SAR studies of 1,5-diarylpyrazole-based CCK1 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2007.09.093

A high throughput screening campaign revealed compound 1 as a potent antagonist of the human CCK₁ receptor. Here, we report the syntheses and SAR studies of 1,5-diarylpyrazole analogs with various structural modifications of the alkane side cha

Full text of DOI:10.1016/j.bmcl.2007.09.093

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 17 (2007) 6493–6498
SAR studies of 1,5-diarylpyrazole-based CCK1 receptor antagonists
Laurent Gomez,* Michael D. Hack, Kelly McClure, Clark Sehon, Liming Huang,
Magda Morton, Lina Li, Terrance D. Barrett, Nigel Shankley and J. Guy Breitenbucher
Johnson & Johnson Pharmaceutical Research and Development L.L.C., 3210 Merryfield Row, San Diego, CA 92121, USA
Received 9 May 2007; revised 8 September 2007; accepted 27 September 2007
Available online 1 October 2007
Abstract—A high throughput screening campaign revealed compound 1 as a potent antagonist of the human CCK1 receptor. Here,
we report the syntheses and SAR studies of 1,5-diarylpyrazole analogs with various structural modifications of the alkane side chain
of the molecule. The difference in affinity between the two enantiomers for the CCK1 receptor and the flexible nature of the linker led
to the design of constrained analogs with increased potency.
Ó 2007 Elsevier Ltd. All rights reserved.
Cholecystokinin (CCK) was originally identified in the
gastrointestinal tract where it was shown to mediate
contraction of the gallbladder.1 Subsequent purification
and peptide sequencing of a cholecystokinin containing
extract from the hog intestinal mucosa2 revealed that
CCK is a 33 amino-acid peptide identical to the hor-
mone implicated in pancreatic enzyme secretion.3 The
physiological effects of CCK are mediated by two
G-protein coupled receptors, the CCK1 (formerly,
CCK-A) and CCK2 (formerly, gastrin/CCK-B) recep-
tors, for which it expresses similar affinity and potency.
CCK has several activities in addition to the two that led
to its discovery. It is reported to cause satiety, inhibit
gastric emptying and acid secretion by indirect mecha-
nisms, and appears to play a major role in peristalsis.
Various selective CCK1 receptor antagonists have been
examined in the clinic for potential application in irrita-
ble bowel syndrome (IBS), non-ulcer dyspepsia, biliary
colic, chronic constipation, and pancreatic cancer.4
MeO
A
B
N
N
C
O
OH
1
Me
Figure 1. Lead CCK1 antagonist identified from HTS.
uate various structural modifications of the linker por-
tion of the molecule.
Initial investigations were aimed at studying the influ-
ence of the substitution around the C-ring on activity
at the CCK1 receptor. It is important to note that we
had previously quantitatively determined additive rela-
tionships between both the A- and C-ring and B- and
C-ring. Therefore, a one-dimensional optimization of
the C-ring substitution was undertaken knowing that
the results could be applied to a combinatorial array
of A- and B-rings. A solution-phase synthesis was
adopted using the procedure described by Murray
et al. (Scheme 1).6 This approach has the advantage of
introducing diversity at the last stage of the synthesis
allowing for rapid access to various analogs. Condensa-
tion of substituted acetophenone 2 with diethyl oxalate
under basic conditions afforded the desired b-ketoester
as the lithium salt 3 in quantitative yield. The pyrazole
ring formation was accomplished using 4-meth-
oxyphenylhydrazine to yield the 1,5-diarylpyrazole in
good yield as a 10:1 mixture of regioisomers in favor
of compound 4. The resulting ester was then reduced
in the presence of DIBAL-H to afford the desired
Recently, we reported the results of a high throughput
screening campaign leading to the identification of com-
pound 1 as a novel pyrazole-based CCK1 receptor
antagonist (Fig. 1).5 Previous communications from
our laboratories disclosed the concomitant influence of
the substitution around three aromatic rings on activity
at the CCK1 receptor.5 In this paper, we describe the
syntheses and SAR studies that were undertaken to eval-
Keywords: CCK1; Cholecystokinin; Pyrazole; Alkene analogs.
