Structure-activity relationship studies toward the discovery of selective apelin receptor agonists

Article abstract of DOI:10.1021/jm401789v

Apelin is the endogenous ligand for the previously orphaned G protein-coupled receptor APJ. Apelin and its receptor are widely distributed in the brain, heart, and vasculature, and are emerging as an important regulator of body fluid homeostasis and cardiovascular functions. To further progress in the pharmacology and the physiological role of the apelin receptor, the development of small, bioavailable agonists and antagonists of the apelin receptor, is crucial. In this context, E339-3D6 (1) was described as the first nonpeptidic apelin receptor agonist. We show here that 1 is actually a mixture of polymethylated species, and we describe an alternative and versatile solid-phase approach that allows access to highly pure 27, the major component of 1. This approach was also applied to prepare a series of derivatives in order to identify the crucial structural determinants required for the ligand to maintain its affinity for the apelin receptor as well as its capacity to promote apelin receptor signaling and internalization the study of the structure-activity relationships led to the identification of ligands 19, 21, and 38, which display an increased affinity compared to that of 27 the latter and 19 behave as full agonists with regard to cAMP production and apelin receptor internalization, whereas 21 is a biased agonist toward cAMP production. Interestingly, the three ligands display a much higher stability in mouse plasma (T_1/2 > 10 h) than the endogenous apelin-17 peptide 2 (T_1/2 < 4 min).

Full text of DOI:10.1021/jm401789v

Article
pubs.acs.org/jmc
Structure−Activity Relationship Studies toward the Discovery of
Selective Apelin Receptor Agonists
Jean-Franco̧
is Margathe,†,§ Xavier Iturrioz,‡,§ Rodrigo Alvear-Perez,‡ Claire Marsol,† Stephanie Riche,†
́ ́
Hadjila Chabane,‡ Nassera Tounsi,† Maxime Kuhry,† Denis Heissler,† Marcel Hibert,†
Catherine Llorens-Cortes,*,‡,∥ and Dominique Bonnet*,†,∥
†Laboratoire d’Innovation Ther
du Rhin, 67401 Illkirch, France
‡Laboratory of Central Neuropeptides in the Regulation of Body Fluid Homeostasis and Cardiovascular Functions, Center for
Interdisciplinary Research in Biology (CIRB), College de France, INSERM U1050, CNRS UMR7241, 75005 Paris, France
́ ́ ́
apeutique, UMR7200 CNRS/Universite de Strasbourg, Labex Medalis, Faculte de Pharmacie, 74 route
̀
S
* Supporting Information
ABSTRACT: Apelin is the endogenous ligand for the
previously orphaned G protein-coupled receptor APJ. Apelin
and its receptor are widely distributed in the brain, heart, and
vasculature, and are emerging as an important regulator of
body fluid homeostasis and cardiovascular functions. To
further progress in the pharmacology and the physiological
role of the apelin receptor, the development of small,
bioavailable agonists and antagonists of the apelin receptor,
is crucial. In this context, E339−3D6 (1) was described as the
first nonpeptidic apelin receptor agonist. We show here that 1
is actually a mixture of polymethylated species, and we
describe an alternative and versatile solid-phase approach that
allows access to highly pure 27, the major component of 1.
This approach was also applied to prepare a series of derivatives in order to identify the crucial structural determinants required
for the ligand to maintain its affinity for the apelin receptor as well as its capacity to promote apelin receptor signaling and
internalization. The study of the structure−activity relationships led to the identification of ligands 19, 21, and 38, which display
an increased affinity compared to that of 27. The latter and 19 behave as full agonists with regard to cAMP production and apelin
receptor internalization, whereas 21 is a biased agonist toward cAMP production. Interestingly, the three ligands display a much
higher stability in mouse plasma (T1/2 > 10 h) than the endogenous apelin-17 peptide 2 (T1/2 < 4 min).
plasma.7,8 They exhibit a high affinity for the human and the rat
INTRODUCTION
■
apelin receptors, although 2 has a 10 times higher affinity than
3 for the human or rat ApelinR.9−11 They similarly inhibit
forskolin-induced cAMP production in cells expressing either
the human5,10 or the rat ApelinR.4,12 Both peptides are potent
inducers of ApelinR internalization in a clathrin-dependent
manner with a higher potency for 2 than for 3.10,13−15
Apelin and its receptor are widely distributed in the
brain4,6,16−18 where they colocalize with arginine vasopressin
(AVP) in magnocellular neurons.8,19,20 Central injection of 2 in
lactating rats inhibits the electrical activity of AVP neurons,
thereby decreasing AVP release into the bloodstream and
increasing aqueous diuresis.8 Moreover, AVP and apelin levels
are conversely regulated by osmotic and volemic stimuli in
humans and rodents to maintain body fluid homeostasis.7,8,20
Apelin and its receptor are also present in the cardiovascular
system.21 Several studies pointed out the role of apelin in the
Apelin is a neuro-vasoactive peptide, isolated from bovine
stomach extracts,1 and identified as the endogenous ligand of
the human APJ orphan receptor (subsequently named ApelinR
for apelin receptor) that shares 31% sequence identity with the
angiotensin II receptor type 1 (AT1R).2 The ApelinR is 380
amino acid long and was identified as a member of the seven
transmembrane-domain G protein-coupled receptors (GPCR)
family. It has also been cloned in mice3 and rats.4 Apelin derives
from a single 77-amino acid precursor named preproapelin.1,5,6
The alignment of preproapelin amino acid (aa) sequences from
mammalian species demonstrates a fully conserved C-terminal
17-aa sequence 2 (Lys-Phe-Arg-Arg-Gln-Arg-Pro-Arg-Leu-Ser-
His-Lys-Gly-Pro-Met-Pro-Phe called apelin-17 or K17F),5
including apelin-13 (Gln-Arg-Pro-Arg-Leu-Ser-His-Lys-Gly-
Pro-Met-Pro-Phe). The N-terminal glutamine residue of the
latter peptide might be lactamized to produce the pyroglutamyl
form 3 (pGlu-Arg-Pro-Arg-Leu-Ser-His-Lys-Gly-Pro-Met-Pro-
Phe called pE13F).5 Both peptides 2 and 3 are the predominant
molecular forms of apelin in the rat brain and in rat and human
Received: November 18, 2013
Published: March 13, 2014
© 2014 American Chemical Society
2908
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
control of cardiovascular functions: apelin increases the
contractile force of the myocardium by a positive inotropic
effect,22−24 and apelin gene-deficient mice subjected to chronic
pressure overload developed severe and progressive heart
failure.25 Moreover, intravenous injection of 2 or 3 in rodents
decreases arterial blood pressure (BP),6,13,15 through nitric
oxide production26 with a more marked effect for 2 than for
3.13 In contrast, the deletion of the C-terminal Phe in 2 or the
Ala substitution of the C-terminal Phe in 3 suppressed the
ability of 2 and 3 to decrease BP,13,27 suggesting that the
hypotensive activity of apelin required the presence of the C-
terminal Phe residue. Since we recently showed that the C-
terminal Phe residue of 2 plays a crucial role in ApelinR
internalization but not in apelin binding or in Gαi-protein
coupling,9 this suggests that the decrease in blood pressure
induced by 2 requires ApelinR internalization and/or involves
another yet to be discovered signaling pathway.
access a series of compound 1 derivatives, evaluated for their
affinity for the ApelinR. The highest affinity compounds were
assessed for their ability to inhibit cAMP production and to
trigger ApelinR internalization in CHO cells stably expressing
the rat ApelinR tagged at its C-terminal part with EGFP. The
selectivity of these compounds toward the AT1R was also
investigated together with their plasma stability.
CHEMISTRY
■
The synthesis of 1 was originally achieved by a solid-phase
strategy based on the regenerative Michael acceptor (REM)
resin.12,35 This approach involves the release of the final
product from the solid-support via a Hofmann elimination,
which requires the treatment of the resin with an excess of
methyl iodide (40 equiv). The RP-HPLC analysis of purified 1
showed a single peak (Supporting Information, Figure S1). We
thus expected to have a compound with structure 27 (Figure
1). However, ES-MS analysis of the purified product clearly
In addition, apelin has been shown to modulate the
adipoinsular axis: a systemic injection of apelin in insulin-
resistant mice decreases glycemia and enhances glucose uptake
in skeletal muscle and adipose tissue, contributing to
homeostatic control of blood glucose.28 However, 2 displays
a short half-life in the blood circulation in mice (in the minute
range; personal data), and 3 is metabolized. Therefore, to
facilitate the in vivo evaluation of apelin effects, it is of
importance to discover more stable small molecular probes,
agonists and antagonists of the ApelinR to further investigate
the physiopathological roles played by apelin and its receptor.
