Discovery of potent cholecystokinin-2 receptor antagonists: Elucidation of key pharmacophore elements by X-ray crystallographic and NMR conformational analysis

Article abstract of DOI:10.1016/j.bmc.2008.01.059

A novel series of cholecystokinin-2 receptor (CCK-2R) antagonists has been identified, as exemplified by anthranilic sulfonamide 1 (pK_i = 7.6). Pharmacokinetic and stability studies indicated that this series of compounds suffered from metabolic degradation, and that both the benzothiadiazole and piperidine rings were rapidly oxidized by liver enzymes. A combination of synthesis, computational methods, ¹H NMR conformational studies, and X-ray crystallographic analyses were applied to elucidate key pharmacophore elements, and to discover analogs with improved pharmacokinetic profiles, and high receptor binding affinity and selectivity.

Full text of DOI:10.1016/j.bmc.2008.01.059

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 3917–3925
Discovery of potent cholecystokinin-2 receptor antagonists:
Elucidation of key pharmacophore elements by X-ray
crystallographic and NMR conformational analysis
Mark D. Rosen,a,* Michael D. Hack,a Brett D. Allison,a Victor K. Phuong,a
Craig R. Woods,a Magda F. Morton,a Clodagh E. Prendergast,a Terrance D. Barrett,a
Carsten Schubert,b Lina Li,a Xiaodong Wu,a Jiejun Wu,a Jamie M. Freedman,a
Nigel P. Shankleya and Michael H. Rabinowitza
aJohnson & Johnson Pharmaceutical Research and Development, L.L.C., Drug Discover, 3210 Merryfield Row, San Diego,
CA 92121, USA
bJohnson & Johnson Pharmaceutical Research and Development, L.L.C., Structural Biology, 665 Stockton Drive, Exton,
PA 19341, USA
Received 3 October 2007; revised 15 January 2008; accepted 23 January 2008
Available online 5 February 2008
Abstract—A novel series of cholecystokinin-2 receptor (CCK-2R) antagonists has been identified, as exemplified by anthranilic sul-
fonamide 1 (pKi = 7.6). Pharmacokinetic and stability studies indicated that this series of compounds suffered from metabolic deg-
radation, and that both the benzothiadiazole and piperidine rings were rapidly oxidized by liver enzymes. A combination of
synthesis, computational methods, 1H NMR conformational studies, and X-ray crystallographic analyses were applied to elucidate
key pharmacophore elements, and to discover analogs with improved pharmacokinetic profiles, and high receptor binding affinity
and selectivity.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
largely due to poor physicochemical properties, and the
associated unfavorable and variable pharmacokinetic
behavior.
The cholecystokinin-2 receptor (CCK-2R) is present in
gastric tissue as well as the cortical region of the brain.1
Recent interest in CCK-2R antagonists has been height-
ened by the implication of gastrin, a primary endoge-
nous ligand for the CCK-2 receptor, as a potent
trophic factor for gastrointestinal adenocarcinomas
such as Barrett’s metaplasia and pancreatic tumors.2–4
The therapeutic potential of gastrin suppression has
been demonstrated recently in clinical trials of the gas-
trin vaccine GastrimmuneTM as a treatment for stomach
and pancreatic cancer.5,6 Although numerous non-pep-
tide CCK-2R antagonists have been forwarded as
potential therapeutic agents for gastrointestinal disease,
to date their promise remains unrealized.7 This has been
Recently this laboratory disclosed a series of anthra-
nilic sulfonamides that are potent and selective CCK-
2R antagonists, and which exemplify a novel chemo-
type in this area of research.8 Representatives of this
series are benzo[1,2,5]thiadiazole-4-sulfonic acid, [5-
bromo-2-(piperidine-1-carbonyl)phenyl] amide (1) and
the 5-chloro congener (2), which display nanomolar
binding affinity and high selectivity over the related
CCK-1 receptor subtype (1: CCK-2R pKi = 7.6;
CCK-1R pKi < 5; 2: CCK-2R pKi = 7.4; CCK-1R
pKi < 5; Fig. 1). Investigation of in vitro ADME pre-
dictors and in vivo pharmacokinetic properties show
that 1 and 2 have good GI absorption, although these
compounds have a high propensity for oxidative
metabolism (Table 1). This was evidenced by poor sta-
bility toward human liver microsomes (HLM), with
15% and 3.9% of 1 and 2 remaining after 30 min,
respectively.8 These data correlate to an unfavorable
Keywords: Cholecystokinin-2 receptor antagonists; Gastrin receptor
antagonists; CCK-2; CCK-B.
*
Corresponding author. Tel.: +1 858 320 3395; fax: +1 858 784
3267.; e-mail: mrosen1@prdus.jnj.com
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.01.059

3918
M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
O
O
O
>100 commercially
available RSO2Cl
N
N
N
R = aromatic,
heteroaromatic
(all pKI < 7.0)
H2N
Br
HN
R
Br
HN
S
X
O
6
O
Scheme 1. Parallel synthesis of benzothiadiazole replacements.
N
S
N
1: X = Br (pKi = 7.6)
2: X = Cl (pKi = 7.4)
tance. Parallel synthesis and screening of a large
number of analogs generated from a diverse set of com-
mercially available sulfonyl chlorides and amine 6 failed
to yield a compound with high CCK-2R binding affin-
ity, demonstrating the extreme sensitivity to substitution
in this region (Scheme 1).10 Therefore, we undertook a
more detailed examination of the physical properties
and conformational preferences of this series, in order
to help elucidate the role of the benzothiadiazole ring
and identify the narrowly defined pharmacophore ele-
ments that are required for high binding affinity.
