted.
The beginning (Lys-227-Val-242) and the end (Phe-264-Glu-268)
of the C3 loop were replaced by the corresponding parts of the
crystal structure 2R4R55 (the free state of these positions is probably
better represented by the complex with a Fab than by a covalent
construct). Also the C terminus was added from the 2R4R file. The
remainder of the E2 and C3 loops, the E3 loop and the motif
between TM7 and TM8 were constructed by Sybyl loop searches.
The side chains differing between the ꢀ2-adrenoceptor and the
gpH2R were mutated.
The resulting rough model was prepared for optimization by
adding hydrogens and providing atoms with Amber-FF99 charges.56
Bad contacts of side chains were removed using the Lovell rotamer
library.57 Subsequently, the model was roughly minimized by the
steepest descent method (200 cycles, Amber-FF99 force field,56
distant dependent dielectric constant of 4). In the first 100 steps,
the backbone of the TMs was fixed. Compound 56 was manually
docked into the model in an extended, energetically favorable
conformation, considering results of previous in vitro mutagenesis
and theoretical studies.6,16,17,39,41 The complex was first minimized
with the Amber-FF99 force field (ligand fixed and provided with
Gasteiger-Hueckel charges, distant dependent dielectric constant
of 4, 20 cycles steepest descent, then Powell algorithm) down to
an rms gradient of less than 0.05 kcal mol-1 Å-1. The final
minimization considered compound 56 and a region of 6 Å around
(combined organic-protein MMFF94 force field58 and charges,
distant dependent dielectric constant of 1, Powell method, rms
gradient criterion 0.05 kcal mol-1 Å-1). The rms deviation of the
TM backbones between the resulting model and the ꢀ2-adrenoceptor
crystal structure 2RH1 amounts to 0.68 Å. All calculations were
performed with SYBYL 7.3 (Tripos, St. Louis, MO) on a SGI
Octane workstation.
(6) Dove, S.; Elz, S.; Seifert, R.; Buschauer, A. Structure-Activity
Relationships of Histamine H2 Receptor Ligands. Mini-ReV. Med.
Chem. 2004, 4, 941–954.
(7) Felix, S. B.; Buschauer, A.; Baumann, G. Therapeutic value of H2-
receptor stimulation in congestive heart failure. Hemodynamic effects
of BU-E-76, BU-E-75 and arpromidine (BU-E-50) in comparison to
impromidine. Agents Actions Suppl. 1991, 33, 257–269.
(8) Seifert, R.; Hoer, A.; Schwaner, I.; Buschauer, A. Histamine increases
cytosolic Ca2+ in HL-60 promyelocytes predominantly via H2 receptors
with an unique agonist/antagonist profile and induces functional
differentiation. Mol. Pharmacol. 1992, 42, 235–241.
(9) Burde, R.; Buschauer, A.; Seifert, R. Characterization of histamine
H2-receptors in human neutrophils with a series of guanidine analogues
of impromidine. Naunyn-Schmiedeberg’s Arch. Pharmacol. 1990, 341,
455–461.
(10) Baumann, G.; Felix, S. B.; Heidecke, C. D.; Rieꢀ, G.; Loher, U.;
Ludwig, L.; Blo¨mer, H. Apparent superiority of H2-receptor stimulation
and simultaneous ꢀ-blockade over conventional treatment with ꢀ-sym-
pathomimetic drugs in post-acute myocardial infarction: Cardiac effects
of impromidinesa new specific H2-receptor agonistsin the surviving
catecholamine-insensitive myocardium. Inflamm. Res. 1984, 15, 216–
228.
(11) Baumann, G.; Permanetter, B.; Wirtzfeld, A. Possible value of H2-
receptor agonists for treatment of catecholamine-insensitive congestive
heart failure. Pharmacol. Ther. 1984, 24, 165–177.
(12) Mo¨rsdorf, P.; Engler, H.; Schickaneder, H.; Buschauer, A.; Schunack,
W.; Baumann, G. Cardiohistaminergicssnew developments in hista-
mine H2-agonists. Drugs Future 1990, 15, 919–933.