*
Corresponding author. Tel.: +1 858 320 3383; fax: +1 858 450
2089; e-mail: lgomez4@prdus.jnj.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.09.093

6494
L. Gomez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6493–6498
MeO
After studying the effect of C-ring substitutions, efforts
were made toward evaluating the role of the pendant
carboxylic acid. A series of hydrogen-bond donors and
acceptors were investigated. The synthetic route used
to access analogs represented by 8 involved activation
of the carboxylic acid followed by amide-bond forma-
tion (Scheme 2). Alternatively, compounds 10 and 11
were synthesized by alkylation of the pyrazole bromide
9 with the desired nucleophiles (Scheme 3). A subse-
quent reduction of compound 10 using DIBAL-H affor-
ded the corresponding primary alcohol 13 in good yield.
Bromination of compound 13 followed by cyanation
using sodium cyanide gave the desired product in 65%
yield. Hydrolysis of the corresponding nitrile under
acidic conditions afforded compound 14 in 82% yield.
The binding data (pKi) are plotted in Figure 3.
O
OLi O
O
N
OEt
N
b
a
Me
R1
R1
OEt
O
R1
2
3
4
MeO
MeO
e, f
c, d
R2
N
N
N
N
Br
R1
R1
6
O
OH
5
Scheme 1. Reagents and conditions: (a) diethyl oxalate, LiHMDS,
ether, ꢀ78 °C to rt; (b) 4-methoxyphenylhydrazine, THF, TsOH,
40 °C; (c) DIBAL, THF, ꢀ78 °C to rt; (d) CBr4, PPh3, DCM, 0 °C to
rt; (e) NaH, phenyl acetates, DMF, 0 °C to rt; (f) LiOH, THF, MeOH,
H2O, 50 °C.
Analysis of the data suggests that highly ionized groups,
such as carboxylic acids or tetrazoles, are well tolerated
and lead to compounds with pKi values ranging from 7.6
to 8.1. However, incorporation of less acidic functional-
ities such as primary alcohols results in analogs with
reduced affinity. Primary, secondary or bulky tertiary
amides are not tolerated. Interestingly, a small tertiary
amide (hydrogen-bond acceptor only) displayed similar
activity as the carboxylic acid. However, the presence
of the carboxylic acid functionality greatly improved
the solubility properties of the resulting analogs and
was therefore selected for further studies.
alcohol in quantitative yield. A subsequent bromination
of the primary alcohol using CBr4/PPh3 yielded the key
intermediate 5 in excellent yield. The final steps of the
sequence were run in parallel and consisted in alkylating
compound 5 using NaH and various phenyl acetic acid
esters followed by hydrolysis of the resulting esters.
The final compounds were purified by automated
reverse-phase preparative HPLC.
The binding data (pKi) of a subset of compounds are
shown in graphical form in Figure 2. Analysis of this
data set shows similar trends across the two series sug-
gesting an additive relationship between both the
B- and C-rings. This investigation shows that meta-
substituted phenyl rings consistently display high affinity
for the CCK1 receptor. Altering the positioning of the
substituent around the ring to an ortho or para position
leads to a 10-fold decrease in activity. This result was
further supported by the low affinity observed with the
naphthyl substitution.
The binding data reported thus far have been for the
racemic mixture. It became important to evaluate the
sensitivity of the CCK1 receptor to the absolute stereo-
MeO
MeO
Me
Me
N
N
N
N
a
O
O
1
2
OH
NR R
7
8
Me
Me
Scheme 2. Reagents and condition:(a) NHR1R2, EDC, HOBT, DMF, rt.