Figure 1. Benzothiadiazole sulfonamides as CCK-2R antagonists.
Table 1. Pharmacokinetic and HLM parameters for 1, 2, and 28
Compound
Rat pharmacokinetic parameters
( SE)a
HLMb
CL (L/kg/h)
t1/2(h)
%F
1
0.70 0.06
1.25 0.26
0.83 0.27
0.36 0.05
0.26 0.03
0.16 0.03
0.34 0.11
0.62 0.05
68
8
15
3.9
21
60
2
42 17
50 16
55 13
The structure of 1 obtained by single crystal X-ray dif-
fraction clearly shows a cup-shaped conformation in
the solid state (Fig. 3A), where the carbonyl oxygen
of the piperidine amide and N3 of the benzothiadiazole
ring are engaged in a proximal relationship with the
acidic proton of the sulfonamide group, in agreement
with the known conformational preferences of anthra-
nilic amides.11–13 In order to arrive at a geometry more
representative of the solution state than the crystal
17
28
a Calculated from at least three animals. Clearance values (CL) and
half-lives (t1/2) measured from iv doses of 2 lmol/kg. Relative frac-
tion absorbed (%F) measured from 2 lmol/kg po doses.
b Percent remaining after 30 min in the presence of pooled human liver
microsome preparations (n = 20).
structure,
1
was energy minimized (B3LYP/6-
31++G(2d,2p)) beginning with the solid state confor-
mation. The resulting coordinates are very similar to
the crystal structure, as shown in Figure 3B. Although
NMR studies of 1 were complicated by rotameric
broadening, NOESY experiments conducted on the re-
lated benzylamide 7 demonstrated that the conforma-
tion placing the carbonyl oxygen atom of the amide
in a proximal position to the sulfonamide NH is highly
populated in solution (i.e., Ha–Hb NOE, Fig. 4), in
agreement with the solid state findings.11–13 In addi-
tion, a small reciprocal NOE enhancement observed
between Hc and Hd suggests that the phenyl portion
of the benzothiadiazole ring is preferentially neighbor-
ing the phenyl ring of the anthranilic core, with the
ring nitrogen atoms oriented toward the sulfonamide
nitrogen. This is in agreement with the solid state
structure (Fig. 3) and provides further evidence that
in vivo pharmacokinetic profile, with rapid clearance
and a half-life of only 16 and 10 min in the rat for 1
and 2. Qualitative HPLC and mass spectral analysis
of the HLM incubate of 1 suggested facile oxidation
of the sulfur atom of the benzothiadiazole ring to the
S-oxide (3), and subsequent hydration to the S,S-diox-
ide (4), in agreement with reported metabolic pathways
of this ring system (Fig. 2).9 In addition, oxidation of
the carbon atom adjacent to the nitrogen atom in the
piperidine ring was also observed (i.e., 5, see Supple-
mentary material).
2. Structural and conformational analysis
With these facts in evidence, replacement of the benzo-
thiadiazole ring became a goal of paramount impor-
O
O
OH O
N
N
N
HN
Br
HN
S
Br
HN
S
Br
O
O
O
S
O
O
O
N
S
O
NH
N
S
N
HN
N
S
O
O
3
4
5
Figure 2. Major metabolites of 1 identified by HPLC/MS analysis of HLM incubate.

M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
3919
Figure 3. (A) Solid state conformation of 1 as determined by X-ray crystallography and (B) energy minimized conformation (B3LYP/6-
31++G(2d,2p), bromide removed to simplify calculation). Distances are in angstroms.
3. Chemistry
NOE
We designed a series of heterocyclic sulfonamides as po-
Ha
N
Hb
tential replacements for the benzothiadiazole ring sys-
tem, including 4-benzoxadiazolyl, 4-benzotriazolyl, 5-
quinoxalyl, in order to test our hypotheses about the
conformational requirements for high CCK-2R binding
affinity. For compounds 12–21, aniline 6 or 8 was cou-
pled with known or commercially available sulfonyl
chlorides (Scheme 2).8
Cl
O
HN
Br
NOE
O S O Hc
Hd
N
N
S
For compounds 9–11, the required sulfonyl chlorides
22–24 were obtained by direct chlorosulfonation of the
parent heterocycle, directly followed by coupling with
amine 6 (Scheme 3). Any sulfonamide regioisomers pres-
ent were separated at this stage by chromatography or
crystallization, and the regiochemisty was established
by 1H NMR. In the case of triazole 11, the regiochemist-
ry was further established by observation of reciprocal
NOE enhancements between the N-methyl group of
the triazole and the neighboring aromatic proton. Benz-
imidazole analog 26 was prepared by reduction of ben-
zothiadiazole 1with zinc metal in acetic acid, and
treating the resulting crude phenylenediamine intermedi-
ate 25 with CH(OMe)3,14 while benzylamide 7 was ob-
tained by HATU mediated amide formation from acid
27 (Scheme 4).8
7
Figure 4. Diagnostic NOE enhancements observed during NMR
conformational analysis of 7 (600 MHz, DMSO-d6).
