Synthesis and pharmacological characterization of novel fluorescent histamine H2-receptor ligands derived from aminopotentidine

Article abstract of DOI:10.1016/j.bmcl.2006.05.039

In an effort to develop a non-radioactive alternative to the [³H]tiotidine and [¹²⁵I]iodoaminopotentidine binding assays for the histamine H₂-receptor (H₂R), primary amines related to aminopotentidine were prepared and coupled with the succinimidyl esters of the bulky fluorescent dyes S0536 and BODIPY 650/665-X. The primary amines exhibited different degrees of antagonistic potency at the human and guinea pig H₂R. Surprisingly, one compound (5) coupled to the cyanine dye S0536 acted as potent partial agonist/antagonist at the H₂R (K_B ～ 50 nM; EC₅₀ ～ 100-150 nM). Compounds coupled to the BODIPY dye exhibited moderately high H₂R-affinity, too. Thus, the H₂R accommodates bulky fluorophores, probably through interaction with extracellular receptor domains. The compounds presented herein provide a starting point for the optimization of fluorescent H₂R ligands with respect to affinity and fluorescence as valuable tools to analyze the molecular mechanisms of H₂R activation.

Full text of DOI:10.1016/j.bmcl.2006.05.039

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3886–3890
Synthesis and pharmacological characterization of novel fluorescent
histamine H2-receptor ligands derived from aminopotentidine
Sheng-Xue Xie,a Georgiana Petrache,b Erich Schneider,b Qi-Zhuang Ye,a
Gunther Bernhardt,b Roland Seifertb,c and Armin Buschauerb,*
¨
aHigh Throughput Screening Laboratory, The University of Kansas, Lawrence, KS 66045, USA
bInstitute of Pharmacy, University of Regensburg, D-93040 Regensburg, Germany
cDepartment of Pharmacology and Toxicology, The University of Kansas, Lawrence, KS 66045, USA
Received 24 April 2006; revised 11 May 2006; accepted 13 May 2006
Available online 30 May 2006
Abstract—In an effort to develop a non-radioactive alternative to the [3H]tiotidine and [125I]iodoaminopotentidine binding assays
for the histamine H2-receptor (H2R), primary amines related to aminopotentidine were prepared and coupled with the succinimidyl
esters of the bulky fluorescent dyes S0536 and BODIPYꢀ 650/665-X. The primary amines exhibited different degrees of antagonistic
potency at the human and guinea pig H2R. Surprisingly, one compound (5) coupled to the cyanine dye S0536 acted as potent partial
agonist/antagonist at the H2R (KB ꢀ 50 nM; EC50 ꢀ 100–150 nM). Compounds coupled to the BODIPY dye exhibited moderately
high H2R-affinity, too. Thus, the H2R accommodates bulky fluorophores, probably through interaction with extracellular receptor
domains. The compounds presented herein provide a starting point for the optimization of fluorescent H2R ligands with respect to
affinity and fluorescence as valuable tools to analyze the molecular mechanisms of H2R activation.
ꢁ 2006 Elsevier Ltd. All rights reserved.
The histamine H2-receptor (H2R) couples to Gs-proteins
and mediates activation of adenylyl cyclase.1
Functionally, the H2R can be analyzed by measuring the
high-affinity GTPase activity of fusion proteins of the
H2RwithGsa,adenylylcyclaseactivityorpositivechrono-
tropy in the spontaneously beating guinea pig right atri-
um.2–4 All these assays yield robust signals and have
provided the basis for the development of both potent
H2R agonists and antagonists. Unfortunately, in terms
of direct analysis of the ligand-binding properties of the
H2R, the situation is more problematic. The tritiated radi-
oligand [3H]tiotidine suffers from a relatively low affinity
(Kd ꢀ 35 nM) and labels only a subpopulation of the
program directed toward the development of fluorescent
H2R ligands. To this end, we have prepared aminopoten-
tidine-derived compounds substituted with fluorescein,
acridine, dansyl, nitrobenzoxadiazole or indolo[2,3-
a]quinolizine at the primary amino group.6 These
compounds are H2R antagonists with pA2 values in the
range of 7.5–8.0, that is, comparable to the affinity of tiot-
idine.2,6 Meanwhile, according to the same strategy
Malan et al.7 prepared aminopotentidine-like H2R
ligands labeled with some of the aforementioned and sev-
eral other fluorophores such as N-methylanthranilamide,
1-cyanoisoindole and 1-cyanoindolizine-2-carboxamide.
