1,4-Benzodiazepine peripheral cholecystokinin (CCK-A) receptor agonists

Article abstract of DOI:10.1016/S0960-894X(01)00164-0

A series of 1,4-benzodiazepines, N-1-substituted with an N-isopropyl-N-phenylacetamide moiety, was synthesized and screened for CCK-A agonist activity. In vitro agonist activity on isolated guinea pig gallbladder along with in vivo induction of satiety fo

Full text of DOI:10.1016/S0960-894X(01)00164-0

Bioorganic & Medicinal ChemistryLetters 11 (2001) 1145–1148
1,4-Benzodiazepine Peripheral Cholecystokinin (CCK-A)
Receptor Agonists
Ronald G. Sherrill,a,* Judd M. Berman,a Lawrence Birkemo,c Dallas K. Croom,c
c
c
Milana Dezube,a GregoryN. Ervin, MaryK. Grizzle, Michael K. James,b
Michael F. Johnson,c KennedyL. Queen, Thomas J. Rimele,b
b
Frank Vanmiddleswortha and Elizabeth E. Sugga
aDepartment of Medicinal Chemistry, GlaxoWellcome Research and Development, 5 Moore Drive, Research Triangle Park,
NC 27709, USA
bDepartment of Receptor Biochemistry, GlaxoWellcome Research and Development, 5 Moore Drive, Research Triangle Park,
NC 27709, USA
cDepartment of Pharmacology, GlaxoWellcome Research and Development, 5 Moore Drive, Research Triangle Park,
NC 27709, USA
Received 5 January2001; accepted 26 February2001
Dedicated to the memoryof Marc Rodriguez, a pioneer in CCK research, an exceptional scientist and a dear friend
Abstract—A series of 1,4-benzodiazepines, N-1-substituted with an N-isopropyl-N-phenylacetamide moiety, was synthesized and
screened for CCK-A agonist activity. In vitro agonist activity on isolated guinea pig gallbladder along with in vivo induction of satiety
following intraperitoneal administration in a rat feeding assaywas demonstrated. # 2001 Elsevier Science Ltd. All rights reserved.
Obesity, defined by the World Health Organization as a
bodymass index exceeding 30 kg m À2, is the most pre-
valent metabolic disorder in western nations. Increased
relative risk of coronaryheart disease, type 2 diabetes
mellitus, hypertension, and cholelithiasis is positively
correlated with excess adiposity.1 Current pharma-
cotherapeutic intervention strategies seek to either reg-
ulate nutrient intake, absorption or metabolism.
Cholecystokinin (CCK) is a gastrointestinal peptide
hormone and neurotransmitter that regulates satiety,
biliaryand pancreatic secretion, gallbladder contrac-
tion, and gut motilitythrough the peripheral CCK-A
seven transmembrane G-coupled protein receptor.2
Administration of CCK to obese humans has been
shown to reduce meal duration and size suggesting this
mechanism as a means of drug intervention in the obese
population.3
response at the peripheral CCK-A receptor (Fig. 1).
These analogues were potent fullyefficacious agonists
with anorectic efficacyin rat feeding models. As both
the 1,5- and 1,4-benzodiazepine scaffolds have been uti-
lized as CCK-A antagonists,5 we envisioned that these
same modifications, which imparted agonist functional
activityin the former series, could be adopted onto the
1,4-benzodiazepine pharmacophore (2). The results of
these studies are detailed below.
4
We have previouslydescribed several series of 1,5-ben-
zodiazepine CCK ligands (1a–d), which all incorporated
an N-isopropyl-N-phenylacetamide agonist ‘trigger’
attached to N-1 to modulate an agonist functional
*Corresponding author. Tel.: +1-919-483-6281; fax: +1-919-483-
6053; e-mail: rgs13254@glaxowellcome.com
Figure 1.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00164-0

