Discovery, Structure-Activity Relationship, and Biological Characterization of a Novel Series of 6-((1 H-Pyrazolo[4,3- b]pyridin-3-yl)amino)-benzo[ d]isothiazole-3-carboxamides as Positive Allosteric Modulators of the Metabotropic Glutamate Receptor 4 (mG

Article abstract of DOI:10.1021/acs.jmedchem.8b00994

This work describes the discovery and characterization of novel 6-(1H-pyrazolo[4,3-b]pyridin-3-yl)amino-benzo[d]isothiazole-3-carboxamides as mGlu₄ PAMs. This scaffold provides improved metabolic clearance and CYP1A2 profiles compared to previo

Full text of DOI:10.1021/acs.jmedchem.8b00994

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Article
The discovery, SAR and biological characterization of a
novel series of 6-((1H-pyrazolo[4,3-b]pyridin-3-yl)amino)-
benzo[d]isothiazole-3-carboxamides as positive allosteric
modulators of the metabotropic glutamate receptor 4 (mGlu4)
Sean Bollinger, Darren W. Engers, Joseph D. Panarese, Mary West, Julie L Engers,
Matthew T. Loch, Alice L Rodriguez, Anna L. Blobaum, Carrie K. Jones, Analisa Thompson
Gray, P. Jeffrey Conn, Craig W Lindsley, Colleen M Niswender, and Corey R. Hopkins
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00994 • Publication Date (Web): 24 Sep 2018
Downloaded from http://pubs.acs.org on September 25, 2018
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 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 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.
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 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The discovery, SAR and biological characterization of a
novel series of 6-((1H-pyrazolo[4,3-b]pyridin-3-yl)amino)-
benzo[d]isothiazole-3-carboxamides as positive allosteric
modulators of the metabotropic glutamate receptor 4
(mGlu4)
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,b
a,b
a,b
a,b
a,b
Sean R. Bollinger , Darren W. Engers , Joseph D. Panarese , Mary West , Julie L. Engers ,
a,b
a,b
a,b
a,b
Matthew T. Loch , Alice L. Rodriguez , Anna L. Blobaum , Carrie K. Jones , Analisa
a,b
a,b,c
a,b,d,e
a,b,c
Thompson Gray , P. Jeffrey Conn , Craig W. Lindsley
, Colleen M. Niswender , Corey
a,b,d,*
R. Hopkins
a
Vanderbilt Center for Neuroscience Drug Discovery, Vanderbilt University, Nashville, TN 37232, USA
Department of Pharmacology, Vanderbilt University, Nashville, TN 37232, USA
Vanderbilt Kennedy Center, Vanderbilt University Medical Center, Nashville, TN 37232, USA
Department of Chemistry, Vanderbilt University, Nashville, TN 37232, USA
b
c
d
e
Department of Biochemistry, Vanderbilt University, Nashville, TN 37232, USA
Abstract:
This work describes the discovery and characterization of novel 6ꢀ((1Hꢀpyrazolo[4,3ꢀb]pyridinꢀ
3
ꢀyl)aminoꢀbenzo[d]isothiazololeꢀ3ꢀcarboxamides as mGlu PAMs. This scaffold provides
4
improved metabolic clearance and CYP1A2 profiles compared to previously discovered mGlu4
PAMs. From this work, 27o (VU6001376) was identified as a potent (EC = 28 nM, 49% Glu
5
0
Max) and selective mGlu PAM with an excellent rat DMPK profile (in vivo rat CL = 3.1
4
p
mL/min/kg, t1/2 = 445 min, CYP1A2 IC50 > 30 µM). 27o was also active in reversing
haloperidol induced catalepsy in a rodent preclinical model of Parkinson’s disease.
INTRODUCTION
Parkinson’s disease (PD), first described 200 years ago, is the second most common
1
neurodegenerative disease (Alzheimer’s disease) and affects >10 million adults, globally. The
2
estimated economic burden of PD is estimated at $25 billion per year in the US alone. The
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 58
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
prevalence increases with age and PD affects ~1% of the population above the age of 60. PD is
a movement disorder that is due to the loss of dopaminergic neurons in the substantia nigra pars
compacta (SNpc) – expressing as a reduction in voluntary movement. The hallmark signs of PD
include tremor at rest, bradykinesia, rigidity and disturbance of posture; these signs have not
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
4
changed since the disease was first established in 1817. The cause of PD is unknown, with a
genetic factor being identified in a small portion of patients and environmental factors being
1
, 3
associated with an increased risk of PD.
Unfortunately, there are no diseaseꢀmodifying
treatments for PD, with only palliative treatments with dopamine replacement strategies (Lꢀ
5
ꢀ7
DOPA) as the main options.
Although LꢀDOPA does provide significant improvement in
symptoms early in the treatment protocol, there are multiple side effects associated with longꢀ
5ꢀ8
term treatment.
Within the basal ganglia (BG), connections are arranged as a set of direct and indirect
9
ꢀ11
pathways which act in a proper balance in a normal functioning BG.
However, an upset of
this balance of the direct and indirect pathways results in motor dysfunctions, with a stronger
11
signal coming from the indirect pathway (inhibiting and/or slowing movement). Thus, multiple
receptor classes that are present along the indirect pathway have been studied in order to identify
novel targets that may provide diseaseꢀmodifying treatment options for PD. One such receptor is
the metabotropic glutamate receptor 4 (mGlu ), which is located within striatoꢀpallidal synapses
4
1
2, 13
and is thought to function presynaptically on GABAergic neurons.
Due to its location, it is
believed that activation of mGlu decrease GABAs release at this synapse, which would, in turn,
4
14,
decrease output of the indirect pathway, and ultimately, reduce or eliminate the PD symptoms.
1
5
One finding underlying the hypothesis that mGlu activation or potentiation might have
4
neuroprotective effects stems from findings that patients undergoing deep brain stimulation
ACS Paragon Plus Environment

