As part of our ongoing small-molecule metabotropic glutamate (mGlu) receptor positive allosteric modulator (PAM) research, we performed structure-activity relationship (SAR) studies around a series of group II mGlu PAMs. Initial analogues exhibited weak activity as mGlu2 receptor PAMs and no activity at mGlu3. Compound optimization led to the identification of potent mGlu2/3 selective PAMs with no in vitro activity at mGlu1,4-8 or 45 other CNS receptors. In vitro pharmacological characterization of representative compound 44 indicated agonist-PAM activity toward mGlu2 and PAM activity at mGlu3. The most potent mGlu2/3 PAMs were characterized in assays predictive of ADME/T and pharmacokinetic (PK) properties, allowing the discovery of systemically active mGlu2/3 PAMs. On the basis of its overall profile, compound 74 was selected for behavioral studies and was shown to dose-dependently decrease cocaine self-administration in rats after intraperitoneal administration. These mGlu2/3 receptor PAMs have significant potential as small molecule tools for investigating group II mGlu pharmacology.
As part of our ongoing small-molecule metabotropic glutamate (mGlu) receptor positive allosteric modulator (PAM) research, we performed structure-activity relationship (SAR) studies around a series of group II mGlu PAMs. Initial analogues exhibited weak activity as mGlu2 receptor PAMs and no activity at mGlu3. Compound optimization led to the identification of potent mGlu2/3 selective PAMs with no in vitro activity at mGlu1,4-8 or 45 other CNS receptors. In vitro pharmacological characterization of representative compound 44 indicated agonist-PAM activity toward mGlu2 and PAM activity at mGlu3. The most potent mGlu2/3 PAMs were characterized in assays predictive of ADME/T and pharmacokinetic (PK) properties, allowing the discovery of systemically active mGlu2/3 PAMs. On the basis of its overall profile, compound 74 was selected for behavioral studies and was shown to dose-dependently decrease cocaine self-administration in rats after intraperitoneal administration. These mGlu2/3 receptor PAMs have significant potential as small molecule tools for investigating group II mGlu pharmacology.
Glutamate is the major
excitatory neurotransmitter
in the mammalian central nervous system (CNS), mediating fast synaptic
transmission through ion channels, primarily the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic
acid (AMPA) and kainate ionotropic glutamate receptor subtypes.[1] The metabotropic glutamate (mGlu) receptors are
a family of eight G protein-coupled receptors that are activated by
glutamate and perform a modulatory function in the nervous system.[2−4] The group II mGlu receptors include the mGlu2 and mGlu3 receptor subtypes, which couple with Gi/o proteins
to negatively regulate the activity of adenylyl cyclase.[3,5] Localization studies suggest that mGlu2 receptors act
predominantly as presynaptic autoreceptors to modulate the release
of glutamate into the synaptic cleft.[6] On
the other hand, mGlu3 receptors exhibit a broad distribution
in the brain and have been shown to be present on astrocytes.[7] In addition, it has been shown that activation
of mGlu3 receptors is required for the neuroprotective
effects of mGlu2/3 agonists toward N-methyl-d-aspartate (NMDA) neurotoxicity in mixed cultures of astrocytes
and neurons, whereas activation of mGlu2 receptors may
be harmful.[8]Various brain regions,
including the cerebral cortex, hippocampus, striatum, amygdala, frontal
cortex, and nucleus accumbens, display high levels of mGlu2 and mGlu3 receptor binding.[9,10] This distribution
pattern suggests a role for the mGlu2/3 receptor subtypes
in the pathology of neuropsychiatric disorders such as anxiety,[11] depression,[12,13] schizophrenia,[14,15] drug dependence,[16−22] neuroprotection,[23,24] Alzheimer’s disease,[25] and sleep/wake architecture.[26] Thus, there is significant potential for the development
of selective group II mGlu receptor activators, including agonists
and positive allosteric modulators (PAMs), for the treatment of CNS
diseases caused by aberrant glutamatergic signaling.Orthosteric
(glutamate site) mGlu2/3 agonists such as LY379268[27] are constrained amino acid analogues that are
typically equipotent at both mGlu2 and mGlu3, presumably because of the high degree of sequence homology at the
glutamate-binding site for these two receptors.[3] Although LY541850, an orthosteric ligand with mixed mGlu2 agonist/mGlu3 antagonist activity, has been reported,[28,29] currently there are no truly mGlu2 selective orthosteric
ligands available. The systemically active mGlu2/3 receptor
agonist LY379268 has been shown to reduce glutamate release from presynaptic
terminals in many brain regions and thus to decrease glutamatergic
neurotransmission.[30,31] LY379268 is active in several
different rodent models of CNS disorders including anxiety,[32] schizophrenia,[33] Huntington’s
disease,[34−36] and drug dependence, where it attenuates cocaine
self-administration both in rats[20,37] and in squirrel
monkeys.[18] However, LY379268 also inhibits
responding for food and food-seeking behavior,[20,21,37,38] suggesting
that mGlu2/3 receptor agonists exhibit nonselective actions
on responding for drug and nondrug reinforcers.In addition
to orthosteric agonists of Group II mGlu receptors, multiple reports
have recently described the synthesis and characterization of selective
mGlu2 receptor PAMs.[5,39−46] These compounds invariably resulted from the optimization of hits
obtained from high-throughput screening (HTS) of small molecule libraries.
PAMs, through their interaction at allosteric sites on the mGlu receptor,
positively modulate (i.e. potentiate) the effects
of the endogenous orthosteric mGlu agonist glutamate. The advantages
of PAMs compared with orthosteric agonists includes enhanced subtype
selectivity, the potential for spatial and temporal modulation of
receptor activation, and ease of optimization and fine-tuning of druglike
properties.[47] We recently reported the
design, synthesis, and in vitro and in vivo characterization of a series of potent and selective
mGlu2 receptor PAMs.[48,49] The optimized compounds 2 and 3 (Figure 1), which
were developed using the prototypical mGlu2 receptor PAM
BINA (1) as a starting point, were employed to determine
the effects of selectively activating mGlu2 receptors on
cocaine or nicotine dependence.[48,49] In these studies we
showed that compound 3, unlike the mGlu2/3 orthosteric agonist LY379268, decreased cocaine self-administration
in rats at doses that did not affect responding for food.[48] We also showed that compound 2 dose-dependently
decreased nicotine self-administration in rats following oral (po)
administration.[49] Taken together, our data
suggest that mGlu2 receptor PAMs have the potential for
therapeutic utility in the treatment of drug dependence.
Figure 1
Structures
of mGlu2 receptor PAMs developed from BINA (1).
Structures
of mGlu2 receptor PAMs developed from BINA (1).As noted above, there have been
many accounts in the literature describing selective mGlu2 receptor PAMs, whereas very little has been reported on compounds
that potentiate the effects of glutamate at mGlu3 receptors.
This is somewhat surprising given the significant sequence homology
(approximately 75%) within the transmembrane regions of mGlu2 and mGlu3 receptors. A single disclosure by Schann and
co-workers described compound 4 (Figure 2) which was reported to be a mixed mGlu2 receptor
negative allosteric modulator (NAM)/mGlu3 receptor PAM.[50] We hypothesized that it might be possible to
design and synthesize compounds that activate both mGlu2 and mGlu3 receptors through an allosteric mechanism.
Considering the dearth of information on mixed mGlu2/3 receptor
PAMs, the development of such compounds would provide valuable pharmacological
tools. For example, a CNS penetrant mGlu2/3 receptor PAM
could facilitate investigations into whether effects on food responding
in rats are due to general activation of mGlu3 receptors
or an effect specific to direct activation of the mGlu receptor by
agonists that act at the mGlu orthosteric binding site.
Figure 2
Structure of
recently reported mGlu2 receptor NAM/mGlu3 receptor
PAM 4.
