New potent calcimimetics: I. Discovery of a series of novel trisubstituted ureas

Article abstract of DOI:10.1016/j.bmcl.2013.01.078

Starting from Fendiline and R-568, we identified a novel series of urea compounds as positive allosteric modulators of the calcium sensing receptor (CaSR), as part of a program to identify novel therapeutics for secondary hyperparathyroidism. Initially id

Full text of DOI:10.1016/j.bmcl.2013.01.078

Accepted Manuscript
New Potent Calcimimetics: I. Discovery of a series of novel trisubstituted ureas
Taoues Temal, Hélène Jary, Marielle Auberval, Sarah Lively, Denis Guédin,
Jean-Paul Vevert, Pierre Deprez
PII:
S0960-894X(13)00111-X
http://dx.doi.org/10.1016/j.bmcl.2013.01.078
BMCL 20066
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry Letters
Please cite this article as: Temal, T., Jary, H., Auberval, M., Lively, S., Guédin, D., Vevert, J-P., Deprez, P., New
Potent Calcimimetics: I. Discovery of a series of novel trisubstituted ureas, Bioorganic & Medicinal Chemistry
Letters (2013), doi: http://dx.doi.org/10.1016/j.bmcl.2013.01.078
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

New Potent Calcimimetics: I. Discovery of a series of novel
trisubstituted ureas
Taoues Temal a, Hélène Jary a, Marielle Auberval a,
Sarah Lively b, Denis Guédin, Jean-Paul Vevert, Pierre Deprez a,*
a Galapagos SASU, 102 avenue Gaston Roussel, 93230 Romainville, France
b Amgen Inc, 1120 Veteran’s Boulevard, South San Francisco, CA94080, USA
Received Month XX, 2012; Accepted Month XX, 2012
* Corresponding author. Tel.: +33 149424687; fax: +33 149424658.
E-mail address: pierre.deprez@glpg.com
Abstract— Starting from Fendiline and R-568, we identified a novel series of urea compounds as positive allosteric modulators of the
calcium sensing receptor (CaSR), as part of a program to identify novel therapeutics for secondary hyperparathyroidism. Initially
identified disubstituted ureas were converted to trisubstituted urea lead 20e, which was further modified to increase in vivo potency.
Replacing a carbomethoxy substituent by various bioisosteres led to compound 46 which exhibited potent in vitro and in vivo activity
after oral administration. ©2000 Elsevier Science Ltd. All rights reserved.
Hyperparathyroidism is characterized by high
circulating levels of parathyroid hormone (PTH) due to
its increased secretion by parathyroid glands.1,2 The
GPCR Class C calcium-sensing receptor (CaSR)3 is
expressed on the surface of parathyroid cells and up
regulates PTH secretion in response to low extracellular
calcium concentration. The CaSR proved to be an
attractive target for treatment of secondary
hyperparathyroidism, in patients with chronic renal
failure on dialysis, with the discovery of Cinacalcet
that we were able to replace the arylalkylamine with a
trisubstituted urea leading to compounds with in vitro
activity below 100 nM and oral in vivo activity.
N
H
Cl
R-568
1
R
H
O
N
N
A)
O
H
N
2
R
2, R1 = H, R2 = OMe
7, R1 = H, R2 = COOMe
(AMG-073, Sensipar®) as
a
positive allosteric
20e, R1 = (CH2)2-morpholine, R2 = COOMe
modulator (calcimimetic, Figure 1).4
Fendiline
46, R1 = (CH2)2-morpholine, R2 =
Most of the potent calcimimetics reported so far belong
to the arylalkylamine family.5-7 In this paper, we
describe the identification of a new trisubstituted urea
series which is structurally distinct from the known
arylalkylamines positive allosteric modulators. In vitro
and in vivo data of optimized compound 46 are
presented.
O
N
CF3
N
B)
H
AMG-073
Figure 1. A) From R-568 and Fendiline to urea based positive CaSR
allosteric modulators. B) Cinacalcet (AMG-073)
Based on the literature data available at the beginning of
the program, we started from known arylalkylamine
allosteric modulator R-568 and the weakly active
Fendiline (Figure 1).8 The gem diphenyl moiety, which
is present in Fendiline and is known to be a privileged
structure fragment for GPCRs,9 was chosen as a key
feature for designing new calcimimetics. We discovered
Disubstituted ureas 2-17 were obtained by coupling
commercially available aryl isocyanates and 3,3-
diphenyl-propylamine in dichloromethane at room
temperature. Alternatively, isocyanate 1, which was
prepared from 3,3-diphenyl-propylamine, was coupled
with different aromatic amines to provide the desired