Page 3 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
within the indirect pathway, the effect that mGlu activation/potentiation is proposed to mimic
4
via a pharmacological mechanism, appear to have a slower disease course, requiring less Lꢀ
1
6ꢀ18
DOPA than cohort of patients who have not been treated with DBS.
Mechanistically, it has
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
been proposed that mGlu potentiation may serve a neuroprotective role via its presynaptic
4
presence at overactive excitatory synapses projecting from the subthalamic nucleus onto the
vulnerable dopaminergic neurons in the substantia nigra pars compacta. Valenti et al. showed
1
9
that stimulation of presynaptic mGlu at these synapses reduced excessive glutamate release.
4
Battaglia et al. and Betts et al. further demonstrated that administration of mGlu PAMs
4
prevented dopaminergic cell death in the substantia nigra induced by either MPTP or 6ꢀOHDA
1
4, 20
treatment.
Administration of the group III mGlu receptor orthosteric agonist, LꢀAP4, also
21ꢀ23
was neuroprotective in 6ꢀOHDA models.
Additionally, administration of an mGlu PAM
4
2
4
reduced ischemic brain damage and an mGlu PAM has also been shown to reduce
4
2
5
inflammatory responses in microglia , providing another potential mechanism for
neuroprotection.
mGlu is part of the larger metabotropic glutamate family of receptors (mGlu receptors)
4
2
6ꢀ29
which are family C members of the GPCR family.
There are eight cloned mGlu receptors
and these are further subdivided into three subꢀclasses: group I (mGlu and mGlu ), group II
1
5
27
(
mGlu and mGlu ) and group III (mGlu , mGlu , mGlu and mGlu ). Much work has been
2 3 4 6 7 8
30ꢀ32
published over the past several years describing both orthosteric agonists
and positive
3
3ꢀ35
allosteric modulators (PAMs)
for mGlu as novel therapies for PD. A number of PAM in
4
vivo tool compounds have been developed (Figure 1), offering evidence that mGlu PAMs could
4
be novel, diseaseꢀmodifying treatment options. The structural classes of molecules are wideꢀ
3
6
37
38
ranging, from triaryl amines, 1 , cyclohexyl amides, 2 , picolinamides, 3 , and pyrazolo[4,3ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 58
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
39, 40
b]pyridines, 4
, which speaks to the variability in the allosteric binding sites of the receptor.
Although much of the work remains in the preclinical stage, recently, Prexton Therapeutics has
4
1
advanced their lead molecule, 5, into clinical studies in Europe. Although this is an exciting
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
development in the mGlu PAM field, there still remain many fundamental questions
4
surrounding this oxime structural class, and bringing forth multiple structural classes for clinical
evaluation is a high priority for the validation of this receptor in the clinic.
Figure 1. Previously reported mGlu PAMs with in vivo efficacy in preclinical animal models of
4
PD.
RESULTS AND DISCUSSION
Our laboratory has recently discovered and reported on the pyrazolo[4,3ꢀb]pyridine, 4,
3
9, 40
structural class as novel, potent and selective mGlu PAMs.
Unfortunately, this compound
4
class suffered from significant CYP1A2 induction liabilities which led us to develop a new
indazoleꢀbased scaffold, 6 (Figure 2). These compounds were able to overcome the CYP1A2
induction issues (as measured through upstream AhR activation) noted with 4; however,
metabolic stability remained a liability which was not resolved with this structural change. Thus,
ACS Paragon Plus Environment

Page 5 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
we further evaluated other 6,5ꢀheteroaryl moieties and this report describes our work towards the
discovery and characterization of benzo[d]isoxazole and benzo[d]isothiazole compounds as
advanced, preclinical mGlu PAMs.
4
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 2. Identification of 6, an mGlu PAM devoid of CYP1A2 induction liability to
4
benzo[d]isoxazole and benzo[d]isothiazole based PAMs with improved in vivo PK properties
(CL , t ).
p
1/2
42
Utilizing the information that was gained during exploration of the indazole scaffold ,
we first investigated simple methyl and ethyl substituted benzo[d]isoxazole compounds. The 6ꢀ
amineꢀ3ꢀalkylꢀNꢀmethylindazole analogs showed excellent potency and even NHꢀindazole
compounds were active, although there was an ~10ꢀfold loss of potency.
As the
benzo[d]isoxazole does not allow for a substituent in the 1ꢀposition (corresponding to the N1 of
the indazole), we were curious if these analogs would have activity as mGlu PAMs. The initial
4
analogs were synthesized as described in Scheme 1, via a palladiumꢀmediated cross coupling of
the 3ꢀaminoꢀpyrazolopyridine (or pyrazine) and the appropriately substituted, commercially
t
t
t
available benzo[d]isoxazole (NaO Bu, BuXPhos, BuXPhos palladacycle), followed by
o
43
deprotection of the PMB group (TFA, µW, 140 C) to yield 10a-j in low yield.
a
Scheme 1. Synthesis of firstꢀgeneration benzo[d]isoxazoles.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 58
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
The SAR around the benzo[d]isoxazole started with the 3ꢀmethylꢀ6ꢀsubstituted analog,
0a, which showed moderate activity as an mGlu PAM (EC = 1860 nM) (Table 1).
1
4
50
Interestingly, the corresponding 3ꢀmethylꢀ5ꢀsubstitued analog, 10b, was significantly more
active (EC = 76 nM). The activity for the 5ꢀsubstituted benzo[d]isoxazole was equivalent (or
5
0
slight loss of potency) regardless of the pyrazolopyridine isomer (10e, EC = 227 nM) or the
50
pyrazolopyrazine (10d, EC = 137 nM). Addition of a fluorine on the pyrazolopyridine moiety
50
also provided an active compound (10f, EC = 219 nM), although there was a loss in potency
5
0
for for the 6ꢀfluoro analog (10c, EC = 1,570 nM). Addition of a fluorine to the 7ꢀposition of
5
0
the benzo[d]isoxazole moiety (10g-i) led to an increase in potency, resulting in the most potent
compound in the series, 10i (EC = 23 nM). Unfortunately, the 5ꢀsubstituted analogs did not
50
provide any opportunity to improve the compounds as any additional substitution at the 3ꢀ
position led to reduced (or inactive) activity (e.g., 10j, EC = >10,000 nM) – a phenomenon that
5
0
4
2
has been seen in previous work.
Table 1. Structures and activities of new 5ꢀ and 6ꢀaminobenzo[d]isoxazole analogs.
ACS Paragon Plus Environment