Based on the experience with [3H]tiotidine,2 the affinity
of most of the fluorescent H2R ligands is not high enough
for spectroscopic applications. In addition, the most
important lesson from investigations with such fluores-
cent probes is that the optical properties of the previously
prepared compounds are not optimal for H2R analysis in
intact cells and tissues.6 Therefore, we prepared a novel
series of ligands with bulky fluorescent dyes emitting light
at wavelengths >650 nm in order to improve their signal-
to-noise ratio.
available H2R molecules.2
[
125I]aminopotentidine
exhibitshighaffinity fortheH2R(Kd 0.3 nM),5 butitshigh
costs and the relatively short t1/2 of 125I (60 days) substan-
tially limit its general use in H2R ligand development.
In an effort to overcome these technical difficulties in
ligand-binding analysis of the H2R, we have initiated a
Keywords: Fluorescent probes; Cyanine; BODIPY; Histamine
H2-receptor agonist; GTPase assay.
*
Primary amines 2a,b related to the H2R antagonist
aminopotentidine5 were prepared from 18 according to
Corresponding author. Tel.: +49 941 943 4827; fax: +49 941 943
4820; e-mail: armin.buschauer@chemie.uni-regensburg.de
0960-894X/$ - see front matter ꢁ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.039

S.-X. Xie et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3886–3890
3887
a known synthetic route6,9 (Scheme 1) and coupled to
the succinimidyl esters 3 and 4 of the fluorescent dyes
S0536 and BODIPYꢀ 650/665-X according to standard
procedures (Scheme 2), yielding the fluorescent H2R
ligands 5, 6 and 7.10 The fluorescent products were
isolated and purified on a C-18 reversed-phase high-
performance liquid chromatography column (Fig. 1).
The primary amines 1, 2a, and 2b derived from amino-
potentidine and the fluorescent compounds 5, 6, and 7
were compared to the reference compound aminopoten-
tidine in the H2R antagonist mode measuring inhibition
of histamine-stimulated high-affinity GTP hydrolysis in
Sf9 insect cell membranes expressing fusion proteins of
the human H2R (hH2R) or guinea pig H2R (gpH2R)
and Gsa according to the previously described proce-
dure.2 The antagonist data are summarized in Table 1.
We also examined the effects of the novel compounds
in the absence of histamine in order to detect possible
partial agonistic and inverse agonistic activity.2 Those
Figure 1. HPLC analysis of 5. Dashed line, reaction mixture (injected
as 1:10 dilution with mobile phase) after addition of the second portion
of 2a. Solid line, chromatogram of the purified compound 5 after
vacuum concentration and dissolution in MeCN. Detection at 640 nm.
Scheme 1. Synthesis of the aminopotentidine-derived primary amines 2a and 2b.
Scheme 2. Fluorescence labeling of the primary amines 1, 2a, and 2b with the succinimidyl ester of S0536 (3), yielding 5, and with the succinimidyl
ester of BODIPYꢀ 650/665-X (4), yielding 6 and 7.