ureas (Scheme 1).
the hCaSR calcium responses. The values indicated in
this paper correspond to an EC50 at 2 mM of calcium.
The most active compounds were then tested in vivo for
their ability to decrease PTH levels in normal rats. Our
two starting points, R-568 and Fendiline, were active at
80 and 1000 nM respectively, and led to compound 46,
active at 60 nM. Cinacalcet was found at 80 nM in this
assay.
N-Substituted ureas 20a-20i were prepared in two steps
from 3,3-diphenyl-propylamine (Scheme 2). This amine
was first alkylated with various alkyl halides under
basic conditions (K2CO3).10,11 The secondary amines
19a-19i were then converted to ureas 20a-20i by
addition of commercially available m-substituted methyl
ester phenyl isocyanate in dichloromethane, or
converted to ureas 28-53 with the corresponding
isocyanate.11
Disubstituted ureas 2-17 were examined for
calcimimetic activity using the luciferase assay with the
results shown in Table 1. Replacement of the -methyl
arylalkyl amino group of R-568 by aryl and benzyl-
substituted ureas was first explored, leading to
compounds with activity ranging from 2-15 M. Meta-
substituted methyl ester 7 was found to be the most
active, but at 2 µM, 85-fold less active than R-568
Esters 22-27 were prepared either by alkylation of
carboxylic acid 21 (obtained by saponification of ester
20e) with various alkyl halides under basic conditions
(K2CO3), or by activation of acid 21 with DCC and
DMAP and addition of various alcohols in
dichloromethane (Scheme 3).
(EC50
= 80 nM). Interestingly, meta-substituted
carboxylic acid 14 was found significantly less active,
indicating that the nature of the meta-substituent is
important for the resulting potency. Benzylic derivatives
15-17 were also prepared and exhibited some positive
CaSR allosteric modulation, but were less active than
compound 7.
a
2
H
H
R
NH2
N
N
n
O
1
R
2-17
n = 0, 1
b
c
N
2
H
H
R
O
N
N
n
O
1
R
1
Table 1. Calcimimetic activity of disubstituted ureas 2-17 a
Scheme 1. Reagents and conditions: (a) Ar(CH)nR1-N=C=O, CH2Cl2, rt,
60-85%; (b) Diphosgene, charcoal, toluene, 120°C; (c) Aryl-NH2/
BenzylNH2 , CH2Cl2, rt, 48% (for 2 steps).
Compound
n
R1
R2
EC50, M
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
m-OMe
p-OMe
o-Cl
p-Cl
m-Cl
m-COOMe
p-COOEt
p-Me
p-Ph
m-Ph
8
15
6
4
4
2
3
4
4
15
2
3
20
6
15
4
1
1
R
R
H
N
N
NH
NH2
b
a
O
2
R
19a-i
20a-i, R1 = see Table 2
R2 = COOMe
28-53 R1 = CH2-morpholine,
R2 = see Table 4
m-COOMe, p-Cl
m-CF3, p-Cl
m-COOH
-H
Scheme 2. Reagents and conditions: (a) R1-CH2-Br, K2CO3, CH3CN,
reflux 60-70%; (b) R2-Ph-N=C=O, CH2Cl2, rt, 75-85%.
(R)-Me
H
H
p-OMe
o-Cl
O
N
O
N
a EC50 values in µM (n ≥ 1) have been determined by Luciferase assay.
H
H
a
N
N
Further substitution of the urea led to the observation
that blocking the aniline NH resulted in significantly
reduced activity (compound 18, EC50 = 10 µM, Figure
2). Substitution of the other NH gave rise to the
compounds shown in Table 2.
N
N
O
O
O
O
RO
O
21, R = H
22-27, R = substituted-alkyl
20e
b or c
Scheme 3. Reagents and conditions: (a) NaOH, MeOH, rt, 92%; (b)
R-X, K2CO3, DMF, reflux, 70-98%; (c) ROH, DCC, DMAP, CH2Cl2, rt,
75-80%.
The compounds were evaluated in CHO cells
transfected with the hCaSR and a 6xTRE luciferase
reporter system.12 Compounds were tested in dose
response, with increasing calcium concentration. The
increasing concentration of
a positive allosteric
modulator induces a dose proportional leftward shift of