Structure of
recently reported mGlu2 receptor NAM/mGlu3 receptor
PAM 4.The strategy for the
design and synthesis of mGlu2/3 receptor PAMs grew out
of our general approach to the development of selective mGlu2 receptor PAMs. In addition to exploring the structure–activity
relationship (SAR) around compound 1, which led to the
series of selective mGlu2 receptor PAMs exemplified by 2 and 3,[48,49] we also investigated
compounds such as 5–7 (Figure 3). The mGlu2 receptor PAMs 5–7 are representative of a series developed by
Pinkerton and co-workers at Merck.[51] The
mGlu2 receptor PAMs in this structural class were reported
to exhibit varying degrees of in vitro potency and efficacy with little
information provided regarding subtype selectivity. Compound 8, a selective mGlu2 receptor PAM that bears a
resemblance to compounds 5–7, was
recently reported by researchers at Eli Lilly.[52] With the exception of the optimized compound 8, most of the compounds in this series were reported to be modestly
potent at mGlu2 receptors in vitro and displayed suboptimal
pharmacokinetic (PK) profiles and brain penetration in vivo. Our initial
goal was to develop group II mGlu receptor PAMs having the potential
for systemic activity, with a focus on maintaining or improving potency
and efficacy at mGlu2 while in parallel investigating the
PAM activity at mGlu3 receptors. Herein we describe the
design, synthesis, and pharmacological characterization of a library
of analogues from which a series of mixed mGlu2/3 receptor
PAMs with unique pharmacology were discovered. Furthermore, because
the selective mGlu2 receptor PAM BINA decreased cocaine-
but not food-maintained responding,[37] we
investigated the in vivo efficacy of one of the synthesized mGlu2/3 receptor PAMs on cocaine- and food-maintained responding
in rats.
Figure 3
Structures of mGlu2 receptor PAMs in the acetophenone
series.
Structures of mGlu2 receptor PAMs in the acetophenone
series.
Chemistry
Examination of compounds 5–7 suggested the presence of common structural
elements. Specifically, they consist of an aryl ether connected via
a butyl ether linker to an acetophenone derivative. Thus, we envisioned
a synthetic strategy wherein the key building blocks were synthesized
and systematically assembled to generate a compound library rapidly
and efficiently. This would allow the production of analogues with
sufficient diversity to investigate the requirements for optimal potency,
efficacy, and pharmacokinetic properties. The structural components
were synthesized or purchased and incorporated into the new group
II mGlu receptor PAMs as shown in Scheme 1.
Briefly, commercially available carboxylic acids were converted to
the corresponding acyl chloride derivatives (9) using
oxalyl chloride in CH2Cl2. After removal of
solvents, the acyl chlorides were employed in a Friedel–Crafts
acylation of substituted phenols (10) using aluminum
chloride to provide the key acetophenone derivatives 11. The phenol derivatives (12) were coupled with 1,4-dibromobutane
by heating with potassium carbonate in acetonitrile to provide the
corresponding bromobutoxybenzoate derivative (13). Finally,
Finkelstein alkylation of intermediate 11 with 13 under microwave conditions delivered the ester derivatives
(14–17 and 18′–75′), which were saponified with potassium
hydroxide to provide the target carboxylic acid derivatives 18–75.
Scheme 1
General Synthesis of New Group II
mGlu Receptor PAMs
Results and Discussion
The initial in vitro evaluation
of analogues was performed in thallium flux assays in human embryonic
kidney 293 (HEK) cells expressing heteromeric G-protein-coupled inwardly
rectifying potassium (GIRK) 1/2 channels and rat mGlu2 or
ratmGlu3 receptors.[41] The potency
and efficacy of compounds were established by measuring the concentration–response
relationship that potentiates the effect of an EC20 concentration
of glutamate. After determination of the concentration–response
relationship, the effect of a maximally effective concentration (% Max)
of each compound was determined. For the allosteric potentiators,
potency is expressed as the EC50 for potentiation of the
glutamate EC20. Table 1 shows the
results of testing the compounds against rat mGlu2 or mGlu3 receptors in vitro.
Table 1
In Vitro Potency
and Efficacy Data at mGlu2 or mGlu3 Receptors
for PAMsa
mGlu2 PAM EC50 (μM)
data and glutamate Max (%) data represent the mean ± SEM for
at least three independent experiments performed in triplicate. ND
= not determined.
mGlu2 PAM EC50 (μM)
data and glutamate Max (%) data represent the mean ± SEM for
at least three independent experiments performed in triplicate. ND
= not determined.Our initial
strategy for the design of the butyl ether derived PAMs library was
to systematically explore different substitution patterns on the left-hand
aryl ether moiety combined with a set of four right-hand acetophenones.
Specifically, acetophenone derivatives were synthesized in which X
was either methyl (e.g., 14) or hydroxyl (e.g., 16) and where R3 was either isopropylmethylene
(e.g., 14 and 16) or cyclopentylmethylene
(e.g., 15 and 17). In the first phase of
library synthesis, analogues in which R1 was a 4-carboxylate
moiety (14–21) were prepared and
tested in vitro. Many of the compounds were inactive or only very
weakly active in the assays. Known compounds 5 and 6 were synthesized and tested as part of the library to provide
a benchmark of activity for the series. Some important observations
were made during this phase regarding the activity of the analogues
tested. First, benchmark compounds 5 and 6 exhibited PAM activity at mGlu2 but not at mGlu3 receptors in the GIRK assay. Moreover, compound 6 was
only very weakly active (EC50 > 10 μM, 45% efficacy)
in the GIRK assay against mGlu2. Second, in general, compounds
containing X = Me were less potent than compounds in which X = OH.
Third, whereas the methyl 4-benzoate derivatives 16 and 17 were inactive, the corresponding free carboxylic acid derivatives 20 and 21 were active (20) or weakly
active (21) as PAMs at both mGlu2 and mGlu3 receptors. Furthermore, compound 20 was at least
an order of magnitude more potent as a PAM at mGlu2 than
benchmark compound 5 and had an EC50 of 2.7
μM (52% efficacy) against mGlu3. Encouraged by the
PAM activity found for compounds 20 and 21, analogues containing free carboxylic acid moieties at other positions
on the aryl ether ring were synthesized and tested (compounds 22–29). In the 3-carboxyl series compounds 22 and 23 (X = Me) were inactive whereas compounds 24 and 25 (X = OH) exhibited PAM activity at
both mGlu2 and mGlu3 receptors. Interestingly,
in the 2-carboxyl series (compounds 26–29), all four compounds were active as mGlu2 receptor PAMs
with no activity at mGlu3. These results prompted us to
synthesize and test the next set of analogues (compounds 30–51) containing carboxylic acid moieties at the
3- or 4-position (R1) in addition to substituents at other
positions around the aryl ether ring (i.e., R2 = F, Cl,
Me, or OMe). In this series, the analogues containing X = Me (30, 31, 34, 35) continued
the trend of very weak (31, EC50 > 10 μM,
36% efficacy) or no mGlu3 activity (30, 34, 35, 12%–14% efficacy), and therefore,
all subsequent analogues were synthesized with X = OH. Furthermore,
several of the analogues in this group (39, 41, 43, 47, and 49) were active
as PAMs at mGlu2 but were essentially inactive at mGlu3 receptors (24%–31% efficacy). However, in addition
to being submicromolar PAMs at mGlu2, compounds 32, 33, 37, 40, 42, 45, 46, 48, and 51 were active as PAMs at mGlu3 receptors in the micromolar
range (EC50 ≈ 1–4 μM, 40%–98%
efficacy). Most promising of all, compounds 36, 44, and 50 possessed submicromolar PAM activity
at mGlu3 receptors as well as good potency at mGlu2. Especially noteworthy were compounds 44 and 50, with EC50 values in the 600–700 nM range
at mGlu3 with excellent efficacy as PAMs (74% and 100%,
respectively). These results suggested that our goal of producing
PAMs with potent activity at both mGlu2 and mGlu3 receptors was attainable. Having identified favorable substitution
patterns for the left-hand aryl ether moiety, we next designed a series
of analogues (compounds 52–75) to
investigate which ketone alkyl group (R3) would impart
the best potency and efficacy at both mGlu2 and mGlu3 receptors. Thus, in this set of analogues, compounds were
synthesized with R3 = methyl (52–56), ethyl (57–61), isopropyl
(62–66), n-propyl
(67–71), or tert-butylmethylene (72–75). Interestingly,
none of the methyl ketone derivatives (52–56) had activity at mGlu3 receptors, while in the
ethyl ketone series (57–61) only
compound 58 (5-CO2H, 2-OMe) had activity at
mGlu3 (EC50 = 6359 nM, 63% efficacy). In the isopropyl ketone series, only compound 63 (5-CO2H, 2-OMe) had activity at mGlu3 (EC50 = 588 nM, 60% efficacy). On the other hand, in the n-propyl series, all compounds demonstrated some level of
activity. While 69 and 71 were weakly active
at mGlu3 (EC50s > 10 μM, 53% and 41%
efficacy, respectively), 67 (4-CO2H, 2-OMe), 68 (5-CO2H, 2-OMe), and 70 (3-CO2H, 4-F) demonstrated more potent mGlu3 activity
(EC50 = 1830 nM, 62% efficacy, EC50 = 496 nM,
64% efficacy, EC50 = 438 nM, 57% efficacy, respectively).