Page 7 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hmGlu pEC
(± SEM)
hmGlu4
%Glu Max
(± SEM)
4
50
a
a
a
Cmpd
R1
R2
R3
X
N
Y
EC (nM)
50
1
0a
0b
H
H
CH
5.73 ± 0.03
1860
86 ± 18
1
N
N
N
CH
CH
N
7.12 ± 0.02
5.80 ± 0.36
6.86 ± 0.06
6.64 ± 0.06
75.9
1,570
137
91 ± 3
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
1
0c
0d
0e
0f
F
H
F
80.6 ± 15.1
85.4 ± 15.8
81.9 ± 12.9
1
1
H
H
CH
N
227
1
H
CH
N
6.66 ± 0.02
219
54 ± 3
1
0g
10h
0i
H
H
H
N
CH
6.56 ± 0.21
274
77.7 ± 18.1
CH
N
N
N
6.91 ± 0.16
7.64 ± 0.03
122
74.7 ± 17.4
70 ± 3
1
22.9
1
0j
H
CH
N
<5.00
>10,000
54 ± 11
a
For assay conditions, see Experimental Section.
Although the SAR was limited around the 5ꢀsubstituted benzo[d]isoxazole framework,
we evaluated these compounds in a battery of Tier 1 in vitro DMPK assays (Table 2). All of the
compounds evaluated displayed moderate/high predicted hepatic clearance in human
microsomes (Q = >50%) and more moderate clearance in rat microsomes (Q <50%). In
H
H
addition, the compounds had good plasma free fraction (%f ~3–8%) in both species.
u
Unfortunately, these small, aromatic compounds were potent CYP1A2 inhibitors (<700 nM), and
showed significant CYP1A2 induction (AhR activation = Emax >50, data not shown). The rat and
human mGlu in vitro potency data tracked well, which is common with most of the mGlu PAM
4
4
scaffolds that we have identified. Lastly, since the in vitro rat liver microsome clearance data
indicated that these compounds would be moderately cleared, we wanted to establish an in
vitro/in vivo correlation (IV/IVC) for this initial scaffold. The compounds were investigated
using a rat IV cassette dosing strategy to evaluate their in vivo pharmacokinetic properties,
4
4
including clearance and halfꢀlife. As can be seen in Table 2, these compounds were all highly
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 58
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
cleared, with the exception of 10d. However, the t1/2 was not ideal (<90 min; MRT <60 min) for
all of the compounds; therefore, none were progressed further into in vivo efficacy studies.
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 2. In vitro DMPK and rat potency data of benzo[d]isoxazole analogs.
a
a
Human
mL/min/kg)
CLINT
14.8
24.2
114
Rat
(mL/min/kg)
CLINT
Plasma protein
binding (%fu)
Rat mGlu EC
4
50
(
(nM);
%Glu Max
d
Cmpd
0b
0c
0d
0e
0f
0g
10h
CLHEP
8.7
11.3
17.7
16.7
11.1
CLHEP Human
Rat
5.8
7.6
6.4
7.9
4.4
CYP1A2 (µM)
1
1
1
1
31.8
33.4
3.0
21.9
22.6
2.9
6.2
5.4
6.6
6.6
5.4
<0.1
<0.1
0.2
0.7
0.5
95.5; 104.8
123; 111.3
65.0; 126.8
209; 107.7
190; 120.6
81.4
23.6
30.1
39.3
21.0
25.2
1
1
31.1
50.2
12.5
14.8
66.4
110
34.1
42.7
2.9
2.8
5.0
3.3
<0.1
53.0; 113.8
169; 113.6
<0.1
a
b
Intrinsic and predicted hepatic clearance based on experiments in liver microsomes. %f = % fraction
unbound. IC determination of CYP1A2 enzyme inhibition in human liver microsomes using specific probe
substrateꢀmetabolite pairs. Thalliumꢀflux mGlu assay.
u
c
5
0
d
4
In vivo PK – Rat Cassette IV Clearance
CL (mL/min/kg)
Vss (L/kg)
2.0
MRT (min)
31.6
t1/2 (min)
54.8
10c
0d
0e
10g
0h
64.6
24.3
38.0
81.1
30.5
1
0.9
37.2
28.6
1
2.5
60.6
106.7
31.2
3.1
38.4
1
1.2
38.0
34.7
Since the introduction of a benzo[d]isoxazole ring system was productive in terms of
potency, we next wanted to evaluate the 6ꢀbenzo[d]isoxazoleꢀ3ꢀsubstituents to assess the impact
on potency, PK and CYP1A2 inhibition. Previously, we showed that extending the steric bulk in
this portion of the indazole scaffold led to more potent compounds, and, importantly, improved
CYP1A2 profiles. The generation of these analogs is shown in Schemes 2 and 3.
The synthesis of the 3ꢀalkylsubstituted benzo[d]isoxazole analogs started with the
commercially available 3ꢀbromophenol, 11, which was acylated under FriedelꢀCrafts conditions
(
AlCl , RCOCl, reflux) to yield 13 in moderate yield. Next, the resulting ketone was converted
3
to the imine, 14 (7M NH in MeOH), which was then subjected to NCSꢀmediated ring closure
3
ACS Paragon Plus Environment

Page 9 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
45
(
NCS, K CO , THF) to yield the penultimate intermediate, 15. The final targets (16a-o) were
2 3
synthesized via palladiumꢀcatalyzed crossꢀcoupling of the 3ꢀaminopyrazolopyridine analogs, 9,
t
t
t
(NaO Bu, BuXPhos, BuXPhos palladacycle) followed by deprotection of the PMB group (TFA,
4
3
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
µW).
a
Scheme 2. Synthesis of 6ꢀbenzo[d]isoxazoleꢀ3ꢀalkyl analogs.
The difluoroethyl analog, 16p, was synthesized as outlined in Scheme 3. The
commercially available ethyl 6ꢀbromobenzo[d]isoxazoleꢀ3ꢀcarboxylate, 17, was reacted with
o
MeMgBr (THF, ꢀ78 C) to yield the ketone, 18. The ketone, 18, was converted to the 1,1ꢀ
o
difluoroethyl analog, 19, via treatment with DAST (dichloroethane, 90 C). The final compound,
1
6p, was realized after palladiumꢀcatalyzed crossꢀcoupling of 9 and 19 (Xantphos, Cs CO , µW)
2
3
4
3
under microwave irradiation followed by PMBꢀdeprotection (TFA, µW).
a
Scheme 3. Synthesis of difluoroalkyl analog, 16p.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 58
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
Extension of the methyl group to an ethyl group, 16a, resulted in similar potency (EC =
5
0
1350 nM vs. 10a, EC = 1860 nM). Further increase in steric bulk to the isopropyl led to an ~2ꢀ
50
fold improvement in potency, 16b (EC = 575 nM). Introduction of small, cycloalkyl groups
5
0
was productive and led to compounds that were equipotent with the isopropyl analog (16cꢀf);
however, moving the pyridine around the pyrazolopyridine head group led to a compound with
much reduced potency (16g, EC = 2,400 nM). Moving the branching point further from the
5
0
ring system did not improve the potency (16h-k). Again, moving the nitrogen of the
pyrazolopyridine around the ring led to a reduction in potency (16k, EC50 = 2,820 nM).
Introduction of longer, branched alkyl groups was a productive change and led to the most potent
compounds in the benzo[d]isoxazole series. The 3ꢀ(pentanꢀ3ꢀyl)benzo[d]isoxazoles were potent
mGlu PAMs (16l, EC = 98 nM; 16m, EC = 129 nM) and, like the previous example, the
4
50
50
pyrazolo[3,4ꢀb]pyridine isomer was much less active (16n, EC = 525 nM). The 3ꢀ(pentanꢀ2ꢀyl)
5
0
analog retained much of the activity (16o, EC = 117 nM) as the racemate. Lastly, as metabolic
5
0
hotꢀspot analysis revealed the methylene (or methine) of the 3ꢀsubstituted (cyclo)alkyl
ACS Paragon Plus Environment

