Discovery of a Novel Class of Negative Allosteric Modulator of the Dopamine D2 Receptor Through Fragmentation of a Bitopic Ligand

Article abstract of DOI:10.1021/acs.jmedchem.5b00585

Recently, we have demonstrated that N-((trans)-4-(2-(7-cyano-3,4-dihydroisoquinolin-2(1H)-yl)ethyl)cyclohexyl)-1H-indole-2-carboxamide (SB269652) (1) adopts a bitopic pose at one protomer of a dopamine D₂ receptor (D₂R) dimer to negatively modulate the binding of dopamine at the other protomer. The 1H-indole-2-carboxamide moiety of 1 extends into a secondary pocket between the extracellular ends of TM2 and TM7 within the D₂R protomer. To target this putative allosteric site, we generated and characterized fragments that include and extend from the 1H-indole-2-carboxamide moiety of 1. N-Isopropyl-1H-indole-2-carboxamide (3) displayed allosteric pharmacology and sensitivity to mutations of the same residues at the top of TM2 as was observed for 1. Using 3 as an "allosteric lead", we designed and synthesized an extensive fragment library to generate novel SAR and identify N-butyl-1H-indole-2-carboxamide (11d), which displayed both increased negative cooperativity and affinity for the D₂R. These data illustrate that fragmentation of extended compounds can expose fragments with purely allosteric pharmacology.

Full text of DOI:10.1021/acs.jmedchem.5b00585

Subscriber access provided by RMIT University Library
Article
Discovery of a Novel Class of Negative Allosteric Modulator of the
2
Dopamine D Receptor Through Fragmentation of a Bitopic Ligand
Shailesh N Mistry, Jeremy Shonberg, Christopher J. Draper-Joyce, Carmen Klein Herenbrink, Mayako
Michino, Lei Shi, Arthur Christopoulos, Ben Capuano, Peter J. Scammells, and J. Robert Lane
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00585 • Publication Date (Web): 10 Aug 2015
Downloaded from http://pubs.acs.org on August 17, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
Discovery of a Novel Class of Negative Allosteric
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Modulator of the Dopamine D Receptor Through
2
Fragmentation of a Bitopic Ligand
†
,^
†,#
‡
Shailesh N. Mistry, Jeremy Shonberg, Christopher J. DraperꢀJoyce, Carmen Klein
‡
§
§, ,
‡
†
∥
⊥
Herenbrink, Mayako Michino , Lei Shi,
Arthur Christopoulos, Ben Capuano, Peter
†,*
‡,*
J. Scammells and J. Robert Lane,
†
Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville
3
052, Victoria, Australia.
‡
Drug Discovery Biology, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville
3
052, Victoria, Australia.
§
Computational Chemistry and Molecular Biophysics Unit, National Institute on Drug Abuse –
Intramural Research Program, National Institutes of Health, Baltimore, MD, USA.
∥Department of Physiology and Biophysics, Weill Medical College of Cornell University, New York,
NY, USA.
⊥Institute for Computational Biomedicine, Weill Medical College of Cornell University, New York,
NY, USA
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
Recently, we have demonstrated that Nꢀ((trans)ꢀ4ꢀ(2ꢀ(7ꢀcyanoꢀ3,4ꢀdihydroisoquinolinꢀ2(1H)ꢀ
yl)ethyl)cyclohexyl)ꢀ1Hꢀindoleꢀ2ꢀcarboxamide (SB269652) (1), adopts a bitopic pose at one protomer
of a dopamine D receptor (D R) dimer, to negatively modulate the binding of dopamine at the other
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
protomer. The 1Hꢀindoleꢀ2ꢀcarboxamide moiety of 1 extends into a secondary pocket between the
extracellular ends of TM2 and TM7 within the D R protomer. To target this putative allosteric site, we
2
generated and characterized fragments that include and extend from the 1Hꢀindoleꢀ2ꢀcarboxamide
moiety of 1. NꢀIsopropylꢀ1Hꢀindoleꢀ2ꢀcarboxamide (3) displayed allosteric pharmacology and
sensitivity to mutations of the same residues at the top of TM2 as was observed for 1. Using 3 as an
‘allosteric lead’, we designed and synthesised an extensive fragment library to generate novel SAR and
identify Nꢀbutylꢀ1Hꢀindoleꢀ2ꢀcarboxamide (11d), which displayed both increased negative
cooperativity and affinity for the D R. These data illustrate that fragmentation of extended compounds
2
can expose fragments with purely allosteric pharmacology.
ACS Paragon Plus Environment

Page 3 of 82
Journal of Medicinal Chemistry
1
2
3
INTRODUCTION
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
G proteinꢀcoupled receptors (GPCRs) are the largest superfamily of cell surface receptors and are
1
targeted by approximately one third of current medicines. There has been an increasing interest in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
targeting allosteric sites within these proteins (sites that are topographically distinct from the
orthosteric binding site of the endogenous hormone or neurotransmitter) to achieve greater selectivity
2
across receptor subtypes. More recently, bitopic ligands, i.e. molecules in which orthosteric and
allosteric pharmacophores have been linked together, have emerged as a new approach to develop
3
,4
selective GPCR ligands. By concomitantly engaging both orthosteric and allosteric sites, bitopic
ligands combine the advantages of selectivity that can result from engagement of an allosteric site, with
the high affinity and wellꢀdefined structureꢀactivity relationships associated with targeting an
3
,5
orthosteric binding pocket. Such an approach has been successfully exploited to target GPCRs within
6ꢀ9
the muscarinic acetylcholine receptor (mAChR) and adenosine receptor families. In addition, recent
studies have demonstrated that ligands that selectively target the M and M mAChRs may achieve
1
2
1
0,11
their receptorꢀsubtypeꢀselectivity through a hitherto unappreciated bitopic mode of interaction.
One
approach to explore the binding mode of bitopic ligands is to isolate both orthosteric and allosteric
1
0,11
fragments from the parent structure.
This approach highlights the possibility that one may discover
novel allosteric ligands for GPCRs by generating fragments of extended ligands that would be expected
to solely interact with an allosteric pocket rather than the orthosteric pocket.
The dopamine D ꢀD receptors mediate the physiological functions of the catecholamine
1
5
1
2
neurotransmitter, dopamine. Many central nervous system diseases are treated with drugs that bind to
the D ꢀlike dopamine receptors (i.e. the D , D D and D dopamine receptors). Drug discovery at
2
2S
2L,
3
4
these receptors has focused on targeting the orthosteric site with agonists used for the treatment of the
motor symptoms of Parkinson’s disease and antagonists or low efficacy partial agonists for the
treatment of schizophrenia. However, allosteric targeting of these receptors may offer several
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
advantages, including maintenance of spatiotemporal patterns associated with endogenous
neurohumoral signalling. However, despite such theoretical advantages this approach has yet to be
exploited therapeutically in part due to the paucity of both allosteric scaffolds for this receptor and
knowledge of the location of allosteric sites that can be targeted with small molecules.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Structures of dopamine receptor ligands; 1 ꢀ negative allosteric modulator of the D R;
2
Eticlopride – selective D R/D R antagonist; Rꢀ22 – selective D R antagonist.
2
3
3
Recently,
we
demonstrated
that
Nꢀ((trans)ꢀ4ꢀ(2ꢀ(7ꢀcyanoꢀ3,4ꢀdihydroisoquinolinꢀ2(1H)ꢀ
yl)ethyl)cyclohexyl)ꢀ1Hꢀindoleꢀ2ꢀcarboxamide (SB269652, 1, Figure 1) is a negative allosteric
modulator of the D R. We presented a novel mechanism of action whereby 1 adopts a bitopic pose at
2
1
3
one protomer of a D R dimer to modulate the binding of dopamine at the other protomer. We
2
generated fragments of 1 that included the 7ꢀcyanoꢀ1,2,3,4ꢀtetrahydroisoquinoline (7ꢀCTHIQ) headꢀ
group of 1 to demonstrate that these fragments exhibited competitive antagonism of dopamine and,
therefore, that this moiety occupies the orthosteric site of the D R. The crystal structure of the D R
2
3
bound to the antagonist: (S)ꢀ3ꢀchloroꢀ5ꢀethylꢀNꢀ((1ꢀethylpyrrolidinꢀ2ꢀyl)methyl)ꢀ6ꢀhydroxyꢀ2ꢀ
methoxybenzamide (eticlopride, Figure 1) and a number of followꢀup studies revealed a secondary
pocket between transmembrane domains (TMs) 2 and 7 in the receptor structure that can be exploited
by extended ligands such as (R)ꢀNꢀ(4ꢀ(4ꢀ(2,3ꢀdichlorophenyl)piperazinꢀ1ꢀyl)ꢀ3ꢀhydroxybutyl)ꢀ1Hꢀ
1
4ꢀ16
indoleꢀ2ꢀcarboxamide (Rꢀ22, Figure 1) to achieve subtype selectivity.
These studies also revealed
ACS Paragon Plus Environment