1146
R. G. Sherrill et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1145–1148
The prerequisite 3-amino-1,4-benzodiazepine core and
agonist trigger were constructed from the previously
published intermediates: 3-Cbz-amino-1,4-benzodiaze-
pines, 3;6 and bromoacetamides, 44a (Scheme 1). The
isopropyl anilide ‘trigger’ was installed by alkylation of
N-1 of benzodiazepine 3 with 4 to generate intermediate
5. Deprotection and acylation of the C-3 amino group
provided analogues 2a,b,e,f,h,j. N-Oxide analogues
(2g,i,k,n) were prepared byoxidation of 5 with m-CPBA
followed bycataltyic removal of the Cbz protecting
group and acylation.
at 1 or 30 mM concentration.8 Full dose–response curves
were obtained for those compounds that induced
>40% of the response produced byCCK-8. Agonist
activityof all compounds was reversed bythe selective
CCK-A antagonist MK-329.9 The satietyeffect of sev-
eral compounds was measured in a rat feeding model4a
following intraperitoneal (ip) administration in fasted
animals. ED50 values reflect the dose of test compound
demonstrating 50% of the maximal response observed
at 10 mmol/kg. ‘%Max’ refers to the percentage reduc-
tion in food intake at 30 min dosed at 10 mmol/kg
(0.5 mmol/kg for CCK-8) relative to control animals.
Ureas 2a,e,h–k,n were prepared from either treatment
with phenyl isocyanate (2a,h–k) or the p-nitrophe-
nylcarbamate of 3-t-butylcarboxy-aniline followed by
TFA deprotection (2e,n). Amides (2b,f,g) were prepared
through carbodiimide coupling with indole-2-carboxylic
acid.
We had previouslyoptimized the substituents necessary
to modulate the conversion of 1,5-benzodiazepine
CCK-A antagonists to CCK-A agonists.4 Accordingly,
we sought to incorporate the identical pharmacophores
into a 1,4-benzodiazepine scaffold. Therefore, all of our
analogues contain the isopropyl anilinoacetamide that
served as the prerequisite agonist ‘trigger’ in the 1,5-
series.4a At the 3-position, phenyl ureas,4a,b anilino
acetamides4c and indazolylmethyl4d,e or indolylmethyl4d
all served as optimal substituents when combined with
the agonist ‘trigger’.
C-3 alkyl analogues were derived from benzodiazepine
intermediate 7, which in turn was prepared byalkyla-
tion of known benzodiazepine7 6 with bromide 4
(Scheme 2). Alkylation of 7 with t-butyl bromoacetate,
TFA deprotection and PyBroP mediated coupling of
aniline provided analogue 2c. Indazole (2l,m) and indole
(2d) derivatives were prepared byalkylation of 7 with
the corresponding N-Boc-protected bromomethyl
aryl intermediate4d,e followed byTFA deprotection.
N-Oxide analogue 2m was prepared byoxidation with
m-CPBA, prior to Boc deprotection.
We were pleased to observe that addition of the agonist
‘trigger’ to a 3-phenylureido substituted 1,4-benzodiaze-
pine (2a) indeed yielded a CCK-agonist, albeit a partial
one with only39% (30 mM) of the maximal CCK-8
response (Table 1). Incorporating the 2-indoleamide
(2b) to the 3-position yielded a weak partial agonist with
only12% (30 mM) of the maximal CCK response.
In vitro CCK-A functional efficacywas measured
through contraction of isolated guinea pig gallbladder
Introduction of a 4-methoxysubstituent to the aniline
present in the ‘trigger’ had produced potent agonists in
the 1,5-benzodiazepine series.4a,d,e Combination of the
Scheme 1. (a) NaH, DMF; (b) HBr, AcOH or Pd/C, H2, EtOH; (c) (i)
3-t-Bu-CO2PhNHCO2-p-NO2-Ph, CH3CN; (ii) TFA or PhNCO,
CH2Cl2 or EDC, indole-2-carboxylic acid, CH2Cl2; (d) m-CPBA,
CH2Cl2.
Scheme 2. (a) NaH, 4, DMF; (b) (i) KHMDS, t-Bu-bromoacetate,
THF; (ii) TFA; (iii) PyBroP; PhNH2, CH2Cl2; (c) (i) LiHMDS,
BrCH2Ar,4d,e THF; (ii) TFA (d) for 2m (i) LiHMDS, BrCH2Ar,4d,e
THF; (ii) m-CPBA, CH2Cl2; (iii) TFA, CH2Cl2.