the solid state conformation is a reasonable representa-
tion of the solution state conformation.11 The clear
conformational preference demonstrated by these crys-
tallographic, computational, and NMR data provided
hypotheses for the pharmacophoric requirements of
high binding affinity. One possible scenario is that
the amide carbonyl group, sulfonamide NH atom,
and benzothiadiazole N3 atom may form a three-point
intramolecular interaction, which enforces a conforma-
tion that is possibly representative of the binding con-
formation. Although the observed geometry is not
completely optimal for a formal three-point hydrogen
bonding array, an energy minimized structure in which
the benzothiadiazole ring was initially rotated 180ꢁ
about the carbon–sulfur bond (and thus unable to par-
ticipate in this putative three-point interaction) was
energetically less favored by 0.8 kcal/mol (B3LYP/6-
311++G(3d,3p)//B3LYP/6-31++G(2d,2p), not shown),
which is consistent with the formation of a weak inter-
action (Fig. 3). A second possibility is that one or both
benzothiadiazole nitrogen atoms are involved in direct
receptor interactions that are critical for high binding
affinity. Although the validity of either possibility was
uncertain at the outset, in searching for benzothiadiaz-
ole replacements it seemed reasonable to use this con-
4. Results and discussion
4.1. Structure–activity relationship
The structure–activity relationship (SAR) of the tar-
geted analogs was evaluated using a human CCK-2R
radioligand competition binding assay (Table 2).16–18
Initially we examined analogs in which the [4.3.0]-fused
heterocyclic motif was preserved. Replacement of the
sulfur atom of the benzothiadiazole ring of 1 with an
N-methyl group resulted in the 2-methylbenzotriazole-
4-yl (9) and -5-yl (10) analogs with over 15- and 100-fold
loss in binding affinity, respectively. A similar loss of
activity was observed for the 1-methylbenzotriazole-4-
yl (11) substitution pattern. Strikingly, the benzoxadiaz-
ole-4-yl congener (12) displayed a 50-fold loss in binding
affinity relative to benzothiadiazole 2. This unexpected
formational guide as
a template. The following
describes the synthesis, biological evaluation, and phys-
ical properties of compounds which address both the
issue of metabolic stability and the questions raised
by the above structural analysis.

3920
M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
O
O
a
N
N
R1
HN
R1
H2N
O
O
S
R2
6: R1 = Br
8: R1 = Cl
9-21
R2 =
N
N
N
N
N
S
S
N
N
R3
MeN
O
N
MeN
N
N
Me
N
9: R1 = Br
10: R1 = Br
11: R1 = Br 12: R1 = Cl
13: R1 = Br 14: R1 = Br; R3 = H
15: R1 = Br; R3 = CH3
N
N
N
S
N
N
N
18: R1 = Cl
19: R1 = Br
20: R1 = Br
21: R1 = Br
16: R1 = Br
17: R1 = Br
Scheme 2. Reagents and condition: (a) R2SO2Cl, pyridine, CH2Cl2, 23 ꢁC, 3–16 h.
X
azole series, the 4-yl linkage (13) allowed for a pKi value
of 7.1. This result was encouraging since the CCK-2R
affinity was, for the first time, only somewhat diminished
from the parent benzothiadiazole 1 (DpKi = À0.5). Fur-
ther examination of the SAR of the benzothiazole ring
showed that transposing the nitrogen and sulfur atoms
(i.e., 7-yl, 14) led to a 10-fold reduction in affinity, while
the 6-yl regioisomer (16) displayed an additional deteri-
oration in binding affinity of similar magnitude. This
further demonstrates the importance of the proximal
relationship between the ring nitrogen atom and the sul-
fonamide linker. The introduction of additional steric
bulk in the form of a methyl group in the 2-position
of the benzothiazole (15) led to a 10-fold decrease in
receptor affinity compared to the des-methyl analog 14.
Y
N
N
a
b
N
N
MeN
MeN
22: X = SO2Cl, Y = H
23: X = H, Y = SO2Cl
4.5:1 22:23
SO2Cl
N
N
N
N
N
N
Me
Me
24
Scheme 3. Reagents and conditions: (a) ClSO3H, reflux, 20 min; (b)
ClSO3H, reflux, 6 h.
The above results demonstrate that the sulfur atom in
the benzothiadiazole and benzothiazole ring systems
plays an important role in CCK-2R binding affinity.
We infer from these data that the role may be primarily
steric rather than electronic, since sulfur and carbon
have approximately equal Pauling electronegativity val-
ues. Therefore, a key difference between high affinity
compounds such as benzothiadiazole 1, where a larger
sulfur atom bridges the two nitrogens, and lower affinity
analogs such as benzimidazole 26 with a smaller carbon
bridge, is the size of the five-membered heterocyclic ring.
To examine this, as well as positional effects of the nitro-
gen atoms, we assessed the isosteric ethylene group as a
sulfur replacement. Toward this end, we were pleased to
find that the quinoxalin-5-yl sulfonamide 17, which is
isosteric to 1, possesses high CCK-2R binding affinity,
with a pKi value of 7.2. This was a particularly encour-
aging result, as 17 was the first compound with this level
of receptor affinity that lacked the divalent sulfur atom.
We then assessed the contribution of each individual
nitrogen atom of the quinoxaline ring to the overall
receptor binding affinity. Replacement of either nitrogen
atom with a methylene group gave the related quinolin-
effect may be attributed to the lowered basicity of the
ring nitrogens due to the greater electronegativity of
the oxygen atom, and the concomitant lowering of the
ability of these atoms to participate in the putative
hydrogen bonding interactions. This is supported by
the calculated pKa values of À2.29 and À3.81 for the
conjugate acid of the ring nitrogen atoms of 2 and 12,
respectively. Furthermore, the greater atomic radius of
sulfur leads to a larger ring size, and the resulting com-
bination of bond angles and nitrogen lone-pair orienta-
tion may favor high receptor binding affinity. The effect
of ring size is explored in greater detail in the [6,6]-fused
ring analogs (vide infra).
Elimination of the central heteratom in the five-mem-
bered ring is illustrated by the benzimidazole and benzo-
thiazole
sulfonamides.