3888
S.-X. Xie et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3886–3890
Table 1. Antagonist potencies of aminopotentidine, 1, 2a, 2b, 5, 6 and 7 at hH2R and gpH2R
Compound
KB hH2R (nM)
KB gpH2R (nM)
KB hH2R/KB gpH2R
Aminopotentidine
1
220 30
15,000 3500
9300 2400
1800 920
47 12
280 40
6800 1700
2300 1200
75 18
0.79
2.20
4.04
24.0
0.81
4.27
1.37
2a
2b
5
58 10
150 95
190 70
6
640 210
260 80
7
Antagonist potencies were determined in the GTPase assay. Reaction mixtures contained 100 nM unlabeled GTP, 0.1 lCi of [c-32P]GTP, 10 lg of Sf9
insect cell membrane protein expressing hH2R-Gsa or gpH2R-Gsa, 1 lM histamine, and antagonists at concentrations between 1 nM and 100 lM to
obtain saturated inhibition curves. Data shown are means SD of three experiments. Non-linear curve fitting was performed using the Prism 4.0
program. KB values were calculated as described.2
data are summarized in Figure 2. In order to confirm the
data obtained with fusion proteins, we also performed
some complementary adenylyl cyclase experiments with
Sf9 membranes expressing non-fused hH2R and
gpH2R.11
than 2a, and previous studies already showed that
gpH2R exhibits higher conformational flexibility than
hH2R.2 Thus, 2b is a particularly useful probe to dissect
the species-differences between hH2R and gpH2R.
It was taken into account that introduction of bulky
fluorophores might result in a decrease in affinity of
the labeled compounds for the H2R since similar obser-
vations had been made for other biogenic amine recep-
tors.12 This presumption is also compatible with the
suggestion that the ligand binding pocket in biogenic
amine receptors is more constrained than in peptide
receptors, allowing only for small structural changes
without compromising affinity. However, to our sur-
prise, the opposite was the case. Specifically, 5 was a
200-fold more potent antagonist at hH2R than 2a, and
for gpH2R the affinity difference was 40-fold. The link-
age of 1 and 2b with the BODIPY dye 4, yielding 6
and 7, resulted in antagonists that showed higher affinity
for hH2R than the corresponding building blocks 1 and
2b. For gpH2R, this difference was also observed for 1
and 6. These data show that biogenic amine receptors
can not only tolerate bulky fluorescent ligands as was
shown already for some cases6,13 but that bulky fluores-
cent groups can even substantially increase ligand-
affinity.
Expectedly, the piperidinomethylphenoxypropylamine
building block 1 was a far less potent H2R antagonist
than the cyanoguanidine aminopotentidine (Table 1)
since the cyanoguanidino group is important for recep-
tor interaction.2 Connection of the piperidinomethyl-
phenoxypropylamine building block with
a
cyanoguanidino moiety linked to the reactant primary
amine via an ethylene group (2a) moderately increased
antagonist affinity. A hexamethylene linker between
the cyanoguanidino moiety and the primary amine
(2b) increased affinity much more substantially, particu-
larly for the gpH2R. In fact, the affinity of 2b for
gpH2R-Gsa is 24-fold higher than for hH2R-Gsa. Aden-
ylyl cyclase experiments with Sf9 membranes expressing
non-fused hH2R and gpH2R confirmed the large affinity
difference of 2b for the two receptor isoforms (KB of
2.5 lM vs 33 nM). These are the largest affinity differ-
ences for a ligand between hH2R and gpH2R described
so far.2,11 Compound 2b possesses a longer linker be-
tween the cyanoguanidino group and the primary amine
Figure 2. Effects of histamine, 1, 2a, 2b, 5, 6 and 7 on GTPase activity in Sf9 insect cell membranes expressing hH2R-GsaS or gpH2R-GsaS. Reaction
mixtures contained 100 nM unlabeled GTP, 0.1 lCi [c-32P]GTP, 10 lg of membrane protein expressing the fusion proteins, and ligands at the
concentrations indicated on the abscissa. Data shown are means SD of three experiments. Non-linear curve fitting was performed using the Prism
4.0 program.