MeO
R2
N
N
MeO
Me
MeO
Me
a
c
R1
O
N
Me
N
N
N
OH
Br
R
9
8
R = COOEt
10
N
b
N
11 R = CN
12 R =
N
7.5
N
H
pKI
MeO
MeO
d-g
7
Me
Me
N
N
N
N
6.5
13
14
OH
COOH
Me
Me
6
R1
Scheme 3. Reagents and conditions: (a) NaH, ArCH2CO2Et or
ArCH2CN, DMF, 0 °C to rt; (b) NaN3, NH4Cl, DMF, 100 °C; (c)
DIBAL, THF, ꢀ78 °C to rt; (d) PBr3, DCM, 0 °C to rt; (e) NaCN,
DMF, 100 °C; (f) H2SO4, MeOH, reflux; (g) LiOH, THF, MeOH,
H2O, 50 °C.
4-Cl
4-Cl
4-Me
3-Cl
2-OMe
3-OMe
4-OMe
R2
3-Me
4-Me
H
2-napht
Figure 2. CCK1 binding data from C-ring substitutions.

L. Gomez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6493–6498
6495
74% yield (Scheme 5). Alternatively, the R-enantiomer
could be obtained using a similar route starting from
R-(3a-cis)-(+)-3,3a,8,8a-Tetrahydro-2H-indeno[1,2-d]-
oxazol-2-one.
MeO
Me
Me
N
N
R
Alternatively, the pure enantiomers could be isolated by
chiral chromatography using a racemic synthesis as
shown in Scheme 6. Alkylation of phenyl acetates with
propargyl bromide under basic conditions followed by
a Sonogashira coupling reaction under standard condi-
tions afforded the desired ynones, 20, in 40–60% overall
yields. A subsequent pyrazole ring formation was
accomplished using substituted phenyl hydrazines. It
was found that optimal dehydration of the cyclized
intermediate required the addition of p-TsOH to afford
the desired 1,5-substituted pyrazoles 21 in 75–80% yield
along with a small amount of the 1,3-substituted regio-
isomer (10–15%, not shown). Hydrolysis of the ester fol-
lowed by chiral separation using supercritical-fluid
chromatography (SFC) gave pure enantiomers 22 and
23 in 80% yield. It should be noted that the absolute
configuration of the products was assigned by compar-
ing the analytical data with products isolated from the
enantioselective synthesis discussed previously.
8.5
8
7.5
7
pKI
6.5
6
5.5
O
N
O
O
OH
O
OH
NMe2
OH
O
NH2
NMe
O
O
O
CN
N
N
N
NHOH
NHMe
N
H
N
H
OH
R
Figure 3. CCK1 binding data of carboxylic acid replacements.
chemistry at the chiral center. To accomplish this inves-
tigation, an initial enantioselective synthesis of pyrazole
was developed using the Evans asymmetric alkylation
procedure as shown in Scheme 4.7 Iodination of com-
pound 16 was accomplished in two steps, which
involved activation of the primary alcohol using MsCl
followed by a Finkelstein reaction to provide the pyra-
zole iodide in 97% overall yield. The final bond connec-
tion of the sequence involved alkylation of 15 with the
pyrazole iodide using NaHMDS as a base which affor-
ded compound 18 in 83% yield. The optimal conditions
for the release of the chiral auxiliary required the use of
hydrogen peroxide under basic conditions. Under these
conditions, the desired S-enantiomer 19 was isolated in
82% yield (99% ee) with modest recovery of the chiral
auxiliary (66%, unoptimized). The preparation of
compound 15 involved alkylation of S-(3a-cis)-(-)-
3,3a,8,8a-Tetrahydro-2H-indeno[1,2-d]-oxazol-2-one with
m-tolyl-acetic acid using Mukaiyama’s reagent under
basic conditions to afford the desired product 15 in
The binding data of selected compounds are summa-
rized in Figure 4. The data suggest a strong preference
for the S-enantiomer which consistently displays about
1.5 log unit increase in binding affinity at the CCK1
receptor over the R-isomer.