The
benzimidazole-4-yl
compound (26) possessed poor CCK-2R affinity
(pKi = 5.8), which may be attributable to the contrac-
tion of ring size due to replacement of sulfur with a car-
bon atom having a smaller atomic radius, as well as the
change in nitrogen hybridization state. In the benzothi-

M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
3921
O
O
N
N
a
b
HN
Br
HN
Br
1
O
O
S
S
O
O
NH2
N
NH2
HN
25
26
O
O
HO
N
H
Cl
HN
S
Br
HN
Br
c
O
O
S
O
O
N
S
N
S
N
N
27
7
Scheme 4. Reagents and conditions: (a) Zn, AcOH, 50 ꢁC, 20 min; (b) CH(OEt)3, TsOHÆH2O, 100 ꢁC, 15 min, 44% (two steps); (c) 4-
chlorobenzylamine, HATU, pyridine, DMF, 23 ꢁC, 67%.
Table 2. Human CCK-1 and CCK-2 receptor binding affinities
clearly demonstrate that the quinoxaline nitrogen atom
CCK-1Ra (pKi)
CCK-2Ra (pKi)
proximal to the sulfonamide group is advantageous for
receptor binding, while a distal heteroatom (cf. 13, 14,
17, 19) is absolutely required.
Compound
1
<5
<5
<5
<5
<5
<5
7.6
7.4
6.3
6.1
<5
5.7
2
7
Having discovered the quinoxaline ring system as a suit-
able replacement for the metabolically labile benzothia-
diazole ring of 1, we were free to address the piperidine
amide portion of this molecule, which had also been
identified as a site of oxidative degradation. We rea-
soned that insertion of an electron withdrawing atom
into this ring should decrease the propensity for oxida-
tion. This rationale led to morpholine analog 28
(pKi = 7.5, Fig. 5), which has a markedly improved
pharmacokinetic profile relative to 1 (Table 1), while
achieving receptor affinity slightly higher than the re-
lated chloro and bromo analogs (data not shown). The
synthesis and biological properties of this molecule have
been described elsewhere.8
9
10
11
12
13
14
15
16
17
<5
<5
<5
<5
<5
<5
5.8
7.1
6.1
5.3
5.5
7.2
18
19
20
21
26
28
<5
<5
<5
<5
<5
<5
<5
6.1
<5
<5
5.8
7.5
a pKi, negative logarithm of the antagonist equilibrium dissociation
constant calculated from the antagonist concentration required to
displace 50% of [125]I-CCK-8S binding (pIC50) by the method of
Cheng and Prussoff.19 All values are 0.3 log units unless otherwise
stated.
4.2. X-ray crystallographic and conformational analysis
In order to further interpret our results and elucidate the
critical relationship of the heterocyclic sulfonamide moi-
O
8-yl (18) and quinolin-5-yl (19) sulfonamides, while
elimination of both provided the napthyl-1-yl sulfon-
amide (21). As expected from our pharmacophore
hypotheses, removal of either quinoxaline nitrogen was
detrimental to CCK-2R binding affinity, and 18 and
19 had pKi values of <5 and 6.1, respectively. Moving
the nitrogen atom of 19 to the adjacent position gave
the isoquinolin-5-yl (20), resulting in total loss of recep-
tor affinity. Removal of both nitrogens (i.e., 21) also led
to complete abolition of binding affinity. These results
N
O
HN
S
I
O
O
N
N
28
Figure 5. Quinoxaline sulfonamide 28.

3922
M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
ety to binding affinity, we subjected select compounds to
similar structural and conformational analyses that we
applied to 1. As a high affinity CCK-2R ligand, quinox-
oline 28 was subjected to single crystal X-ray diffraction
studies. For comparison, the X-ray crystal structure of
quinoline 19 was obtained, which is a ligand of much
lower affinity. Quinoline 19 lacks the nitrogen atom
proximal to the sulfonamide group, which we believe
is required to either interact directly with the receptor,
or conversely to form the intramolecular interaction that
enforces a preference for the postulated cup-shaped
binding conformation. Examining the superimposed so-
lid state structures of 1, 19, and 28 reveals conforma-
tions that are qualitatively similar to that which was
observed for 1, yet which differ somewhat in the place-
ment of the quinoline and quinoxaline heterocycles
(Fig. 6A). However, on allowing these structures to re-
lax by energy minimization (B3LYP/6-31++G(2d,2p)),
the geometries of 19 and 28 more closely align with 1
with respect to the heterocycles (Fig. 6B). A deviation
is observed for 19, where the quinoline ring deviates
about the dihedral angle aNSCC by 12ꢁ relative to 1
and 28, and has twisted such that the carbon atom anal-
ogous to the nitrogen atom in 28 points outwards rela-
tive to the conformations of 1 and 28 (Table 3).
Energy minimized geometries of 19 and 28 were also ob-
tained in which the respective heterocycles were initially
rotated 180ꢁ about the carbon–sulfur bond, as was done
in the case of 1 (vide supra). This alternate geometry of
28 was less favored by 1.4 kcal/mol, while, surprisingly,
the alternate geometry of 19 was also less favored by
1.0 kcal/mol. Thus, it is questionable as to how much
an intramolecular hydrogen bond contributes to the sta-
bilization of the low-energy conformations observed,
and a more likely explanation for the importance of
the nitrogen atom proximal to the sulfonamide group
is that it makes a direct interaction with the receptor
that is favorable, but not required, for CCK-2R binding
affinity. The SAR data clearly indicates that distal nitro-
gen atom is involved in an interaction that is crucial for
binding.