Page 11 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
substituents as a potential site of metabolism, we synthesized the 3ꢀ(1,1ꢀdifluoroethyl) derivative
in order to block this site. The difluoroethyl analog was slightly more potent than the
corresponding ethyl analog (16p, EC = 678 nM vs. 16a, EC = 1350 nM).
5
0
50
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
Table 3. Structures and activities of select 6ꢀaminoꢀ3ꢀalkylbenzo[d]isoxazole analogs.
hmGlu pEC
(± SEM)
hmGlu EC
%Glu Max
(± SEM)
4
50
4
50
a
a
a
Cmpd
6a
R
X
N
Y
(nM)
1
CH
5.87 ± 0.03
1350
72 ± 4
1
6b
6c
N
N
N
CH
N
6.24 ± 0.01
575
23 ± 1
1
6.15 ± 0.04
6.07 ± 0.01
708
851
102 ± 11
88 ± 11
16d
6e
6f
16g
CH
1
N
N
N
CH
N
6.29 ± 0.07
6.14 ± 0.11
5.62 ± 0.01
514
723
85.5 ± 23.1
87.6 ± 22.0
60 ± 7
1
CH
2400
1
6h
N
CH
6.09 ± 0.03
813
53 ± 2
1
6i
N
N
N
CH
N
6.07 ± 0.12
5.89 ± 0.25
5.55 ± 0.03
7.01 ± 0.01
6.89 ± 0.02
6.28 ± 0.005
860
1,300
2820
98
83.5 ± 13.2
68.9 ± 9.5
84 ± 5
1
6j
16k
6l
6m
6n
CH
N
1
N
106 ± 3
92 ± 2
1
N
CH
N
129
1
CH
525
97 ± 2
1
6o
N
CH
6.93 ± 0.04
117
92 ± 3
1
6p
N
CH
6.17 ± 0.01
678
58.8 ± 15.1
a
For assay conditions, see Experimental Section.
This new set of 3ꢀalkylꢀ6ꢀbenzo[d]isoxazole analogs were evaluated in Tier 1 in vitro
DMPK assays as well as in vivo rat PK assays (Table 4). Most of the compounds that were
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 58
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
evaluated showed high intrinsic (and high predicted hepatic) clearance in both human and rat
liver microsomes (Q >50%). It should be noted that introduction of the difluoroethyl group
H
significantly reduced the intrinsic clearance, especially in rat liver microsomes (16p).
Introduction of the extended (cyclo)alkyl groups in the 3ꢀposition led to a reduction across the
board in plasma free fraction, as compared to the simple 3ꢀmethyl compounds (Table 2).
However, the level of CYP1A2 inhibition was significantly reduced – mimicking the trend that
we previously reported. In addition, many of these compounds had reduced AhR activation (16i
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
(VU0652264), AhR activation = 6.4). A selection of these compounds was evaluated in vivo in a
4
4
rat IV cassette clearance experiment. The compounds were moderately cleared; however, they
showed good CNS penetration when evaluated in a rat cassette plasma:brain distribution assay
(
16i, K = 4.7).
p
Table 4. In vitro and in vivo DMPK for 6ꢀaminoꢀ3ꢀalkylbenzo[d]isoxazole analogs.
Rat
Plasma unbound
Rat mGlu4
Human (mL/min/kg)
(mL/min/kg)
(%fu)
Human
2.7
CYP1A2
(µM)
<0.1
1.8
EC (nM);
%GluMax
ND
50
Cmpd
16a
16d
16e
16f
CLINT
69.9
83.2
42.3
30.7
7.2
CLHEP
16.1
16.8
14.0
12.5
5.4
CLINT
CLHEP
52.1
66.2
56.0
59.5
33.0
38.8
61.2
9.5
Rat
3.7
1.6
0.8
0.7
2.3
2.2
1.9
1.7
204
1219
281
398
62.5
86.9
488
11
2.6
1.3
1.0
2.3
1.6
0.9
0.3
ND
3.0
1.2
3.2
1.1
1.0
0.5
518; 101.9
680; 97.9
501; 98.7
542; 84.8
16i
16j
7.3
5.4
e
16m
16p
84.2
24.6
16.8
11.3
ND
1600; 83.8
a
b
Intrinsic and predicted hepatic clearance based on experiments in liver microsomes. %f = % fraction
u
c
unbound. IC determination of CYP1A2 enzyme inhibition in human liver microsomes using specific
probe substrateꢀmetabolite pairs. Thalliumꢀflux mGlu assay. ND = not determined
5
0
d
e
4
In vivo PK
Rat IV Cassette
CLp
Cassette PBL
Plasma
(ng/mL)
117
Brain
(ng/g)
145
Cmpd
16e
16f
(mL/min/kg)
23.2
t1/2 (min)
45.9
Kp
1.2
N/A
4.7
62.1
37.8
100.0
87.4
ND
60.5
ND
285
16i
16j
16p
43.2
27.8
79.9
90.8
ND
ND
ND
ND
N/A
N/A
ACS Paragon Plus Environment