S.-X. Xie et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3886–3890
3889
The results for compounds 1, 2a, 2b, 5, 6, and 7 in the
GTPase assay in the absence of histamine were striking,
too (Fig. 2). Histamine was a full agonist at hH2R-Gsa
and gpH2R-Gsa and served as a reference compound.
All compounds except for 5 and 2a at hH2R-Gsa slightly
reduced basal GTPase activity, indicating that the H2R
exhibits a certain degree of constitutive activity and that
the compounds act as inverse agonists.2 In marked
contrast, 5 acted as partial agonist both at hH2R and
gpH2R. At the hH2R the EC50 of 5 was 160 nM, and
the efficacy was 58% of that of histamine. The EC50 of
5 at gpH2R was 90 nM, and the efficacy was 65% of that
of histamine. The partial agonist effects of 5 on hH2R
and gpH2R were clearly H2R-mediated since the H2R
antagonists tiotidine, cimetidine, ranitidine, famotidine,
and zolantidine inhibited the effects of 5 on GTP hydro-
lysis with very similar KB values as for histamine (data
not shown).2 We also conducted adenylyl cyclase assays
with Sf9 membranes expressing non-fused H2R from hu-
man and guinea pig11 and confirmed the partial agonis-
tic activity of 5 (data not shown). Finally, we examined
the potential antagonistic effects of the compounds at
the human and guinea pig H1R measuring inhibition
of histamine-stimulated GTP hydrolysis in Sf9 mem-
branes expressing recombinant H1-receptors.11 Com-
pounds 1, 2a, 2b, 5, 6, and 7 exhibited only very low
affinity for the H1R (KB > 30 lM, data not shown).
the high degree of conservation between biogenic amine
receptors,14 the results of the present study suggest that
it will ultimately also become possible to develop high-
affinity fluorescent ligands for b-adrenergic receptors
and muscarinic cholinoceptors.
Acknowledgments
We thank Dr. G. Georg (Department of Medicinal
Chemistry, University of Kansas, KS) for continuous
support and encouragement, and Ms. S. Bollwein for
technical assistance.We are also grateful to the Deut-
scher Akademischer Austauschdienst (DAAD) for a
predoctoral fellowship to G.P. This work was supported
by the National Institutes of Health Center of Biomed-
ical Research Excellence (COBRE) award 1 P20
RR15563 and matching support from the State of Kan-
sas and the University of Kansas (R.S. and Q.-Z.Y.) and
the Graduate Training Program (Graduiertenkolleg)
GRK 760 ‘Medicinal Chemistry: Molecular Recogni-
tion—Ligand–Receptor Interactions’ of the Deutsche
Forschungsgemeinschaft.
References and notes
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. Pharmacol. Rev. 1997, 49, 253.
It is unlikely that transmembrane domains are involved
in conferring the relatively high affinity of the hH2R and
the gpH2R for 5 for two reasons. First, the fluorescent
group in compound 5 is too bulky to be inserted into
the ligand binding pocket of the H2R.2,11 Second, cya-
nine dye moiety is charged, that is, it bears a negatively
charged sulfate group and a resonance-stabilized cation
(Scheme 2). Accordingly, it is reasonable to assume that
the fluorophore of 5 interacts with extracellular domains
of the H2R, that is, the N-terminus and/or the extracel-
lular loops, both of which contain positively and nega-
tively charged amino acids.2 It is conceivable that ionic
interactions between the fluorescent group and extracel-
lular portions of the receptor protein increase the affinity
of the H2R for fluorescent ligands. In addition, we as-
sume that conformational changes in extracellular por-
tions of the H2R can result in at least partial
activation. So far, conformational changes related to
biogenic amine receptor activation were assumed to be
restricted to the transmembrane domains.14 Thus, fluo-
rescent ligands, particularly 5, may serve as probes to
examine the molecular mechanisms underlying biogenic
amine receptor activation. For actual fluorescence anal-
ysis of the H2R, the affinity of 5 is still too low and
should be increased by at least one order of magnitude.