Page 5 of 82
Journal of Medicinal Chemistry
1
2
3
that such ligands extend into a similar secondary pocket in the D R. Using mutagenesis in combination
2
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with derivatives of 1 we demonstrated that the indoleꢀ2ꢀcarboxamide moiety of 1 extends into a
secondary pocket between the extracellular ends of TM2 and TM7 of the D R. We hypothesize that this
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
secondary pocket may an allosteric site that can targeted by allosteric ligands that do not concomitantly
engage the orthosteric site unlike the bitopic ligand 1. In this study we design and synthesise a library
of fragments derived from 1 that include and extend from the indoleꢀ2ꢀcarboxamide moiety of 1 to
probe the SAR of this putative allosteric site. NꢀIsopropylꢀ1Hꢀindoleꢀ2ꢀcarboxamide (3), the smallest
fragment of 1 containing the indoleꢀ2ꢀcarboxamide moiety, displays allosteric pharmacology with
micromolar affinity for the D R. We used mutagenesis to demonstrate that this fragment binds in a
2
similar allosteric pocket as the indoleꢀ2ꢀcarboxamide moiety of 1. We then performed an SAR analysis
using 3 as our lead compound to identify Nꢀbutylꢀ1Hꢀindoleꢀ2ꢀcarboxamide (11d), which displayed
both 10ꢀfold increased negative cooperativity and 7ꢀfold increased affinity for the D R. As such, we
2
report indoleꢀ2ꢀcarboxamides as a novel scaffold, and attractive starting point for the development of
negative allosteric modulators of the D R that target an allosteric pocket within the D R between TM2
2
2
and TM7. These data further confirm the bitopic mode of interaction of 1 and illustrate that
fragmentation of extended compounds such as 1 can reveal novel allosteric scaffolds.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
Chemistry. The synthesis of the smallest fragment of 1, Nꢀisopropylꢀ1Hꢀindoleꢀ2ꢀcarboxamide (3)
was achieved in a good yield, through coupling of commercially available 1Hꢀindoleꢀ2ꢀcarboxylic acid
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(2) and isopropylamine, employing Oꢀ(1Hꢀ6ꢀchlorobenzotriazoleꢀ1ꢀyl)ꢀ1,1,3,3ꢀtetramethyluronium
hexafluorophosphate (HCTU) as a coupling reagent (Scheme 1).
a
Scheme 1. Synthesis of fragments of 1
O
O
(
a)
HO
HN
N
H
HN
2
3
R2
R2
(f)
N
R1
R2
(d)
O
N
R2
O
N
NHBoc
NHR3
H
HN
1
2
3
2
4
5
6
R = OH
7a R = Me; R = Boc; HCl salt
7b R = Pr; R = Boc
9a R = Me; HCl salt
(b)
2
3
2
1
9b R = Pr
R = OEt
(e)
(c)
2
3
1
8a R = Me; R = H; diꢀTFA salt
b R = Pr; R = H; diꢀTFA salt
R = H
2
3
8
a
Reagents and conditions: (a) HCTU, isopropylamine, DMF, 75%; (b) i. EtI, K CO , MeCN, rt to
2
3
reflux, ii. 50 °C, iii. reflux, 100%; (c) i. dry toluene, ꢀ78 °C, ii. DIBALꢀH, ꢀ78 °C, iii. MeOH, ꢀ78 °C,
2
1
00%; (d) (R ) NH, NaBH(OAc) , 1,2ꢀdichloroethane, rt, 50ꢀ56%; in the case of 7a, the product was
2 3
converted to the corresponding HCl salt form using 1 M HCl/Et O (e) TFA/DCM (1:3), rt, 96ꢀ99%; (f)
DIPEA, BOP, DCM, 36ꢀ49%; in the case of 9a, the product was converted to the corresponding HCl
2
salt form using 1 M HCl (aq)
.
The larger fragments of 1 (compounds 9aꢀb) were accessible through a fiveꢀstep procedure, initially
starting with 2ꢀ((trans)ꢀ4ꢀ((tertꢀbutoxycarbonyl)amino)cyclohexyl)acetic acid (4). Carboxylic acid 4
underwent alkylation with EtI in the presence of K CO under reflux conditions in MeCN, to form the
2
3
corresponding ethyl ester 5 in quantitative yield. The partial reduction of 5 was achieved cleanly in the
ACS Paragon Plus Environment