5. Conclusion
We have described the identification of a series of potent
CCK-2R antagonists based on an anthranilic sulfon-
amide structural motif. Metabolic instability of the ben-
zothiadiazole and piperidine rings was found to be a
major liability of this series, and parallel synthesis tech-
niques failed to discover a suitable replacement for the
benzothiadiazole ring.10 Extensive characterization of
the physical and conformational properties of sulfon-
amide 1 was carried out using X-ray crystallography,
NMR spectroscopy, and computational methods, and
the synthesis and evaluation of compounds suggested
Figure 6. Superposition of 1 (green), 19 (cyan), and 28 (yellow): (A) X-ray crystal structures and (B) energy minimized structures (B3LYP/6-
31++G(2d,2p), halide atoms removed to simplify calculation).
Table 3. Bond angles and distances for X-ray and minimized structures of 1, 19, and 28
O
dXH H
N
aCNS
O
O
X
C
S
aNSC
C
aNSCC
˚
dXH (A)
Compound
aCNS
aNSC
aNSCC
X-ray
Min
X-ray
Min
X-ray
Min
X-ray
Min
1
119.4
122.0
119.0
126.8
125.1
126.1
103.8
107.8
107.2
106.4
106.5
107.2
À72.5
À73.6
À64.2
À61.1
À73.1
À61.3
3.1
3.1
2.5
3.0
3.2
2.6
19
28

M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
3923
by this analysis was carried out, leading to the discovery
of quinoxaline sulfonamide 28. Structural analysis of 28,
along with related compounds 13 and 19, was performed
using X-ray crystallography similar to that used for 1.
Comparison of the conformations of these compounds
revealed subtle differences in the orientation of the het-
erocyclic ring of the sulfonamide relative to the anthra-
nilic amide core that can help explain the remarkable
differences in CCK-2R affinity observed with relatively
minor structural changes. The design and synthesis of
conformationally-restricted analogs to further explore
the structural and pharmacophoric requirements for
high levels of CCK-2R binding, as well as further opti-
mization toward reducing oxidative metabolism and
improving in vivo efficacy, will be the subject of future
disclosures from these laboratories.
1H NMR (400 MHz, CDCl3, rotameric broadening): d
11.56 (s, 1H), 8.37 (d, J = 7.0 Hz, 1H), 8.22 (d,
J = 8.8 Hz, 1H), 7.90 (d, J = 1.7, 1H), 7.72 (dd,
J = 8.8, 7.1 Hz, 1H), 7.34 (d, J = 8.4 Hz, 2H), 7.24 (d,
partially obscured by CHCl3, 2H), 7.16 (d, J = 8.4 Hz,
1H), 7.07 (dd, J = 8.4, 1.7 Hz, 1H), 6.25 (br t, 1H),
4.53 (d, J = 5.8 Hz, 2H).
6.3. General procedure for sulfonamide formation (9–11,
13–16)
The appropriate sulfonyl chloride was added to a solu-
tion of anthranilic piperidine amide 6 or 8,8 pyridine
(ca. 1.5 equiv) and DCM (0.1 M final concentration),
and the reaction mixture was maintained overnight at
ambient temperature. The volatiles were removed in va-
cuo, then the residue was taken up in EtOAc and
washed with 1 N HCl. The organic layer was dried
(Na2SO4), filtered, concentrated, and purified as indi-
cated to provide the titled sulfonamides.
6. Experimental
6.1. General
6.3.1. 2-Methyl-2H-benzotriazole-4-sulfonic acid [5-bro-
mo-2-(piperidine-1-carbonyl)phenyl]amide (9) and 2-
methyl-2H-benzotriazole-5-sulfonic acid [5-bromo-2-
(piperidine-1-carbonyl)phenyl]amide (10). The titled com-
pounds were prepared from a crude mixture of sulfonyl
chlorides 22 and 23 (0.29 g, 1.3 mmol, 4.5:1 22:23) and
amine 6 (0.30 g, 1.1 mmol),8 according to the above gen-
eral procedure. The crude products were separated by
silica gel chromatography (hexane/EtOAc 20:80 to
70:30). The more rapidly eluting fractions were collected
to provide 10 as a solid (0.031 g, 6%). Continued elution
provided 9 as a solid (0.14 g, 28%). Compound 9: MS
(ESI): m/z 476.0 [MÀH]À. 1H NMR (400 MHz, CDCl3,
Except where indicated, materials and reagents were used
as supplied. Reaction monitoring was performed with
EMD Silica Gel 60 F254 250 lm pre-coated TLC plates
and visualized with UV light. Routine chromatographic
purifications were performed via semi-automated MPLC
using prepacked RediSep 35–60 lm silica gel columns on
ISCO Sg100 systems with UV peak detection. Semipre-
parative HPLC purification was performed on a Gilson
automated HPLC system running Gilson Unipoint LC
software with UV peak detection at 220 nm and fitted
with a reverse phase YMC-Pack ODS-A 250 · 30 mm
column; mobile phase gradients consisted of mixtures of
MeCN and water with 0.05% TFA and flow rates of 10–
20 mL/min. Mass spectra were recorded on an Agilent
single quadrupole electrospray MSD mass detection. Pro-
ton NMR spectra were recorded at either 400 or 500 MHz
using Bruker DPX-400 and DPX-500 spectrometers.
Melting point determinations were made using a Stanford
Research Systems OptiMelt system and are uncorrected.
Combustion analyses were performed by NuMega Reso-
nance Labs, Inc., San Diego, CA. Bioanalytical quantita-
tion was performed via mass spectrometry using either a
Finnigan Quantum Ultra system (Thermo Electron
Corp.), or an API-4000 system (Applied Biosystems), un-
less otherwise indicated. The syntheses and characteriza-
tion of compounds 12, 17–21, and 28 have been previously
described.8
rotameric broadening):
d 8.62 (s, 1H), 8.02 (d,
J = 8.4 Hz, 1H), 7.90 (d, J = 6.8 Hz, 1H), 7.80 (d,
J = 2.0 Hz, 1H), 7.39 (dd, J = 8.8, 7.6 Hz, 1H), 7.10
(dd, J = 8.4, 2.0 Hz, 1H), 6.85 (d, 8.0 Hz, 1H), 4.55 (s,
3H), 2.70–3.45 (br s, 4H), 1.10–1.50 (br m, 6H). Anal.