2. Kelley, M. T.; Burckstummer, T.; Wenzel-Seifert, K.;
¨
¨
Dove, S.; Buschauer, A.; Seifert, R. Mol. Pharmacol. 2001,
60, 1210.
3. Houston, C.; Wenzel-Seifert, K.; Burckstummer, T.;
¨
¨
Seifert, R. J. Neurochem. 2002, 80, 678.
4. Black, J. W.; Duncan, W. A. M.; Durant, G. J.; Ganellin,
C. R.; Parsons, M. E. Nature 1972, 236, 385.
5. Ruat, M.; Traiffort, E.; Bouthenet, M. L.; Schwartz, J. C.;
Hirschfeld, J.; Buschauer, A.; Schunack, W. Proc. Natl.
Acad. Sci. U.S.A. 1990, 87, 1658.
6. Li, L.; Kracht, J.; Peng, S.; Bernhardt, G.; Elz, S.;
Buschauer, A. Bioorg. Med. Chem. Lett. 2003, 13, 1717.
7. Malan, S. F.; van Marle, A.; Menge, W. M.; Zuliana, V.;
Hoffman, M.; Timmerman, H.; Leurs, R. Bioorg. Med.
Chem. 2004, 12, 6495.
8. Buschauer, A.; Postius, S.; Szelenyi, I.; Schunack, W.
Arzneim.-Forsch. 1985, 35, 1025.
9. Hirschfeld, J.; Buschauer, A.; Elz, S.; Schunack, W.; Ruat,
M.; Traiffort, E.; Schwartz, J.-C. J. Med. Chem. 1992, 35,
2231.
10. (a) Preparation of 4-[2-[5-[1-[6-[2-[2-cyano-3-[3-[3-(piperi-
din-1-ylmethyl)phenoxy]-propyl]guanidino]ethylamino]-6-
oxohexyl]-3,3-dimethylindolin-2-ylidene]penta-1,3-dienyl]-
3,3-dimethyl-3H-indolium-1-yl]butane-1-sulfonate (5). To
a mixture of 20 lL of anhydrous pyridine and 200 lL of a
1 mMsolution of 4-[2-[5-[1-[6-(2,5-dioxopyrrolidin-1-yloxy)-
6-oxohexyl]-3,3-dimethylindolin-2-ylidene]penta-1,3-dienyl]-
3,3-dimethyl-3H-indolium-1-yl]butane-1-sulfonate (3) (FEW
Chemicals) in MeCN three portions of 5 lL of a 10 mM
solution of 2a were added under argon atmosphere
within 2 h. After 17 h, the product was isolated by
RP-HPLC: Thermo Separation Products (ThermoQuest),
SN4000 controller, P4000 pump, AS3000 autosampler,
UV–vis detector Spectra Focus, fluorescence detector
FL3000, SCM400 solvent degasser; column: Luna 3 lm
C18(2), (150 mm · 4 mm) equipped with a guard column
It should be noted that the identification of 5 as a
relatively high-affinity partial H2R agonist was
fortuitous. Our present data show that the H2R also
tolerates BODIPY dyes linked to cyanoguanidines (6
and 7) quite well. Accordingly, future studies will
have to systematically combine various cyanoguanidines
with various fluorophores emitting at wavelengths
> 650 nm. We assume that more potent fluorescent
H2R ligands than 5 can be developed. Moreover, given

3890
S.-X. Xie et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3886–3890
(Phenomenex); mobile phase: MeOH/aqueous TFA (0.1%),
MeCN and purified by RP-HPLC: Kontron Instruments:
HPLC pump 420, UV–vis detector: HPLC detector 430,
fluorescence detector: Merck–Hitachi F1000; column
(250 · 4 mm, 5 lm) and precolumn: Nucleosil 300-5 C18
(Macherey & Nagel); eluent: MeCN/0.1% TFA (6: 50/50; 7:
60/40), flow rate: 1.0 mL/min; detection: 649 nm, fluores-
cence detection: kex:275 nm, kem: 305 nm. k0 values: 6, 4.36;
7, 1.40. Compound 6 (C44H51BF2N6O4), yield 33%, EI-MS:
m/z = 777 (MH+, 100%); 7 (C52H65BF2N10O4), yield 27%,
EI-MS: m/z = 943 (MH+, 100%). For pharmacological
investigations, the pooled fractions of 6 and 7, respectively,
were aliquoted in portions of 1 nmol after photometric
determination of the concentration (e646 (in MeOH) =
106,600 M ꢁ1 cmꢁ1). (c) The fluorescence properties of
compounds 5–7 in phosphate-buffered saline, pH 7.4
(Dulbecco’s solution B), containing 5% (vol/vol) DMSO
(corresponding to the conditions of the pharmacological
assay) were analyzed with an Aminco Bowman AB2
spectrofluorometer at 20 ꢂC. Both emission and excitation
slits were set to 4 nm. kmax (nm):5, ex 648, em 666; 6 and 7, ex
651, em 670.