Page 7 of 82
Journal of Medicinal Chemistry
1
2
3
presence of DIBALꢀH, with quenching at ꢀ78°C, to afford aldehyde 6 in quantitative yield. Reductive
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
amination of 6 with either dimethylamine or dipropylamine using NaBH(OAc) in 1,2ꢀdichloroethane,
3
gave tertiary amine derivatives 7a and 7b, respectively. Compounds 7aꢀb underwent NꢀBoc
deprotection in the presence of TFA/DCM (1:3), to afford diamines 8aꢀb in the bisꢀtrifluoroacetate salt
form. Finally, coupling of 8aꢀb with 1Hꢀindoleꢀ2ꢀcarboxylic acid (1), employing BOP as the coupling
reagent, furnished the desired fragment compounds 9aꢀb (Scheme 1). In the case of 9a, formation of
the HCl salt in 1 M HCl(aq) was carried out prior to pharmacological evaluation.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 2. Synthesis of Nꢀsubstitutedꢀ1Hꢀindoleꢀ2ꢀcarboxamides
k azetidinꢀ1ꢀyl
pyrrolidinꢀ1ꢀyl
m piperidinꢀ1ꢀyl
n morpholinꢀ4ꢀyl
o cyclopropylamino
p cyclobutylamino
q cyclopentylamino
O
O
O
4
R =
(a)
l
a NH2
b NHEt
c NH Pr
d NH Bu
N
OH
N
H
H
n
2
10
n
e NHCH CH=CH
2
2
(
c)
(b)
f
NHCH CH OH
r
cyclohexylamino
s (piperidinꢀ1ꢀyl)amino
phenylamino
2
2
g NHCH(Et)
h NH Bu
2
O
t
t
i
j
N(Me)
N( Pr)
2
u (cyclopropylmethyl)amino
v (cyclohexylmethyl)amino
w benzylamino
R4
2
N
i
H
11a-w
a
Reagents and conditions: (a) MeOH, concentrated H SO , reflux, 98%; (b) amine/amine solution,
2
4
EtOH, rt, 34ꢀ76%; (c) HCTU, DMF, excess amine or stoichiometric amine and DIPEA, rt, 34ꢀ81%.
Our initial library of compounds focused on analogues of 3 varied at the carboxamide moiety.
Synthesis of target amides 11aꢀw was achieved by either HCTUꢀmediated coupling (11gꢀh, jꢀw), with
a series of commercially available amines (in a similar fashion to the synthesis of 3), or by direct
aminolysis of the intermediate methyl ester 10 (Scheme 2). Methyl ester 10 was obtained from
carboxylic acid 2 in nearꢀquantitative yield, through esterification by reflux in MeOH, in the presence
of catalytic concentrated H SO . Subsequent direct aminolysis of 10 carried out in EtOH at rt, in the
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
presence of excess amine or amine solution, afforded 11aꢀf and 11i in modest to good yield. This
extensive library of 1Hꢀindoleꢀ2ꢀcarboxamides was comprised of compounds designed to interrogate
the effect of introducing increasing size of both linear and branched alkyl chains, saturated and
aromatic carbocycles and also amides with varying levels of Nꢀsubstitution.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The systematic introduction of the electronegative element fluorine into various positions on
aromatic ring scaffold, known as a “fluorine walk”, has in the past uncovered derivatives of the parent
compound with improved potency or physicochemical characteristics; in particular this strategy was
employed in SAR studies of the prototypical positive allosteric modulator of the M mAChR –
1
1
7,18
BQCA.
As a second area of investigation, we sought to generate a small library of monoꢀfluoroꢀ
substituted Nꢀisopropylꢀ1Hꢀindoleꢀ2ꢀcarboxamides (13aꢀe), in addition we synthesized the
corresponding 5ꢀmethoxy analogue of 3 (Scheme 3).
a
Scheme 3. Synthesis of substituted Nꢀisopropylꢀ1Hꢀindoleꢀ2ꢀcarboxamide
F
O
O
(a)
N
HN
N
H
HN
H
3
13a
O
O
(b)
R5
R5
N
OH
N
H
HN
H
1
1
1
2b R = 4ꢀF
2c R = 5ꢀF
2d R = 6ꢀF
13b R = 4ꢀF
13c R = 5ꢀF
13d R = 6ꢀF
13e R = 7ꢀF
12e R = 7ꢀF
14 R = 5ꢀOMe
15 R = 5ꢀOMe
a
Reagents and conditions: (a) i. Sat. NaHCO3 (aq), MeCN, 0 °C, ii. 1ꢀChloromethylꢀ4ꢀfluoroꢀ1,4ꢀ
diazoniabicyclo[2.2.2]octane bis(tetrafluoroborate), rt, 46%; (b) HCTU, isopropylamine, DMF, rt, 7ꢀ
80%.
ACS Paragon Plus Environment

Page 9 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
3
ꢀFluoroꢀNꢀisopropylꢀ1Hꢀindoleꢀ2ꢀcarboxamide (13a) was accessed in moderate yield, through
fluorination of
3
using the electrophilic fluorine source 1ꢀchloromethylꢀ4ꢀfluoroꢀ1,4ꢀ
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
diazoniabicyclo[2.2.2]octane bis(tetrafluoroborate), adapting literature methodology. The remaining
fluoroꢀ and methoxyꢀsubstituted analogues (13bꢀe, 15) were obtained through HCTUꢀmediated
coupling the corresponding commercially available carboxylic acid precursors (12bꢀe, 14) with
isopropylamine as described above.
a
Scheme 4. Synthesis of Nꢀisopropyl carboxamideꢀsubstituted heterocycles
X
Y
O
X
Y
O
(a)
OH
HN
Z
Z
16 X = Z = CH; Y = NMe
17 X = Z = CH; Y = NMe
19 X = N; Y = NH; Z = CH
21 X = N; Y = O; Z = CH
23a X = Z = CH; Y = O
1
2
2
2
2
8
0
X = N; Y = NH; Z = CH
X = N; Y = O; Z = CH
2a X = Z = CH; Y = O
4a X = Z = CH; Y = S
6
25a X = Z = CH; Y = S
X = CH; Y = NH; Z = N
27 X = CH; Y = NH; Z = N
O
O
OH
NH
(a)
Y
Y
2
2
2b Y = O
4b Y = S
23b Y = O
25b Y = S
O
O
O
O
(b)
(a)
N
HO
N
NH
30a-d
N
H
H
H
2
8a-d
29a-d
Substituent position: a = 4; b = 5; c = 6; d = 7
a
Reagents and conditions: (a) HCTU, isopropylamine, DMF, rt, 22ꢀ90%; (b) LiOH.H O, THF/water
2
(1:1), rt, 78ꢀ100%;
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Our final series of fragment analogues used a series of aromatic core replacements to ascertain
whether the allosteric modulatory activity of 3 could be affected by the electronic nature of the bicyclic
heteroaromatic ring, or the presence of the hydrogenꢀbond donor moiety of the indole (Scheme 4). In
addition, we also wanted to investigate positional isomers of 3, where the attachment point of the Nꢀ
isopropylcarboxamide was varied around the indole scaffold (Scheme 4).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The Nꢀmethylꢀ1Hꢀindole, benzo[d]imidazole, benzo[d]oxazole, benzofuran, benzo[b]thiophene and
pyrrolo[2,3ꢀb]pyridine analogues of 3 (compounds 17, 19, 21, 23a, 25a and 27, respectively) were once
again easily synthesized through HCTUꢀmediated coupling of the parent carboxylic acid with
isopropylamine. Considered to be of interest also, were the 3ꢀsubstituted benzofuran,
benzo[b]thiophene analogues (23b and 25b) which were obtained in the same manner.
Finally, synthesis of the series of positional isomers of 3 was carried out in a twoꢀstep procedure.
Attempted direct aminolysis of commercially available methyl esters 28aꢀd proved unsuccessful,
therefore, initial basic hydrolysis of the esters to give the corresponding carboxylic acids 29aꢀd was
carried out. Final HCTUꢀmediated coupling of 29aꢀd with isopropylamine gave the desired
carboxamides 30aꢀd (Scheme 4).
Pharmacological analysis of the series of carboxamide fragments revealed the Nꢀbutylꢀ1Hꢀindoleꢀ2ꢀ
carboxamide 11d, to have higher affinity for the D R and greater negative cooperativity with dopamine
2
as compared to the parent fragment 3 (see below). With this in mind, we sought to synthesise an
analogue of 1 incorporating the structure of 11d (Scheme 5), anticipating that combining the 7ꢀCTHIQ
and 11d moieties would proffer a higher affinity D R bitopic ligand. The target compound 36 was
2
2
0
prepared as described previously (Scheme 5).
ACS Paragon Plus Environment