To date, very few nonendogenous ligands have been
described in the literature.29,10 The synthesis of cyclic analogues
of the minimal fragment apelin-12 (Arg-Pro-Arg-Leu-Ser-His-
Lys-Gly-Pro-Met-Pro-Phe) was described as a strategy to
potentially improve the bioactivity, selectivity, and bioavail-
ability of apelin.30 The Arg-Pro-Arg sequence was also explored
to design and to synthesize antagonist cyclic peptides with
micromolar affinity for the ApelinR.31,32 Nevertheless, these
ligands exhibit high molecular weights and low affinity and/or
efficacy for the ApelinR. More recently, 4-oxo-6-(pyrimidin-2-
ylthiomethyl)-4H-pyran-3-yl 4-nitrobenzoate (ML221), coming
from the screening of the NIH small molecule collection, was
identified and described as an apelin functional antagonist with
regard to the inhibition of cAMP production.33
Prior to these disclosures, in order to accelerate the discovery
of ligands for orphan GPCR, we have set up an original FRET-
based assay that consists of screening fluorescent small organic
compound libraries on enhanced green fluorescent protein
(EGFP)-fused GPCR.34 Applied to ApelinR, this approach
enabled the successful discovery of the first nonpeptidic
ApelinR agonist, E339−3D6 (1).12 The latter behaves as a
partial agonist with regard to cAMP production and as a full
agonist with regard to ApelinR internalization. As apelin, 1 was
found to induce vasorelaxation of rat aorta precontracted by
noradrenaline, and when centrally injected, potently inhibited
AVP release in the blood circulation in water-deprived mice.
The pharmacological interest of 1 prompted us to study
further the structure−activity relationships around this
compound with the aim (1) to identify the important structural
determinants of the ligand required for its affinity for the
ApelinR as well as for its capacity to promote ApelinR signaling
and internalization; (2) to develop more potent ApelinR
agonists; (3) to increase the binding selectivity toward the
ApelinR versus AT1R; (4) to decrease its size and improve its
metabolic stability. Herein, we report a solid-phase strategy to
Figure 1. Chemical structure of 27, the major component of 1.12
showed the presence, in addition to 27, of several
polymethylated compounds that could not be isolated by RP-
HPLC, regardless of the nature of the column and the mobile
phase (Figure 2A). The presence of three major methylated
forms was confirmed by analytical capillary zone electro-
phoresis (Supporting Information, Figure S2). MS/MS experi-
ments performed on 1 followed by analysis of the various
fragments allowed us to identify the number and the location of
the different methyl groups. The analysis clearly showed that no
methyl group was incorporated on Lissamine Rhodamine B
(LRB), whereas there was one methyl on the iminothiazolyl,
one or none on the 3-benzyl imidazolyl, and one or two on the
piperidyl ring (Supporting Information, Figure S3). These
results prompted us to devise an alternative and versatile solid-
phase approach to access fully defined and characterized 27, as
well as a series of derivatives to study the structure−activity
relationships around 27 (Scheme 1).
Starting from commercially available 2-(3,5-dimethoxy-4-
formylphenoxy)ethyl (DFPE)-PS resin, N-Boc-4-aminopiper-
idine was anchored to the solid support by reductive amination
in the presence of NaBH3CN.36 Then, the acylation of the
resulting secondary amine with Fmoc-L-His(Bzl)-OH led to
resin 9. A selective methylation of the imidazole ring at the N-3
2909
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
Figure 2. ES-MS profiles: (A) 1 obtained on REM resin and (B) 27 obtained on DFPE resin.
position in the presence of MeI was followed by the removal of
the N-Boc protecting group by treatment with trimethylsilyl
trifluoromethanesulfonate (TMSOTf)/lutidine.36,37 Twice 30
min was found to be the optimal duration in order to drive the
piperidine deprotection to completion while avoiding any
premature cleavage of the product from the resin. Reductive
alkylation of the deprotected piperidine with formaldehyde led
to resin 11. Following a classical Fmoc/t-butyl approach, Fmoc-
L-Lys(ivDde)-OH and (2-(Boc)imino-3-methyl-2,3-dihydro-
thiazol-4-yl)acetic acid (8) were incorporated on solid support
via (benzotriazol-1-yl-oxy)trispyrrolidinophosphonium hexa-
fluorophosphate (PyBOP) activation. Then, the 1-(4,4-
dimethyl-2,6-dioxocyclohex-1-ylidene)-3-methylbutyl (ivDde)
protecting group was removed in the presence of 2% hydrazine
in DMF providing resin 13. The deprotected ε-amino group
was then reacted with p-nitrophenyl chloroformate followed by
the addition of an excess of 1,12-dodecanediamine to provide
resin 23. Sulfonylation of this resin with Lissamine Rhodamine
B sulfonyl chloride (LRB-Cl), obtained by treatment of Acid
Red 52 sodium salt with oxalyl chloride,38 yielded resin 25. A
final treatment with TFA enabled the cleavage of compound 27
from the solid support, followed by its isolation by RP-HPLC.
In contrast to 1, ES-MS analysis of the purified 27 showed that
it was obtained as a unique and well-defined trimethylated form
(Figure 2B). Thanks to this convenient and flexible solid-phase
approach, compounds derived from 27 were thus obtained in
moderate to good yields. The purity and the identity of the
compounds were checked by analytical RP-HPLC, liquid
chromatography−mass spectrometry (LC-MS), and high
resolution mass spectrometry (HRMS).
As depicted in Scheme 2, compound 8 was obtained in a
three-step process starting from commercially available ester 4.
2910
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
a
Scheme 1. General Solid-Phase Strategy for the Synthesis of 27 and Derivatives
a
Reagents: (i) 4-Amino-1-Boc-piperidine, NaBH3CN, DMF/MeOH/AcOH, 60 °C, 24 h; (ii) Fmoc-His(1-Bzl)-OH, HATU, DIEA, CH2Cl2/DMF,
rt, overnight; (iii) MeI, DMF, 40 °C, overnight; (iv) TMSOTf, 2,6-lutidine, rt, 2 × 30 min; (v) H2CO, THF/HC(OMe)3, rt, 5 h then NaBH(OAc)3,
1,2-DCE, rt, overnight; (vi) PhCHO, THF/HC(OCH3)3, rt, 5 h then NaBH(OAc)3, ClCH2−CH2Cl, rt, overnight; (vii) Piperidine, DMF, rt, 2 × 15
min; (viii) Fmoc-Lys(ivDde)-OH, PyBOP, HOBt, DIEA, DMF, rt, 2 h; (ix) piperidine, DMF, rt, 2 × 15 min; (x) 8, PyBOP, HOBt, DIEA, CH2Cl2/
DMF, rt, overnight; (xi) N2H4 hydrate, DMF, rt, 3 × 5 min; (xii) LRB-Cl, DIEA, CH2Cl2, rt, 5 h; (xiii) PB-Cl, DIEA, CH2Cl2, rt, 5 h; (xiv) Ac2O,
DIEA, CH2Cl2, rt, 2 × 10 min; (xv) TFA/H2O/i-Pr3SiH, rt, 3 h; (xvi) p-NO2−C6H4−OCOCl, DIEA, THF/CH2Cl2, rt, 2 × 45 min then
H2NCH2(CH2)nCH2NH2, DIEA, CH2Cl2, 30 °C, overnight.
a
acylation, which occurred in preliminary experiments during
the solid-phase incorporation of 6. Basic hydrolysis of the
protected compound 7 led to acid 8 in 61% yield, ready for
introduction on solid-phase.
Scheme 2. Synthesis of Compound 8
RESULTS AND DISCUSSION
■
Following our recent discovery of 1, the first nonpeptidic
agonist for ApelinR, a structure−activity relationship study was
undertaken. As 1 was shown to be actually a mixture of
diversely methylated compounds, our first objective was to
identify the contribution of five of them to the activity of the
mixture. Thanks to the novel solid-phase approach developed
on DFPE-PS resin, compounds 27 and 29−32 were produced
by changing the number and the position of the methyl groups.
Their affinities for the ApelinR were then evaluated by a
radioligand binding assay (Table 1).
a
Reagents: (i) MeI, THF, rt, 3 d, 96%; (ii) 1 N HCl, dioxane/H2O, 40
°C, overnight, 88%; (iii) Boc2O, 4-DMAP, pyridine, CH2Cl2, 0 °C to
rt, 2 h, 44%; (iv) LiOH, dioxane/H2O, 40 °C, 1 h, 61%.