R. G. Sherrill et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1145–1148
Table 1. In vitro activityof 1,4-benzodiazepine CCK-A agonists
1147
b
2
R1
R2
R3
X
GPGB % CCKa ED50
2
R1
R2
R3
X
N
GPGB % CCK ED50
(nM)
(mM)
(nM)
(mM)
a
b
c
d
e
f
H
H
Ph
Ph
Ph
Ph
Ph
N
N
N
N
N
N
39 (30)
ND
h
i
OMe C6H11
56 (30)
233
12 (30)
43 (30)
94 (30)
27 (1)
ND
1100
41
OMe C6H11 N-O
87 (30)
90 (30)
100 (30)
27 (1)
342
296
202
ND
13
OMe
H
j
OMe
OMe
OMe
OMe
Me
Me
Ph
N
N-O
N
k
l
OMe
ND
ND
131
OMe C6H11
0 (30)
m
n
Ph
N-O
100 (30)
98 (30)
g
OMe C6H11 N-O
43 (30)
OMe C6H11 N-O
3
aFunctional activityin the isolated guinea pig gallbladder following incubation with the test ligand for 30 min at 37 ꢀC; relative efficacyas deter-
mined bythe maximal contraction observed at 1 or 30 mM standardized to CCK-8 (1 mM)=100%.
bED50, concentration at which 50% of the maximal contraction was observed for % CCK.
4-methoxytrigger modification with the C-3 anilino-
acetamide (2c) and indolylmethyl (2d) groups produced
partial agonists in the 1,4-benzodiazepine series (43 and
24%, respectively; 30 mM). We were initiallysomewhat
disappointed with the efficacydemonstrated in the 1,4-
benzodiazepine template relative to the 1,5-benzodiaze-
pines. The analogous 1,5-benzodiazepine homologues4
of analogues 2a–d demonstrated 86, 49, 78, and 70% of
the maximal CCK response at 30 mM, respectively;
essentially2–3 times more efficacious. However, addi-
tion of the 3-carboxyfunctionalityto the phenyl urea
(2e) produced a full CCK-agonist with 40 nM potency
in our in vitro assay. The 1,5-homologue4b of analogue
2e had demonstrated identical potencyand efficacy
(100% at 30 uM, ED50=40 nM).
CCK-A agonist activityin the guinea pig gallbladder
assay(34% at 1 mM, unpublished data).
We hypothesized that the carbonyl functionality present
at the 4-position in the 1,5-series, but absent in the 1,4-
series, played a critical role in regulating the functional
activityof the ligand within the receptor. Oxidation of
the 1,4-benzodiazepine to its corresponding N-oxide
might approximate the carbonyl pharmacophore pre-
sent in the 1,5-series. We were pleased to find that oxi-
dation of analogue 2f to 2g restored the CCK-A agonist
Table 2. Anoretic profilea of 1,4-benzodiazepine CCK-A agonists
Compounds
ED50 (nM)b
%Maxc
2i
2m
2n
CCK-8
25
150
80
75
79
69
98
We then chose to modifythe 5-position of our series
substituting cyclohexyl and methyl for the C-5 phenyl
group of our earlier analogues. However, to our sur-
prise, the combination of the 5-cyclohexyl group along
with the C-3 indole amide (2f) produced a CCK-A
antagonist as evidenced byinhibition of the CCK-8
functional response to the test tissue (data not shown).
This was particularlyunexpected considering the analo-
gous 1,5-benzodiazepine analogue had demonstrated
27
aAnoretic potencyin Long–Evans rats conditioned to a palatable
liquid diet and fasted for 2 h.
bED50, dose of test compound that produces a half-maximal response
(nM/kg).
c%Max, maximal response at 0.5 mmol/kg, ip for CCK-8 and 10 mmol/
kg, ip for test compound.