Page 11 of 82
Journal of Medicinal Chemistry
1
2
3
a
Scheme 5. Synthesis of a prospective higher affinity bitopic ligand incorporating fragment 11d
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
c)
R6
R7
N
R7
NC
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
7
7
3
3
1 R = OH; R = NH
34 R = NHBoc
2
(
a)
b)
(d)
6
7
7
2 R = OH; R = NHBoc
35 R = NH2
(
6
7
33 R = OMs; R = NHBoc
(e)
N
O
NC
N
H
HN
3
6
a
Reagents and conditions: (a) Boc O, TEA, DCM, rt, 82%; (b) MsCl, TEA, DCM, rt, 93%; (c) 7ꢀ
CTHIQ, DIPEA, DCM, reflux, 12%; (d) i. TFA/DCM (1:5), rt; ii. neutralization; (e) 2, HCTU, DIPEA,
2
DMF, rt, 69%.
Pharmacology.
Fragments of 1 containing the indoleꢀ2ꢀcarboxamide moiety display allosteric pharmacology. In
previous work we demonstrated that 1 binds as a bitopic ligand to one protomer of a homodimer of the
1
3
D R to modulate the binding of a ligand to the orthosteric site of a second protomer. We previously
2
identified purely orthosteric fragments of 1 containing the 7ꢀCTHIQ moiety that displayed competitive
pharmacology with dopamine. This core group comprises the key elements expected to interact with
the orthosteric binding site of aminergic receptors (i.e. an aromatic ring in close proximity to a charged
nitrogen at physiological pH). In the current study we have derived fragments extending from the
indoleꢀ2ꢀcarboxamide moiety of 1 (distal to the 7ꢀCTHIQ orthosteric headꢀgroup), with the aim to
provide further validation of the bitopic mode of action of 1. In addition we wanted to examine whether
the approach of dividing bitopic ligands into discrete fragments can unmask novel purely allosteric
scaffolds.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
5
a)
Figure 2
[
S] GTPγS
[3] (ꢁM)
1
1
20
00
80
0
3
10
30
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
4
2
0
0
0
0
100
00
3
ꢀ
9
ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3 ꢀ2
Log[dopamine] (M)
b) cAMP
c)
e)
g)
pERK1/2
[3] (ꢁM)
8
0
0
3
60
1
3
1
0
0
00
4
2
0
0
0
300
ꢀ
11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4
Log[dopamine] (M)
d)
Radioligand binding
Radioligand binding
1
1
20
00
80
[3] (ꢀM)
1
20
0
3
10
30
1
100
8
6
4
0
0
0
1
3
6
4
2
0
0
0
0
00
00
3
20
0
ꢀ8
ꢀ7
ꢀ6
ꢀ5
ꢀ4
ꢀ3
ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3 ꢀ2 ꢀ1
Log[dopamine] (M)
Log[ligand] (M)
f) Gα activation: D R
Gα activation: D R
o1 3
o1
2
[3] (ꢀM)
1
20
120
100
80
60
40
20
0
[
3] (ꢀM)
0
3
10
30
100
300
1
00
0
3
1
3
1
8
6
0
0
0
0
00
40
20
0
300
ꢀ
12 ꢀ11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4
Log[dopamine] (M)
ꢀ12 ꢀ11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4
Log[dopamine] (M)
ACS Paragon Plus Environment

Page 13 of 82
Journal of Medicinal Chemistry
1
2
3
Figure 2. A fragment of 1 containing the indoleꢀ2ꢀcarboxamide pharmacophore (3) inhibited the action
of dopamine D R in a nonꢀcompetitive manner, in functional assays measuring: (a) D R mediated
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
35
[ S]GTPγS binding; (b) inhibition of forskolin stimulated cAMP; (c) ERK1/2 phosphorylation. Data
35
were fitted using an allosteric ternary complex model ([ S]GTPγS and cAMP, Equation 4) or an
operational model of allosterism (pERK1/2, Equation 5), to derive values of affinity and cooperativity
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
Table 1). (d) Both 1 and 3 caused partial displacement of the binding of [ H]spiperone in an
equilibrium binding assay consistent with an allosteric mode of action. (e) Dopamineꢀmediated
3
inhibition of the equilibrium binding of [ H]spiperone in the absence or presence of 3. Data points
represent the mean ± SEM of three experiments performed in duplicate. Curves drawn through the data
points represent the best fit of a twoꢀsite allosteric ternary complex model (Equation 1, Table 1).
Compound 3 interacts with the D R (f) and the D R (g) in an allosteric manner in an assay measuring
2
3
Ga G protein activation.
oB
Table 1. Characterization of fragments of 1 containing the indoleꢀ2ꢀcarboxamide moiety in terms
a
of binding and function at the D R
2
35
3
[
S]GTPγS
Logαβ
(αβ)
cAMP
Logαβ
(αβ)
pERK1/2
Logα
[ H]spiperone binding
pK
B
pK
B
pK
B
pK
B
Logα
Logα
′
Logβ
ꢀ100
ꢀ100
(
K
B
, ꢁM)
(K
B
, ꢁM)
(K
B
, ꢁM)
(α)
(K
B
, ꢁM)
(α)
(α
′
)
4
.89 ± 0.29 ꢀ0.89 ± 0.17 4.72 ± 0.29 ꢀ0.54 ± 0.09 4.35 ± 0.06 ꢀ0.53 ± 0.13
4.88 ± 0.17
(13)
ꢀ0.51 ± 0.04
(0.31)
ꢀ0.53 ± 0.04
(0.30)
3
(13)
.54 ± 0.4 ꢀ0.84 ± 0.15
3) (0.14)
(0.13)
(19)
ND
(0.28)
ND
(45)
5.54 ± 0.4 ꢀ0.61 ± 0.14
(3) (0.25)
(0.30)
5
9
a
ꢀ
ꢀ
ꢀ
(
a
The ability of 3 or 9a to modulate the functional effect of agonists or their affinity for the D R
2
35
expressed in FlpꢀIN CHO cells was tested at various signaling endpoints. For [ S]GTPγS and cAMP
assays data was fitted to an allosteric ternary complex model (Equation 4) whereas for pERK1/2
assays data was fit to an operational model of allosterism (Equation 5) to derive values of affinity
(
pK ) and cooperativity. Radioligand binding data were fitted with Equation 1 to derive values of
B
3
cooperativity with dopamine (Logα) and [ H]spiperone binding (Logα′)
Our previous study demonstrated that the indoleꢀ2ꢀcarboxamide moiety of 1 extended into a
13
secondary ‘allosteric’ pocket within the D R. Accordingly, we first synthesised the fragment 3,
2
(Scheme 1, Figure 2a) with the hypothesis that this fragment would act as a modulator of dopamine
binding and/or function. Our characterization of 3 revealed this small molecule inhibited the action of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dopamine in a manner that was best fit by an allosteric model in both functional and ligand binding
35
assays. When an interaction study with dopamine was performed in either a [ S]GTPγS assay (Figure
2a) or an assay measuring D Rꢀmediated inhibition of forskolinꢀstimulated cAMP production (Figure
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
b), 3 caused a limited 8ꢀfold or 3ꢀfold reduction in agonist potency, respectively (Figure 2b).
Application of an allosteric ternary complex model to the concentration response data yielded estimates
3
5
of affinity for the unoccupied receptor ([ S]GTPγS assay, K = 13 ꢁM: cAMP assay, K = 19 ꢁM) and
B
B
35
its cooperativity with dopamine ([ S]GTPγS assay, αβ = 0.13: cAMP assay, αβ = 0.29; where α
represents cooperativity with dopamine affinity, β represents modulatory effect upon dopamine
efficacy and values less than 1 indicate negative cooperativity, Table 1). This pattern was distinct from
that observed in an assay measuring ERK1/2 phosphorylation, in which 3 caused both a reduction in
dopamine potency and a significant decrease in maximal response (Figure 2c). This latter observation
21
is consistent with negative modulation of orthosteric ligand efficacy. Accordingly, data were fitted
using an operational model of allosterism whereby Logβ was fixed to ꢀ100 to reflect the high negative
modulatory effect of 3 upon dopamine efficacy. This analysis allowed estimation of an affinity value
(
K = 46 ꢁM) for 3 and its cooperativity with dopamine binding (α = 0.29) (Table 1). However,
B
because of the low affinity of 3 and its relatively weak modulatory effect upon dopamine potency we
sought additional evidence of an allosteric mode of interaction. Therefore, we extended our
characterization to a radioligandꢀbinding assay. We first tested the ability of 3 to antagonize
3
[ H]spiperone and found that increasing concentrations of 3 produced only a partial displacement of
this radioligand consistent with an allosteric mode of interaction (Figure 2d). We extended these
studies to perform a binding interaction study to investigate the effect of increasing concentrations of 3
3
on the affinity of both dopamine and the radiolabeled antagonist [ H]spiperone. The value of affinity
derived for 3 (K = 13 ꢁM) was consistent with that determined in each functional assay (Figure 2e,
B
3
Table 1). In addition, 3 was a negative modulator of both dopamine (α = 0.31) and [ H]spiperone (α =
ACS Paragon Plus Environment