It is noteworthy that compound 67 (4-CO2H,
2-OMe) was the most potent PAM and selective for mGlu2 with
good efficacy (EC50 = 34 nM, 86% efficacy). Finally, in
the tert-butylmethylene series (compounds 72–75) all four analogues exhibited good potency
and excellent efficacy for both mGlu2 and mGlu3 receptors. Compound 72 was the most potent against
mGlu3 (EC50 = 151 nM, 106% efficacy) with similar
potency against mGlu2, while compounds 73–75 fell into the 258–310 nM potency range (102%–110%
efficacy) against mGlu3 receptors. Thus, a combination
of favored left-hand aryl ether moieties with R3 = tert-butylmethylene provides compounds with unique PAM activity
at both mGlu2 and mGlu3 receptors.With
these results in hand, we tested some of the most promising new PAMs
to provide an estimation of their druglike properties in vitro using
absorption, distribution, metabolism, and excretion (ADME) assays.
The results of these studies are shown in Table 2. The permeability across artificial membranes was assessed via the
parallel artificial membrane permeability assay (PAMPA) assay and
showed a range of results, with most compounds showing some degree
of permeability. Plasma and microsomal stability was likewise acceptable,
with only four of the analogues showing less than 20% remaining after
1 h. In addition, an algorithm was used to provide an estimate of
plasma protein binding for 11 of the analogues. The data, which are
provided in the Supporting Information (Table
S1), suggest that the compounds in this series are all >90% plasma
protein bound.
Table 2
In Vitro ADME Data for Group II mGlu PAMs
Permeability is
monitored by measuring the amount of compound that can diffuse through
a polar brain lipid membrane to predict BBB permeability.[42]
Percent remaining after incubation for 60 min at 37.5 °C.
Compounds 20, 25, 33, 44, 67, 72, and 73 were profiled against the remaining mGlu receptor subtypes
to determine their selectivity relative to mGlu2 and mGlu3 (Table 3). With the exception of 67, which displayed weak antagonist/NAM activity (IC50 > 10 μM) at mGlu1 and mGlu4, and
compound 73, which showed weak PAM activity at mGlu6 (EC50 > 10 μM), these compounds were
found to be highly selective for mGlu2 and mGlu3. As a representative of this series, compound 72 was
profiled against a representative panel of CNS receptors through the
NIMH Psychoactive Drug Screening Program (PDSP; see http://pdsp.med.unc.edu/indexR.html for details). As shown in Table S2 (see Supporting
Information), at 10 μM, no binding activity was detected
for compound 72 at 45 CNS receptors, suggesting that
the new mGlu2/3 PAMs have a low likelihood of off-target
activity.
Table 3
mGlu Receptor Subtype
Selectivitya
compound
20
25
33
44
67
72
73
mGlu1
inactiveb
inactiveb
inactiveb
inactiveb
antagonist
inactiveb
inactiveb
FS = 0.3
Emin = 3%
Emax = 79%
mGlu2
Ago-PAM
Ago-PAM
Ago-PAM
Ago-PAM
Ago-PAM
Ago-PAM
Ago-PAM
FS = 3.2
FS = 12.5
FS = 13.3
FS = ND
FS = ND
FS = ND
FS = ND
Emin = 54%
Emin = 45%
Emin = 54%
Emin =
90%
Emin = 79%
Emin = 72%
Emin = 79%
Emax = 77%
Emax = 71%
Emax =
77%
Emax = 95%
Emax = 92%
Emax = 95%
Emax = 92%
mGlu3
PAM
PAM
PAM
PAM
PAM
Ago-PAM
Ago-PAM
FS = 2.7
FS = 4.5
FS = 2.5
FS = 3.9
FS = 2.9
FS = 7.1
FS = 8.9
Emin = 9%
Emin = 6%
Emin = 10%
Emin = 4%
Emin = 1%
Emin = 54%
Emin =
30%
Emax = 153%
Emax = 143%
Emax = 143%
Emax = 116%
Emax = 88%
Emax = 95%
Emax = 99%
mGlu4
inactiveb
inactiveb
inactiveb
inactiveb
antagonist
inactiveb
inactiveb
FS = 1.9
Emin = 0%
Emax = 74%
mGlu5
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
mGlu6
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
PAM
FS = 2.0
Emin = 8%
Emax = 93%
mGlu7
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
mGlu8
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
inactiveb
In these selectivity experiments, for all receptors a full concentration–response
of agonist was performed once in triplicate in the presence and absence
of a 10 μM final concentration of each compound. This allows
determination of positive allosteric modulator (PAM) (left shift of
the agonist concentration response curve), antagonist (right shift
in the agonist concentration response with a possible decrease in
maximal agonist response), and agonist (increase in baseline response)
activity in a single experiment. General activity for each compound
at each mGlu is listed (PAM, antagonist, Ago-PAM, inactive) followed
by the fold-shift (FS) of the agonist concentration–response
obtained. Where tested compounds demonstrate activity toward an mGlu
receptor subtype, the maximal (Emax) and
minimal (Emin) responses of the concentration–response
of agonist are indicated. Where 10 μM test compound induced
greater than a 2-fold shift (FS) of the glutamate concentration–response
curve (L-AP4 in the case of mGlu7), full compound concentration–response
curves were performed in triplicate on 3 different days to assess
compound potency. Compound 67 showed weak antagonist/NAM
activity (IC50 > 10 μM) at mGlu1 and
mGlu4, and compound 73 showed weak PAM activity
at mGlu6 (EC50 > 10 μM).
Inactive compounds show no ability to
left- or right-shift the agonist concentration response curve at 10
μM. ND = not determined.
Permeability is
monitored by measuring the amount of compound that can diffuse through
a polar brain lipid membrane to predict BBB permeability.[42]Percent remaining after incubation for 60 min at 37.5 °C.In these selectivity experiments, for all receptors a full concentration–response
of agonist was performed once in triplicate in the presence and absence
of a 10 μM final concentration of each compound. This allows
determination of positive allosteric modulator (PAM) (left shift of
the agonist concentration response curve), antagonist (right shift
in the agonist concentration response with a possible decrease in
maximal agonist response), and agonist (increase in baseline response)
activity in a single experiment. General activity for each compound
at each mGlu is listed (PAM, antagonist, Ago-PAM, inactive) followed
by the fold-shift (FS) of the agonist concentration–response
obtained. Where tested compounds demonstrate activity toward an mGlu
receptor subtype, the maximal (Emax) and
minimal (Emin) responses of the concentration–response
of agonist are indicated. Where 10 μM test compound induced
greater than a 2-fold shift (FS) of the glutamate concentration–response
curve (L-AP4 in the case of mGlu7), full compound concentration–response
curves were performed in triplicate on 3 different days to assess
compound potency. Compound 67 showed weak antagonist/NAM
activity (IC50 > 10 μM) at mGlu1 and
mGlu4, and compound 73 showed weak PAM activity
at mGlu6 (EC50 > 10 μM).Inactive compounds show no ability to
left- or right-shift the agonist concentration response curve at 10
μM. ND = not determined.At a relatively
early stage of the project and on the basis of the overall in vitro
profile of the PAMs (Tables 1 and 2), we selected compounds 20, 36, 44, and 50 for in vivo assessment
of pharmacokinetic (PK) properties in rats. For this initial evaluation
we determined the PK properties of the compounds by oral (po) and
intravenous (iv) routes of administration as shown in Tables 4 and 5, respectively. The
PAMs were found to be systemically bioavailable with half-life (t1/2) values of greater than 90 min when dosed
po and demonstrate a range of maximal plasma levels from a low of
1.05 μM (50) to a high of 12.46 μM (36) (Table 4). All compounds had moderate
volume of distribution at steady state (Vdss) and medium
to high clearance (CL) values, indicating moderate metabolism with
a primary distribution in plasma and extracellular fluids, suggesting
that one or more of the PAMs might have promise as candidates for
in vivo studies (Table 5). The four compounds
exhibited low (20) to good oral bioavailability (50) (% F) albeit at the relatively high
oral dose of 20 mg/kg; however, the brain levels of 20, 36, and 44 were low, resulting in low
brain/plasma ratios. Although the brain/plasma ratios of these compounds
are low, the total brain concentrations of 20 and 44 are 9-fold and 18-fold above the in vitro EC50 for mGlu2, respectively, and close to the in vitro EC50s for mGlu3.