As previously described for the synthesis of 2-imino-3-alkyl
thiazoline,39 methylation at the N-3 position was achieved by
treatment with MeI at room temperature for 3 days, affording
derivative 5 in excellent yield (96%). Ester hydrolysis under
acidic conditions provided acid 6. Alternatively, the imino
moiety of 5 was protected with a Boc group (44%) to avoid
Binding Affinities of the Various Methylated Compo-
nents of 1. Interestingly, with the exception of 32, the affinity
remains in the micromolar range indicating that the
methylation pattern is not the key component of the affinity
2911
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
Table 1. Inhibition Constants of Methylated Components of
1
Table 2. Influence of the Length of the Spacer and the
Presence of LRB Dye on Binding Affinities for the ApelinR
a
cpd
R
Ki (μM)
ab
,
cpd
R1
R2
R3
R4
Ki (μM)
27
28
19
21
22
33
-CONH(CH2)12NH-LRB
0.69 0.12
0.31 0.07
0.40 0.08
0.089 0.014
>100
27
29
30
31
32
-Me
-Me
-Me
-Me
-Me
-H
-Me
-Me
-Me
-Me
-Me
-Me
-H
0.69 0.12
1.4 0.1
3.0 1.0
1.5 0.3
9.0 4.0
-CONH(CH2)6NH-LRB
-Me
-Me
-H
-LRB
-PB
-COMe
n-BuNH-LRB
-H
>100
a
a
The inhibition constants (Ki) of compounds for the ApelinR were
The inhibition constants (Ki) of compounds for the ApelinR were
determined on CHO cell membranes stably expressing the human
ApelinR by a competition binding assay using as radioligand [125I]-3.
Data represent the means SEM of 3 to 10 independent experiments
determined on CHO cell membranes stably expressing the human
ApelinR by a competition binding assay using as radioligand [125I]-3.
Data represent the means SEM of 3 to 10 independent experiments
performed in duplicate.
b
performed in duplicate. Ki value for 1 is 0.43 0.04 μM.
LRB group resulted in a total loss of affinity (22), indicating the
crucial role played by LRB in the binding mode of the ligand.
However, as demonstrated with 33, the LRB moiety alone was
found unable to bind to the ApelinR. Interestingly, the
substitution of LRB for the structurally close Patent Blue
(PB) enabled us to increase the affinity of the resulting
compound 21 to a value of 89 nM.
Influence of the Heterocyclic Substituents. In the next
step, the influence of the 1-methyl-4-piperidyl (X group), 1-
methyl-3-benzyl-imidazolium (Y group), and 2-imino-3-methyl-
thiazol-4-yl (Z group) substituents on the binding affinity of the
resulting compounds for the ApelinR was carefully investigated
(Table 3). The substitution of X for propanamide linked at C-2
of 1 (Table 1). However, some significant trends can be
observed. For instance, the best affinity for ApelinR was found
for the trimethylated derivative 27 (Ki = 0.69 μM). The
tetramethylated compound 29 bearing two methyl groups on
the piperidyl ring displayed a 2-fold decrease in affinity (Ki =
1.4 μM). In the absence of a methyl group on the 3-benzyl-
imidazolyl ring, similar affinities were obtained for the
dimethylated compound 31 (Ki = 1.5 μM) and the
trimethylated species 30 (Ki = 3.0 μM). These results tend
to demonstrate the importance of the imidazolium ring to
retain a submicromolar affinity for the ApelinR. Finally, the
suppression of the methyl group on the thiazoline ring in 32 led
also to an important decrease in affinity (Ki = 9.0 μM). This
result may be ascribed to the influence of the methylation at the
N-3 position of the aminothiazolyl ring on the basicity of the
exocyclic nitrogen group. Indeed, as determined by potenti-
ometry in a water medium, the pKa values of 4 and 5 were
Table 3. Influence of the Nature of the Substituents X, Y,
and Z on the Binding Affinities of Compounds 34−36 for
the ApelinR
found to be 4.83
0.02 and 9.77
0.02, respectively.
Consequently, at physiological pH, the imino moiety is mainly
protonated when the endocyclic nitrogen is methylated. This
result suggests that the more basic imino form combined with
the presence of the methyl group favors the binding of the
ligand to the ApelinR.
Influence of the Length of the Spacer. Having
determined the most potent methylated form, compound 27,
we next examined the influence of the length of the spacer
between the LRB group and the rest of the molecule on the
affinity for the ApelinR. Table 2 shows that shortening the
spacer from 12 to 6 carbon atoms as in 28 resulted in a 2-fold
increase in affinity (from Ki = 0.69 to 0.31 μM). Surprisingly,
the direct incorporation of LRB on the ε-amino group of the
lysinyl core is not detrimental to the affinity. Indeed, the Ki of
19 is still 0.40 μM, even better than that determined for 27.
These data suggest that LRB does not bind ApelinR at the level
of the aromatic pocket as we previously suggested12 but more
likely at the surface of the receptor or less deeply in the
receptor transmembrane bundles. However, the removal of the
a
The inhibition constants (Ki) of compounds for the ApelinR were
determined on CHO cell membranes stably expressing the human
ApelinR by a competition binding assay using as radioligand [125I]-3.
Data represent the means SEM of 3 to 4 independent experiments
performed in duplicate.
2912
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
(34) or of Z for a methyl group (36) led to a decrease in the
affinity for both compounds (2.5- and 3-fold, respectively),
whereas the substitution of Y for a methyl group (35) did not
modify the affinity for the ApelinR (Ki = 0.45 μM). The
imidazolium group seems important for the affinity only when a
long spacer (C12) between the LRB and the trimethylated part
of the compound is present (31). When the spacer is removed,
then the benzyl substituted imidazolium can also be suppressed
without loss of affinity (35, Ki = 0.45 μM).
internalization of the ApelinR/2 complexes (Figure 3A). We
then tested the capacity of 27 to trigger ApelinR internalization.
Incubation of CHO cells with increasing concentrations of 27
(100 μM) for 20 min resulted in ApelinR-EGFP internalization.
Total internalization of the ApelinR occurred for concen-
trations of 27 comprised between 10 and 100 μM (Figure 3A),
suggesting that 27 behaves as a full agonist with regard to
ApelinR internalization.
As demonstrated with 28 and 19, the length of the spacer has
a low impact on the agonist character of ligands. Indeed, both
ligands are still able to induce ApelinR internalization at the
same concentration than 1 and 27 (Figure 3A). In contrast, the
substitution of LRB for the structurally close PB results in the
modification of the ability of the ligand to induce ApelinR
internalization. Indeed, the cells treated with 21 exhibited no or
very weak receptor internalization even at a high concentration
(Figure 3B), showing the importance of LRB to ensure the
agonist character of the ligand related to ApelinR internal-
ization. The ability of 21 to behave as an ApelinR antagonist
with regard to ApelinR internalization was then evaluated.
Competition experiments were performed in the presence of 2
at 1 and 10 nM. Confocal microscopy analysis showed that 21
(100 μM) did not block or modify the ability of 2 to trigger
ApelinR internalization (Figure 3B). This result indicates that
21 is not an ApelinR antagonist on ApelinR internalization.
Other structural determinants are important for the ability of
the ligand to ensure ApelinR internalization. The introduction
at the X position of 1-benzyl-4-piperidyl or (1-methyl-4-
piperidyl)methyl groups (compounds 38 and 20, respectively)
did not affect the ability of the ligands to induce ApelinR
internalization (Supporting Information, Figure S4). The
suppression of the methyl group at N-1 of benzyl-imidazolium
also had a low impact on the potency of 39. However, the total
suppression of the benzyl imidazolium group (35) resulted in
the inability of the compound to induce ApelinR internal-
ization. A similar result was obtained when the piperidyl moiety
was switched to the neutral tetrahydropyran group. Indeed, 37
was no longer able to induce ApelinR internalization
(Supporting Information, Figure S4), demonstrating the crucial
role of the cationic piperidyl group on the potency of the ligand
to trigger ApelinR internalization. Finally, the removal of the
methyl at N-3 of iminothiazoline (40) was also found
detrimental for ligand-induced ApelinR internalization (Sup-
porting Information, Figure S4).
Influence of the Basicity of the Heterocyclic Sub-
stituents. To further evaluate the influence of the basicity of
the heterocyclic substituents on the affinity of 19 (Table 4), 1-
Table 4. Influence of the Basicity of the Heterocyclic
Substituents on Binding Affinities for the ApelinR
a
The inhibition constants (Ki) of compounds for the ApelinR were
determined on CHO cell membranes stably expressing the human
ApelinR by a competition binding assay using as radioligand [125I]-3.
Data represent the means SEM of 3 to 4 independent experiments
performed in duplicate.
Figure 4 summarizes the important features of the structure−
activity relationships around 27. LRB, 1-methyl piperidyl, and
2-imino-3-methylthiazolyl groups play a crucial role in both the
affinity and the potency of 27 to induce ApelinR internalization.
The 3-benzyl imidazolyl group has only an influence on the
potency of 27 to induce ApelinR internalization, whereas its N′
methylation has a low but significant impact on its affinity.
Finally, the length of the spacer between the LRB group and
the rest of the molecule has no influence on both the affinity
and the potency of 27 to induce ApelinR internalization.