(13) Buschauer, A. Synthesis and in vitro pharmacology of arpromidine
and related phenyl(pyridylalkyl)guanidines, a potential new class of
positive inotropic drugs. J. Med. Chem. 1989, 32, 1963–1970.
(14) Buschauer, A.; Baumann, G. Structure-activity relationships of
histamine H2-agonists, a new class of positive inotropic drugs. Agents
Actions Suppl. 1991, 33, 231–256.
(15) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Motoshima,
H.; Fox, B. A.; Trong, I. L.; Teller, D. C.; Okada, T.; Stenkamp, R. E.;
Yamamoto, M.; Miyano, M. Crystal Structure of Rhodopsin: A G
Protein-Coupled Receptor. Science 2000, 289, 739–745.
(16) Nederkoorn, P. H. J.; Gelder, E. M.; Donne´-Op den Kelder, G. M.;
Timmerman, H. The agonistic binding site at the histamine H2 receptor.
II. Theoretical investigations of histamine binding to receptor models
of the seven R-helical transmembrane domain. J. Comput.-Aided Mol.
Des. 1996, 10, 479–489.
(17) Kelley, M. T.; Bu¨rckstu¨mmer, T.; Wenzel-Seifert, K.; Dove, S.;
Buschauer, A.; Seifert, R. Distinct interaction of human and guinea
pig histamine H2-receptor with guanidine-type agonists. Mol. Phar-
macol. 2001, 60, 1210–1225.
Acknowledgment. We are grateful to Dr. Hendrik Preuss
for helpful discussion and Astrid Seefeld, Gertraud Wilberg,
Elvira Schreiber, Christine Braun, and Kerstin Ro¨hrl for expert
technical assistance. This work was supported by the Graduate
Training Program (Graduiertenkolleg) GRK 760, “Medicinal
Chemistry: Molecular RecognitionsLigand-Receptor Interac-
tions”, of the Deutsche Forschungsgemeinschaft.
(18) Cherezov, V.; Rosenbaum, D. M.; Hanson, M. A.; Rasmussen, S. G.;
Thian, F. S.; Kobilka, T. S.; Choi, H. J.; Kuhn, P.; Weis, W. I.;
Kobilka, B. K.; Stevens, R. C. High-resolution crystal structure of an
engineered human beta2-adrenergic G protein-coupled receptor. Sci-
ence 2007, 318, 1258–1265.
(19) Rosenbaum, D. M.; Cherezov, V.; Hanson, M. A.; Rasmussen, S. G.;
Thian, F. S.; Kobilka, T. S.; Choi, H. J.; Yao, X. J.; Weis, W. I.;
Stevens, R. C.; Kobilka, B. K. GPCR engineering yields high-
resolution structural insights into beta2-adrenergic receptor function.
Science 2007, 318, 1266–1273.
Supporting Information Available: Synthetic protocols and
analytical data of the compounds 10,12,13, 19-21, 24a-d, 25a-d,
29, 26a-d, 30, 27a-d, 31, 32a,b,33a,b,34a,b,35a,b,37a,b,39a,b,
40a,b, 42a-c, 43a-c, 44,45, 47a-52a,55a-59a, 61a-77a,
47-52, 55, 57-62, 65-77, HPLC data for the acylguanidines
46-77, tracings of key target compounds, and pharmacological
methods. The supplementary crystallographic data for compound
49a can be obtained free of charge from The Cambridge Crystal-
lographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif,
deposition number CCDC 686506. This material is available free
of charge via the Internet at http://pubs.acs.org.
(20) Buschauer, A.; Friese-Kimmel, A.; Baumann, G.; Schunack, W.
Synthesis and histamine H2 agonistic activity of arpromidine analogues:
replacement of the pheniramine-like moiety by nonheterocyclic groups.
Eur. J. Med. Chem. 1992, 27, 321–330.
(21) Durant, G. J.; Duncan, W. A. M.; Ganellin, C. R.; Parsons, M. E.;
Blakemore, R. C.; Rasmussen, A. C. Impromidine (SKF 92676) is a
very potent and specific agonist for histamine H2 receptors. Nature
(London) 1978, 276, 403–405.