Table 4
In Vivo PK Data for
mGlu2/mGlu3 PAMs in Rats after po Administration
(20 mg/kg)a
compd
Cmax (μM)
Tmax (min)
AUC(0→t) (μmol/L)·min
T1/2 (min)
F (%)
brain (μM)
plasma
(μM)
brain/plasma
20
2.57 ± 0.45
135 ± 15
544.8 ± 75.4
303 ± 55
23.5
1.12 ± 0.43
5.29 ± 1.98
0.20 ± 0.03
36
12.46 ± 6.42
30 ± 0
720.5 ± 283.4
89 ± 17
58.2
0.23 ± 0.06
4.23 ± 1.19
0.06 ± 0.03
44
2.75 ± 0.47
90 ± 17
1274.6 ± 246.3
471 ± 89
41.9
0.81 ± 0.17
16.38 ± 1.13
0.05 ± 0.01
50
1.05 ± 0.15
96 ± 40
466.2 ± 105.2
569 ± 199
60.4
ND
ND
ND
Cmax: maximum concentration of the compound detected in
plasma. Tmax: time at Cmax. AUC: area under the curve. t1/2: terminal half-life. F: oral bioavailability.
Brains and plasma were harvested at or near the Tmax. Compounds were dosed in a volume of 2 mL/kg po (n = 3–4) at 20 mg/kg in 0.6% Tween 80. ND = not determined.
Table 5
In Vivo PK Data for
mGlu2/mGlu3 PAMs in Rats after iv Administration
(2 mg/kg)a
compd
Cmax (μM)
CL (mL·min–1·kg–1)
Vdss (L·kg–1)
AUC(0→t) (μmol/L)·min
T1/2 (min)
20
5.00 ± 0.60
21.09 ± 3.66
0.81 ± 0.14
231.7 ± 42.12
30 ± 2
36
3.52 ± 0.13
38.78 ± 2.55
0.89 ± 0.03
123.7 ± 7.7
18 ± 3
44
6.43 ± 0.39
14.58 ± 1.18
0.55 ± 0.04
304.6 ± 25.8
28 ± 0
50
2.15 ± 0.14
58.29 ± 7.98
2.87 ± 0.60
77.2 ± 8.0
57 ± 12
Cmax: maximum concentration of the compound detected in
plasma. AUC: area under the curve. t1/2: terminal half-life. CL: clearance. Vdss: steady state
volume of distribution. Compounds were injected in a volume of 1 mL/kg
iv (n = 3–4) through an iv catheter at 2 mg/kg
in 0.6% Tween 80 or in 1 M NaOH. pH was adjusted to ∼7.
Cmax: maximum concentration of the compound detected in
plasma. Tmax: time at Cmax. AUC: area under the curve. t1/2: terminal half-life. F: oral bioavailability.
Brains and plasma were harvested at or near the Tmax. Compounds were dosed in a volume of 2 mL/kg po (n = 3–4) at 20 mg/kg in 0.6% Tween 80. ND = not determined.Cmax: maximum concentration of the compound detected in
plasma. AUC: area under the curve. t1/2: terminal half-life. CL: clearance. Vdss: steady state
volume of distribution. Compounds were injected in a volume of 1 mL/kg
iv (n = 3–4) through an iv catheter at 2 mg/kg
in 0.6% Tween 80 or in 1 M NaOH. pH was adjusted to ∼7.Since compound 44 had
shown the best mix of in vitro and in vivo properties at this stage,
including potent activity and efficacy at mGlu3 receptors,
this PAM was selected for comprehensive in vitro pharmacological characterization.
We began with a more detailed investigation of the activity of 44 in the mGlu2 and mGlu3 GIRK thallium
flux assays. The nature of the GIRK assay requires that each compound
is screened for a single mode of pharmacology at a time, since activity
is only detected through the GIRK channel when thallium is added to
the assay. The data presented in Table 1 represent
compounds screened for activity in “PAM mode”, where
a test compound is added, followed 2.5 min later by an EC20 concentration of glutamate in the presence of thallium. We had noted
that the response of 44 in “PAM mode” toward
mGlu2 decreased slightly at higher concentrations of test
compound (Figure 4A). This decrease could be
caused by either receptor desensitization or intrinsic agonist activity
of 44 that was not detected because of the mode in which
the functional assay was performed. To investigate this further, we
carried out the same experiments in the absence of an EC20 concentration of agonist (agonist mode). For these experiments,
test compounds were added in the presence of thallium and GIRK activity
was immediately monitored. We found that 44 displayed
intrinsic agonist activity toward mGlu2 but not mGlu3 in the GIRK assay (Figure 4B). Thus,
this compound is best characterized as having mGlu2 agonist-PAM
activity and mGlu3 PAM activity in the GIRK thallium flux
assays.
Figure 4
Compound 44 displays Ago-PAM activity toward mGlu2 and PAM activity toward mGlu3 in GIRK thallium-flux
assays. A concentration–response of 44 was performed
in the presence (A) and absence (B) of an EC20 of glutamate
in either the mGlu2 GIRK assay (squares) or mGlu3 GIRK assay (triangles). In the mGlu2 assay, 44 displays both agonist and PAM activity and is characterized as an
Ago-PAM. In the mGlu3 assay, only PAM activity is detected.
Data were analyzed using nonlinear regression, providing EC50 values for each curve. Data were obtained from three separate experiments
performed in triplicate, normalized to the response to 100 μM
glutamate in each experiment, and are expressed as the mean ±
SEM.
Compound 44 displays Ago-PAM activity toward mGlu2 and PAM activity toward mGlu3 in GIRK thallium-flux
assays. A concentration–response of 44 was performed
in the presence (A) and absence (B) of an EC20 of glutamate
in either the mGlu2 GIRK assay (squares) or mGlu3 GIRK assay (triangles). In the mGlu2 assay, 44 displays both agonist and PAM activity and is characterized as an
Ago-PAM. In the mGlu3 assay, only PAM activity is detected.
Data were analyzed using nonlinear regression, providing EC50 values for each curve. Data were obtained from three separate experiments
performed in triplicate, normalized to the response to 100 μM
glutamate in each experiment, and are expressed as the mean ±
SEM.We next evaluated 44 in a fold-shift assay, another measure of the potentiating activity
of a PAM toward the orthosteric ligand glutamate (Figure 5). Fold-shift values are calculated by determining
the ratio of the potency of the orthosteric agonist glutamate in the
presence and absence of increasing concentrations of an allosteric
modulator. Increasing fixed concentrations of 44 dose-dependently
shifted the glutamate concentration–response of mGlu2 (Figure 5A) and mGlu3 (Figure 5B) to the left, consistent with an enhancement of
glutamate responses. For these assays, the Ago-PAM activity toward
mGlu2 is readily apparent as the increase in baseline at
low concentrations of glutamate (Figure 5A).
These data are in contrast with the results for mGlu3 (Figure 5B), which does not show a change in baseline of
the glutamate dose–response.
Figure 5
Compound 44 dose-dependently
induces a leftward shift in the glutamate concentration response at
(A) mGlu2 and (B) mGlu3 in GIRK thallium flux
assays. The increase in baseline in the mGlu2 GIRK assay
at higher concentrations of 44 is due to the Ago-PAM
activity of this compound. The leftward shifts induced by 44 indicate a potentiation of the response of mGlu2 and
mGlu3 to glutamate. The maximal fold-shift at mGlu2 is 4.50 ± 0.96 and was derived from the test concentration
of 44 (300 nM) due to Ago-PAM activity. The maximal fold-shift
at mGlu3 is 5.48 ± 0.27 for the 10 μM test concentration.
Concentration–response relationships were generated by adding
a fixed concentration of 44 to cells as indicated, followed
by increasing concentrations of glutamate. Data were analyzed using
nonlinear regression, providing EC50 values for each curve.
Data were obtained from three separate experiments performed in duplicate,
normalized to the response to 100 μM glutamate in each experiment,
and are expressed as the mean ± SEM.