Page 13 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To further establish that the reduction in the AhR activation improved the overall PK
profile of the scaffold, we initiated a subchronic, 4ꢀday repeat dosing PK study with 16i (Table
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
5
A). This compound was dosed onceꢀdaily (15 mg/kg, PO) for four days and the plasma levels
were evaluated on days 1 and 4. Unlike 4, for which we saw a significant reduction in plasma
exposure at day 4, Cmax and AUC levels of 16i were comparable at both day 1 (Cmax = 86.3
ng/mL; AUC = 624 hr･ng/mL) and day 4 (Cmax = 62.8 ng/mL; AUC 690 hr･ng/mL). This result
suggests that by reducing the AhR activation, we would be able to identify, and potentially
develop, an mGlu PAM that would be suitable for chronic dosing applications.
4
Next, we evaluated 16i for its ability to reverse haloperidolꢀinduced catalepsy in a
4
6
preclinical rodent model of PD (Table 5B). In this model, male SpragueꢀDawley rats were
randomly assigned to treatment groups and injected with haloperidol (0.75 mg/kg i.p.) and
returned to their home cages for 60 min. Vehicle, 16i (10mg/kg) or 4 (10mg/kg) were
administered orally 30 minutes prior to testing for catalepsy (N = 11 – 12 rats/group). As shown
in Table 5B, 16i exhibited a robust reversal of cataleptic activity in this model after oral
administration (Oneꢀway ANOVA: F2,32 = 26.9, p <0.0001) – similar to previously evaluated mGlu4
PAMs (4, Table 5B). To better understand this in vivo efficacy (PK/PD), we quantified the
plasma, brain and cerebrospinal fluid (CSF) levels of 16i from the PD study. As shown in Table
5C, the reversal of HIC was observed with CSF levels of 627 nM, levels equivalent to the rat
mGlu EC .
4
50
Table 5. Subchronic PK dosing of 16i and in vivo efficacy studies.
A
Chronic dosing of 16i, (15 mg/kg, PO)
B
Haloperidol efficacy
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 58
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
Statistics:
16i reversal of Haloperidolꢀinduced
Catalepsy in male SpragueꢀDawley rats. Oneꢀway
ANOVA: F2,32 = 26.9, p <0.0001. ***p < 0.0001 vs.
Vehicle
+
Haloperidol (Dunnett’s multiple
comparison test). N = 11 – 12 rats/group.
C
In vivo exposure in rat plasma/brain after PO administration in haloperidolꢀinduced catalepsy
Dose
0 mg/kg
Plasma (nM)
450
Free Plasma (nM)
11.2
Brain (nM)
958
Free Brain (nM)
5.8
CSF (nM)
627
1
Although we have identified compounds that mitigate the CYP1A2 liability, there still
remained issues with modest potency and high clearance in vivo and short t1/2 in rats. Thus, we
next turned our attention to moving from the benzo[d]isoxazole scaffold to the
benzo[d]isothiazole scaffold in an attempt to probe the ring size of the fiveꢀmembered annulated
ring system. As larger substituents on the 3ꢀposition of the benzo[d]isoxazole group were
productive in reducing the CYP1A2 inhibition, we started with these same groups for evaluation.
As these moieties were not commercially available, significant synthetic chemistry effort was
devoted to making the appropriate analogs and these efforts are outlined in Schemes 4–9.
The synthetic procedure for the 3ꢀalkylꢀ6ꢀbromobenzo[d]isothiazole or 3ꢀalkylꢀ6ꢀ
bromoisothiazolo[4,5ꢀb]pyridine analogs (e.g., 27a-i) is shown in Scheme 4. The commercially
available 4ꢀbromoꢀ2ꢀfluorobenzaldehyde, 20, was converted to the secondary alcohol, 21, via
reaction with the appropriate alkyl Grignard. The alcohol was oxidized to the key intermediate
ketone, 22, using DessꢀMartin periodinane. The isothiazolo[4,5ꢀb]pyridine analogs were
synthesized in an analogous manner starting with the 5ꢀbromoꢀ3ꢀfluoropicolinonitrile, 23, which
ACS Paragon Plus Environment

Page 15 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
was reacted with the alkyl Grignard to yield 24. Both 22 and 24 were carried forward under the
same reaction conditions. Thus, they were reacted with BnSH under basic conditions to yield the
benzylthio derivative, 25. The ring closure to the isothiazole ring system was completed using
47
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
SO Cl and NH to yield 26.
The final compounds, 27a-i, were made via the palladiumꢀ
2
2
3
43
catalyzed crossꢀcoupling with 9 followed by deprotection as detailed above.
Scheme 4.
The trifluoromethyl analog, 27j, was synthesized in a similar manner to the previous
t
compounds with a reversal of some of the steps. Thus, 20 was reacted with BnSH (NaO Bu,
THF) to yield 28. Next, the aldehyde was converted to the trifluoromethyl ketone via a twoꢀstep
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 58
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
o
procedure starting with trifluoromethyl addition (TMSꢀCF , TBAF, 0 C) to yield the secondary
3
alcohol which was oxidized to the ketone, 29, with DessꢀMartin periodinane. The thioꢀketone
was cyclized to the isothiazole (SO Cl , rt) followed by the reaction with ammonia (7M in
2
2
4
7
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
MeOH) to yield 30. Finally, the target compound, 27j, was synthesized as described above
43
under palladium catalysis/deprotection conditions.
Scheme 5.
Compound 31 was coupled (Pd (dba) , Xantphos, Cs CO ) with the PMBꢀprotected
2
3
2
3
pyrazolo[4,3ꢀb]pyridineꢀ3ꢀamine, 9, to give the desired compound 32 in modest yield. The ethyl
o
ester, 32, was converted to the methyl ketone, 33, via reaction with MeMgBr (ꢀ78 C, THF),
o
followed by PMBꢀdeprotection (TFA:toluene, 150 C, µW) in good yield. The final tertiary
alcohol, 27k, was obtained in low yields via another addition of MeMgBr to the methyl ketone.
Scheme 6.
ACS Paragon Plus Environment

Page 17 of 58
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
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 synthesis of the amide analogs (27n-v) is shown in Scheme 7. Starting from 32, the
carboxylic acid 34 was realized by saponification of the ester (LiOH, MeOH). Next, the amides,
27n-v, were synthesized via coupling with the appropriate amine under HATU conditions
(DIEA, DMF) followed by PMBꢀdeprotection in low to moderate yield over the three steps.
Scheme 7.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 58
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
The preparation of the 3ꢀacylpiperidine analogs, 27l,m, is shown in Scheme 8. The 1ꢀ
bromoꢀ3ꢀfluorobenzene starting material, 38, was acylated with 39 under FriedelꢀCrafts
o
conditions (AlCl , 80 C) to give 40 in low yield. Next, 40, was reacted with benzylthiol
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
displacing the fluorine to yield the isothiazole precursor, 41. The isothiazole was synthesized as
previously described by reacting the 2ꢀbenzylthio ketone with sulfuryl chloride followed by the
4
7
addition of ammonia which gave the cyclized isothiazole 42 in 55% yield over three steps.
Finally, the target compounds, 27l-m, were synthesized by palladiumꢀcatalyzed crossꢀcoupling
43
followed by PMBꢀdeprotection in good yield.
Scheme 8.
ACS Paragon Plus Environment