(22) Honrubia, M. A.; Vilaro, M. T.; Palacios, J. M.; Mengod, G.
Distribution of the histamine H(2) receptor in monkey brain and its
mRNA localization in monkey and human brain. Synapse 2000, 38,
343–354.
References
(1) Hill, S. J.; Ganellin, C. R.; Timmerman, H.; Schwartz, J.-C.; Shankley,
N. P.; Young, J. M.; Schunack, W.; Levi, R.; Haas, H. L. International
Union of Pharmacology. XIII. Classification of Histamine Receptors.
Pharmacol. ReV. 1997, 49, 253–278.
(2) Hough, L. B. Genomics Meets Histamine Receptors: New Subtypes,
New Receptors. Mol. Pharmacol. 2001, 59, 415–419.
(3) Foord, S. M.; Bonner, T. I.; Neubig, R. R.; Rosser, E. M.; Pin, J.-P.;
Davenport, A. P.; Spedding, M.; Harmar, A. J. International Union of
Pharmacology. XLVI. G Protein-Coupled Receptor List. Pharmacol.
ReV. 2005, 57, 279–288.
(23) Martinez-Mir, M. I.; Pollard, H.; Moreau, J.; Arrang, J.-M.; Ruat, M.;
Traiffort, E.; Schwartz, J.-C.; Palacios, J. M. Three histamine receptors
(H1, H2 and H3) visualized in the brain of human and nonhuman
primates. Brain Res. 1990, 526, 322–327.

7204 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 22
Ghorai et al.
(24) Martinez-Mir, M. I.; Pollard, H.; Moreau, J.; Traiffort, E.; Ruat, M.;
Schwartz, J.-C.; Palacios, J. M. Loss of striatal histamine H2 receptors
in Huntington’s chorea but not in Parkinson’s disease: comparison
with animal models. Synapse 1993, 15, 209–220.
(25) Traiffort, E.; Pollard, H.; Moreau, J.; Ruat, M.; Schwartz, J.-C.;
Martinez-Mir, M. I.; Palacios, J. M. Pharmacological characterization
and autoradiographic localization of histamine H2 receptors in human
brain identified with [125I]iodoaminopotentidine. J. Neurochem. 1992,
59, 290–299.
(26) Brennauer, A. Acylguanidines as bioisosteric groups in argininamide-
type neuropeptide Y, Y1, and Y2 receptor antagonists: synthesis,
stability, and pharmacological activityPh.D. Thesis, University of
Regensburg, Regensburg, Germany, 2006.
(41) Giraldo, J. A pH-dependent model of the activation mechanism of
the histamine H2 receptor. Biochem. Pharmacol. 1999, 58, 343–353.
(42) Burde, R.; Seifert, R.; Buschauer, A.; Schultz, G. Histamine inhibits
activation of human neutrophils and HL-60 leukemic cells via H2-
receptors. Naunyn-Schmiedeberg’s Arch. Pharmacol. 1989, 340, 671–
678.
(43) Shi, L.; Liapakis, G.; Xu, R.; Guarnieri, F.; Ballesteros, J. A.; Javitch,
J. A. Beta2 adrenergic receptor activation. Modulation of the proline
kink in transmembrane 6 by a rotamer toggle switch. J. Biol. Chem.
2002, 277, 40989–40996.
(44) Arrang, J. M.; Garbarg, M.; Schwartz, J. C. Auto-inhibition of brain
histamine release mediated by a novel class (H3) of histamine receptor.
Nature (London) 1983, 302, 832–837.
(27) Brennauer, A.; Dove, S.; Buschauer, A. Structure-activity relationships
of nonpeptide neuropeptide Y receptor antagonists In Handbook of
Experimental Pharmacology; Michel, M. C. , Eds.; Springer: Berlin,
Heidelberg, New York, 2004; Vol. 162, pp 505-546.