(C19H20BrN5O3SÆ0.4 C6H14) C, H, N. Compound 10:
MS (ESI): m/z 476.0 [MÀH]À. 1H NMR (400 MHz,
CDCl3, rotameric broadening): d 8.84 (s, 1H), 8.46 (s,
1H), 7.94 (d, J = 8.8 Hz, 1H), 7.89 (d, J = 2.0 Hz, 1H),
7.77 (dd, J = 8.8, 1.6 Hz, 1H), 7.19 (dd, J = 8.4,
2.0 Hz, 1H), 6.95 (d, J = 8.4 Hz, 1H), 4.56 (s, 3H),
2.52–3.52 (br m, 4H), 1.15–1.55 (br m, 6H). Anal.
(C19H20BrN5O3S) C, H, N.
6.3.2. 1-Methyl-1H-benzotriazole-4-sulfonic acid [5-bro-
mo-2-(piperidine-1-carbonyl) phenyl]amide (11). Com-
pound 11 was prepared from crude sulfonyl chloride 24
(0.15 g, 0.65 mmol) and amine 68 (0.12 g, 0.43 mmol), as
described in the general procedure for sulfonamide for-
mation. The crude product was purified by recrystalliza-
tion (hexanes/EtOAc) to provide 11 (0.040 g, 19%).
Mp = 208–209 ꢁC. MS (ESI): m/z 479.8 [M+H]+. 1H
NMR (400 MHz, CDCl3, rotameric broadening): d 8.96
(s, 1H), 7.96 (dd, J = 7.3, 0.8 Hz, 1H), 7.86 (d,
J = 1.8 Hz, 1H), 7.75 (dd, J = 8.4, 0.8 Hz, 1H), 7.56 (dd,
J = 8.4, 7.6 Hz, 1H), 7.12 (dd, J = 8.2, 1.8 Hz, 1H), 6.92
(d, J = 8.2 Hz, 1H), 4.34 (s, 3H), 3.10–3.70 (br s, 4H),
1.20–1.90 (br m, 6H). 13C NMR (100 MHz, CDCl3, roto-
meric broadening): d 167.1, 140.7, 136.7, 134.5, 130.0,
6.2. 2-(Benzo[1,2,5]thiadiazole-4-sulfonylamino)-4-bro-
mo-N-(4-chlorobenzyl) benzamide (7)
A solution of HATU (0.038 g, 0.10 mmol) and DMF
(0.20 mL) was added to a solution of acid 278 (0.021 g,
0.050 mmol), pyridine (0.012 mL, 0.15 mmol), and
DMF (0.20 mL), with stirring. The mixture was main-
tained for 1 h, then 4-chlorobenzylzamine (0.012 mL,
0.10 mmol) was added. The reaction was allowed to pro-
ceed for 1 h, then TFA (0.050 mL) and DMF (1.0 mL)
was added. The crude reaction mixture was directly sub-
jected to preparative HPLC purification, providing 7 as
a solid (0.018 mg, 67%). MS (ESI): m/z 536.8 [MÀH]À.

3924
M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
129.9, 126.7, 126.6, 125.1, 124.6, 124.3, 123.8, 114.9, 34.7,
25.7, 24.5. Anal. (C19H20BrN5O3S) C, H, N.
2.65–3.60 (br m, 4H), 1.00–1.55 (br m, 6H). Anal.
(C19H18BrN3O3S2) C, H, N.
6.3.3. Benzothiazole-4-sulfonic acid [5-bromo-2-(piperi-
dine-1-carbonyl)phenyl]amide (13). The titled compound
was prepared from benzothiazole-4-sulfonyl chloride15
(0.12 g, 0.53 mmol) and 68 (0.075 g, 0.27 mmol), as de-
scribed in the general procedure for sulfonamide forma-
tion. Silica gel chromatography (EtOAc/hexanes 5:95 to
75:25) provided 13 (0.12 g, 91%). MS (ESI): m/z 482.0
[M+H]+. 1H NMR (500 MHz, CDCl3, rotameric broad-
ening): d 9.26 (s, 1H), 8.87 (s, 1H), 8.19 (dd, J = 8.1,
0.9 Hz, 1H), 8.14 (dd, J = 7.6, 0.9 Hz, 1H), 7.81 (d,
J = 1.2 Hz, 1H), 7.56 (t, J = 7.8 Hz, 1H), 7.14 (dd,
J = 8.2, 1.8 Hz, 1H), 6.92 (d, J = 8.2 Hz, 1H), 2.80–
3.65 (br m, 4H), 1.35–1.65 (br m, 6H). Anal.
(C19H18BrN3O3S2) C, H, N.
6.4. 2-Methyl-2H-benzotriazole-4-sulfonyl chloride (22)
and 2-methyl-2H-benzotriazole-5-sulfonic acid (23)
A solution of 2-methylbenzotriazole (0.75 g, 5.6 mmol)
and chlorosulfonic acid (10 mL) was heated to reflux
for 30 min. The reaction mixture was allowed to cool
to 23 ꢁC, and was carefully poured over ice (caution!
highly exothermic!). The resulting mixture was extracted
with DCM, dried, and concentrated, providing 1.2 g of a
4.5:1 mixture of 22 and 23 as a solid (1.2 g, 93%). This
mixture was used directly without purification.