From the structure−activity relationship study around 27
emerged 19 and 21 that exhibited a lower molecular weight and
improved binding affinities for ApelinR as compared to 27, but
with different pharmacological properties. To further character-
ize these three new ligands and to determine the best
candidates for further in vivo evaluation, we decided to
investigate (1) their binding selectivity toward angiotensin II
receptor type 1 (AT1R), which exhibits the highest amino acid
sequence identity compared to that of ApelinR in the GPCR
methyl-4-piperidyl was switched to the tetrahydropyran moiety,
not charged at physiological pH (37), leading to a 2.5-fold
decrease in the affinity of the compound for the ApelinR.
Interestingly, switching to 1-benzyl-4-piperidyl (20) or (1-
methyl-4-piperidyl)methyl (38) groups at the X position
resulted in a slight increase in the affinity of these compounds
as compared to 27 (respectively, 0.27 and 0.18 vs 0.40 μM).
Finally, the suppression of the methyl group on imidazole or
iminothiazoline rings (compounds 39 and 40, respectively) led
to a 2-fold decrease in the affinity (0.96 and 0.85 μM,
respectively) of these compounds as compared to 19.
Ability of the Ligands to Trigger ApelinR Internal-
ization. Confocal microscopic analysis of CHO cells stably
expressing the rat ApelinR-EGFP in resting conditions
displayed intense ApelinR-EGFP fluorescence at the plasma
membrane (Figure 3) as described previously.13 Incubation
with 2 (100 nM) for 20 min resulted in a massive
2913
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
Figure 3. Effects of 2, 27, 19, and 21 on rat ApelinR-EGFP internalization in CHO cells. (A) CHO cells stably expressing the rat ApelinR-EGFP
were treated with 2 (100 nM), 27 (100 μM), or 19 (100 μM) for 20 min at 37 °C. Cells were fixed, and receptor internalization was analyzed by
measuring EGFP fluorescence by confocal microscopy. (B) CHO cells stably expressing the rat ApelinR-EGFP were treated with either 2 (10 nM, 1
nM) or 21 (100 μM) or both for 20 min at 37 °C. Cells were fixed, and receptor internalization was analyzed by confocal microscopy.
Figure 4. Summary of the structure−activity relationships around 27: important structural determinants for both the affinity and the potency of the
ligand to trigger ApelinR internalization.
family; (2) their effects on forskolin-induced cAMP production;
and (3) their stability in mouse plasma.
Incubation of CHO cells stably expressing the rat ApelinR in
the presence of 5 μM forskolin with increasing concentrations
of 2, 27, 19, and 21 (10 nM to 100 μM) resulted in a
concentration-dependent inhibition of forskolin-induced cAMP
production (Figure 5) with IC50 values of 1.4 × 10−3, 4.3, 3.0,
and 1.7 μM, respectively (Table 5). The maximal inhibitory
effects of these compounds on forskolin-induced cAMP
production were 85 13%, 100 7%, 77 15%, and 69
14% for 2, 27, 19, and 21, respectively. Since 21 displaces the
radioligand [125I]-3 in a competitive manner and since it
inhibits cAMP production as apelin does, this suggests that 21
likely binds to the ApelinR at the same binding site as the
endogenous peptides. The molecular mechanism of the biased
Binding Selectivity and Effects on Forskolin-Induced
cAMP Production. The selectivity of 27, 19, and 21 toward
AT1R was evaluated by a radioligand binding assay. Whereas
27 displayed only a low 3-fold binding selectivity for both
receptors, 19 exhibited a >2500-fold binding selectivity
(AT1R/ApelinR). These data suggest that the shorter the
spacer, the more selective is the compound. This was confirmed
with 21 that displayed a 675-fold binding selectivity toward the
ApelinR as compared to that of the AT1R. To investigate
further their pharmacological profile, the effect of the three
ligands on forskolin-induced cAMP production was evaluated.
2914
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
ApelinR internalization, whereas 21 behaves as a biased agonist
toward cAMP production. Finally, 19 and 21 display a much
greater stability in mouse plasma than the endogenous apelin
peptide 2 and represent therefore suitable pharmacological
tools for in vivo studies. Considering their fluorescent
properties, we anticipate that these probes could not only
constitute convenient tools for in vivo labeling of the ApelinR
but also improve our knowledge on its physiopathologic roles.
EXPERIMENTAL SECTION
■
General Methods. Reagents were obtained from commercial
sources and used without any further purification. DFPE-PS resin
(100−200 mesh, 0.92 mmol/g) was purchased from Novabiochem.
Acid Red 52 sodium salt (Lissamine Rhodamine B sodium sulfonate)
was obtained from TCI Europe (Zwijndrecht, Belgium). Patent Blue
VF [129-17-9] was obtained from ACROS with a 50% dye content. Its
conversion into activated sulfonyl chloride (PB-Cl) was performed
following the experimental conditions previously reported.42 Solid-
phase reactions conducted at room temperature were performed in
polypropylene tubes equipped with polyethylene frits and poly-
propylene caps using an orbital shaking device. Solid-phase reactions at
higher temperatures were conducted in sealed glassware tubes using a
Chemflex rotating oven from Robbins Scientific as the shaking device.
Thin-layer chromatography was performed on silica gel 60F254 plates.
Merck silica gel (Kieselgel 60; 230−400 mesh) was used for
chromatography columns. Flash chromatography was performed on
silica gel (40 μm, Grace) or RP18 (25−40 μm, Merck) prepacked
columns on a SpotII ultimate from Armen or a PLC 2020 from Gilson.
1H NMR spectra were recorded at 300 MHz or 400 MHz on a Bruker
Advance spectrometer. Chemical shifts are reported in parts per
million (ppm), and coupling constants (J) are reported in hertz (Hz).
Analytical reverse-phase high performance liquid chromatography
(RP-HPLC) separations were performed on C18 Ascentis Express (2.7
μm, 4.6 mm × 75 mm) using a linear gradient (5% to 100% of solvent
B in solvent A in 7.5 min, flow rate of 1.6 mL·min−1, detection at 220
nm; solvent A, water/0.1% TFA; solvent B, acetonitrile/0.1% TFA).
Semipreparative reverse phase high performance liquid chromatog-
raphy (RP-HPLC) separations were performed on a Waters XBridge
RP-C18 column (5 μm, 19 × 100 mm) using a linear gradient (solvent
B in solvent A; solvent A, water/0.1% TFA; solvent B, acetonitrile/
0.1% TFA; flow rate of 20 mL·min−1; detection at 220 nm). Purified
final compounds eluted as single and symmetrical peaks (thereby
confirming a purity of ≥95%) at the retention times (tR) given below.
High resolution mass spectra (HRMS) were acquired on a Bruker
MicroTof mass spectrometer, using electrospray ionization (ESI) and
a time-of-flight analyzer (TOF).
Figure 5. Effects of 2, 27, 19, and 21 on forskolin-induced cAMP
production in CHO cells stably expressing the rat ApelinR-EGFP.
cAMP production was induced by treatment of cells with 5 μM
forskolin. The effects of various concentrations of 2, 27, 19, and 21 on
forskolin-induced cAMP production were evaluated by HTRF
(Cisbio). Sigmoidal dose−response curves are representative of
three independent experiments performed in duplicate.
agonism is probably related to subtle differences in the
interaction network at the orthosteric binding cavity.
Altogether, these data suggest that 27 and 19 are full agonists
with regard to cAMP production and ApelinR internalization,
whereas 21 is a biased agonist toward cAMP production. Biased
agonists trigger conformational changes of the signaling
effectors such as G-protein and β-arrestin, different from
those induced by unbiased agonists.40,41 The development of
such compounds could be therapeutically useful since biased
agonists, by stabilizing the receptor in a defined conformation,
could induce a specific apelin biological action.
Plasma Stability of Compounds 19, 21, and 27 in
Mouse Plasma. The nonpeptidic character of 19, 21, and 27
prompted us to evaluate their stability in mouse plasma at 37
°C. Whereas the half-life stability of 2 is around 4.6 min, the
three ApelinR ligands 27, 19, and 21 exhibit a much higher
stability in mouse plasma (T1/2 > 10 h), which opens the way to
the in vivo evaluation of these compounds.