O
N
active (20 fold-less active in vitro on the CaSR (EC50
=
2 µM)). This was confirmed by in vivo stability
measurements of the ester 20e after oral administration
in rat, indicating that only the corresponding acid could
be detected after 30 minutes in plasma. In order to
address this issue, other esters (22-27) were prepared
from carboxylic acid 21 (Scheme 3) and the in vitro rat
plasma stability was measured after 30 and 60 minutes
(Table 3). Most of the esters were in the same range of
activity as methyl ester 20e. However, tert-butyl ester
25 and isopropyl ester 22 were much more stable in
plasma after 1 h (50 and 95% remaining, respectively)
and thus proved to be the best compromise between
potency and ester stability. These two esters were then
evaluated in vivo in normal rats but did not result in
better decrease in PTH levels, likely due to too low
exposure after oral administration.
Another approach was then investigated to increase in
vivo potency of compound 20e. The methyl ester of 20e
was replaced by various known ester bioisosteres14 and
other substituents. The activities of these compounds
listed in Table 4 ranged from 60 nM to 1 µM. N-(Me)-
substituted tetrazole was an especially potent surrogate
of the methyl ester. Compound 52, 2-N-(Me)
substituted-tetrazole (EC50 = 60 nM) was 2-fold more
active than ester 20e. The corresponding NH-tetrazole
51 was only weakly active (EC50 = 1 µM) comparable
to the carboxylic acid 21. Ethyl ketone 28 was also a
good surrogate of the methyl ester (EC50 = 150 nM).
Another potent compound also identified was the
oxazole 46 (EC50= 60 nM). Also of interest were meta
substituted-phenyls. Substitution with 5-membered
rings (furans 44, 45 and oxazole 46) led to compounds
H
N
N
O
O
O
18
Figure 2. Substitution on the aniline NH.
Compound 20a, with a nitrogen bearing an isobutyl
group, was as active as the non-substituted urea 7 (EC50
= 1.5 µM), indicating that it is possible to modulate the
physicochemical properties of the series by modifying
the nitrogen substituent on the urea. Taking into account
the low solubility and high lipophilicity of compound
20a, solubilising groups were then introduced at this
position. Among them, the morpholinoethyl group not
only improved solubility in acidic media but also
increased potency up to 100 nM (compound 20e), at a
level comparable to R-568 (80 nM) in our assay. The
fact that the corresponding cyclohexylethyl compound
20d is 100-fold less active indicated that additional H
bonding with morpholine was critical for improved
potency.
We explored the side chain length on the nitrogen as
well as other substitutions. We identified morpholine
20f and piperazines 20g and 20h which all have a 3
carbon atom linker as potent compounds.
R1
H
N
N
O
O
O
Table 2. Calcimimetic activity of trisubstituted urea 20a-i a
with potency similar to ester 20e.
Compound
R1
-iPr
-Ph
EC50, M
O
20a
20b
20c
20d
1.5
2
0.8
10
N
H
-COOtBu
N
N
O
20e
20f
0.1
RO
O
Table 3. Potency and plasma stability of esters 20e-27 a
O
N
Compound
R
EC50
M
Plasma
Plasma
0.15
stability b
stabilityc
O
N
30 min
60 min
1
20g
20h
0.06
0.1
20e
21
22
23
24
25d
26
27
-Me
-H
0.1
2
0.15
0.2
1.0
0.3
0.15
0.4
18
nd
75
5
80
95
nd
nd
NH
N
nd
50
1
80
95
nd
nd
-iPr
-CH2Ph
N
N
-CH(CH3)C(CH3)3
-tBu
20i
0.4
O
O
N
N
-(CH2)2OH
-(CH2)3CF2CF3
a EC50 values in µM (n ≥ 1) have been determined by Luciferase assay.
a EC50 values in µM (n = 2) have been determined by Luciferase assay.
Percentage of compound remaining in rat plasma (in vitro) after 30
b
minutes.
Compound 20e was evaluated in vivo (30 mg/kg) for its
ability to decrease PTH levels in normal Sprague
Dawley rats following oral administration.13 20e
lowered PTH levels (-60%) 30 minutes post dose, but
became weakly active after 2 hours. We postulated that
the weak effect was related to rapid hydrolysis of the
ester group to the corresponding acid 21, which was less
c
Percentage of compound remaining in rat plasma (in vitro) after 60
minutes. Compound 25 was obtained by addition of triphosgene and
diisopropylethylamine to commercially available 3-aminobenzoic acid
tertbutyl ester in CH2Cl2 followed by addition of amine 19e.
d