Page 19 of 58
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
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
Lastly, the tertiary amine analog, 27w, was synthesized as outlined in Scheme 9. The
starting material, 35, which was made as outlined in Scheme 4 for 26, was monoꢀbrominated on
48
the 3ꢀmethyl position (NBS, benzoylperoxide, CCl ) to yield 36.
The newly made 3ꢀ
4
o
bromomethyl was displaced by 4,4ꢀdifluoropiperidine (NMP, µW, 150 C) to give the
48
penultimate intermediate, 37 in 27% yield over two steps.
The final compound, 27w, was
completed via palladiumꢀcatalyzed crossꢀcoupling with 9 as outlined above, followed by PMBꢀ
4
3
deprotection in 43% yield over the two steps.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 58
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
Scheme 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
Having established synthetic procedures to access the required analogs, we evaluated the
new analogs in the human mGlu potency assay (Table 6). Utilizing the knowledge gained from
4
the benzo[d]isoxazole series, we started with sterically hindered 3ꢀalkyl groups since these,
generally, gave the best CYP1A2 induction profile. The pyrazolo[4,3ꢀb]pyridine analog, 27a
(EC50 = 269 nM), led to an ~2ꢀfold increase in potency compared to the benzo[d]isoxazole
analog, 16j. Addition of a nitrogen to the benzo[d]isothiazole ring system, 27b, led to a
significant increase in potency (EC = 74 nM). The pyrazolo[3,4ꢀb]pyridine regioisomer, 27c,
5
0
resulted in a reduction in potency of ~3ꢀfold compared to 27a (EC50 = 759 nM). The
pyrazolo[3,4ꢀb]pyrazine with a nitrogen in the benzo[d]isothiazole system was also a potent
compound (27d, EC50 = 105 nM). In general, the benzo[d]isothiazole ring system was
significantly more potent than the corresponding benzo[d]isoxazole scaffold. Examples of this
increase in potency can be seen in a number of analogs, such as 27e (EC = 40 nM) vs. 16a
50
(EC = 1350 nM). In addition, the trend of the pyrazolo[3,4ꢀb]pyridine leading to a loss of
5
0
potency was continued for the ethyl substituent, 27f (EC = 102 nM). Small cycloalkyl groups
50
were well tolerated (27g,h) as well as bulkier groups such as tertꢀbutyl (27i); however, the 3ꢀ
ACS Paragon Plus Environment

Page 21 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
trifluoromethyl, 27j, was less active (EC = 794 nM). The tertiary alcohol was synthesized to
5
0
introduce a group that may improve the water solubility; however, this moiety was much less
active (27k, EC50 = 1,050 nM). To further explore the functional group tolerance off the 3ꢀ
position, we made acylated piperidine analogs, 27l,m, which retained activity (EC = 200 and
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
5
0
3
89 nM, respectively). Next, we turned to the amide moiety which proved to be a fruitful
functional group in the indazole series. A variety of secondary and tertiary amides were
synthesized (27n-v) and evaluated. The 3,3ꢀdifluoropyrrolidine analog, 27n, was active (EC =
5
0
3
31 nM); however, potency was improved by expanding the ring to the 6ꢀmembered piperidine
(
4,4ꢀdifluoropiperidine), 27r (EC = 52 nM). However, other 6ꢀmembered tertiary amides were
5
0
not as potent (27s-v), suggesting the gemꢀdifluoro moiety is important in the binding. The most
potent amides that were investigated with the secondary 3,3ꢀdifluorocyclobutane derivatives 27o
(
EC = 50 nM) and 27p (EC = 26 nM) being the most potent analogs. Methylation of the
50 50
secondary amide was productive and led to an equivalent compound in terms of potency (27q,
EC = 68 nM). Removal of the amide group, resulting in the tertiary amine compound, 27w
50
(EC50 = 81 nM) was active; however, all of these amine compounds that were synthesized
suffered from very high in vitro liver microsomal clearance (data not shown).
Table 6. Structures and activities of 6ꢀaminoꢀ3ꢀsubstituted benzo[d]isothiazole analogs.
hmGlu pEC
(± SEM)
hmGlu EC
%Glu Max
(± SEM)
4
50
4
50
a
a
a,b
Cmpd
7a
7b
7c
7d
7e
R
X
N
N
CH
N
N
Y
CH
CH
N
N
CH
Z
CH
N
CH
N
CH
(nM)
2
2
2
2
6.57 ± 0.04
7.13 ± 0.01
6.12 ± 0.04
6.98 ± 0.05
7.40 ± 0.07
269
74
759
105
40
70 ± 2
51 ± 3
74 ± 1
54 ± 4
77 ± 4
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 58
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
2
7f
7g
7h
CH
N
N
N
CH
N
CH
CH
CH
6.99 ± 0.05
7.09 ± 0.02
7.17 ± 0.04
102
81
68
57 ± 4
76 ± 4
73 ± 3
2
2
2
7i
N
CH
N
CH
N
CH
CH
CH
6.49 ± 0.08
6.10 ± 0.04
5.98 ± 0.09
324
794
64 ± 5
50 ± 4
98 ± 8
2
7j
CF
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
2
7k
CH
1050
2
7l
N
N
N
CH
CH
CH
CH
CH
CH
6.70 ± 0.05
6.41 ± 0.04
6.48 ± 0.01
200
389
331
79 ± 4
111 ± 6
92 ± 7
2
7m
27n
27o
N
N
CH
N
CH
CH
7.30 ± 0.08
7.58 ± 0.07
50.1
26.0
50.5 ± 7.5
2
7p
68.1 ± 15.6
27q
N
CH
CH
7.17 ± 0.21
68.1
76.7 ± 17.6
2
7r
7s
N
N
CH
CH
CH
CH
7.28 ± 0.06
6.56 ± 0.05
52.4
275
71.0 ± 16.6
62 ± 12
2
2
7t
N
N
CH
CH
CH
CH
6.49 ± 0.06
6.35 ± 0.03
324
447
44 ± 9
63 ± 3
2
7u
2
7v
N
N
CH
CH
CH
N
6.40 ± 0.04
7.09 ± 0.05
398
81
80 ± 2
84 ± 6
2
7w
a
For assay conditions, see Experimental Section.
ACS Paragon Plus Environment