CONCLUSIONS
■
A new convenient synthetic route was developed to allow the
study of the structure−activity relationships of the first
described nonpeptidic ApelinR agonist 1. This approach
enabled the access to 27, the major component of 1. The
structural determinants of its activity have been characterized
leading to the discovery of 19 and 21, two novel ligands of the
ApelinR displaying lower molecular weights and higher
affinities than 27. In addition, 19 and 21 show an increased
selectivity toward the AT1R, as compared to 27. Interestingly,
19 is a full agonist with regard to cAMP production and
Methyl (2-Imino-3-methyl-2,3-dihydrothiazol-4-yl)acetate hydro-
iodide (5). Methyl iodide (38 mL, 0.610 mol) was added to a solution
of commercially available compound 4 (10.50 g, 0.061 mol) in dry
THF (110 mL), and the mixture was stirred at room temperature for
72 h. The mixture was filtered, and the precipitate was washed several
times with cold THF and dried in vacuo to give the title compound
1
(17.94 g, 93%) as a white solid. tR = 0.49 min; mp 153−155 °C; H
Table 5. Summary of Pharmacological Properties of 2, 27, 19, and 21
a
a
binding affinity (Ki, μM)
binding selectivity
AT1R/ApelinR
b
c
cpd
ApelinR
AT1R
>100
inhibition of cAMP production (IC50, μM)
plasma half-life (T1/2
)
2
0.04 0.002 × 10−3
0.69 0.12
>105
1.4 0.67 × 10−3
4.3 1.9
4.6 min
>10 h
>10 h
>10 h
27
19
21
2.23 0.7
>1000
3
0.40 0.08
2630
675
3.0 0.7
0.089 0.014
98 30
1.7 0.8
a
The inhibition constants (Ki) of the compounds for the ApelinR or AT1R were determined on CHO cell membranes expressing either the human
b
ApelinR or the human AT1R by competition binding assays using as radioligand either [125I]-3 or [125I]-angiotensin II. cAMP was quantified using
the cAMP dynamic 2 assay kit (Cisbio Bioassays, see Experimental Section). Data represent the means SE of 3 to 4 independent experiments
c
performed in duplicate. Stability was evaluated in mouse plasma at 37 °C (see Experimental Section).
2915
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
NMR (DMSO-d6, 400 MHz) δ 9.39 (br s, 2H), 6.94 (s, 1H), 3.97 (s,
2H), 3.67 (s, 3H), 3.44 (s, 3H); 13C NMR (DMSO-d6, 100 MHz) δ
168.8, 168.4, 134.4, 105.8, 52.4, 33.1, 32.5.
for another 30 min. The solution was drained, and the resin was
washed first with dry CH2Cl2 and then with dry THF. To the resin,
swollen in dry THF/HC(OCH3)3, 1/1 (20 mL/g of resin),
formaldehyde (37 wt % in water, 40 equiv) was added, and the
mixture was shaken at room temperature for 5 h. The solution was
drained, and the resin was washed with dry 1,2-dichloroethane (1,2-
DCE). A suspension of sodium triacetoxyborohydride (20 equiv) in
dry 1,2-DCE (20 mL/g of resin) was added, and the mixture was
shaken at room temperature overnight. The solution was drained, and
the resin washed with DMF, CH2Cl2, and MeOH. To ensure
completion of the reaction, a small sample of resin was treated with
TFA/H2O/TIS, 95/2.5/2.5, the solution was collected, evaporated in
vacuo, and analyzed by LC-MS.
Synthesis of Resin 12. Resin 12 was prepared as resin 11, except
that formaldehyde was replaced by benzaldehyde (20 equiv).
General Experimental Procedure for the Synthesis of Resins
13 and 14 from Resins 11 and 12, Respectively. The resin was
swollen in DMF, and the excess solvent was removed by filtration. A
solution of piperidine in DMF (20% v/v−15 mL/g of resin) was
added, and the mixture was shaken at room temperature for 10 min.
The solution was drained, and the operation was repeated for 15 min.
The solution was drained and the resin was washed with DMF,
CH2Cl2, and MeOH. The resin was again swollen in DMF, and the
excess solvent was removed by filtration. A solution of Fmoc-
Lys(ivDde)-OH (5 equiv), PyBOP, (5 equiv), and HOBt (5 equiv) in
DMF (15 mL/g of resin) was added followed by N,N-diisopropyle-
thylamine (DIEA) (12.5 equiv), and the mixture was shaken at room
temperature for 2 h. The solution was drained, and the resin was
washed with DMF, CH2Cl2, and MeOH. The resin was again treated
with piperidine in DMF as described above. A solution of 8 (4 equiv),
PyBOP, (4 equiv), and HOBt (4 equiv) in CH2Cl2/DMF, 4/1 (15
mL/g of resin), was added followed by DIEA (10 equiv), and the
mixture was shaken at room temperature overnight. The solution was
drained, and the resin was washed with DMF, CH2Cl2, and MeOH.
The resin was swollen in DMF, and the excess solvent was removed by
filtration. A solution of hydrazine hydrate in DMF (5% v/v−15 mL/g
of resin) was then added, and the mixture was shaken at room
temperature for 5 min. The solution was drained, and the hydrazine
treatment was repeated twice. The resin was washed with DMF,
CH2Cl2, and MeOH.
General Experimental Procedure for the Synthesis of Resins
15 and 17 from Resin 13 and of Resin 16 from Resin 14. The
resin was swollen in dry CH2Cl2, and the excess solvent was removed
by filtration. A solution of LRB-Cl or PB-Cl (2 equiv) in dry CH2Cl2
(25 mL/g of resin) was added followed by DIEA (3 equiv). The
mixture was shaken at room temperature for 5 h, and the solution was
drained. The resin was washed successively with 10% DIEA in DMF,
THF/H2O, 3/2, DMF, CH2Cl2, and MeOH, and those washings were
repeated at least 5 times. The resin was then washed with Et2O and
dried under vacuum.
Synthesis of Resin 18. Acetic anhydride (10 equiv) and DIEA (10
equiv) were added to resin 13 and swollen in dry CH2Cl2 (10 mL/g of
resin), and the mixture was shaken at room temperature for 10 min.
The solution was drained, and the treatment was repeated once. The
solution was drained, the resin washed with DMF, CH2Cl2, and
MeOH, and dried under vacuum.
General Experimental Procedure for the Synthesis of Resins
23 and 24. Resin 13 was swollen in dry CH2Cl2 and the excess
solvent was removed by filtration. A solution of 4-nitrophenyl
chloroformate (5 equiv) in dry THF/CH2Cl2, 1/1 (40 mL/g of
resin), was added followed by DIEA (5 equiv). The mixture was
shaken at room temperature for 45 min, and the solution was drained.
The procedure was repeated once. The resin was washed with CH2Cl2
and Et2O before being transferred into a microwave tube. A
suspension of diamine H2NCH2(CH2)nCH2NH2 (n = 4 or 10, 20
equiv) in 1,2-DCE (40 mL/g of resin) was added, the tube was sealed,
and the mixture was shaken at 30 °C overnight. The resin was filtered
and washed with DMF, CH2Cl2, and MeOH.
(2-Imino-3-methyl-2,3-dihydrothiazol-4-yl)acetic Acid Hydro-
chloride (6). 1 N HCl (6.6 mL, 6.6 mmol) was slowly added to a
solution of 5 (82.0 mg, 0.26 mmol) in dioxane/water, 2/1 (4.1 mL),
and the mixture was stirred at 40 °C for 19 h. It was then concentrated
in vacuo, and EtOAc was added to the crude. A precipitate formed,
which was collected by filtration, washed with cold EtOAc, and dried
under vacuum to give the title compound (47.6 mg, 88%) as a white-
1
gray solid. mp 172−174 °C. H NMR (400 MHz, DMSO-d6) δ 9.53
(br s, 1H), 6.90 (s, 1H), 3.84 (s, 2H), 3.46 (s, 3H); 13C NMR (100
MHz, DMSO-d6) δ 169.9, 168.4, 135.2, 105.2, 33.2, 33.0. HRMS
(ESI) calcd for C6H9N2O2S ([M + H]+) 173.0384; found, 173.0377.
Methyl 2-(tert-Butoxycarbonylimino-3-methyl-2,3-dihydrothia-
zol-4-yl)acetate (7). Compound 5 (4.0 g, 12.7 mmol) was added
portionwise to a solution of di-tert-butyl dicarbonate (13.9 g, 63.7
mmol), DMAP (476 mg, 3.9 mmol), and triethylamine (4.6 mL, 32.9
mmol) in dry THF (80 mL), under an argon atmosphere. The mixture
was heated to reflux for 4 h then cooled to room temperature and
filtered. The filtrate was concentrated and the crude product purified
by column chromatography (5% methanol in CH2Cl2), affording the
title compound (3.26 g, 61%) as a white solid. tR = 3.08 min; mp 126−
1
128 °C; H NMR (CDCl3, 400 MHz) δ 6.37 (s, 1H), 3.68 (s, 3H),
3.59 (s, 2H), 3.53 (s, 3H), 1.49 (s, 9H); 13C NMR (CDCl3, 100 MHz)
δ 170.7, 168.7, 162.3, 130.8, 106.4, 79.9, 52.8, 33.9, 33.2, 28.5.
2-(tert-Butoxycarbonylimino-3-methyl-2,3-dihydrothiazol-4-yl)-
acetic Acid (8). Lithium hydroxide (1.44 g, 60.1 mmol) was added to a
solution of 7 (2.50 g, 8.7 mmol) in dioxane/water, 2/1 (60 mL), and
the mixture was stirred at 40 °C for 1 h, then cooled to room
temperature. The pH of the solution was adjusted to pH 4 by the
addition of citric acid, and the aqueous layer was extracted several
times with CH2Cl2. The combined organic extracts were washed with
brine, dried over sodium sulfate, and concentrated. The residue was
dissolved in the minimum amount of Et2O, and n-heptane was added.