Devazepide (pKi = 8.74)8 and Lintitript (pIC50 = 9.25)
are two well-known CCK1 antagonists. There are visual
similarities between these two antagonists, in particular,
the indole amide moiety and pendant phenyl ring. There
are also similarities between our compounds and these
O
O
O
O
O
OH
Me
HN(S)
H
O
a
N(S)
H
+
H
(R)
(R)
H
Me
15
Scheme 5. Reagents and condition: (a) 2-chloro-1-methylpyridium
iodide, TEA, DMAP, DCM, 0 °C to rt.
MeO
MeO
N
N
N
a, b
c
O
N
1
2
R
R
O
N
OEt
OH
I
a, b
c, d, e
N
O
OEt
3
16
17
2
R
OH
Cl
Cl
R
3
Cl
Cl
O
R
3
20
21
R
MeO
MeO
1
1
R
R
O O
N
f
H
N
N
N
O
N
N
N
H
O
O
d
(S)
H
(R)
H
N
(S)
O
+
N
(S)
(S)
(S)
H
OH
2
2
OH
OH
R
R
H
(R)
Me
3
3
22
23
R
Me
Cl
R
Cl
Cl
18
Cl
19
Scheme 6. Reagents and conditions: (a) LDA, propargyl bromide,
ꢀ78 °C to rt; (b) ArCOCl, N-methylmorpholine, PdCl2(PPh3)2, CuI,
THF/toluene, rt; (c) arylhydrazine, THF, rt; (d) p-TsOH, THF, rt; (e)
LiOH, THF/MeOH/H2O, 35 °C; (f) SFC, AD column.
Scheme 4. Reagents and conditions: (a) CH3SO2Cl, TEA, THF, rt; (b)
NaI, acetone, reflux; (c) 15, NaHMDS, THF, ꢀ78 °C to rt; (d) 30%
H2O2, LiOH, H2O, 0 °C to rt.

6496
L. Gomez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6493–6498
increasing the potency of our CCK1 receptor antago-
MeO
N
nists? An alignment of such a constrained analog with
Devazepide and Lintitript is shown in Figure 6. It can
be seen that the constrained linker does not visually
disrupt the alignment between the molecules. The syn-
thetic route to access this series is shown in Scheme 7.
N
N
N
N
O
O
2
OH
OH
R
3
27
R
Me
2
3
24, R = 3,4-(OCH O) and R = 2-napht
2
2
3
25, R = 4-Me and R = 2-napht
Oxidation of 1,5-diarylpyrazole alcohols using Dess–
Martin periodinane followed by a Perkin condensation
of the resulting aldehydes with phenyl acetic acids affor-
ded the desired alkenes 29 as a single E-isomer in 85–
90% yields. It was found that upon exposure to intense
focused light in a quartz tube, this material equilibrated
to a thermodynamic mixture of E- and Z-isomers in 1:1
ratio after 12 h. Both isomers were easily separated by
column chromatography to afford 29 and 30 in 70%
combined yields. The geometry of the olefin was as-
signed using 2D NMR spectroscopy and nOe studies.
Also, it should be noted that photochemical stability
2
3
26, R = 4-OMe and R = 3-OMe
8.5
8
7.5
pKI
7
6.5
rac.
6
S
19
24
25
R
26
27
Figure 4. CCK1 binding data: influence of the chiral center.
N
H
N
N
N
N
antagonists, especially with Devazepide, which appears
to array its aromatic rings in a similar way. A three-
dimensional molecular alignment of Devazepide, Linti-
tript, and one of our antagonists is shown in Figure 5.9
(S)
N
NH
OO
O
S
O
O
N
OH
Cl
N
H
Me
OH
Pyrazole alkene
Lintitript
Devazepide
The figure shows good overlap between the three lipo-
philic aromatic rings as well as the carbonyl groups of
Devazepide and Lintitript and the acid moiety of our
pyrazole alkane, suggesting a similar binding mode for
all three classes of compounds. More importantly, it
appears that the flexible linker of the pyrazole overlays
with the more rigid amide functionality of both litera-
ture CCK1 antagonists. These observations led us to
propose another question regarding the geometry of
the linker: can we incorporate the rigidity found in
Devazepide and Lintitript into our side chain and per-
haps lock it into an active conformation, thereby
Figure 6. Molecular alignment of a pyrazole alkene (yellow) with
known CCK1 antagonists. Devazepide (blue) and Lintitript (magenta).