Compound 44 dose-dependently
induces a leftward shift in the glutamate concentration response at
(A) mGlu2 and (B) mGlu3 in GIRK thallium flux
assays. The increase in baseline in the mGlu2 GIRK assay
at higher concentrations of 44 is due to the Ago-PAM
activity of this compound. The leftward shifts induced by 44 indicate a potentiation of the response of mGlu2 and
mGlu3 to glutamate. The maximal fold-shift at mGlu2 is 4.50 ± 0.96 and was derived from the test concentration
of 44 (300 nM) due to Ago-PAM activity. The maximal fold-shift
at mGlu3 is 5.48 ± 0.27 for the 10 μM test concentration.
Concentration–response relationships were generated by adding
a fixed concentration of 44 to cells as indicated, followed
by increasing concentrations of glutamate. Data were analyzed using
nonlinear regression, providing EC50 values for each curve.
Data were obtained from three separate experiments performed in duplicate,
normalized to the response to 100 μM glutamate in each experiment,
and are expressed as the mean ± SEM.In the final set of pharmacological characterization experiments,
we evaluated the potent PAM 44 in an orthogonal assay
of mGlu3 and mGlu2 activity. For mGlu3 we utilized the TREx tetracycline-inducible system (Invitrogen).
We developed a cell line in which the expression of mGlu3 is dose-dependently induced by tetracycline (Tet) and functionally
coupled to calcium mobilization by the promiscuous G protein Gα15 (Figure 6). In the absence
of Tet, no measurable expression of mGlu3 is detected either
by Western blot (Figure 6A) or by functional
response to calcium mobilization (Figure 6B).
The optimal calcium mobilization response for this cell line was achieved
at 20 ng/mL Tet for 20 h prior to assay. This concentration of Tet
was then utilized for further characterization of 44 in
the TREx293 mGlu3 Gα15 calcium assay (Figure 7B), which shows that 44 demonstrates
mGlu3 PAM activity in this orthogonal assay whereas the
mGlu2 selective PAM BINA remains inactive. These compounds
were also evaluated in calcium assays utilizing HEK293A mGlu2 Gα15 cells as shown in Figure 7A. Unlike the mGlu2 GIRK assay where it displays
Ago-PAM activity, compound 44 behaves as a PAM in the
calcium assay. In future experiments, it will be essential to determine
the potency and selectivity of 44 in native tissue preparations
to determine whether the observed profile in vitro is replicated in
vivo.
Figure 6
Development of a cell line with inducible mGlu3 expression
coupled to calcium mobilization via the promiscuous G protein Gα15. (A) TREx293 mGlu3 Gα15 cells were stimulated with the indicated concentrations of tetracycline
(Tet) for 20 h. Protein lysates were prepared. Equivalent amounts
of protein were loaded for all lanes, and mGlu2/3 expression
was detected by Western blot. Both monomeric and dimeric mGlu3 were detected as indicated. (B) TREx293 mGlu3 Gα15 cells were stimulated with the indicated concentrations
of Tet for 20 h, and a calcium mobilization assay was performed. Tet
dose-dependently induced a glutamate-simulated calcium response that
was maximal at 20 ng/mL Tet. Data were analyzed using nonlinear regression.
Data were obtained from three separate experiments performed in triplicate,
normalized to the response to 100 μM glutamate in each experiment,
and are expressed as the mean ± SEM.
Figure 7
Compound 44 displays PAM activity toward mGlu and mGlu3 in calcium assays utilizing the
promiscuous G protein Gα15. A concentration–response
of 44 (triangles) and the control mGlu2 selective PAM
BINA (squares) was performed in the presence of an EC20 of glutamate in either the (A) HEK293A mGlu2 Gα15 calcium assay or (B) TREx293 mGlu3 Gα15 calcium assay. In both assays, 44 displays PAM activity.
BINA displays PAM activity in the mGlu2 calcium assay but
is inactive in the mGlu3 calcium assay. For this assay,
mGlu3 expression was induced with 20 ng/mL Tet for 20 h
prior to assay. Data were analyzed using nonlinear regression, providing
EC50 values for each curve. Data were obtained from three
separate experiments performed in triplicate, normalized to the response
to 100 μM glutamate in each experiment, and are expressed as
the mean ± SEM.
Development of a cell line with inducible mGlu3 expression
coupled to calcium mobilization via the promiscuous G protein Gα15. (A) TREx293 mGlu3 Gα15 cells were stimulated with the indicated concentrations of tetracycline
(Tet) for 20 h. Protein lysates were prepared. Equivalent amounts
of protein were loaded for all lanes, and mGlu2/3 expression
was detected by Western blot. Both monomeric and dimeric mGlu3 were detected as indicated. (B) TREx293 mGlu3 Gα15 cells were stimulated with the indicated concentrations
of Tet for 20 h, and a calcium mobilization assay was performed. Tet
dose-dependently induced a glutamate-simulated calcium response that
was maximal at 20 ng/mL Tet. Data were analyzed using nonlinear regression.
Data were obtained from three separate experiments performed in triplicate,
normalized to the response to 100 μM glutamate in each experiment,
and are expressed as the mean ± SEM.Compound 44 displays PAM activity toward mGlu and mGlu3 in calcium assays utilizing the
promiscuous G protein Gα15. A concentration–response
of 44 (triangles) and the control mGlu2 selective PAM
BINA (squares) was performed in the presence of an EC20 of glutamate in either the (A) HEK293A mGlu2 Gα15 calcium assay or (B) TREx293 mGlu3 Gα15 calcium assay. In both assays, 44 displays PAM activity.
BINA displays PAM activity in the mGlu2 calcium assay but
is inactive in the mGlu3calcium assay. For this assay,
mGlu3 expression was induced with 20 ng/mL Tet for 20 h
prior to assay. Data were analyzed using nonlinear regression, providing
EC50 values for each curve. Data were obtained from three
separate experiments performed in triplicate, normalized to the response
to 100 μM glutamate in each experiment, and are expressed as
the mean ± SEM.Following the in vitro pharmacological characterization of 44 as a representative analogue of this class of mGlu2/3 PAMs, we wished to evaluate a member of this series in
efficacy studies in rats. Given the low brain levels achieved by po
dosing for compounds 20, 36, 44, and 50 (Table 4), we evaluated
nine compounds by intraperitoneal (ip) dosing in order to avoid first
pass metabolism and to cast a wider net for a compound suitable for
rat efficacy studies (Table 6). All compound
plasma levels were determined, but only those with the highest plasma
concentrations (i.e., 44, 73, 74, and 75) were evaluated for brain levels. On the basis
of its combination of potency, selectivity, and PK properties, compound 74 was selected for efficacy studies in rats.
Table 6
In Vivo PK for mGlu2/3 PAMs in Rats after ip Administration
(10 mg/kg)a
compd
plasma (μM)a
plasma t1/2 (min)
brain (μM)a
brain/plasma
44
10.57 ± 2.52
43
0.23 ± 0.10
0.03 ± 0.01
50
3.29 ± 1.27
20
ND
ND
60
4.49 ± 0.49
22
ND
ND
65
5.44 ± 2.92
30
ND
ND
68
2.18 ± 0.58
28
ND
ND
72
2.44 ± 2.19
27
ND
ND
73
6.81 ± 0.84
132
0.47 ± 0.35
0.02 ± 0.01
74
17.05 ± 0.19
106
0.56 ± 0.10
0.03 ± 0.01
75
6.52 ± 0.54
80
0.22 ± 0.04
0.01 ± 0.01
Maximum concentration of the compound detected in plasma
or brain. t1/2: terminal half-life. Compounds
were dosed ip (n = 3) at 10 mg/kg in 10% EtOH/1%
Tween 80. pH was adjusted to ∼7. Brains and plasma were harvested
at the Tmax (30 min for all tested). ND = not determined.
Maximum concentration of the compound detected in plasma
or brain. t1/2: terminal half-life. Compounds
were dosed ip (n = 3) at 10 mg/kg in 10% EtOH/1%
Tween 80. pH was adjusted to ∼7. Brains and plasma were harvested
at the Tmax (30 min for all tested). ND = not determined.Taking all the relevant data
into account, we determined that the mGlu2/3 receptor PAM 74 would be a good candidate for evaluation in a rat model
of cocaine dependence. When assessed in vivo, compound 74 dose-dependently decreased cocaine- and food-maintained responding
[compound 74 dose main effect: F3,51 = 14.55; p < 0.0001] (Figure 8). Interestingly, however, cocaine-maintained responding
was decreased to a greater extent than food-maintained responding
at the highest dose tested (40 mg/kg; p < 0.05).