Page 15 of 82
Journal of Medicinal Chemistry
1
2
3
0
.30) affinity. These values indicate that, whilst the magnitude of negative allosteric modulation of
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dopamine affinity is consistent across all assays, the decrease of dopamine Emax in the pERK1/2 assay
is consistent with additional effects of 3 upon dopamine efficacy. Such a distinct pattern of action may
indicate that 3 differentially antagonizes the effect of dopamine at different signalling endpoints.
However, the sensitivity of the respective assays and the different assay conditions of the various
assays may also underlie this effect although it is interesting to note that the parent compound, 1,
demonstrated a profile of modulation whereby the degree of negative cooperativity with dopamine
could be attributed solely to an effect on dopamine affinity and was not significantly different across
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4
these different pathways. Compound 1 has been shown to display approximately 50ꢀfold selectivity
22,23
for the D R over the D R.
Furthermore, it was postulated that this subtype selectivity was driven
3
2
1
5
through interaction of the indoleꢀ2ꢀcarboxamide moiety of 1 with an extended pocket. The D R has
3
been shown to selectively couple to the Gα G protein whereas the D R can couple to all members of
o1
2
2
4
the Gα G protein family. Accordingly, we assessed the action of 3 at the D R (Figure 2f) compared
i/o
2
to the D R (Figure 2g) in an assay measuring activation of Gα G protein using a bioluminescence
3
o1
25
resonance energy transfer (BRET) biosensor. This revealed that 3 displays both affinity (D R pK =
2
B
4
.84 ± 0.27 (K = 14 ꢁM): D R, pK = 5.05 ± 0.21 (K = 9 ꢁM)) and negative cooperativity (D R, logα
B 3 B B 2
=
ꢀ0.77 ± 0.14 (α = 0.17): D R, logα = ꢀ1.13 ± 0.16 (α = 0.07) that is not significantly different between
3
the D R and D R.
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 3
a)
35
[ S] GTPγS
[
9a] (ꢁM)
1
1
20
00
80
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
6
4
0
0
10
30
100
20
0
ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3
Log[dopamine] (M)
b) pERK1/
c)
Radioligand binding
[
9a] (ꢀM)
[9a] (ꢁM)
2
1
20
80
0
1
0
3
100
60
40
20
0
8
6
4
2
0
0
0
0
0
3
1
3
1
3
0
0
00
00
10
3
0
100
ꢀ
11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4
Log[dopamine] (M)
ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3 ꢀ2 ꢀ1
Log[dopamine] (M)
Figure 3. A fragment of 1 containing the indoleꢀ2ꢀcarboxamide pharmacophore and extending to a
tertiary amine (emulating the position of that present on the 7ꢀCTHIQ moiety) interacts with the D R in
a mixed orthosteric/allosteric manner, in functional assays measuring: (a) D R mediated [ S]GTPγS
binding, and (b) ERK1/2 phosphorylation, 9a inhibited the action of dopamine in a nonꢀcompetitive
manner. Data were fitted using an allosteric ternary complex model ([ S]GTP S, Equation 4) or an
operational model of allosterism (pERK1/2, Equation 5) (Table 1). (c) Dopamineꢀmediated inhibition
of the equilibrium binding of [ H]spiperone in the absence or presence of 9a. Data points represent the
mean ± SEM of three experiments performed in duplicate. Curves drawn through the data points
2
35
2
35
γ
3
2
6
represent the best fit of a twoꢀsite binding model.
One would also expect that extension of the purely allosteric pharmacophore to include structural
features of the orthosteric moiety, incorporating a suitable linker moiety in between, would result in the
generation of ligands that displayed competitive behaviour with an orthosteric ligand. We generated
additional fragments of 1 that extended away from the indoleꢀ2ꢀcarboxamide moiety and included a
ACS Paragon Plus Environment

Page 17 of 82
Journal of Medicinal Chemistry
1
2
3
tertiary amine group (emulating that present on the 7ꢀCTHIQ moiety of 1). Characterisation of 9a
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
5
(
Figure 3, Scheme 1) was consistent with either an allosteric mode of interaction ([ S]GTPγS (Figure
3a) and pERK1/2 (Figure 3b) or competitive mode of interaction (radioligand binding) (Figure 3c,
Table 1). This discrepancy is most likely due to the larger molecular size of spiperone as compared to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
dopamine whereby 9a may be able to occupy the D R simultaneously with dopamine but not with
2
spiperone. Of interest, the allosteric mode of interaction suggests that the protonated amino group of 9a
does not project into the area occupied the protonated amino of dopamine. If one considers the
competitive action of fragments of 1 that contain the 7ꢀCTHIQ moiety, this suggests that the orientation
of the amino group of 9a and the amino group within the 7ꢀCTHIQ moiety of 1 must be different.
These data also highlight the proximity of the orthosteric and allosteric pockets.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a) pERK1/2
cAMP
b)
d)
[
9b] (ꢁM)
8
6
4
2
0
0
0
0
0
0
1
3
1
3
1
0
0
00
ꢀ
11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4
Log[dopamine] (M)
c)
Radioligand binding
Radioligand binding
[9b] (ꢁM)
120
00
120
0
1
3
10
1
1
00
8
0
8
6
4
2
0
0
0
0
0
60
1
3
3
1
0
00
4
0
9
a
20
0
9b
ꢀ
9
ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3 ꢀ2
Log[ligand] (M)
ꢀ
9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3 ꢀ2 ꢀ1
Log[dopamine] (M)
Figure 4. A more extensive fragment of 1 containing the indoleꢀ2ꢀcarboxamide pharmacophore and
extending to a tertiary amine (emulating the position of that present on the 7ꢀCTHIQ moiety, but with
longer alkyl chains compared to 9a) is a partial agonist at the D R. To investigate the mode of
interaction of 1 with the D R we synthesised compound 9b. In functional assays measuring: (a) D Rꢀ
mediated inhibition of forskolin stimulated cAMP production, and (b) ERK1/2 phosphorylation, 9b
behaved as a partial agonist. In the pERK1/2 assay, 9b inhibited the action of dopamine in a nonꢀ
competitive manner decreasing Emax but also displayed agonism in its own right. In contrast in the
cAMP assay, 9b behaved as a competitive partial agonist. (c) Dopamineꢀmediated inhibition of the
2
2
2
3
equilibrium binding of [ H]spiperone in the absence or presence of 9b. Data points represent the mean
±
SEM of three experiments performed in duplicate. Curves drawn through the data points represent
2
6
the best fit of a twoꢀsite binding model (Supplementary Table 1). (d) The ability of 1 and fragments
3
of 1 to displace [ H]spiperone at the D R. Both 1 (pK = 6.29 ± 0.07, Logα = ꢀ0.46 ± 0.02) and 3 (pK
2
B
B
=
4.75 ± 0.23, LogαCompound 3 = ꢀ0.23 ± 0.03) demonstrate an inability to completely displace
Compound 3
ACS Paragon Plus Environment