1
1
1
R
R
2
R
OH
O
N
N
a
b
N
H
N
(E)
N
N
N
N
O
OH
OH
O
N
N
N
(S)
(S)
N
NH
OO
2
2
H
S
R
O
O
R
R
N
Cl
N
H
3
28
29
R
Me
OH
Pyrazole alkane
Lintitript
Devazepide
1
1
R
R
3
c
N
R
N
(Z)
N
+
N
(E)
OH
2
2
R
O
O
R
OH
3
30
29
R
R
1
N
d
N
O
2
OH
R
3
R
31
Scheme 7. Reagents and conditions: (a) Dess–Martin periodinane,
DCM, rt; (b) ArCH2COOH, NEt3, Ac2O; (c) uv light, CHCl3, rt; (d) p-
TosNHNH2, NaOAc, EtOH, 100 °C.
Figure 5. Molecular alignment of a pyrazole alkane (yellow) with
known CCK1 antagonists Devazepide (blue) and Lintitript (magenta).

L. Gomez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6493–6498
6497
studies were conducted on both the E- and Z-isomers. It
1
was found that the compounds were extremely stable
when stored as dry powder and subjected to intense fo-
cused light for several days. However, when stored in
solution (0.01 and 1 mg/mL at pH 11) the compounds
were moderately stable with approximately 80% recov-
ery after 3 days. A subsequent reduction of the alkene
functionality could be performed using p-tosyl hydra-
zine under basic conditions to obtain the corresponding
alkane 31 in 75% yield.
R
N
N
OH
O
2
R
Cl
8.4
8.2
8
7.8
7.6
7.4
7.2
7
A direct comparison of the binding affinity for both the
E- and Z-isomers clearly indicates that the relative ori-
entation of the carboxylic acid and the aromatic group
with respect to the pyrazole ring is crucial for good
interaction with key residues in the active site, consistent
with the alignment illustrated in Figure 6. Regardless of
substitutions, analogs bearing the Z-configured double
bond consistently display greater activity than their
E-isomer counterparts (Fig. 7).10 More importantly,
when compared to the fully saturated linker (racemic al-
kane), constrained side-chain analogs with Z-configura-
tion display greater activity as shown in Figure 8. Even
considering the small difference in activity between the
racemate and the pure S-enantiomer (which does not ex-
ceed 0.3 log unit), the improved activity is significant
and consistent across a set of five compounds. The
amplitude of the gain in activity (0.3–0.6 log unit) sug-
gests that reducing the overall entropy of the ligands
by constraining the flexible linker did have the expected
effect. These results are consistent with calculations
conducted on the reduction of entropy of a single
carbon–carbon bond. The gain was estimated to be
about 0.4 kcal/mol, which corresponds to approximately
0.3 log unit of activity at 298 K.11 It appears that the
increase in activity is mainly a reflection of the gain in
entropy, which suggests that both the alkane and alkene
analogs adopt a similar binding mode.
pKI
Z -isomer
alkane
R1
R2
4-OEt
2,5-diCl
2-napht
4-OMe
2,5-diCl
3,4-diCl
pyridyl
4-OEt
2-napht
3,4-diCl
3,4-diCl
3,4-diCl
Figure 8. CCK1 binding data: comparison between alkane and
Z-isomer.
the molecule with a strong preference for S-enantiomers.
Further studies showed that the flexible linker could suc-
cessfully be replaced by a more rigid alkene which
resulted in an increase affinity for the CCK1 receptor.
The chemistry developed to access both E- and Z-iso-
mers involved equilibration of E-isomers using intense
focused light. This process allowed us to rapidly access
a large numbers of analogs. In addition, rat pharmaco-
kinetic studies conducted on compound 19 show excel-
lent oral bioavailability (2 lmol/kg; %F = 100) and
good half-life (t1/2 = 5.6 0.7 h). These potent pyra-
zole-based CCK1 receptor antagonists,12 with good
pharmacokinetic properties and a good selectivity pro-
file over other targets,13 represent a novel class of com-
pounds for the potential treatment of IBS, pancreatitis
and other GI disorders.