62/38; flow rate: 0.5 mL/min; column temperature: 40 ꢂC;
injection: 20 lL push loop; detection of 5 (2a): absorbance
640 nm, fluorescence ex = 640 (275) nm; em = 664
(305) nm. For pharmacological investigations, the pooled
fractions of compound 5 were aliquoted in portions of
1 nmol after photometric determination of the concentra-
tion (e646 (in MeOH) = 127,000 Mꢁ1 cmꢁ1). After removal
of the solvent in a vacuum concentrator at room temper-
ature, the purity of 5 was confirmed by RP-HPLC under the
aforementioned conditions (k0 value = 2.70). Yield: 29%.
ESI-MS (Finnigan Thermo Separations Q 7000) of com-
pound 5 (C54H72N8O5S): m/z = 945 (MH+, 100%). (b)
Preparation of 5,5-difluoro-3-[4-[2-oxo-2-[6-oxo-6-[3-[3-
(piperidin-1-ylmethyl)phenoxy]propylamino]hexylamino]-
ethoxy]styryl]-7-(1H-pyrrol-2-yl)-5H-dipyrrolo[1,2-c:10,20-
f][1,3,2]diazaborinin-4-ium-5-uide (6) and 3-[4-[18-(cyanoi-
mino)-2,9-dioxo-22-[3-(piperidin-1-ylmethyl)phenoxy]-3,10,
17,19-tetraazadocosyloxy]styryl]-5,5-difluoro-7-(1H-pyrrol-
2-yl)-5H-dipyrrolo[1,2-c:10,20-f][1,3,2]diazaborinin-4-ium-
5-uide (7). A solution of 2 lmol of the pertinent amine (1 or
2b) in 200 lL of a 1:1 mixture of pyridine and anhydrous
DMF was added to 1.04 lmol (0.67 mg) of 6-[[[4,4-diflu-
oro-5-(2-pyrrolyl)-4-bora-3a,4a-diaza-s-indacene-3-yl]sty-
ryloxy]acetyl]aminohexanoic acid succinimidyl ester (4,
BODIPYꢀ 650/665-X SE) (Invitrogen) and stirred for 2
day under argon atmosphere. The solvents were removed in
a vacuum concentrator, the residue was dissolved in
11. Xie, S. X.; Ghorai, P.; Ye, Q. Z.; Buschauer, A.; Seifert, R.
J. Pharmacol. Exp. Ther. 2006, 317, 139.
12. Middleton, R. J.; Kellam, B. Curr. Opin. Chem. Biol. 2005,
9, 517.
13. Li, L.; Kracht, J.; Peng, S.; Bernhardt, G.; Buschauer, A.
Bioorg. Med. Chem. Lett. 2003, 13, 1245.
14. Gether, U. Endocr. Rev. 2000, 21, 90.