Page 19 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
[
H]spiperone in agreement with an allosteric mode of interaction (Equation 2). In contrast, data of
both 9a and 9b, which possess a tertiary amine group (positioned atomically the same distance as that
2
6
present on the 7ꢀCTHIQ of 1), were best fitted by a oneꢀsite competitive model, with estimated
affinity of pKI_Compound 9b = 4.70 ± 0.03.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Extension of the fragment further to generate 9b (Figure 4, Scheme 1) yielded a low affinity partial
agonist with a strictly competitive mode of interaction with dopamine in all assays, with the exception
of the pERK1/2 assay, in which a concentrationꢀdependent decrease in maximal agonist effect was
observed (Figure 4). This suggests that the extended alkyl chains of 9b either reorientate this ligand
relative to the position of 9a so that the protonated amino moiety now occupies and competes directly
with the amino moiety of dopamine. Alternatively, if the orientation of 9b is equivalent to that of 9a
whereby the protonated amine does not compete with that of dopamine the additional length of the
alkyl chain substitutions may extend into the binding pocket of dopamine. In either case agonism
appears to be concomitant with competition with dopamine.
Compound 3 is modulated by the mutation of residues within a secondary binding pocket of the
D R. While the ability to separate orthosteric and allosteric fragments from 1 is consistent with its
2
bitopic mode of interaction at the D R, it does not necessarily follow that the allosteric fragment 3 will
2
occupy the same binding pocket as that of the corresponding indoleꢀ2ꢀcarboxamide moiety of 1.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a)
1
E95A: decrease in affinity &
cooperativity
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
V91A:
decrease
in
cooperativity
c)
b)
V91A
WT
[3] (ꢀM)
[
3] (ꢁM)
8
0
8
6
4
2
0
0
0
0
0
0
3
0
3
60
40
20
0
1
3
0
0
1
3
1
0
0
00
100
300
300
ꢀ
11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4
Log[dopamine] (M)
ꢀ11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3
Log[dopamine] (M)
d)
E95A
[
3] (ꢀM)
8
0
0
3
10
30
100
300
6
4
2
0
0
0
0
ꢀ
11 ꢀ10 ꢀ9 ꢀ8 ꢀ7 ꢀ6 ꢀ5 ꢀ4 ꢀ3
Log[dopamine] (M)
Figure 5. Mutation of residues at the top of TM2 modulates the allosteric action of 3; (a) When docked
into a homology model of the D R the 7ꢀCTHIQ core of 1 occupies the orthosteric pocket. The 1Hꢀ
indoleꢀ2ꢀcarboxamide moiety of 1 extends into a secondary pocket between TM2 and TM7, where the
indolic NH makes a hydrogen bond interaction with Glu95
a hydrophobic interaction with Val91
published ). Mutation of these residues decreases the affinity or negative cooperativity of 1.
2
2.65
and the 1,4ꢀcyclohexylene group makes
,
2
.61
(Figure recreated from modeling studies previously
1
4
ACS Paragon Plus Environment

Page 21 of 82
Journal of Medicinal Chemistry
1
2
3
Interaction studies between 3 and dopamine in an assay measuring D R mediated pERK1/2
phosphorylation using CHO cells expressing the (b) WT, (c) Val91 Ala or (d) Glu95 Ala.
2
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
.61
2.65
Table 2: Characterization of binding and function of Nꢀisopropylꢀ1Hꢀindoleꢀ2ꢀcarboxamide (3) at
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
the D R
2
1
3
c
c
pKB
Logα
(α)
Logβ
(β)
pKB
Logαβ
(αβ)
(K , ꢁM)
B
(
K , ꢁM)
B
Wild Type
6.26 ± 0.09
0.55)
ꢀ1.22 ± 0.11
(0.06)
4.35 ± 0.06
(45)
ꢀ0.53 ± 0.13
(0.29)
ꢀ100
(
2
.61
Val91 Ala
6.26 ± 0.09
(0.55)
ꢀ0.48 ± 0.16
(0.33)
b
ꢀ
ꢀ
ꢀ
2
.65
Glu95 Ala
5.31 ± 0.19
ꢀ0.32 ± 0.14
(0.48)
6.38 ± 0.31
(0.04)
0.12 ± 0.12
(1.31)
ꢀ0.81 ± 0.11
(0.15)
(4.90)
a
The ability of 3 to modulate the functional effect of agonists or their effect at either the WT D R or
2
mutant D Rs expressed in FlpꢀIN CHO cells was tested in a pERK1/2 assay and data was fit to an
2
operational model of allosterism (Equation 5) to derive values of affinity (pK ) and cooperativity with
B
b
c
14
dopamine. no effect of 3 was observed up to concentration of 300 ꢁM. Values taken from .
2
.61
2.65
Our previous study identified two residues at the top of TM2 (Val91
and Glu95 ) that interact
with the indoleꢀ2ꢀcarboxamide moiety of 1 and mutation of these residues to alanine reduced the
1
3
affinity and/or negative cooperativity of 1 (Figure 5a, Table 2). We hypothesised that if 3 occupies
the same binding pocket as the 1Hꢀindoleꢀ2ꢀcarboxamide moiety of 1, then one would expect a similar
2
.61
sensitivity to mutation of these residues. Mutation of Val91 Ala removed any effect of 3 up to a
concentration of 300 ꢁM upon a dopamine in a pERK1/2 assay, indicating this residue is key for 3
binding or function (Figure 5b, Table 2). This mutation caused no change in affinity of 1 but caused a
5
ꢀfold decrease in negative cooperativity with dopamine. This indicates that 1 makes additional
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
interactions, most likely within the orthosteric pocket of the D2R, which are the main contibutors to
2
.65
affinity. In contrast, the mutation of Glu95 Ala caused a 100ꢀfold increase in affinity and a complete
abolition of negative cooperativity with dopamine affinity (Figure 5d, Table 2). In addition, while 3 at
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the WT D R (Figure 2c) displayed high negative modulation of dopamine efficacy (Logβ = ꢀ100) this
2
2.65
modulatory effect was more limited at the Glu95 Ala (β = 0.15) (Figure 5b & d, Table 2). This is
quite different to the effect of this mutation upon 1, where both an 8ꢀfold decrease in negative
cooperativity and a 9ꢀfold decrease in affinity was observed. It is important to note that these
mutations caused no change in the affinity of dopamine. Therefore, these results suggest that 3
occupies a similar binding site between TM2 and TM7 of the D R to the 1Hꢀindoleꢀ2ꢀcarboxamide
2
moiety of 1 although it is important to note the individual contribution of the different residues to
binding affinity and cooperativity differ between 1 and 3. Therefore, we must acknowledge the
possibility that the orientation of 3 within this pocket may be different to that of 1Hꢀindoleꢀ2ꢀ
carboxamide moiety of 1. Furthermore, because we cannot measure binding of 3 directly (i.e with a
radiolabelled derivative) then we must also consider the possibility that the effects of such mutations,
rather than indicating a direct residueꢀligand interaction, may result from an allosteric effect i.e.
mediated by a global change in receptor conformation.
We then conducted a detailed SAR study to understand the structural determinants of the allosteric
action of 3 and to generate derivatives with higher affinity and/or negative cooperativity.
Preliminary investigation into the SAR of indole and related heterocycle carboxamide derivatives
The most complex allosteric behaviour for compound 3 was observed in the pERK1/2 phosphorylation
assay in which increasing concentrations of the modulator caused both a decrease in dopamine affinity
and maximal effect. Moreover, this assay provides a robust, mediumꢀthroughput measurement of D R
2
activation. As such, all derivatives of compound 3 were screened using this assay. We observed four
ACS Paragon Plus Environment