Page 23 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
These new benzo[d]isothiazole analogs were evaluated in a battery of Tier 1 in vitro DMPK
assays (Table 7). Although many of the compounds showed moderate clearance in human liver
microsomes (CLHEP), they were more highly cleared in rat liver microsomes (e.g., 27b). As
noted above, 27w, was highly cleared in rat liver microsomes and was unstable in rat plasma,
which was a common trend in all of the amines that were synthesized for the project. The
plasma protein binding was high to moderate for all compounds that were evaluated, with no
clear trend emerging. The CYP1A2 inhibition was also examined and showed a similar trend to
previously reported scaffolds. The 3ꢀneopentyl derivatives (27b,d) showed significantly reduced
inhibition of the CYP1A2 enzyme; however, the smaller 3ꢀcyclopropyl (27g,h) group did show
nanomolar inhibition (similar to what was seen in the indazole scaffold). As the substitution at
the 3ꢀposition increased in size, the CYP1A2 inhibition decreased and all of the amides that were
tested did not show any inhibition of the enzyme (>30 µM). In fact, 27o, was inactive against all
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
CYP enzymes studied (CYP1A2, 3A4, 2C9, 2D6: >30 µM). Lastly, the two amide compounds
(
27o,p) were the most potent against the rat mGlu receptor (~82 nM).
4
Table 7.
Human
Rat
Plasma protein
Rat mGlu4
(
mL/ming/kg)
(mL/min/kg)
binding (%fu)
CYP1A2
(µM)
22.1
EC (nM);
50
Cmpd
CLINT
CLHEP
CLINT
CLHEP
Human
Rat
%GluMax
e
2
2
7b
7d
24.7
11.4
277
55.9
0.6
1.6
ND
40.2
35.4
28.9
41.2
15.8
19.0
2.60
298
13.8
13.2
12.2
13.9
9.00
10.0
2.30
19.6
15.1
14.2
15.6
11.9
189
127
176
59.4
21.0
34.2
8.30
2938
393
51.1
45.1
50.1
32.1
16.2
23.0
7.40
68.4
59.4
49.8
53.6
65.8
1.3
3.5
0.9
1.2
4.2
2.4
0.4
0.6
2.1
7.7
1.1
3.0
1.4
0.8
3.9
0.5
0.7
<0.1
2.9
5.9
>30e
ND
0.4
158; 112.3
e
2
7e
NDe
2
2
7g
7h
NDe
0.7
ND
2
7l
21.7
>30
>30e
ND
>30e
ND
166; 98.9
82.7; 76.8
82.2; 86.6
27o
27p
27q
e
ND
2
2
7r
7s
54.0
44.3
60
139; 102.8
e
172
229
1085
NDe
e
2
7t
1.3
3.4
ND
ND
ND
f
f
e
2
7w
27.5
>30
a
b
Intrinsic and predicted hepatic clearance based on experiments in liver microsomes. %f = % fraction
u
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 58
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
c
unbound. IC determination of CYP1A2 enzyme inhibition in human liver microsomes using specific
5
0
d
e
f
probe substrateꢀmetabolite pairs. Thalliumꢀflux mGlu assay. ND = not determined. unstable
4
Next, based on their overall profile (potency, in vitro DMPK), a select few compounds
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
were chosen for administration in a rat cassette to determine the plasma:brain partition ratio
4
4
(
Table 8). All of the compounds evaluated showed excellent plasma levels; however, only one
compound, 27o, showed appreciable brain levels (K = 0.4). Very minor changes to the
p
structure, such as the pyrazolo[4,3ꢀb]pyrazine, 27p (K = 0.08) and the methylated amide, 27q
p
(
K = BLQ), led to a significant reduction in brain penetration. Since the K of 27o was
p p
acceptable, we next evaluated this compound in a rat IV cassette experiment for in vivo plasma
clearance and halfꢀlife (Table 8). This IV dosed experiment showed that 27o was a low
clearance compound in rat (CL = 3.1 mL/min/kg) with an excellent halfꢀlife (t = 445 min).
p
1/2
Table 8. In vivo cassette plasma:brain evaluation and IV PK cassette study for select
compounds.
Cassette PBL
Plasma
Brain
(ng/g)
(
ng/mL)
Kp
2
7o
7p
27q
7r
284
107
55.6
BLQ
35.5
0.4
2
690
511
0.08
N/A
0.14
2
249.0
Rat IV Cassette Clearance, 27o
CL (mL/min/kg)
3.1
p
V (L/kg)
1.6
445
510
ss
t1/2 (min)
MRT (min)
The difluorocyclobutane, 27o, was evaluated in our mGlu selectivity panel and it was
inactive against all seven members of the mGlu receptor family (Table 9). In addition, we sent
the compound for a larger selectivity panel screen (Eurofins Lead Profiling), which is a panel of
ACS Paragon Plus Environment

Page 25 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
49
~
70 GPCRs, transporters, and ion channels. The results show that 27o is a selective compound
as the only receptor that was identified was Adenosine A (76% inhibition of radioligand binding
3
at 10 µM).
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
Table 9. Selectivity of 27o.
Cmpd
hmGlu1
rmGlu2
rmGlu3
rmGlu5
Inactive
Eurofins Lead Profiling
rmGlu7
rmGlu8
2
7o
Receptor
% Inhibition @ 10 µM
2
7o
Adenosine A
3
76
Finally, since mGlu has emerged as a target for the possible treatment of PD, we
4
assessed the efficacy of 27o (10 mg/kg) in the rat haloperidolꢀinduced catalepsy (HIC) model,
4
6
(
Table 10A). This compound showed robust activity in this model after oral administration
(Oneꢀway ANOVA: F2,27 = 101.8, p <0.0001) – comparable to previously evaluated mGlu PAMs (4,
4
Table 10A) and previously published adenosine A2A antagonists. To better understand in vivo
efficacy in relation to exposure (PK/PD), we quantified the plasma, brain and cerebrospinal fluid
(CSF) levels of 27o. As shown in Table 10B, the reversal of HIC was observed with total brain
levels of 252 nM (K = 0.23) and CSF levels of 46 nM. We also evaluated 27o for potential
p
efflux (Table 10) but no significant efflux was noted (MDCKꢀPgp ER: 1.9) (Table 10C). These
levels of drug vs. efficacy in HIC is similar to other mGlu PAMs developed in our laboratory
4
and further supports the use of total brain and CSF levels of drug to correlate PK/PD.
Table 10. In vivo evaluation after haloperidol assay.
A
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 58
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
Statistics: 27o, Reversal of Haloperidolꢀinduced Catalepsy in male
SpragueꢀDawley rats. Oneꢀway ANOVA: F2,27 = 101.8, p <0.0001. ***p
<
0.0001 vs. Vehicle + Haloperidol (Dunnett’s multiple comparison test).
N = 10 animals/group.
B. In vivo exposure in rat plasma/brain after PO administration in
haloperidolꢀinduced catalepsy
Cmpd
27o
Dose (mg/kg)
10
Plasma (nM)
1120
Brain (nM)
252
CSF (nM)
46.0
ꢀ
6
C. MDCKꢀPgp (Papp, 10 cm/s)
AꢀB
2.9
BꢀA
5.4
Efflux
1.9
27o
CONCLUSION
Herein, we have presented the synthesis and biological characterization of a highly potent
and selective mGlu PAM benzo[d]isothiazoleꢀbased scaffold. These compounds were prepared
4
based on previous work from our laboratories with the design aimed at targeting compounds that
mitigate the CYP1A2 induction issue but yield metabolically stable compounds that can be
evaluated in preclinical models of PD. The lead compound, 27o, exhibited potent PAM activity
against both the human and rat mGlu receptors as well as selectivity against the other mGlu
4
receptors and a wider selectivity panel. Studies with the rodent haloperidolꢀinduced catalepsy
model confirm that 27o is active with sufficient brain and CSF levels (PK/PD). In conclusion,
ACS Paragon Plus Environment