Because of the within-subjects design of the dose response (i.e.,
each rat received each dose of compound 74 using a Latin-square
design), it was not possible to collect brain samples to determine
brain concentrations of compound 74 at 40 mg/kg during
behavioral testing. It is unlikely that brain concentrations of compound 74 differed between cocaine- and food-maintained rats at this
dose, suggesting that the observed differences in behavior were not
a function of group differences in brain pharmacokinetic properties
of compound 74, although this should be confirmed in
future studies. We have previously shown that the mGlu2/3 receptor agonist LY379268 similarly decreased both cocaine- and
food-maintained responding.[37,53] Moreover, the selective
mGlu2 receptor PAM BINA decreased only cocaine-maintained
responding while having no effect on food-maintained responding.[37] Consistent with our previous results, Morishima
et al. (2005) have demonstrated that mGlu2 receptor knockout
mice exhibited increased conditioned place preference for cocaine.[54] While these prior studies focused only on the
role of mGlu2 receptors in responding for drug and natural
rewards, our present findings begin to delineate the individual roles
of mGlu2 and mGlu3 receptors in reward processing.
These patterns of results possibly suggest that activation of mGlu2 receptors selectively modulates drug-reinforced behavior,
whereas activation of mGlu3 receptors either selectively
modulates responding for natural rewards or nonselectively modulates
responding for both drug and natural rewards. Use of an mGlu2/3 receptor PAM, as reported here, provided an initial tool by which
to indirectly test these hypotheses. However, future development and
use of selective mGlu3 receptor PAMs are necessary to test
these hypotheses directly. Moreover, increasing mGlu2/3 receptor activity using a PAM compared to a receptor agonist may
affect behaviors reinforced by natural rewards to a lesser extent
relative to drug-reinforced behaviors. Thus, targeting mGlu2 receptors with a PAM may be an effective strategy for treating drug
dependence without affecting other motivated behaviors.
Figure 8
The mGlu2/3 PAM, 74, decreased cocaine-maintained responding,
and to a lesser extent food-maintained responding, in rats. At the
highest dose tested (40 mg/kg), cocaine-maintained responding was
significantly lower than food-maintained responding. The 40 mg/kg
dose decreased responding for food compared to vehicle only, whereas
the same dose decreased responding for cocaine compared to all other
doses tested. (∗) Responding was significantly different from
0 mg/kg. (#) Responding was significantly different from cocaine.
(@) Responding was significantly different from 0, 10, and 20 mg/kg.
Data are expressed as the mean ± SEM of responding at baseline.
The mGlu2/3 PAM, 74, decreased cocaine-maintained responding,
and to a lesser extent food-maintained responding, in rats. At the
highest dose tested (40 mg/kg), cocaine-maintained responding was
significantly lower than food-maintained responding. The 40 mg/kg
dose decreased responding for food compared to vehicle only, whereas
the same dose decreased responding for cocaine compared to all other
doses tested. (∗) Responding was significantly different from
0 mg/kg. (#) Responding was significantly different from cocaine.
(@) Responding was significantly different from 0, 10, and 20 mg/kg.
Data are expressed as the mean ± SEM of responding at baseline.In conclusion, a ligand–based rational design approach utilizing
certain previously reported selective mGlu2 receptor PAMs
as starting points led to a series of new mGlu2/3 PAMs.
A library of more than 60 analogues was synthesized and tested, providing
compounds with excellent potency and efficacy at mGlu2 and
mGlu3 receptors in vitro. The most promising mGlu2/3 PAMs were profiled in in vitro ADME assays, and on the basis of
these data, several compounds were selected for PK studies in rats.
The lead structures were found to be essentially devoid of activity
at other mGlu receptor subtypes and 45 additional CNS receptors. Representative
PAM 44 was characterized extensively in vitro and was
found to demonstrate ago-PAM activity at mGlu2 and PAM
activity at mGlu3 receptors. Finally, the systemically
active mGlu2/3 PAM 74 dose-dependently decreased
cocaine self-administration in rats following a single intraperitoneal
dose. Future studies will focus on additional characterization of
the new mGlu2/3 PAMs and further optimization of compounds
with enhanced PAM activity at mGlu3 receptors.
Experimental Section
General Chemistry
All reactions
were performed in oven-dried glassware under an atmosphere of argon
with magnetic stirring. All solvents and chemicals used were purchased
from Sigma-Aldrich or Acros, and were used as received without further
purification. Purity of compounds was established by liquid chromatography–mass
spectroscopy (HPLC–MS) and was >95% for all tested compounds.
Silica gel column chromatography was carried out using prepacked silica
cartridges from RediSep (ISCO Ltd.) and eluted using an Isco Companion
system. 1H and 13C NMR spectra were obtained
on a Jeol 400 spectrometer at 400 and 100 MHz, respectively. Chemical
shifts are reported in δ (ppm) relative to residual solvent
peaks or TMS as internal standards. Coupling constants are reported
in Hz. Melting points were obtained using a capillary melting point
apparatus (MEL-TEMP) and are uncorrected. High-resolution ESI-TOF
mass spectra were acquired from the Mass Spectrometry Core at The
Sanford-Burnham Medical Research Institute (Orlando, FL). HPLC–MS
analyses were performed on a Shimadzu 2010EV LCMS instrument using
the following conditions: Kromisil C18 column (reverse phase, 4.6
mm × 50 mm); a linear gradient from 10% acetonitrile and 90%
water to 95% acetonitrile and 5% water over 4.5 min; flow rate of
1 mL/min; UV photodiode array detection from 200 to 300 nm.
General
Methods for the Synthesis of mGlu2/3 Receptor PAMs
General
Method A
To a stirred solution of methyl 4-hydroxybenzoate
(1 mmol, 1 equiv) and 1,4-dibromobutane (3 mmol, 3 equiv) in ACN was
added potassium carbonate (2 mmol, 2 equiv). The reaction mixture
was heated at reflux for 6 h, at which time it was cooled to room
temperature. The crude reaction mixture was diluted with CH2Cl2 and washed twice with 5% aqueous HCl (200 mL). The
organic layers were collected and washed twice with saturated NaHCO3 solution (200 mL). The organic layers were collected, dried
over Na2SO4, and evaporated to dryness. To a
stirred solution of AlCl3 (0.039 mol, 1 equiv) in CH2Cl2 at 0 °C under nitrogen, the acyl chloride
(0.039 mol, 1 equiv) was dissolved in CH2Cl2 and added dropwise to the stirred solution. The phenol (0.039 mol,
1 equiv) was added to the reaction mixture, and the mixture was allowed
to warm to room temperature over 12 h. The reaction was quenched with
HCl (5% aqueous), and CH2Cl2 was added (50 mL).
The organic layer was separated and washed with saturated NaHCO3 solution, then brine and dried over Na2SO4. The solvents were removed by rotary evaporation, and the
products were isolated by flash chromatography [SiO2, hexanes/EtOAc
(4:1)] and concentrated in vacuo. To a crimp top microwave vial were
added the phenol (1 mmol, 1 equiv), bromobutoxy benzoate (1 mmol,
1 equiv), potassium carbonate (2 mmol, 2 equiv), and potassium iodide
(0.1 mmol, 0.1 equiv), all dissolved in ACN (0.2 M). The reaction
mixture was heated in the microwave at 160 °C for 15 min. Following
filtration and evaporation of solvents, the products were isolated
by flash chromatography or reverse phase HPLC and lyophilized to provide
the final compounds which were determined to be >95% pure by HPLC–UV,
HPLC–MS, and 1H NMR.
General Method B
To a stirred solution of the product from “General Method A” (1 mmol, 1 equiv) in dioxane at room
temperature was added KOH (6 mmol, 6 equiv) in water (0.5 mL). The
mixture was stirred continuously for an additional 12 h. The reaction
was quenched with HCl (5% aqueous), and CH2Cl2 (50 mL) was added. The organic layer was separated and dried over
Na2SO4. The solvents were removed by rotary
evaporation, and the products were isolated by flash chromatography
[SiO2, hexanes/EtOAc (1:1)] or reverse phase HPLC and lyophilized
to provide the final compounds which were determined to be >95%
pure by HPLC–UV, HPLC–MS, and 1H NMR. Compounds 4, 5, and 6 were synthesized according
to the published procedures.[50,51] The following compounds
were prepared using the general procedures A and B from the appropriate
starting materials.