1148
R. G. Sherrill et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1145–1148
activity(43% at 30 mM, 131 nM). This result suggested
that the N-oxide moietymight be an essential pharma-
cophore for optimal agonist response. However, 5-
cyclohexyl phenyl urea analogue 2h and its corresponding
N-oxide 2i were similarlyefficacious (56 and 87% at
30 mM, respectively) and equipotent (233 vs 342 nM,
respectively). Further, 5-methyl phenyl urea analogue 2j
and its corresponding N-oxide 2k were also essentially
equallyefficacious (90 and 100% at 30 mM, respectively)
and equipotent (296 vs 202 nM, respectively). We had
already shown the indazolylmethyl substituent at C-3 to
be an optimal CCK-A agonist in the 1,5-series.4d,e The
analogous 1,4-analogue (2l) was a weak partial agonist
(27% at 1 mM). However, oxidation to 2m provided an
extremelypotent fullyefficacious agonist (98% at
30 mM, 13 nM). The potencyof this 1,4-analogue actu-
template to give compounds whose potency, in some
cases, equalled or exceeded their 1,5-homologues. These
agents demonstrated efficacyin an in vitro guinea pig
gallbladder assayand anoretic activityin a rat feeding
model.
References
1. Kopelman, P. G. Nature 2000, 404, 635.
2. Crawley, J. N.; Corwin, R. L. Peptides 1994, 15, 731.
3. Pi-Sunyer, X.; Kissileff, H. R.; Thornton, J.; Smith, G. P.
Physiol. Behav. 1982, 29, 627.
4. (a) Aquino, C. J.; Armour, D. R.; Berman, J. M.; Birkemo,
L. S.; Carr, R. A. E.; Croom, D. K.; Dezube, M.; Dougherty,
R. W.; Ervin, G. N.; Grizzle, M. K.; Head, J. E.; Hirst, G. C.;
James, M. K.; Johnson, M. F.; Long, J. E.; Miller, L. J.;
Queen, K. L.; Rimele, T. J.; Smith, D. N.; Sugg, E. E. J. Med.
Chem. 1996, 39, 562. (b) Hirst, G. C.; Aquino, C.; Birkemo,
L.; Croom, D. K.; Dezube, M.; Dougherty, R. W.; Ervin,
G. N.; Grizzle, M. K.; Henke, B.; James, M. K.; Johnson,
M. F.; Momtahen, T.; Queen, K. L.; Sherrill, R. G.; Szewczyk,
J.; Willson, T. M.; Sugg, E. E. J. Med. Chem. 1996, 39, 5236.
(c) Willson, T. M.; Henke, B. R.; Momtahen, T. M.; Myers, P.
L.; Sugg, E. E.; Unwalla, R. J.; Croom, D. K.; Grizzle, M. K.;
Johnson, M. F.; Queen, K. L.; Rimele, T. R.; Yingling, J. D.;
James, M. K. J. Med. Chem. 1996, 39, 3030. (d) Henke, B. R.;
Willson, T. M.; Sugg, E. E.; Croom, D. K.; Dougherty, R. W.,
Jr.; Queen, K. L.; Birkemo, L. S.; Ervin, G. N.; Grizzle, M.
K.; Johnson, M. F.; James, M. K. J. Med. Chem. 1996, 39,
2655. (e) Henke, B. R.; Aquino, C. J.; Birkemo, L. S.; Croom,
D. K.; Dougherty, R. W., Jr.; Ervin, G. N.; Grizzle, M. K.;
Hirst, G. C.; James, M. K.; Johnson, M. F.; Queen, K. L.;
Sherrill, R. G.; Sugg, E. E.; Suh, E. M.; Swewczyk, J. W.;
Unwalla, R. J.; Yingling, J.; Willson, T. M. J. Med. Chem.
1997, 40, 2706.
5. de Tuillio, P.; Delarge, J.; Pirotte, B. Curr. Med. Chem.
1999, 6, 433.
6. Sherrill, R. G.; Sugg, E. E. J. Org. Chem. 1995, 60, 730.
7. Bock, M. G.; DiPardo, R. M.; Evans, B. E.; Rittle, K. E.;
Veber, D. F.; Freidinger, R. M.; Hirshfield, J.; Springer, J. P.
J. Org. Chem. 1987, 52, 3232.
8. Sugg, E. E.; Kimery, M. J.; Ding, J. M.; Kenakin, D. C.;
Miller, L. J.; Queen, K. L.; Rimele, T. J. J. Med. Chem. 1995,
38, 207.
9. Evans, B. E.; Bock, M. G.; Rittle, K. E.; Dipardo, R. M.;
Whitter, W. L.; Veber, D. L.; Anderson, P. S.; Freidinger,
R. M. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 4918.
4d
allyexceeded the analogous 1,5-homologue (109 nM).
From the SAR available, the nature of the functional
group attached to C-3 of the benzodiazepine ring dic-
tated whether oxidation to the N-oxide was required for
agonist activity. Analogue 2n incorporated all of the
functional groups which had demonstrated superior
efficacy and potency in this series: the 5-cyclohexyl
group, the 4-methoxysubstituted ‘trigger’ and the
3-carboxyphenyl urea at the 3-position. As expected,
this compound was a full agonist (98% at 30 mM) with
exceptional potency(3 nM).
Finally, we wished to evaluate the in vivo anoretic effi-
10
cacyof these agents in a rat conditioned-feeder assay.
(Table 2). Analogues 2i, 2m, and 2n were selected for
further screening based on potencyand efficacydemon-
strated in the in vitro assay. Following intraperitoneal
(ip) administration, all three agents demonstrated dra-
matic reduction in food intake, similar in both potency
and efficacyto CCK-8.
In summary, we have described a novel series of potent
and efficacious 1,4-benzodiazepine and 1,4-benzodiaze-
pine-4-N-oxide CCK-A agonists that incorporate an
N-isopropyl-N-phenylacetamide agonist ‘trigger’ at the
N-1 ring position. Modifications to the C-3 position,
which had demonstrated agonist activityin the 1,5-ben-
zodiazepine series, were succesfullyadapted to this