Page 27 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
this new scaffold represents a new and improved mGlu PAM molecule for further study in
4
additional preclinical models.
EXPERIMENTAL SECTION
1
13
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
Reagents and General Methods. All H & C NMR spectra were recorded on Bruker
AVꢀ400 (400 MHz) or Bruker AVꢀNMR (600 MHz) instrument. Chemical shifts are reported in
ppm relative to residual solvent peaks as an internal standard set to δH 7.26 or δC 77.0 (CDCl3)
and δH 3.31 or δC 49.0 (CD OD). Data are reported as follows: chemical shift, multiplicity (s
3
=singlet, d = doublet, t = triplet, q = quartet, br = broad, m = multiplet), integration, coupling
constant (Hz). IR spectra were recorded as thin films and are reported in wavenumbers (cmꢀ1).
Low resolution mass spectra were obtained on an Agilent 1200 LCMS with electrospray
ionization. High resolution mass spectra were recorded on a Waters QtofꢀAPIꢀUS plus Acquity
system. The value ꢀ is the error in the measurement (in ppm) given by the equation ꢀ = [(ME –
MT)/ MT] × 106, where ME is the experimental mass and MT is the theoretical mass. The
HRMS results were obtained with ES as the ion source and leucine enkephalin as the reference.
Low resolution mass spectra were obtained on an Agilent 1200 LCMS with electrospray
ionization, with a gradient of 5ꢀ95% MeCN in 0.1% TFA water over 1.5 min. Analytical thin
layer chromatography was performed on 250 ꢁM silica gel 60 F254 plates. Visualization was
accomplished with UV light, and/or the use of ninhydrin, anisaldehyde and ceric ammonium
molybdate solutions followed by charring on a hotꢀplate. Chromatography on silica gel was
performed using Silica Gel 60 (230ꢀ400 mesh) from Sorbent Technologies. Solvents for
extraction, washing and chromatography were HPLC grade. All reagents were purchased from
Aldrich Chemical Co. and were used without purification. Flameꢀdried (under vacuum)
glassware was used for all reactions. All reagents and solvents were commercial grade and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 58
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
purified prior to use when necessary. Mass spectra were obtained on a Micromass QꢀTof APIꢀUS
mass spectrometer was used to acquire highꢀresolution mass spectrometry (HRMS) data.
Microwave synthesis was performed on an Initiator+ by Biotage.
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
Compounds were purified on a Gilson preparative reversedꢀphase HPLC system
comprised of a 333 aqueous pump with solventꢀselection valve, 334 organic pump, GXꢀ271 or
GXꢀ281 liquid hander, two column switching valves, and a 155 UV detector. UV wavelength
for fraction collection was userꢀdefined, with absorbance at 254 nm always monitored. Column:
Phenomenex Axiaꢀpacked Gemini C18, 30 x 50 mm, 5 µm. For Acidic Method: Mobile phase:
CH CN in H O (0.1% TFA). Gradient conditions: 0.75 min equilibration, followed by userꢀ
3
2
defined gradient (starting organic percentage, ending organic percentage, duration), hold at 95%
CH CN in H O (0.1% TFA) for 1 min, 50 mL/min, 23° C. For Basic Method: Mobile phase:
3
2
CH CN in H O (0.05% v/v NH OH). Gradient conditions: 0.75 min equilibration, followed by
3
2
4
userꢀdefined gradient (starting organic percentage, ending organic percentage, duration), hold at
5% CH CN in H O (0.05% v/v NH OH) for 1 min, 50 mL/min, 23° C. The purity of all
9
3
2
4
compounds was determined by LCMS to be >95% (or as stated).
General synthetic procedure for (21) or (24). To a solution of 4ꢀbromoꢀ2ꢀ
o
fluorobenzaldehyde, 20, (1 eq.) or 5ꢀbromoꢀ3ꢀfluoropicolinonitrile, 23, in THF (1 mL) at 0 C
was added RMgBr (2 eq.). The reaction was removed from the ice bath and allowed to stir at rt
for 24 h. The reaction was neutralized with a solution of NH Cl (2 mL) and diluted with EtOAc
4
(10 mL). The organic layers were washed with water (2x20 mL). The aqueous layer was
extracted with EtOAc (2x20 mL). Organic layers were combined, and solvent was removed
under reduced pressure to yield 21 or 24. The product yield was assumed to be 100% and was
used for further reactions without purification.
ACS Paragon Plus Environment

Page 29 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
-(4-bromo-2-fluoro-phenyl)-3,3-dimethyl-butan-1-ol (21a). To a solution of 4ꢀbromoꢀ
o
2
ꢀfluorobenzaldehyde (1.00 g, 4.93 mmol) in THF (1 mL) at 0 C was added chloro(2,2ꢀ
dimethylpropyl)magnesium (9.91 mL, 9.91 mmol). The reaction was removed from the ice bath
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
and allowed to stir at rt for 24 h. The reaction was neutralized with a solution of NH Cl (2 mL)
4
and diluted with EtOAc (10 mL). The organic layers were washed with water (2x20 mL). The
aqueous layer was extracted with EtOAc (2x20 mL). Organic layers were combined, and solvent
was removed under reduced pressure to yield 1ꢀ(4ꢀbromoꢀ2ꢀfluoroꢀphenyl)ꢀ3,3ꢀdimethylꢀbutanꢀ
1
ꢀol, 21a, as a yellow oil. The product yield was assumed to be quantitative and was used for
further reactions without purification.
General synthetic procedure for (22). To a crude suspension of 21 (1 eq.) in DCM (10
mL) at rt was added DessꢀMartin periodinane (1 eq.). After 1 h, the reaction was diluted with
DCM (10 mL) and washed with a saturated solution of Na S O (25 mL) and NaHCO (25 mL).
2
2
3
3
The extract was filtered through a hydrophobic filter and concentrated under reduced pressure to
yield 22. The product yield was assumed to be quantitative and was used for further reactions
without purification.
1
-(4-bromo-2-fluoro-phenyl)-3,3-dimethyl-butan-1-one (22a). To a crude suspension
of 21a (1.36 mg, 4.92 mmol) in DCM (10 mL) at rt was added DessꢀMartin periodinane (2.09
mg, 4.92 mmol). After 1 h, the reaction was diluted with DCM (10 mL) and washed with a
saturated solution of Na S O (25 mL) and NaHCO (25 mL). The extract was filtered through a
2
2
3
3
hydrophobic filter and concentrated under reduced pressure to yield 1ꢀ(4ꢀbromoꢀ2ꢀfluoroꢀ
phenyl)ꢀ3,3ꢀdimethylꢀbutanꢀ1ꢀone, 22a. The product yield was assumed to be quantitative and
was used for further reactions without purification. LCMS: R = 1.324 min, m/z = 273.2 [M +
T
+
H] .
ACS Paragon Plus Environment