(28) Schneider, E.; Keller, M.; Brennauer, A.; Hoefelschweiger, B. K.;
Gross, D.; Wolfbeis, O. S.; Bernhardt, G.; Buschauer, A. Synthesis
and characterization of the first fluorescent nonpeptide NPY Y1
receptor antagonist. ChemBioChem 2007, 8, 1981–1988.
(29) Feichtinger, K.; Sings, H. L.; Baker, T. J.; Matthews, K.; Goodman,
M. Triurethane-Protected Guanidines and Triflyldiurethane-Protected
Guanidines: New Reagents for Guanidinylation Reactions. J. Org.
Chem. 1998, 63, 8432–8439.
(30) Stark, H.; Purand, K.; Hu¨ls, A.; Ligneau, X.; Garbarg, M.; Schwartz,
J.-C.; Schunack, W. [125I]Iodoproxyfan and Related Compounds: A
Reversible Radioligand and Novel Classes of Antagonists with High
Affinity and Selectivity for the Histamine H3 Receptor. J. Med. Chem.
1996, 39, 1220–1226.
(31) Mitsunobu, O.; Yamada, M.; Mukaiyama, T. Preparation of esters of
phosphoric acid by the reaction of trivalent phosphorus compounds
with diethyl azodicarboxylate in the presence of alcohols. Bull. Chem.
Soc. Jpn. 1967, 40, 935–939.
(32) Go¨tte, C. Synthese, Enantiomerentrennung und Struktur-Wirkungs-
beziehungen neuer Histamin H2 Rezeptoragonisten des Arpromidin
TypsPh.D.Thesis, University of Regensburg, Regensburg, Germany,
2001.
(33) Preuss, H.; Ghorai, P.; Kraus, A.; Dove, S.; Buschauer, A.; Seifert,
R. Point mutations in the second extracellular loop of the histamine
H2 receptor do not affect the species-selective activity of guanidine-
type agonists. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2007, 376,
253–264.
(34) Preuss, H.; Ghorai, P.; Kraus, A.; Dove, S.; Buschauer, A.; Seifert,
R. Mutations of Cys-17 and Ala-271 in the human histamine H2
receptor determine the species selectivity of guanidine-type agonists
and increase constitutive activity. J. Pharmacol. Exp. Ther. 2007, 321,
975–982.
(35) Preuss, H.; Ghorai, P.; Kraus, A.; Dove, S.; Buschauer, A.; Seifert,
R. Constitutive activity and ligand selectivity of human, guinea pig,
rat, and canine histamine H2 receptors. J. Pharmacol. Exp. Ther. 2007,
321, 983–995.
(36) Xie, S. X.; Petrache, G.; Schneider, E.; Ye, Q. Z.; Bernhardt, G.;
Seifert, R.; Buschauer, A. Synthesis and pharmacological characteriza-
tion of novel fluorescent histamine H2-receptor ligands derived from
aminopotentidine. Bioorg. Med. Chem. Lett. 2006, 16, 3886–3890.
(37) Xie, S.-X.; Ghorai, P.; Ye, Q.-Z.; Buschauer, A.; Seifert, R. Probing
Ligand-Specific Histamine H1- and H2-Receptor Conformations with
NG-Acylated Imidazolylpropylguanidines. J. Pharmacol. Exp. Ther.
2006, 317, 139–146.
(38) Xie, S.-X.; Kraus, A.; Ghorai, P.; Ye, Q.-Z.; Elz, S.; Buschauer, A.;
Seifert, R. N1-(3-Cyclohexylbutanoyl)-N2-[3-(1H-imidazol-4-yl)pro-
pyl]guanidine (UR-AK57), a Potent Partial Agonist for the Human
Histamine H1- and H2-Receptors. J. Pharmacol. Exp. Ther. 2006, 317,
1262–1268.
(39) Gantz, I.; DelValle, J.; Wang, L. D.; Tashiro, T.; Munzert, G.; Guo,
Y. J.; Konda, Y.; Yamada, T. Molecular basis for the interaction of
histamine with the histamine H2 receptor. J. Biol. Chem. 1992, 267,
20840–20843.