The mixture was sonicated and filtered, and the precipitate was washed
with cold n-heptane and dried in vacuo to give the title compound
1
(1.47 g, 61%) as a white solid. tR = 2.96 min; mp 107−109 °C; H
NMR (CDCl3, 400 MHz) δ 9.05 (sl, 1H), 6.45 (s, 1H), 3.61 (s, 2H),
3.55 (s, 3H), 1.48 (s, 9H); 13C NMR (CDCl3, 100 MHz) δ 17.1.6,
170.1, 161.0, 132.2, 107.2, 80.5, 34.3, 33.7, 28.4; HRMS (ESI) calcd
for C11H16N2O4S ([M]+) 272.0830; found, 272.0818.
Synthesis of Resin 9 from DFPE Resin. A mixture of DFPE resin
(loading: 0.92 mmol/g), 4-amino-1-Boc-piperidine (5 equiv), and
sodium cyanoborohydride (5 equiv) in DMF/MeOH/AcOH, 80/19/
1 (10 mL/g of resin), was shaken at 60 °C overnight. The solution was
drained, and the resin was washed in sequence with DMF, CH2Cl2,
and MeOH, then dried under vacuum for 24 h. The resin was swollen
in CH2Cl2, and the excess solvent was removed by filtration. In a
separate vial, DIEA (12.5 equiv) was added to a solution of Fmoc-
His(1-Bzl)-OH (5 equiv) and HATU (4.95 equiv) in DMF/CH2Cl2,
4/1 (15 mL/g of resin). After 1 min, the resulting solution was added
to the resin, and the mixture was shaken at room temperature
overnight. The solution was drained, and the resin was washed with
DMF, CH2Cl2, MeOH, and dried under vacuum.
Synthesis of Resin 10. To resin 9, swollen in DMF (10 mL/g of
resin), was added methyl iodide (40 equiv), and the mixture was
shaken at 40 °C overnight. The solution was drained, and the resin was
washed in sequence with DMF, CH2Cl2, and MeOH. The resin was
then swollen in CH2Cl2, the excess solvent was removed by filtration,
and the resin dried under vacuum for no less than 24 h. To ensure
completion of the reaction, a small sample of resin was treated with
TFA/H2O/triisopropylsilane (TIS), 95/2.5/2.5, and the solution was
collected, evaporated in vacuo, and analyzed by LC-MS.
Synthesis of Resin 11. Resin 10 was swollen in dry CH2Cl2, and the
excess solvent removed by filtration. In a separate vial, freshly prepared
solutions of TMSOTf (0.4 M in dry CH2Cl2, 10 mL/g of resin) and
2,6-lutidine (0.8 M in dry CH2Cl2, 10 mL/g of resin) were combined
and added to the resin. The mixture was shaken at room temperature
for 30 min. The solution was drained, and the reaction was repeated
2916
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
Synthesis of Resins 25 and 26 from Resins 23 and 24,
Respectively. Resins 23 and 24 were labeled with LRB-Cl as described
previously for resins 15 and 16.
General Experimental Procedure for the Cleavage of the
Products from the Resin. The dried resin was treated with TFA/
H2O/TIS, 95/2.5/2.5 (25 mL/g of resin), and the mixture was shaken
at room temperature for 3 h. The solution was collected and the beads
washed with TFA. The solution was evaporated in vacuo and the crude
product purified by semipreparative RP-HPLC. Lyophilization
afforded the expected product.
Compound 19. Prepared on a 690 μmol scale. Pink solid (44.0 mg,
4%). tR = 3.71 min. (>95% purity [220 nm]); HRMS (ESI) calcd for
C59H76N11O9S3 ([M]+) 1178.4989; found, 1178.4967.
Compound 20. Prepared on a 50 μmol scale. Pink solid (23.9 mg,
30%). tR = 3.88 min. (>97% purity [220 nm]); HRMS (ESI) calcd for
C65H81N11O9S3 ([M + H]2+) 627.7690; found, 627.7702.
Quentin, France) and incubated for 30 min at 37 °C. Then cells were
incubated for 90 min at 37 °C with 90 μM of cycloheximide (Sigma-
Aldrich, St. Quentin, France). Internalization was performed by
incubating the cells for 20 min at 37 °C with 1 nM, 10 nM, and 100
nM of 2 or 100 mM of 19, 21, and 27 as described previously.13 Cells
were then mounted in Aquapolymount (Polysciences, Warrington, PA,
USA) for confocal microscopic analysis.
Confocal Microscopy. The cells were examined with a Leica TCS
SP5MP (Leica Microsystems, Heidelberg, Germany) equipped with an
argon ion laser adjusted to 488 nm for GFP and a helium neon laser
adjusted to 568 nm for LRB. Sections were scanned, and double-
labeled images were obtained with two photomultipliers and
appropriate filter settings for the separate detection of fluorescence
from GFP and LRB. Optical sections (1024 × 1024) of individual cells
were taken at the equatorial level (level of the nucleus), using a 63xNA
oil-immersion objective.
Compound 21. Prepared on a 115 μmol scale. Blue solid (17.8 mg,
10%). tR = 3.47 min. (>98% purity [220 nm]); HRMS (ESI) calcd for
C59H78N11O8S3 ([M]+) 1164.5197; found, 1164.5193.
Compound 22. Prepared on a 55 μmol scale. White solid (8.2 mg,
15%). tR = 2.00/2.06 min. (>98% purity [220 nm]); MS (ESI) m/z
680.2 ([M]+), 340.8 ([M + H]2+).
ASSOCIATED CONTENT
■
S
* Supporting Information
General procedure for the preparation of compounds 29−40;
pKa determination for 4 and 5; RP-HPLC, CZE, and MS/MS
analysis of 1 (Figures S1, S2, and S3, respectively), and confocal
microscopy monitoring of the ability of 20, 37, 38, and 40 to
trigger rat ApelinR internalization (Figure S4). This material is
available free of charge via the Internet at http://pubs.acs.org.
Compound 27. Prepared on a 560 μmol scale. Pink solid (8.8 mg,
2%). tR = 4.82 min. (>95% purity [220 nm]); HRMS (ESI) calcd for
C72H102N13O10S3 ([M + H]2+) 702.3517; found, 702.8568.
Compound 28. Prepared on a 110 μmol scale. Pink solid (14.6 mg,
8%). HRMS (ESI) calcd for C66H92N13O10S3 ([M + H]2+) 660.8087;
found, 660.8062.
Stability in Mouse Plasma. Stability of compounds 2, 27, 19, and
21 was determined in mouse plasma at 37 °C. For each compound,
the stock solution (100 μM in water) was diluted in plasma to a final
incubation concentration of 5 μM. The incubation at 37 °C was
stopped at t0 and 24 h by adding one volume of ice cold acetonitrile
containing 0.1% trifluoroacetic acid. The sample was vortexed for 1
min and then centrifuged at 4 °C before LC-MS injection of the
supernatant. Analyses were performed on a Kinetex RP-C18 column
(2.6 μm, 100 Å, 50 × 4.6 mm) using a linear gradient (solvent B in
solvent A, solvent A, water/0.05% TFA; solvent B, acetonitrile; flow
rate of 2 mL·min−1; detection at 358 nm). The percentage of the
remaining test compound relative to t0 was measured by monitoring
the peak area of the chromatogram.
AUTHOR INFORMATION
■
Corresponding Authors
*(C.L.-C.) Phone: +33 1 44 27 16 63. E-mail: c.llorens-cortes@
college-de-france.fr.
*(D.B.) Phone: +33 368 85 42 36. E-mail: dominique.bonnet@
unistra.fr
Author Contributions
§J.-F.M. and X.I. contributed equally to this work.
Author Contributions
∥C.L.-C. and D.B. are co-last authors.