Page 23 of 82
Journal of Medicinal Chemistry
1
2
3
distinct patterns of ligand action across the entire series of compounds (Table 3 & 4). A number of
compounds caused both a rightward displacement of the dopamine concentrationꢀresponse curve and a
decrease in the maximal response of dopamine that did not reach a limit within the concentration range
used. These data were analyzed according to an operational model of allosterism (as for compound 3,
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1) to determine a value of affinity (K ) and binding cooperativity parameter (α) between
B
dopamine and the allosteric ligand (where Logβ was constrained to ꢀ100 to represent high negative
modulation of dopamine efficacy). Other compounds caused a limited rightward shift of the dopamine
doseꢀresponse curve but no decrease in Emax. These data were fit using an operational model of
allosterism where Logβ was constrained to 0 to represent no effect upon dopamine efficacy. A number
of compounds produced an unlimited rightward shift of a dopamine doseꢀresponse curve within the
concentration range of test compound used. Accordingly, data were fit with a competitive model to
allow an estimate of affinity for the D R. However, because the concentration range used to test these
2
compounds was restricted by compound solubility we cannot exclude the possibility that these
compounds also act by binding to an allosteric pocket while maintaining high negative cooperativity
with dopamine. Finally, some compounds had no effect upon the dopamine doseꢀresponse curve within
the range of compound tested.
Table 3. Characterization of analogues of compound 3, varied at the carboxamide Nꢀsubstituent
a
and indoleꢀsubstitution in terms of binding and function at the D R
2
1
2
No.
R
R
pK (K , ꢁM)
Logα (α)
B
B
b
3
H
H
4.35 ± 0.07 (45)
ND
ꢀ0.53 ± 0.09 (0.29)
ND
1
1a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
b
1
1b
H
H
H
H
H
3.67 ± 0.14 (214)
4.21 ± 0.06 (62)
5.21 ± 0.07 (6)
5.33± 0.21 (5)
ND
ꢀ0.35 ± 0.21 (0.45)
ꢀ0.60 ± 0.09 (0.25)
ꢀ1.63 ± 0.11 (0.02)
ꢀ0.26 ±0.06 (0.55)
ND
b
1
1c
b
11d
b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
11e
11f
d
b
1
1g
H
4.41 ± 0.14 (39)
NA
1
1h
H
H
4.30 ± 0.06 (50)
ND
ꢀ0.58 ± 0.13 (0.26)
ND
1
1i
b
1
1j
H
4.72 ± 0.10 (19)
ꢀ1.39 ± 0.11 (0.04)
1
1k
H
H
ND
ND
NA
d
11l
3.08 ± 0.11 (831)
1
1m
1n
H
ND
ND
1
H
H
H
H
ND
ND
d
1
1o
3.81 ± 0.08 (154)
4.77 ± 0.13 (17)
4.72 ± 0.26 (19)
NA
c
11p
ꢀ1.77 ± 0.27 (0.02)
ꢀ0.63 ± 0.11 (0.23)
b
1
1q
1
1r
H
H
ND
ND
b
1
1s
4.04 ± 0.10 (91)
ꢀ0.53 ± 0.13 (0.30)
d
1
1t
H
H
H
4.73 ± 0.34 (19)
4.53 ± 0.07 (30)
ND
NA
ꢀ1.01 ± 0.08 (0.10)
ND
b
1
1u
11v
1
1w
H
ND
ND
b
1
3a
3ꢀF
4.35 ± 0.29 (45)
ꢀ0.72 ± 0.35 (0.19)
ACS Paragon Plus Environment

Page 25 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
b
13b
4ꢀF
5ꢀF
4.74 ± 0.10 (18)
4.04 ± 0.08 (91)
4.26 ± 0.17 (55)
4.74 ± 0.11 (18)
3.98 ± 0.20 (104)
ꢀ1.14 ± 0.10 (0.07)
d
1
3c
NA
d
13d
6ꢀF
NA
ꢀ0.96 ± 0.12 (0.11)
NA
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b
1
3e
7ꢀF
d
15
5ꢀOMe
a
Interaction studies between compounds and dopamine were performed using an ERK1/2
phosphorylation assay upon FlpIn CHO cells stably expressing the D R as in Figure 1c. For
compounds that caused a decrease in the maximal response of dopamine, data were analysed according
to an operational model of allosterism where K is the equilibrium dissociation constant of the
allosteric ligand, α is the binding cooperativity parameter between the orthosteric and allosteric ligand,
and β (Logβ constrained to ꢀ100 to represent high negative cooperativity with dopamine efficacy)
denotes the magnitude of the allosteric effect of the modulator on the efficacy of the orthosteric
agonist. For compounds that caused a limited rightward shift of the dopamine doseꢀresponse curve but
no decrease in Emax, data were fit using an operational model of allosterism where Logβ was
b
2
L
B
c
d
constrained to 0, to represent neutral cooperativity with dopamine efficacy. For compounds that
produced an unlimited rightward shift of a dopamine doseꢀresponse curve within the concentration
range of test compound used data were fit using model of competitive antagonism (Equation 6) . As
such a value of cooperativity could not be derived (denoted NA). ND signifies no effect of compound
was observed even at the highest concentration of 300 ꢁM. Data represent the mean ± SEM of three
experiments performed in duplicate apart from compounds with no effect, which were tested in two
separate experiments.
The subtle differences and stepwise structural changes introduced to members of the fragment library
reveal some interesting SAR in terms of both their binding and function at the D R (Table 3).
2
Considering analogues of 3 varied only at the carboxamide Nꢀsubstituent, substitution appears to be
essential for binding to the receptor, as the primary amide derivative 11a does not exhibit receptor
binding at even the highest concentrations employed. Furthermore, the degree of Nꢀsubstitution of the
carboxamide moiety is also a key determinant of affinity (K ), with monoꢀsubstituted derivatives (3,
B
1
1bꢀe, gꢀh, oꢀq, tꢀu) exhibiting activity, whereas the majority of diꢀsubstituted derivatives (11i, k, mꢀn)
did not.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Comparing the monoꢀsubstituted carboxamide analogues of 3 bearing nonꢀcyclic Nꢀsubstituents
(
11bꢀh), increasing the size of linear alkyl substituents from Nꢀethyl (11b K = 214 ꢁM, α = 0.45), to
B
Nꢀpropyl (11c K = 62 ꢁM,α = 0.25), to Nꢀbutyl (11d K = 6 ꢁM, α = 0.02), improved both affinity and
B
B
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
increased negative cooperativity in a progressive fashion. This demonstrates the importance of Nꢀ
substituent chain length to both binding and negative cooperativity. Interestingly, the Nꢀallyl analogue
1
1e (K = 5 ꢁM, α = 0.55) is of comparable size to 11c, but differs only in the degree of saturation in
B
the Nꢀsubstituent. However, this subtle structural difference results in around a 10ꢀfold increase in
affinity, but approximately 2ꢀfold decrease in negative cooperativity; suggesting that affinity and
negative cooperativity are driven by different structural features of these allosteric fragments, i.e.
simply increasing fragment affinity does not automatically result in increased negative cooperativity.
The inactivity of the ethanolamide analogue 11f indicates that polarity in the carboxamide Nꢀ
substituent is not tolerated, suggesting that this region of the molecule may largely interact with
hydrophobic residues. The two branched amide analogues 11g (K = 39 ꢁM) and 11h (K = 50 ꢁM, α =
B
B
0.26) both displayed similar affinity to 3, however Nꢀ(pentanꢀ3ꢀyl) analogue 11g lacked determinable
negative cooperativity. These data, when combined with the activity profile of Nꢀethyl analogue 11b,
suggest that an optimal degree of branching may be present on the Nꢀisopropyl substituent of 3 both in
terms of degree of substitution of the carbon atom adjacent to the carboxamide nitrogen in addition to
branch length. Further branching to the Nꢀtertꢀbutyl present on 11h does not appear to impact affinity
or cooperativity. Extending the branch length of the Nꢀisopropyl group of 3 to the Nꢀ(pentanꢀ3ꢀyl)
group of 11g did not affect affinity and the effect on negative cooperativity was indeterminable. In
comparison, reducing branching (i.e. 3 vs 11b) causes both affinity and negative cooperativity to
reduce.
Diꢀsubstituted carboxamide derivatives largely comprised of aliphatic heterocycle bearing analogues
(
11kꢀn). Of these, only pyrrolidyl analogue 11l (K = 831 ꢁM) displayed some, albeit weak affinity,
B
ACS Paragon Plus Environment