(40) Ballesteros, J. A.; Weinstein, H. Integrated methods for the construction
of three-dimensional models and computational probing of structure-
function relations in G protein-coupled receptors. Methods Neurosci.
1995, 25, 366–428.
(45) Arrang, J. M.; Garbarg, M.; Schwartz, J. C. Autoinhibition of histamine
synthesis mediated by presynaptic H3-receptors. Neuroscience 1987,
23, 149–157.
(46) Liu, C.; Ma, X.; Jiang, X.; Wilson, S. J.; Hofstra, C. L.; Blevitt, J.;
Pyati, J.; Li, X.; Chai, W.; Carruthers, N.; Lovenberg, T. W. Cloning
and pharmacological characterization of a fourth histamine receptor
(H(4)) expressed in bone marrow. Mol. Pharmacol. 2001, 59, 420–
426.
(47) Lim, H. D.; van Rijn, R. M.; Ling, P.; Bakker, R. A.; Thurmond,
R. L.; Leurs, R. 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. 2005, 314, 1310–1321.
(48) Dove, S.; Michel, M. C.; Knieps, S.; Buschauer, A. Pharmacology
and quantitative structure-activity relationships of imidazolylpropy-
lguanidines with mepyramine-like substructures as non-peptide neu-
ropeptide Y Y1 receptor antagonists. Can. J. Physiol. Pharmacol. 2000,
78, 108–115.
(49) Ziemek, R.; Schneider, E.; Kraus, A.; Cabrele, C.; Beck-Sickinger,
A. G.; Bernhardt, G.; Buschauer, A. Determination of affinity and
activity of ligands at the human neuropeptide Y Y4 receptor by flow
cytometry and aequorin luminescence. J. Recept. Signal Transduct.
Res. 2007, 27, 217–233.
(50) Schmuck, C.; Lex, J. Acetate Binding within a Supramolecular
Network Formed by a Guanidinocarbonyl Pyrrole Cation in the Solid
State. Org. Lett. 1999, 1, 1779–1781.
(51) Furchgott, R. F. The classification of adrenoceptors (adrenergic
receptors). An evaluation from the standpoint of receptor theory In
Catecholamines, Blaschko, H.; Muscholl, E. , Eds.;Springer: Berlin,
Heidelberg, New York,1972; Vol. 33, pp 283-335.
(52) Arunlakshana, O.; Schild, H. O. Some quantitative uses of drug
antagonists. Br. J. Pharmacol. Chemother. 1959, 14, 48–58.
(53) Van Rossum, J. M. Cumulative dose-response curves. II. Technique
for the making of dose-response curves in isolated organs and the
evaluation of drug parameters. Arch. Int. Pharmacodyn. Ther. 1963,
143, 299–330.
(54) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition constant
(K1) and the concentration of inhibitor which causes 50% inhibition
(I50) of an enzymatic reaction. Biochem. Pharmacol. 1973, 22, 3099–
3108.
(55) Rasmussen, S. G.; Choi, H. J.; Rosenbaum, D. M.; Kobilka, T. S.;
Thian, F. S.; Edwards, P. C.; Burghammer, M.; Ratnala, V. R.;
Sanishvili, R.; Fischetti, R. F.; Schertler, G. F.; Weis, W. I.; Kobilka,
B. K. Crystal structure of the human beta2 adrenergic G-protein-
coupled receptor. Nature (London) 2007, 450, 383–387.
(56) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A.
Development and testing of a general amber force field. J. Comput.
Chem. 2004, 25, 1157–1174.
(57) Lovell, S. C.; Word, J. M.; Richardson, J. S.; Richardson, D. C. The
penultimate rotamer library. Proteins 2000, 40, 389–408.
(58) Halgren, T. A. Maximally diagonal force constants in dependent angle-
bending coordinates. II. Implications for the design of empirical force
fields. J. Am. Chem. Soc. 1990, 112, 4710–4723.
(59) Houston, C.; Wenzel-Seifert, K.; Bu¨rckstu¨mmer, T.; Seifert, R. 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. 2002,
80, 678–696.
JM800841W