O
N
were pleased to see that compound 20e, (100 nM),
substituted with a methyl ester on the aniline ring, was
also efficacious in vivo and was able to decrease PTH
levels up to 60% 10 minutes post dose. An SAR
optimization around the meta substitution on the aniline
ring, to replace the labile ester group, was performed to
improve in vivo efficacy. In particular, replacing the
ester group by bioisosteres led to oxazole 46, which
showed good in vitro potency and in vivo efficacy after
oral administration.
These data paved the way for the identification of other
compounds around the trisubstitued urea series with
comparable in vitro/in vivo potency and improved
ADMET properties. These will be described in the
following paper.16
H
N
N
O
R2
Table 4. Calcimimetic activity of analogues 28-53 a
Compound
R2
-COCH2CH3
-COMe
-CH2OH
-CH2OMe
-SMe
-SCF3
-SOMe
-SO2Me
EC50, µM
0.15
0.55
0.8
28
29
30
31
32
33
34
35
0.4
0.13
0.35
0.8
1
0.4
36
37
-SO2CF3
-OCF3
0.2
38
-CN
0.8
39
40
41
-CF3
-OMe
-NMe2
0.4
0.3
0.3
O
N
O
N
O
N
42
43
44
45
46
47
48
49
50
51
52
53
- phenyl
0.2
0.2
H
H
H
N
N
N
N
N
N
-3-pyridinyl
-2-Furanyl
-1-Furanyl
-5-oxazolyl
-5-(3-Me)-isoxazolyl
-2-thiazolyl
-2-NH-benzimidazolyl
-2-benzothiazolyl
-5-NH-tetrazolyl
-5-(2-NMe)- tetrazolyl
-5-(1-NMe)- tetrazolyl
O
O
O
0.15
0.1
0.06
0.15
0.3
0.6
0.15
1
O
O
N
N
N
N
N
46
28
52
70
60
50
40
30
20
10
0
0.06
0.2
a EC50 values in nM (n ≥ 2) have been determined by Luciferase assay.
Three potent compounds tetrazole 52, oxazole 46 and
ketone 28 were evaluated in vivo in normal rats at 30
mg/kg po and PTH levels measured over time.13 As
anticipated, these three compounds exhibited much
longer duration of action in vivo than the methyl ester
analogue 20e (Figure 3). Oxazole 46 decreased PTH
levels almost completely (-95 %) for at least 2 hours
post dose (30mg/kg given orally).15
**
*
*
**
***
***
**
**
***
0 10 20 30 40 50 60 70 80 90 100 110 120
Time (min)
Based on its promising in vitro and in vivo profile,
further characterization of oxazole 46 was carried out.
Overall, oxazole 46 had an acceptable safety (Ames)
and selectivity profile (Cerep screen, <50% inhibition at
Figure 3. In vivo serum PTH 10, 30 and 120 minutes after oral
administration of compounds (30 mg/kg) in male Sprague-Dawley rats.
(Control vehicle (♦), cpd 46 (■), cpd 28 (▲), cpd 52 (●)).
Values are means of two experiments (n = 7-15). *p<0.05, **p<0.01,
***p<0.0001 significantly different from ‘control vehicle’ group (Anova,
t-test)
10 µM on
a
panel of 30 receptors/ions
channels/enzymes) but resulted in hERG inhibition (-
73% at 1 µM in the patch clamp assay) which prevented
its further development.
Acknowledgments
The authors thank Séverine Hebbe from Galapagos and
Paul Harrington from Amgen for careful review of the
manuscript.
In conclusion, a novel series of potent trisubstituted urea
derivatives, as positive allosteric modulators of the
calcium sensing receptor, exemplified by compound 46,
was developed. Starting from known Fendiline and R-
568, we discovered that we were able to replace the
alkylamine with an aryl urea leading to compounds
having good in vitro potency. Substitution of the NH
urea by the morpholinoethyl group not only improved
the physicochemical properties but resulted in
compounds with good in vitro potency. In particular, we
References and Notes
1. Nemeth, E.F. Principles of Bone Biology, 2nd ed.;
Bilezikian, J. P.; Raisz, L. G.; Rodan, G. A. Eds.;
Academic Press: San Diego, 2002, 1339.
2. Wuthrich, R. P.; Martin, D.; Bilezikian, J. P. Eur. J.
Clin. Invest., 2007, 37 (12), 915.