Page 27 of 82
Journal of Medicinal Chemistry
1
2
3
with no detectable negative cooperativity. Although the diꢀsubstituted amide derivatives generally
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
lacked affinity, the Nꢀdiisopropyl analogue 11j (K = 19 ꢁM, α = 0.04) possessed both higher affinity
B
and almost 10ꢀfold higher negative cooperativity than 3. This activity is difficult to rationalize amidst
the data for other compounds in the library. However, given the fragmentꢀsize nature of these
compounds, the possibility of alternative binding poses should not be discounted.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The remaining analogues in this series comprised cyclic moieties with the linking nitrogen atom
external to the ring. In the Nꢀcarbocycle series, ring size influenced both affinity and negative
cooperativity when comparing Nꢀcyclopropyl 11o (K = 154 ꢁM), Nꢀcyclobutyl 11p (K = 17 ꢁM, α =
B
B
ꢀ
0.02), Nꢀcyclopentyl 11q (K = 19 ꢁM, α = 0.23) and Nꢀcyclohexyl 11r (no determinable activity)
B
analogues. The optimal size in terms of affinity and highest level of negative cooperativity is exhibited
by the Nꢀcyclobutyl bearing 11p. Incrementing ring size by a single methylene unit to give 11q does
not affect affinity, but causes approximately 10ꢀfold reduction in negative cooperativity. This data
supports the SAR discussed above, of the acyclic substituents, with 11oꢀp being most similar in size
and shape to 3, whereas 11qꢀr are more comparable to 11g. The inactivity of 11r may initially seem
counterintuitive, given a common cyclohexyl moiety present on the relatively high affinity parent
compound 1. This apparent discrepancy can be reconciled if the bitopic nature of 1 is taken into
account, i.e. the high affinity of 1 is derived from concomitant saltꢀbridge interaction with the
orthosteric binding site of the D R. Without the strong ‘orthosteric anchor’ interaction, the allosteric
2
fragments are more sensitive to subtle structural changes.
Though substituent size is important for both the affinity and negative cooperativity exhibited by
these compounds, the nature of the carboxamide NH may also be key. Both the acidity of the NH bond
(i.e. its ability to donate a hydrogen bond) and steric crowding around the NH bond is also important.
The Nꢀ(1ꢀpiperidyl) analogue 11s (K = 91 ꢁM, α = 0.30) differs from 11r only by replacement of the
B
ring methine by a nitrogen atom but bears an allosteric activity profile similar to 3. Also, the Nꢀphenyl
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 82
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
analogue 11t (K = 19 ꢁM) regains affinity compared to 11r, where the saturated cyclohexyl group is
B
replaced with a planar phenyl ring. In both 11s and 11t, the carboxamide NH would be expected to be a
better hydrogen bond donor (due to substituent electronꢀwithdrawing effect) than that of 11r. Finally,
compounds 11uꢀw also support the importance of access to the carboxamide NH combined with
sensitivity to substituent size. Insertion of a methylene spacer on 11o to give cyclopropylmethyl
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
analogue 11u (K = 30 ꢁM, α = 0.10) improves affinity around 10ꢀfold, as well as imparting
B
determinable negative cooperativity. In comparison, no activity for 11vꢀw was observed up to a
concentration of 300 ꢁM, possibly due to these compounds bearing larger substituents causing a
sterically unfavourable interaction with the receptor.
Compounds 13aꢀe and 15 (Table 3) were based on the structure of 3, however introduced either
fluoroꢀ or methoxyꢀ substituents on to the indole core. Whilst all analogues in this series retained
activity at the D R, the 5ꢀfluoro analogue 13c (K = 91 ꢁM) and 5ꢀmethoxy analogue 15 (K = 104
2
B
B
ꢁ
M) exhibited reduced affinity. In comparison the 4ꢀfluoro analogue 13b (K = 18 ꢁM, α = 0.07) and
B
7ꢀfluoro analogue 13e (K = 18 ꢁM, α = 0.11) both displayed slightly improved affinity and displayed
B
higher negative cooperativity with dopamine compared to 3. The SAR of this series is difficult to
extrapolate without examining a wider range of compounds, though substitution at the 5ꢀ and 6ꢀ
positions of the ring appears to be less well tolerated. The effect of substitution at other ring positions is
likely to be determined by both regional and electronic influence of each substituent.
Table 4. Characterization of analogues of compound 3, varying the nature of the bicyclic
heteroaromatic ring, and substitution position of the carboxamide, in terms of binding and
a
function at the D R
2
ACS Paragon Plus Environment

Page 29 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
No.
Heterocycle
pK (K , ꢁM)
Logα (α)
B
B
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
3
4.35 ± 0.07 (45)
ꢀ0.53 ± 0.09 (0.29)
1
7
ND
ND
c
19
5.69 ± 0.38 (2)
ꢀ1.07 ± 0.22 (0.08)
2
1
ND
ND
ND
ND
23a
3b
2
ND
4.31 ± 0.20 (49)
ND
ND
NA
ND
d
2
5a
5b
2
2
7
ND
ND
ND
3
0a
ND
b
3
0b
3.91 ± 0.5 (123)
5.16 ± 0.5 (7)
ꢀ1.27 ± 0.3 (0.05)
ꢀ1.1 ± 0.3 (0.08)
c
3
0c
30d
ND
ND
a
Interaction studies between compounds and dopamine were performed using an ERK1/2
phosphorylation assay upon FlpIn CHO cells stably expressing the D R as in Figure 1c. For
compounds that caused a decrease in the maximal response of dopamine, data were analysed according
to an operational model of allosterism where K is the equilibrium dissociation constant of the
allosteric ligand, α is the binding cooperativity parameter between the orthosteric and allosteric ligand,
and β (Logβ constrained to ꢀ100 to represent high negative cooperativity with dopamine efficacy)
denotes the magnitude of the allosteric effect of the modulator on the efficacy of the orthosteric
b
2L
B
ACS Paragon Plus Environment