Notes
Radioligand Binding Experiments. Membrane preparation from
CHO cells stably expressing the human ApelinR or the human
angiotensin II receptor type 1 were purchased from Perkin-Elmer
(Perkin-Elmer, Wellesley, MA, USA). Membrane preparations (0.5 μg
of total mass of membranes/assay) were incubated for 1 h at 20 °C
with 0.2 nM [125I]-3 (Perkin-Elmer) or with 0.1 nM [125I]-angiotensin
II (Perkin-Elmer) in binding buffer (50 mM Hepes, 5 mM MgCl2 at
pH 7.5, BSA 1%) alone or in the presence of the different compounds
to be tested at various concentrations. The reaction was stopped by
adding 4 mL of cold binding buffer and filtered on Whatman GF/C
filters. After washing, radioactivity was counted with a Wizard 1470
Wallac gamma counter (Perkin-Elmer, Turku, Finland).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
J.-F.M. was supported by the Agence Nationale pour la
Recherche (ANR Physique et Chimie du Vivant 2008), the
Centre National de la Recherche Scientifique, the Universite
Strasbourg, and the Institut National de la Sante et de la
Recherche Medicale. We are grateful to Patrick Wehrung for
ES-MS analyses, to Dr. Agnes Hagege for CZE analysis, and to
́
de
́
́
̀
̀
Dr. Patrick Gizzi for stability studies.
cAMP Assay. cAMP was quantified using the cAMP dynamic 2
assay kit (Cisbio Bioassays, Codolet, France). CHO cells stably
expressing ApelinR-EGFP (1 × 106 cells/well) were seeded in 96 wells
plates previously coated with polylysine (WI; 0.01%) (Sigma-Aldrich,
St Quentin, France) and incubated for 24 h before treatment. After the
following day, cells were stimulated with 100 μL of stimulation buffer
(HBSS, 5 mM Hepes, 0.1% BSA stabilizer, and 1 mM IBMX, pH 7.4)
containing 5 μM forskolin (FSK) and increasing concentrations of the
test compound (10 nM to 100 μM) for 30 min at 37 °C. Cells were
then washed 3 times with 100 μL of stimulation buffer. Finally, for
each well we added 50 μL of stimulation buffer, 25 μL of anti-cAMP
antibody cryptate, and 25 μL of cAMP D2. After 1 h of incubation at
20 °C, fluorescence was measured using an EnVision Xcite Multilabel
Reader (Perkin-Elmer, Wellesley, MA, USA).
ABBREVIATIONS USED
■
AT1R, angiotensin II receptor type 1; ApelinR, apelin receptor;
AVP, arginine-vasopressine; DFPE, 2-(3,5-dimethoxy-4-
formylphenoxy)ethyl; EGFP, enhanced green fluorescent
protein; FRET, fluorescence resonance energy transfer;
GPCRs, G protein-coupled receptors; HBTU, N-[(1Hbenzo-
triazol-1-yl)(dimethylamino)methylene]-N-methylmethanami-
nium hexafluorophosphate N-oxide; HOBt, N-hydroxybenzo-
triazole; HRMS, high resolution mass spectrometry; LRB,
Lissamine Rhodamine B; PB-Cl, Patent Blue sulfonyl chloride;
MS, mass spectrometry; PyBOP, (benzotriazol-1-yl-oxy)-
trispyrrolidinophosphonium hexafluorophosphate; REM, re-
generative Michael acceptor; TIS, triisopropylsilane; TMSOTf,
trimethylsilyl trifluoromethanesulfonate
ApelinR Internalization Studies. CHO cells stably expressing the
rat ApelinR-EGFP were seeded at 40% confluency on glass coverslips
coated with polylysine (weight/volume 0.01%) (Sigma-Aldrich, St.
2917
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
(16) Brailoiu, G. C.; Dun, S. L.; Yang, J.; Ohsawa, M.; Chang, J. K.;
Dun, N. J. Apelin-immunoreactivity in the rat hypothalamus and
pituitary. Neurosci. Lett. 2002, 327, 193−1997.
(17) O’Carroll, A. M.; Selby, T. L.; Palkovits, M.; Lolait, S. J.
Distribution of mRNA encoding B78/apj, the rat homologue of the
human APJ receptor, and its endogenous ligand apelin in brain and
peripheral tissues. Biochim. Biophys. Acta 2000, 1492, 72−80.
(18) Reaux, A.; Gallatz, K.; Palkovits, M.; Llorens-Cortes, C.
Distribution of apelin-synthesizing neurons in the adult rat brain.
Neuroscience (Oxford, U.K.) 2002, 113, 653−662.
(19) O’Carroll, A. M.; Don, A. L.; Lolait, S. J. APJ receptor mRNA
expression in the rat hypothalamic paraventricular nucleus: regulation
by stress and glucocorticoids. J. Neuroendocrinol. 2003, 15, 1095−1101.
(20) Reaux-Le Goazigo, A.; Morinville, A.; Burlet, A.; Llorens-Cortes,
C.; Beaudet, A. Dehydration-induced cross-regulation of apelin and
vasopressin immunoreactivity levels in magnocellular hypothalamic
neurons. Endocrinology 2004, 145, 4392−4400.
(21) Japp, A. G.; Cruden, N. L.; Amer, D. A.; Li, V. K.; Goudie, E. B.;
Johnston, N. R.; Sharma, S.; Neilson, I.; Webb, D. J.; Megson, I. L.;
Flapan, A. D.; Newby, D. E. Vascular effects of apelin in vivo in man. J.
Am. Coll. Cardiol. 2008, 52, 908−913.
(22) Ashley, E. A.; Powers, J.; Chen, M.; Kundu, R.; Finsterbach, T.;
Caffarelli, A.; Deng, A.; Eichhorn, J.; Mahajan, R.; Agrawal, R.; Greve,
J.; Robbins, R.; Patterson, A. J.; Bernstein, D.; Quertermous, T. The
endogenous peptide apelin potently improves cardiac contractility and
reduces cardiac loading in vivo. Cardiovasc. Res. 2005, 65, 73−82.
(23) Berry, M. F.; Pirolli, T. J.; Jayasankar, V.; Burdick, J.; Morine, K.
J.; Gardner, T. J.; Woo, Y. J. Apelin has in vivo inotropic effects on
normal and failing hearts. Circulation 2004, 110, 187−193.
(24) Szokodi, I.; Tavi, P.; Foldes, G.; Voutilainen-Myllyla, S.; Ilves,
M.; Tokola, H.; Pikkarainen, S.; Piuhola, J.; Rysa, J.; Toth, M.;
Ruskoaho, H. Apelin, the novel endogenous ligand of the orphan
receptor APJ, regulates cardiac contractility. Circ. Res. 2002, 91, 434−
440.
(25) Kuba, K.; Zhang, L.; Imai, Y.; Arab, S.; Chen, M.; Maekawa, Y.;
Leschnik, M.; Leibbrandt, A.; Markovic, M.; Schwaighofer, J.; Beetz,
N.; Musialek, R.; Neely, G. G.; Komnenovic, V.; Kolm, U.; Metzler, B.;
Ricci, R.; Hara, H.; Meixner, A.; Nghiem, M.; Chen, X.; Dawood, F.;
Wong, K. M.; Sarao, R.; Cukerman, E.; Kimura, A.; Hein, L.;
Thalhammer, J.; Liu, P. P.; Penninger, J. M. Impaired heart
contractility in Apelin gene-deficient mice associated with aging and
pressure overload. Circ. Res. 2007, 101, 32−42.
(26) Tatemoto, K.; Takayama, K.; Zou, M. X.; Kumaki, I.; Zhang, W.;
Kumano, K.; Fujimiya, M. The novel peptide apelin lowers blood
pressure via a nitric oxide-dependent mechanism. Regul. Pept. 2001, 99,
87−92.
(27) Lee, D. K.; Saldivia, V. R.; Nguyen, T.; Cheng, R.; George, S. R.;
O’Dowd, B. F. Modification of the terminal residue of apelin-13
antagonizes its hypotensive action. Endocrinology 2005, 146, 231−236.
(28) Boucher, J.; Masri, B.; Daviaud, D.; Gesta, S.; Guigne, C.;
Mazzucotelli, A.; Castan-Laurell, I.; Tack, I.; Knibiehler, B.; Carpene,
C.; Audigier, Y.; Saulnier-Blache, J. S.; Valet, P. Apelin, a newly
identified adipokine up-regulated by insulin and obesity. Endocrinology
2005, 146, 1764−1771.
REFERENCES
■
(1) Tatemoto, K.; Hosoya, M.; Habata, Y.; Fujii, R.; Kakegawa, T.;
Zou, M. X.; Kawamata, Y.; Fukusumi, S.; Hinuma, S.; Kitada, C.;
Kurokawa, T.; Onda, H.; Fujino, M. Isolation and characterization of a
novel endogenous peptide ligand for the human APJ receptor.
Biochem. Biophys. Res. Commun. 1998, 251, 471−476.
(2) O’Dowd, B. F.; Heiber, M.; Chan, A.; Heng, H. H.; Tsui, L. C.;
Kennedy, J. L.; Shi, X.; Petronis, A.; George, S. R.; Nguyen, T. A
human gene that shows identity with the gene encoding the
angiotensin receptor is located on chromosome 11. Gene 1993, 136,
355−360.
(3) Devic, E.; Rizzoti, K.; Bodin, S.; Knibiehler, B.; Audigier, Y.
Amino acid sequence and embryonic expression of msr/apj, the mouse
homolog of Xenopus X-msr and human APJ. Mech. Dev. 1999, 84,
199−203.
(4) De Mota, N.; Lenkei, Z.; Llorens-Cortes, C. Cloning,
pharmacological characterization and brain distribution of the rat
apelin receptor. Neuroendocrinology 2000, 72, 400−407.
(5) Habata, Y.; Fujii, R.; Hosoya, M.; Fukusumi, S.; Kawamata, Y.;
Hinuma, S.; Kitada, C.; Nishizawa, N.; Murosaki, S.; Kurokawa, T.;
Onda, H.; Tatemoto, K.; Fujino, M. Apelin, the natural ligand of the
orphan receptor APJ, is abundantly secreted in the colostrum. Biochim.