6.5. 1-Methyl-1H-benzotriazole-4-sulfonyl chloride (24)
A
mixture of 1-methyl-1H-benzotriazole (1.0 g,
6.3.4. Benzothiazole-7-sulfonic acid [5-bromo-2-(piperi-
dine-1-carbonyl)phenyl]amide (14). Compound 14 was
prepared from benzothiazole-7-sulfonyl chloride15
(0.12 g, 0.53 mmol) and 68 (0.075 g, 0.27 mmol), as de-
scribed in the general procedure for sulfonamide forma-
tion. Silica gel chromatography (EtOAc/hexanes 5:95 to
75:25) provided a solid (0.13 g, 100%). MS (ESI): m/z
7.5 mmol) and chlorosulfonic acid (2.5 mL) was heated
to reflux for 6 h, then was allowed to cool to rt and
poured over ice. The mixture was extracted with EtOAc,
dried, and concentrated to a tan oil (1.05 g, 61%). This
crude material was used directly without purification.
6.6. 2,3-Diaminobenzenesulfonic acid [5-bromo-2-(piperi-
dine-1-carbonyl)-phenyl]-amide (25)
1
482.0 [M+H]+. H NMR (500 MHz, CDCl3, rotameric
broadening): d 9.13 (s, 1H), 9.03 (s, 1H), 8.30 (dd,
J = 8.1, 0.7 Hz, 1H), 7.91-7.92 (m, 2H), 7.60 (t,
J = 7.8 Hz, 1H), 7.22 (dd, J = 8.2, 1.9 Hz, 1H), 6.89 (d,
J = 8.2 Hz, 1H), 2.55–3.45 (br m, 4H), 1.10–1.55 (br
m, 6H). Anal. (C19H18BrN3O3S2) C, H, N.
A mixture of 18 (0.10 g, 0.21 mmol), zinc powder (0.15 g,
2.1 mmol), and acetic acid (2 mL) was heated to 50 ꢁC
for 20 min. The mixture was allowed to cool to rt, and
was then filtered through Celite, rinsing with MeOH.
The resulting solution was concentrated to a solid. This
mixture of 25 and zinc acetate was used directly in the
subsequent step. An analytical sample was prepared by
partitioning a portion of the crude material between
EtOAc and aqueous NaHCO3. The organic layer was
dried and concentrated, and the residue was chromato-
graphed (5:95 MeOH/DCM). MS (ESI): m/z 455.0
[M+H]+. 1H NMR (500 MHz, CDCl3, rotameric broad-
ening): d 8.60–8.90 (br s, 1H), 7.81 (d, J = 1.9 Hz, 1H),
7.17–7.22 (m, 2H), 6.96 (d, J = 8.2 Hz, 1H), 6.82 (dd,
J = 7.6,1.3 Hz, 1H), 6.60 (t, J = 7.9, 1H), 4.65–4.90 (br
s, 2H), 2.9–3.8 (br m, 6H), 1.4–1.65 (br m, 6H).
6.3.5. 2-Methyl-benzothiazole-7-sulfonic acid [5-bromo-2-
(piperidine-1-carbonyl)-phenyl]-amide (15). The titled
compound was prepared from 2-methylbenzothiazole-
7-sulfonyl chloride15 (0.10 g, 0.35 mmol) and 68
(0.15 g, 0.53 mmol), as described in the general proce-
dure for sulfonamide formation. Silica gel chromatogra-
phy (EtOAc/hexanes 0:100 to 75:25) provided 15 as a
1
solid (0.13 g, 75%). MS (ESI): m/z 495.8 [M+H]+. H
NMR (500 MHz, CDCl3, rotameric broadening): d
9.00 (s, 1H), 8.10 (dd, J = 8.1, 0.9 Hz, 1H), 7.90 (d,
J = 1.9 Hz, 1H), 7.80 (dd, J = 7.7, 1.0 Hz, 1H), 7.51 (t,
J = 7.9 Hz, 1H), 7.21 (dd, J = 8.2, 1.9 Hz, 1H), 6.90 (d,
J = 8.2 Hz, 1H), 2.87 (s, 3H), 2.55–3.55 (br m, 4H),
1.05–1.60 (br m, 6H). 13C NMR (125 MHz, CDCl3,
rotomeric broadening): d 170.2, 167.1, 154.9, 137.1,
133.2, 133.0, 128.7, 127.4, 126.9, 126.9, 125.9, 125.1,
124.8, 124.5, 25.7–25.9 (broad), 24.2, 9.9. Anal.
(C20H20BrN3O3S2) C, H, N.
6.7. Benzimidazole-4-sulfonic acid [5-bromo-2-(piperi-
dine-1-carbonyl)phenyl]amide (26)
Crude 25 (0.12 g, 0.21 mmol based on amount of 1 from
previous step, vide supra), triethyl orthoformate
(1.0 mL), and p-toluenesulfonic acid monohydrate
(2.0 mg, 0.011 mmol) was heated to 100 ꢁC for 15 min.
The solution was allowed to cool to rt, and then was
diluted with EtOAc, washed with saturated aqueous
NaHCO3, dried, and concentrated. The residue was
chromatographed (MeOH/DCM 0:100 to 5:95) to pro-
vide the titled compound as a solid (0.043 g, 44% over
two steps). MS (ESI): m/z 464.9 [M+H]+. 1H NMR
(500 MHz, CDCl3, rotameric broadening): d 11.12 (br
s, 1H), 9.07 (br s, 1H), 8.01 (br s, 2H), 7.92 (d,
J = 1.6 Hz, 1H), 7.60 (d, J = 7.5 Hz, 1H), 7.25–7.30
(m, 2H), 6.93 (d, J = 8.2 Hz, 1H), 3.05–3.40 (br m,
2H), 2.40–2.75 (br m, 2H), 1.15–1.55 (br m, 6H). Anal.