Humanembryonic kidney (HEK-293) cell lines coexpressing
rat mGlu receptors 2, 3, 4, 6, 7, or 8 and G protein-coupled inwardly
rectifying potassium (GIRK) channels[55] were
grown in growth medium containing 45% DMEM, 45% F-12, 10% FBS, 20
mM HEPES, 2 mM l-glutamine, antibiotic/antimycotic, nonessential
amino acids, 700 μg/mL G418, and 0.6 μg/mL puromycin at
37 °C in the presence of 5% CO2. Cells expressing
rat mGlu1 and mGlu5 receptor were cultured as
described in Hemstapat et al.[56] All cell
culture reagents were purchased from Invitrogen Corp. (Carlsbad, CA)
unless otherwise noted. Calcium assays were used to assess activity
of compounds at mGlu1 and mGlu5, as previously
described by Engers et al.[57] Calcium assays
at mGlu3 were performed as described for mGlu5 with the exception that TREx293 mGlu3 Gα15 cells were treated with tetracycline at 20 ng/mL for 20 h prior
to assay.Compound activity at group II (mGlu2 and
mGlu3) and group III (mGlu4, mGlu6, mGlu7, and mGlu8) was assessed using thallium
flux through GIRK channels, a method that has been described in detail.[55] Briefly, cells were plated into 384-well, black-walled,
clear-bottomed poly-d-lysine-coated plates at a density of
(15 000 cells/20 μL)/well in DMEM containing 10% dialyzed
FBS, 20 mM HEPES, and 100 units/mL penicillin/streptomycin (assay
medium). Plated cells were incubated overnight at 37 °C in the
presence of 5% CO2. The following day, the medium was exchanged
from the cells to assay buffer [Hanks’ balanced salt solution
(Invitrogen) containing 20 mM HEPES, pH 7.3] using an ELX405 microplate
washer (BioTek), leaving 20 μL/well, followed by the addition
of 20 μL/well FluoZin2-AM (330 nM final concentration) indicator
dye (Invitrogen, prepared as a stock in DMSO and mixed in a 1:1 ratio
with Pluronic acid F-127) in assay buffer. Cells were incubated for
1 h at room temperature, and the dye was exchanged to assay buffer
using an ELX405, leaving 20 μL/well. Test compounds were diluted
to 2 times their final desired concentration in assay buffer (0.3%
DMSO final concentration). Agonists were diluted in thallium buffer
[125 mM sodium bicarbonate (added fresh on the morning of the experiment),
1 mM magnesium sulfate, 1.8 mM calcium sulfate, 5 mM glucose, 12 mM
thallium sulfate, and 10 mM HEPES, pH 7.3] at 5 times the final concentration
to be assayed. Cell plates and compound plates were loaded onto a
kinetic imaging plate reader (FDSS 6000 or 7000; Hamamatsu Corporation,
Bridgewater, NJ). Appropriate baseline readings were taken (10 images
at 1 Hz; excitation, 470 ± 20 nm; emission, 540 ± 30 nm),
and test compounds were added in a 20 μL volume and incubated
for approximately 2.5 min before the addition of 10 μL of thallium
buffer with or without agonist. After the addition of agonist, data
were collected for approximately an additional 2.5 min. Data were
analyzed using Excel (Microsoft Corp, Redmond, WA). The slope of the
fluorescence increase beginning 5 s after thallium/agonist addition
and ending 15 s after thallium/agonist addition was calculated, corrected
to vehicle and maximal agonist control slope values, and plotted using
either XLfit (ID Business Solutions Ltd.) or Prism software (GraphPad
Software, San Diego, CA) to generate concentration–response
curves. Potencies were calculated from fits using a four-point parameter
logistic equation. For concentration–response curve experiments,
compounds were serially diluted 1:3 into 10 point concentration–response
curves and were transferred to daughter plates using an Echo acoustic
plate reformatter (Labcyte, Sunnyvale, CA). Test compounds were applied
and followed by EC20 concentrations of glutamate. For selectivity
experiments, full concentration–response curves of glutamate
or L-AP4 (for mGlu7) were obtained in the presence of a
10 μM concentration of compound, and compounds that affected
the concentration–response by less than 2-fold in terms of
potency or efficacy were designated as inactive.
TREx293 mGlu3 Gα15 Cell Line Creation
In order
to generate a tetracycline (Tet) inducible ratmGlu3 stable
cell line to be used for a calcium mobilizaion assay, TREx293 cells
(Invitrogen) were transfected with mouse Gα15-pCMV6
plasmid (Origene) using Fugene6 (Promega). The cells were selected
for Gα15 expression with 1 mg/mL G418 in the presence
of 10 μg/mL blasticidins to maintain Tet repressor expression.
Two weeks after the selection, polyclonal TREx293 Gα15 cells were obtained. The entire coding sequence of ratmGlu3 was amplified by polymerase chain reaction (PCR) and cloned
into the Tet-inducible expression plasmid pcDNA5/TO (Invitrogen).
RatmGlu3-pcDNA5/TO was transfected into TREx293 Gα15 cells and selected for mGlu3 expression
with 200 μg/mL hygromycin in the presence of G418 and blasticidins.
The resulting polyclonal TREx293 mGlu3 Gα15 cells were plated for monoclonal selection, and positive monoclones
were identified in the calcium mobilization assay. Cells were maintained
in growth medium containing DMEM, 10% Tet-tested FBS (Atlanta Biogicals),
20 mM HEPES, 2 mM l-glutamine, antibiotic/antimycotic, nonessential
amino acids, 500 μg/mL G418, 100 μg/mL hygromycin, and
5 μg/mL blasticidin S at 37 °C in the presence of 5% CO2.
HEK293A mGlu2 Gα15 Cell Line Creation
In order to generate a rat mGlu2 stable cell line to
be used for a calcium mobilization assay, HEK293A cells (ATCC) were
transfected with mouse Gα15-pCMV6 plasmid (Origene)
using Fugene6 (Promega). The cells were selected for Gα15 expression with 1 mg/mL G418. Two weeks after the selection, polyclonal
HEK293A Gα15 cells were obtained. The entire coding
sequence of rat mGlu2 was amplified by PCR and cloned into
the expression plasmid pIRESpuro3 (Invitrogen). Rat mGlu2-pIRESpuro3 was transfected into HEK293A Gα15 cells
and selected for mGlu2 expression with 0.6 μg/mL
puromycin in the presence of G418. The resulting polyclonal HEK293A
mGlu2 Gα15 cells were then utilized for
calcium mobilization assays. Cells were maintained in growth medium
containing DMEM, 10% FBS, 20 mM HEPES, 2 mM l-glutamine,
antibiotic/antimycotic, nonessential amino acids, 700 μg/mL
G418, and 0.6 μg/mL puromycin at 37 °C in the presence
of 5% CO2.
Western Blotting
Western blotting
was performed as detailed previously[58] utilizing
10% polyacrylamide gels and a rabbit polyclonal mGlu2/3 antibody (Millipore, catalog no. 06-676) for detection of mGlu3.
Microsomal Stability in Vitro Assay
Pooled rat liver
microsomes (BD Biosciences, no. 452701) were preincubated with test
compounds at 37.5 °C for 5 min in the absence of NADPH. The reaction
was initiated by addition of NADPH, and the mixture was incubated
under the same conditions. The final incubation concentrations were
4 μM test compound, 2 mM NADPH, and 1 mg/mL (total protein)
liver microsomes in phosphate buffered saline (PBS) at pH 7.4. One
aliquot (100 μL) of the incubation mixture was withdrawn at
15 min time points and combined immediately with 100 μL of ACN/MeOH.
After mixing, the sample was centrifuged at approximately 13 000
rpm for 12 min. The supernatant was filtered and transferred into
an autosampler vial, and the amount of test compound was quantified
using a Shimadzu LCMS 2010EV mass spectrometer. The change of the
AUC (area under the curve) of the parent compound as a function of
time was used as a measure of microsomal stability. Test compounds
were run in duplicate with a positive control.
Plasma Stability in Vitro
Assay
A 20 μL aliquot of a 10 mM solution in DMSO of
the test compound was added to 2.0 mL of heparinized rat plasma (Lampire,
P1-150N) to obtain a 100 μM final solution. The mixture was
incubated for 1 h at 37.5 °C. Aliquots of 100 μL were taken
at 15 min intervals and diluted with 100 μL of MeOH/ACN. After
mixing, the sample was centrifuged at approximately 13 000
rpm for 12 min. The supernatant was filtered and transferred into
an autosampler vial, and the amount of test compound was quantified
using the Shimadzu LCMS-2010EV system. The change of the AUC of the
parent compound as a function of time was used as a measure of plasma
stability.