3. Brown, E. M.; Gamba, G.; Riccardi, D.; Lombardi, M.;
Butters, R.; Kifor, O.; Sun, A.; Hediger, M. A.; Lytton, J.;
Herbert, S. C. Nature, 1993, 366, 575.
4. Sorbera, L.A.; Castaner, R. M.; Bayes, M. Drugs Future
2002, 27 (9), 831.
5. Cohen, J. H.; Combs, D. W.; Rybczynski, P. J. U.S.
Patent 6,172,091, 2001.
6. Kessler, A.; Faure, H.; Petrel, C.; Ruat, M.; Dodd, R. H.
Bioorg. Med. Chem. Lett., 2000, 10, 2001.
at a volume of 5 ml/kg. Compounds were formulated in
methylcellulose 0.5%. Vehicle-treated controls rats
received the same volume of vehicle. Rats were sacrificed
by decapitation at various times after dosing (0, 10, 30,
120 minutes). Blood samples were collected in Sarstedt Z
tubes, allowed to clot for 20 minutes and centrifuged (3000
rpm, Jouan CR422) at 4°C. Serums were removed and
stored at -20°C until assayed. Serum PTH was quantified
according to the provider’s instructions using a rat IRMA
kit (Immutopics, 50-2000).
7. Miyazaki, H.; Tsubakimoto, J.; Yasuda, K.; Takamuro,
I.; Sakurai, O.; Yanagida, T.; Hisada, Y. W.O. Patent
115,975, 2005.
14. Wermuth, C. G. The Practice of Medicinal Chemistry
1996, Chapter 13.
8. Nemeth, E.F.; Van Wagenen, B.C.; Balandrin, M.F.;
Delmar, E.G.; Moe, S.T. U.S. Patent 6,031,003, 2002.
9. Patchett, A. A.; Nargund, R. P. Annu. Rep. Med. Chem.
2000, 35, 289.
10. Deprez, P.; Patek, M. W.O. Patent 059,102, 2002.
11. Deprez, P.; Jary, H.; Temal, T. W.O. Patent 060,026,
2007.
15. The effect starts to be significant at 10 minutes post
dose, even with a drop in the control group at 10
minutes possibly due to the effect of stress on PTH level
after oral gavage. At 30 minutes and 2 hours, the
vehicle group is back to normal range and PTH level
remains very low in the treated group with compound
46.
12. The human parathyroid cell Ca2+ receptor cDNA was
subcloned into the mammalian expression vector PECE
(Ref. 1). The luciferase reporter was subcloned into the
mammalian expression vector pGL3basic (Promega).
Resistance to neomycin (pSV2-neo) and resistance to
puromycin (pSG5-puro) were used as selection markers.
All these plasmids were simultaneously transfected into
CHO cells by calcium phosphate precipitation. Transfected
cells were grown in F12 medium containing 7.5% foetal
bovine serum, 100 U/ml penicillin and 100 µg/ml
streptomycin (as 1% Pen-Strep, BioWithaker), neomycin
(750 µg/ml) and puromycin (5 µg/ml). Neomycin and
puromycin resistant colonies were subcloned and assayed
for activation against a range of calcium concentration.
Clone 8-5-5 was used to assess the effects of compounds
on [Ca2+]i. This stably transfected cell line is termed ET8-
5-5.
16. Deprez, P.; Temal, T.; Jary, H.; Auberval, M.; Lively,
S.; Guédin, D.; Vevert, J.-P. Bioorg. Med. Chem. Lett.
2013, 23, submitted.
For measurements of [Ca2+]i, the cells were recovered
from tissue culture flasks by brief treatment with Trypsin-
EDTA (Invitrogen; containing 0.53 mM EDTA•4Na in
HBSS) and then seeded in Culture treated 96-well plates
(Greiner) at 50K cells per well in the growth media (same
as above, except neomycin 400 µg/ml). Cells were grown
in 37°C TC incubator for 24 hours. The culture medium
was then removed and replaced with F12 medium, 1%
Pen-Strep for an overnight foetal bovine serum starvation
in 37°C TC incubator. Then the starvation medium was
removed and replaced with a test buffer (20 mM HEPES
pH 7.4, 125 mM NaCl, 5 mM KCl, 1 mM MgCl2, 5.5 mM
Glucose,
supplemented with
2
g/L lysosyme and 0.3 mM CaCl2)
range of test compound
a
concentrations crossed against a superadded range of
CaCl2 concentrations. The cells were incubated with the
test compounds for 5 hours in 37°C TC incubator. Then
the test buffer was discarded, and cells were added with
the substrate for Luciferase from SteadyLite Kit (Perkin-
Elmer).The luminescence was recorded.
13. in vivo mesurements : Overnight- fasted male Sprague-
Dawley rats (200-220g BW, CERJ, France) were orally
dosed by gavage with calcimimetics (30 mg/kg) or vehicle

O
N
O
N
H
H
H
N
N
H
N
N
N
N
O
O
O
O
O
O
O
2
N
20e
46
CaSR EC50 = 8 µM
CaSR EC50 = 0.1 µM
low oral activity in rats
CaSR EC50 = 0.06 µM
potent oral activity in rats