(C19H19BrN4O3S) C, H, N.
6.3.6. Benzothiazole-6-sulfonic acid [5-bromo-2-(piperi-
dine-1-carbonyl)phenyl]amide (16). Compound 16 was
prepared from benzothiazole-6-sulfonyl chloride15
(0.12 g, 0.53 mmol) and 68 (0.075 g, 0.27 mmol), as de-
scribed in the general procedure for sulfonamide forma-
tion. Silica gel chromatography (EtOAc/hexanes 5:95 to
75:25) afforded a solid (0.12 g, 93%). MS (ESI): m/z
1
482.0 [M+H]+. H NMR (500 MHz, CDCl3, rotameric
broadening): d 9.20 (s, 1H), 8.87 (s, 1H), 8.51 (d,
J = 1.8 Hz, 1H), 8.19 (d, J = 8.6 Hz, 1H), 7.93 (dd,
J = 8.6, 1.8 Hz, 1H), 7.91 (d, J = 1.9 Hz, 1H), 7.22
(dd, J = 8.2, 1.9 Hz, 1H), 6.95 (d, J = 8.2 Hz, 1H),

M. D. Rosen et al. / Bioorg. Med. Chem. 16 (2008) 3917–3925
3925
5. Durrant, L. G.; Spendlove, I. Expert Opin. Emerg. Drugs
2003, 8, 489–500.
6. Watson, S. A.; Gilliam, A. D. Expert Opin. Biol. Ther.
2001, 1, 309–317.
6.8. Human CCK-1 and CCK-2 radioligand binding
assays
Human CCK-1 and CCK-2 receptor radioligand bind-
ing assays were performed as previously described.16–18
7. Herranz, R. Med. Res. Rev. 2003, 23, 559–605.
8. Allison, B. D.; Phuong, V. K.; McAtee, L. C.; Rosen, M.
D.; Morton, M.; Prendergast, C.; Barrett, T.; Laguad, G.;
Freedman, J.; Li, L.; Wu, X.; Venkatesan, H.; Pippel, M.;
Woods, C.; Rizzolio, M. C.; Hack, M.; Hoey, K.; Deng,
X.; King, C.; Shankley, N. P.; Rabinowitz, M. H. J. Med.
Chem. 2006, 49, 6371–6390.
9. Koch, P.; Boelsterli, J. J.; Hirst, D. R.; Walkinshaw, M.
D. J. Chem. Soc., Perkin Trans. 1990, 2, 1705–1708.
10. Woods, Craig R.; Hack, Michael D.; Allison, Brett D.;
Phuong, Victor K.; Rosen, Mark D.; Morton, Magda F.;
Prendergast, Clodagh E.; Barrett, Terrance D.; Shankley,
Nigel P.; Rabinowitz, Michael H. Bioorg. Med. Chem
Lett. 2007, 17, 6905–6909.
Acknowledgments
The authors thank Heather McAllister and D.J. Tog-
narelli for analytical support, and Wensheng Lang, Wil-
liam Jones, and John Masucci for in vitro metabolism
analysis.
Supplementary data
11. De Luca, S.; Saviano, M.; Lassiani, L.; Yannakopoulou,
K.; Stefanidou, P.; Aloj, L.; Morelli, G.; Varnavas, A.
J. Med. Chem. 2006, 49, 2456–2462.
12. Feigel, M.; Lugert, G.; Manero, J.; Bremer, M. Z.
Naturforsch. 1990, 45b, 258–266.
13. Feigel, M.; Lugert, G.; Manero, J.; Bremer, M. Z.
Naturforsch. 1989, 44b, 1109–1116.
14. Sakurai, H.; Ritonga, M. T. S.; Shibatani, H.; Hirao, T.
J. Org. Chem. 2005, 70, 2754–2762.
Protocols for HLM metabolite identification, tabular
combustion analysis data, and X-ray crystallographic
data are available online at www.sciencedirect.com.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmc.2008.01.059.
15. Cremlyn, R. J.; Bassin, J. P.; Farouk, S.; Potterton, M.;
Mattu, T. Phosphorus, Sulfur Silicon Relat. Elem. 1992, 73,
107–120.
References and notes
1. Dufresne, M.; Seva, C.; Fourmy, D. Physiol. Rev. 2006,
86, 805–847.
2. Konturek, P. C.; Nikiforuk, A.; Kania, J.; Raithel, M.;
Hahn, E. G.; Muhldorfer, S. Dig. Dis. Sci. 2004, 49, 1075–
1083.
3. Colucci, R.; Blandizzi, C.; Tanini, M.; Vassalle, C.;
Breschi, M. C.; Del Tacca, M. Br. J. Pharmacol. 2005,
144, 338–348.
16. Morton, M. F.; Liu, P. Q.; Reik, A.; de la Rosa, R.;
Mendel, M.; Li, X. Y.; Case, C.; Pabo, C.; Moreno, V.;
Pyati, J.; Shankley, N. P. Regul. Pept. 2005, 129, 227–232.
17. Morton, M. F.; Pyati, J.; Dai, H.; Li, L.; Moreno, V.;
Shankley, N. P. Br. J. Pharmacol. 2005, 374–384.
18. Ulrich, C. D.; Ferber, I.; Holicky, E.; Hadac, E.; Buell, G.;
Miller, L. J. Biochem. Biophys. Res. Commun. 1993, 193,
204–211.
4. Reubi, J. C.; Schaer, J. C.; Waser, B. Cancer Res. 1997, 57,
1377–1386.
19. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22,
3099–3310.