In conclusion, investigations presented in this article
show a clear sensitivity to the chiral center present on
MeO
N
N
OH
Acknowledgments
O
Cl
3
We thank Jiejun Wu and Heather McAllister for analyt-
ical chemistry support and Michele Rizzolio for the pho-
tochemical studies conducted on the alkene analogs.
R
Cl
8.5
8.3
8.1
7.9
7.7
7.5
7.3
7.1
6.9
6.7
6.5
Supplementary data
pKI
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.09.093.
Z-isomer
E-isomer
H
4-Cl
3-Cl
4-OMe
R3
3,4-diCl
3-Me
4-Me
References and notes
Figure 7. CCK1 binding data: influence of the geometry of the double
bond.
1. Ivy, A. C.; Olderg, E. Am. J. Physiol. 1928, 86, 599.
2. Jorpes, E.; Mutt, V. Acta Physiol. Scand. 1966, 66, 196.

6498
L. Gomez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6493–6498
3. Harper, A. A.; Raper, H. S. J. Physiol. 1943, 102,
115.
Moreno, V.; Shankley, N. P. Br. J. Pharmacol. 2005,
145, 374.
4. (a) Varga, G. Curr. Opin. Investig. Drugs 2002, 3, 621; (b)
Herranz, R. Med. Res. Rev. 2003, 23, 559; (c) Varga, G.;
Balint, A.; Burghardt, B.; D’Amato, M. Br. J. Pharmacol.
2004, 141, 1275; (d) Peter, S. A. S.; D’Amato, M.;
Beglinger, C. Dig. Dis. 2006, 24, 70.
5. McClure, K.; Hack, M.; Huang, L.; Sehon, C.; Morton,
M.; Li, L.; Barrett, T. D.; Shankley, N.; Breitenbucher, J.
G. Bioorg. Med. Chem. Lett. 2006, 16, 72; Sehon, C.;
McClure, K.; Hack, M.; Morton, M.; Gomez, L.; Li, L.;
Barrett, T. D.; Shankley, N.; Breitenbucher, J. G. Bioorg.
Med. Chem. Lett. 2006, 16, 77.
6. Murray, W. V.; Hadden, S. K.; Wachter, M. P.
J. Heterocycl. Chem. 1990, 27, 1933.
7. (a) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am.
Chem. Soc. 1981, 104, 1737; (b) Ager, D. J.; Prakash, I.;
Schaad, D. R. Chem. Rev. 1996, 96, 835.
9. MOE (The Molecular Operating Environment) Version
2006.08, software available from Chemical Computing
Group Inc., 1010 Sherbrooke Street West, Suite 910,
Montreal, Canada H3A 2R7. hhttp://www.chemcomp.
comi.
10. It should be noted that 3,4-diCl substitutions on the B-ring
display comparable activities than 4-Cl and 4-Me (see
supplementary material).
11. Lazardis, T. Curr. Org. Chem. 2002, 6, 1319.
12. For in vivo and in vitro pharmacology of 1,5-diarylpyraz-
ole-based CCK1 receptor antagonists, see: Morton, M. F.;
Barrett, T.; Yan, W.; Freedman, J. M.; Lagaud, G.;
Prendergast, C. E.; Moreno, V.; Pyati, J.; Figueroa, K.;
Li, L.; Wu, X.; Rizzolio, M.; Breitenbucher, J. G.; McClure,
K.; Shankley, N. P. J. Pharmacol. Exp. Ther. 2007, in press,
doi:10.1124/jpet.107.124578.
8. Data generated using our assay conditions. For details,
see: Morton, M. F.; Pyati, J.; Dai, H.; Li, L.;
13. Biological CCK2 binding data are contained in supple-
mentary material.