Biophys. Acta 1999, 1452, 25−35.
(6) Lee, D. K.; Cheng, R.; Nguyen, T.; Fan, T.; Kariyawasam, A. P.;
Liu, Y.; Osmond, D. H.; George, S. R.; O’Dowd, B. F. Characterization
of apelin, the ligand for the APJ receptor. J. Neurochem. 2000, 74, 34−
41.
(7) Azizi, M.; Iturrioz, X.; Blanchard, A.; Peyrard, S.; De Mota, N.;
Chartrel, N.; Vaudry, H.; Corvol, P.; Llorens-Cortes, C. Reciprocal
regulation of plasma apelin and vasopressin by osmotic stimuli. J. Am.
Soc. Nephrol. 2008, 19, 1015−1024.
(8) De Mota, N.; Reaux-Le Goazigo, A.; El Messari, S.; Chartrel, N.;
Roesch, D.; Dujardin, C.; Kordon, C.; Vaudry, H.; Moos, F.; Llorens-
Cortes, C. Apelin, a potent diuretic neuropeptide counteracting
vasopressin actions through inhibition of vasopressin neuron activity
and vasopressin release. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 10464−
10469.
(9) Iturrioz, X.; Gerbier, R.; Leroux, V.; Alvear-Perez, R.; Maigret, B.;
Llorens-Cortes, C. By interacting with the C-terminal Phe of apelin,
Phe255 and Trp259 in helix VI of the apelin receptor are critical for
internalization. J. Biol. Chem. 2010, 285, 32627−32637.
(10) Medhurst, A. D.; Jennings, C. A.; Robbins, M. J.; Davis, R. P.;
Ellis, C.; Winborn, K. Y.; Lawrie, K. W.; Hervieu, G.; Riley, G.; Bolaky,
J. E.; Herrity, N. C.; Murdock, P.; Darker, J. G. Pharmacological and
immunohistochemical characterization of the APJ receptor and its
endogenous ligand apelin. J. Neurochem. 2003, 84, 1162−1172.
(11) Zhou, N.; Fan, X.; Mukhtar, M.; Fang, J.; Patel, C. A.; DuBois,
G. C.; Pomerantz, R. J. Cell-cell fusion and internalization of the CNS-
based, HIV-1 co-receptor, APJ. Virology 2003, 307, 22−36.
(12) Iturrioz, X.; Alvear-Perez, R.; De Mota, N.; Franchet, C.;
Guillier, F.; Leroux, V.; Dabire, H.; Le Jouan, M.; Chabane, H.;
Gerbier, R.; Bonnet, D.; Berdeaux, A.; Maigret, B.; Galzi, J. L.; Hibert,
M.; Llorens-Cortes, C. Identification and pharmacological properties
of E339−3D6, the first nonpeptidic apelin receptor agonist. FASEB J.
2010, 24, 1506−1517.
(13) El Messari, S.; Iturrioz, X.; Fassot, C.; De Mota, N.; Roesch, D.;
Llorens-Cortes, C. Functional dissociation of apelin receptor signaling
and endocytosis: implications for the effects of apelin on arterial blood
pressure. J. Neurochem. 2004, 90, 1290−1301.
(14) Evans, N. A.; Groarke, D. A.; Warrack, J.; Greenwood, C. J.;
Dodgson, K.; Milligan, G.; Wilson, S. Visualizing differences in ligand-
induced beta-arrestin-GFP interactions and trafficking between three
recently characterized G protein-coupled receptors. J. Neurochem.
2001, 77, 476−485.
(29) Murza, A.; Parent, A.; Besserer-Offroy, E.; Tremblay, H.;
Karadereye, F.; Beaudet, N.; Leduc, R.; Sarret, P.; Marsault, E.
Elucidation of the structure-activity relationships of apelin: influence of
unnatural amino acids on binding, signaling, and plasma stability.
Chem. Med. Chem. 2012, 7, 318−325.
(30) Hamada, J.; Kimura, J.; Ishida, J.; Kohda, T.; Morishita, S.;
Ichihara, S.; Fukamizu, A. Evaluation of novel cyclic analogues of
apelin. Int. J. Mol. Med. 2008, 22, 547−552.
(31) Macaluso, N. J.; Pitkin, S. L.; Maguire, J. J.; Davenport, A. P.;
Glen, R. C. Discovery of a competitive apelin receptor (APJ)
antagonist. Chem. Med. Chem. 2011, 6, 1017−1023.
(32) Macaluso, N. J. M.; Glen, R. C. Exploring the ’RPRL’ Motif of
Apelin-13 through Molecular Simulation and Biological Evaluation of
Cyclic Peptide Analogues. Chem. Med. Chem. 2010, 5, 1247−1253.
(15) Reaux, A.; De Mota, N.; Skultetyova, I.; Lenkei, Z.; El Messari,
S.; Gallatz, K.; Corvol, P.; Palkovits, M.; Llorens-Cortes, C.
Physiological role of a novel neuropeptide, apelin, and its receptor
in the rat brain. J. Neurochem. 2001, 77, 1085−1096.
2918
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919

Journal of Medicinal Chemistry
Article
(33) Maloney, P. R.; Khan, P.; Hedrick, M.; Gosalia, P.; Milewski, M.;
Li, L.; Roth, G. P.; Sergienko, E.; Suyama, E.; Sugarman, E.; Nguyen,
K.; Mehta, A.; Vasile, S.; Su, Y.; Stonich, D.; Nguyen, H.; Zeng, F. Y.;
Novo, A. M.; Vicchiarelli, M.; Diwan, J.; Chung, T. D.; Smith, L. H.;
Pinkerton, A. B. Discovery of 4-oxo-6-((pyrimidin-2-ylthio)methyl)-
4H-pyran-3-yl 4-nitrobenzoate (ML221) as a functional antagonist of
the apelin (APJ) receptor. Bioorg. Med. Chem. Lett. 2012, 22, 6656−
6660.
(34) Hibert, M.; Franchet, C. A., Galzi, J. L.; Pattus, F.; Guillier, F.
Collections of Traceable Compounds and Uses Thereof. Patent WO
2006003329 A1, 2006.
(35) Toy, P. H.; Reger, T. S.; Janda, K. D. Soluble polymer bound
cleavage reagents: A multipolymer strategy for the cleavage of tertiary
amines from REM resin. Org. Lett. 2000, 2, 2205−2207.
(36) Bonnet, D.; Riche, S.; Loison, S.; Dagher, R.; Frantz, M. C.;
Boudier, L.; Rahmeh, R.; Mouillac, B.; Haiech, J.; Hibert, M. Solid-
phase organic tagging resins for labeling biomolecules by 1,3-dipolar
cycloaddition: Application to the synthesis of a fluorescent non-
peptidic vasopressin receptor ligand. Chem.Eur. J. 2008, 14, 6247−
6254.
(37) Zhang, A. J.; Russell, D. H.; Zhu, J. P.; Burgess, K. A method for
removal of N-BOC protecting groups from substrates on TFA-
sensitive resins. Tetrahedron Lett. 1998, 39, 7439−7442.
(38) Yang, H.; Vasudevan, S.; Oriakhi, C. O.; Shields, J.; Carter, R. G.
Scalable synthesis of lissamine rhodamine B sulfonyl chloride and
incorporation of xanthene derivatives onto polymer supports. Synthesis
2008, 6, 957−961.
(39) Papadopouloudaifotis, Z.; Costakis, E.; Tsatsas, G.; Varonos, D.
Synthesis and pharmacological activity of some 2-imino-3-alkyl
thiazoline derivatives. Eur. J. Med. Chem. 1979, 14, 253−256.
(40) Sauliere, A.; Bellot, M.; Paris, H.; Denis, C.; Finana, F.; Hansen,
J. T.; Altie, M. F.; Seguelas, M. H.; Pathak, A.; Hansen, J. L.; Senard, J.
M.; Gales, C. Deciphering biased-agonism complexity reveals a new
active AT(1) receptor entity. Nat. Chem. Biol. 2012, 8, 622−630.
(41) Shukla, A. K.; Violin, J. D.; Whalen, E. J.; Gesty-Palmer, D.;
Shenoy, S. K.; Lefkowitz, R. J. Distinct conformational changes in beta-
arrestin report biased agonism at seven-transmembrane receptors.
Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 9988−9993.
(42) Bonnet, D.; Ilien, B.; Galzi, J. L.; Riche, S.; Antheaune, C.;
Hibert, M. A rapid and versatile method to label receptor ligands using
“click” chemistry: Validation with the muscarinic M1 antagonist
pirenzepine. Bioconjugate Chem. 2006, 17, 1618−1623.
2919
dx.doi.org/10.1021/jm401789v | J. Med. Chem. 2014, 57, 2908−2919