A 96-well microtiter plate (Millipore, MSSACCEPTOR)
was filled with 300 μL of aqueous buffer solution (in general,
phosphate pH 7.2 buffer was used) and covered with a microtiter filterplate
(Millipore, MAIPNTR10) to create a sort of sandwich construction.
The hydrophobic filter material was impregnated with a 10% solution
of polar brain lipid extract in chloroform (Avanti) as the artificial
membrane, and the organic solvent was allowed to completely evaporate.
Permeation studies were started by the transfer of 150 μL of
a 100 μM test compound solution on top of the filter plate.
The maximum DMSO content of the stock solutions was <1.5%. In parallel,
an equilibrium solution lacking a membrane was prepared using the
exact concentrations and specifications but lacking the membrane.
The concentrations of the acceptor and equilibrium solutions were
determined using the Shimadzu LCMS-2010EV and AUC methods. The acceptor
plate and equilibrium plate concentrations were used to calculate
the permeability rate (log Pe) of
the compounds. The log Pe values
were calculated using the following equations:In this equation, VD (cm3)
is the donor volume (0.150 cm3), VA (cm3) is the acceptor volume (0.300 cm3), area (cm2) is the accessible filter area (0.168 cm2), and time (s) is the incubation time. Variables [drug]acceptor and [drug]equilibruim are concentrations
of the test drug for the sample (acceptor) and reference (equilibrium)
solutions in the acceptor compartment.
Behavioral Assessments
Subjects
Male Wistar rats (Charles River Laboratories, Raleigh, NC) weighing
300–350 g at the beginning of each experiment were housed in
pairs in standard rat Plexiglas cages with food and water available
ad libitum except during food training and the food self-administration
experiment (see below). Rats were maintained in a climate-controlled
room at 21 °C on a 12 h reverse light/dark cycle, and all experiments
were conducted during the dark (i.e., active) phase (7:00 to 19:00
h) of the cycle under dim red lighting. All procedures were conducted
in accordance with the guidelines from the National Institutes of
Health and the Association for the Assessment and Accreditation of
Laboratory Animal Care and were approved by the Institutional Animal
Care and Use Committee.
Drugs
Cocaine hydrochloride (National
Institute on Drug Abuse, Bethesda, MD) was dissolved in sterile physiological
saline and filtered through a 0.22 μm syringe filter (Fisher
Scientific, Pittsburgh, PA) for sterilization purposes. 74 was mixed into a 10% EtOH, 1% Tween 80 solution.
Food Training
Details regarding the experimental procedures have been described
previously.[37] All rats were placed under
food restriction (20 g of food/day) and trained during daily 1 h sessions
to lever-press for 45 mg food pellets (Research Diets, New Brunswick,
NJ) under a fixed ratio 1 reinforcement schedule with a 1 s time-out
period (FR1 TO1s). Successful responses were followed by illumination
of a cue light for the duration of the time-out period, when lever
presses had no consequence. Successful acquisition of food responding,
defined as earning 100 pellets during each session, resulted in progression
of the training program to FR1 TO10s and FR1 TO20s. Training lasted
approximately 5 days.
Cocaine Self-Administration Experiment
After successful acquisition of food training, rats (n = 11) were fed ad libitum, surgically prepared with intravenous
catheters inserted into the right jugular vein under isoflurane anesthesia
(1–1.5% isoflurane/oxygen mixture), and allowed 7 days to recover
(see Jin et al. (2010) for details). Rats were then trained to self-administer
cocaine under a FR1 TO20s reinforcement schedule during daily 1 h
sessions. Each response at the active lever resulted in an intravenous
infusion of cocaine (0.5 mg/kg/infusion) over a 2 s period in a volume
of 0.05 μL. Rats were trained for approximately 10 days until
responding for cocaine stabilized (i.e., >10 infusions/session;
<20% variability in number of infusions over three consecutive
sessions). After stabilization of responding, rats were administered 74 (0, 10, 20, 40 mg/kg, ip; 3 mL/kg volume; 60 min pretreatment
time) according to a within-subjects Latin-square design. At least
4 days elapsed between drug/vehicle injections to re-establish stable
self-administration behavior (<20% variability over three consecutive
sessions).
Food Self-Administration Experiment
To assess nonspecific actions of 74, after successful
acquisition of food training and stabilization of responding (<20%
variability over three consecutive sessions), rats (n = 8) were administered 74 (0, 10, 20, 40 mg/kg, ip;
3 mL/kg volume; 60 min pretreatment time) according to a within-subjects
Latin-square design. All test parameters, including the FR1 TO20s
reinforcement schedule, were identical to the parameters under which
cocaine was self-administered.
Statistical Analyses
The number of cocaine infusions/food pellets earned during test
sessions with 74 was calculated as a percentage of the
average number of infusions/pellets earned during the prior three
baseline sessions. Data were then analyzed with a mixed design analysis
of variance (ANOVA) with 74 dose (within-subjects) and
self-administration (i.e., cocaine vs food; between-subjects) as factors.
Significant effects were further analyzed with Tukey post hoc tests.
The level of significance was set at α = 0.05.
Authors: Colleen M Niswender; Kari A Johnson; Qingwei Luo; Jennifer E Ayala; Caroline Kim; P Jeffrey Conn; C David Weaver Journal: Mol Pharmacol Date: 2008-01-02 Impact factor: 4.436
Authors: Kamondanai Hemstapat; Tomas de Paulis; Yelin Chen; Ashley E Brady; Vandana K Grover; David Alagille; Gilles D Tamagnan; P Jeffrey Conn Journal: Mol Pharmacol Date: 2006-04-27 Impact factor: 4.436
Authors: Steven P Govek; Celine Bonnefous; John H Hutchinson; Theodore Kamenecka; Jeffrey McQuiston; Richard Pracitto; Lucy X Zhao; Michael F Gardner; Joyce K James; Lorrie P Daggett; Blake A Rowe; Hervé Schaffhauser; Linda J Bristow; Una C Campbell; Dana E Rodriguez; Jean-Michel Vernier Journal: Bioorg Med Chem Lett Date: 2005-09-15 Impact factor: 2.823
Authors: Carmen Dominguez; Lourdes Prieto; Matthew J Valli; Steven M Massey; Mark Bures; Rebecca A Wright; Bryan G Johnson; Sherri L Andis; Ann Kingston; Darryle D Schoepp; James A Monn Journal: J Med Chem Date: 2005-05-19 Impact factor: 7.446
Authors: Ruggero Galici; Carrie K Jones; Kamondanai Hemstapat; Yi Nong; Nicholas G Echemendia; Lilly C Williams; Tomas de Paulis; P Jeffrey Conn Journal: J Pharmacol Exp Ther Date: 2006-04-11 Impact factor: 4.030
Authors: Craig W Lindsley; Kyle A Emmitte; Corey R Hopkins; Thomas M Bridges; Karen J Gregory; Colleen M Niswender; P Jeffrey Conn Journal: Chem Rev Date: 2016-02-16 Impact factor: 60.622
Authors: Anushree N Karkhanis; Thomas J R Beveridge; Bruce E Blough; Sara R Jones; Mark J Ferris Journal: Drug Alcohol Depend Date: 2016-06-25 Impact factor: 4.492
Authors: Joseph E Pero; Michael A Rossi; Michael J Kelly; Hannah D G F Lehman; Mark E Layton; Robert M Garbaccio; Julie A O'Brien; Brian C Magliaro; Jason M Uslaner; Sarah L Huszar; Kerry L Fillgrove; Cuyue Tang; Yuhsin Kuo; Leo A Joyce; Edward C Sherer; Marlene A Jacobson Journal: ACS Med Chem Lett Date: 2016-01-10 Impact factor: 4.345
Authors: Xiaofei Zhang; Katsushi Kumata; Tomoteru Yamasaki; Ran Cheng; Akiko Hatori; Longle Ma; Yiding Zhang; Lin Xie; Lu Wang; Hye Jin Kang; Douglas J Sheffler; Nicholas D P Cosford; Ming-Rong Zhang; Steven H Liang Journal: ACS Chem Neurosci Date: 2017-06-13 Impact factor: 4.418
Figure 1
Figure 2
Figure 3
Scheme 1
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8