Comfort A Boateng1, Oluyomi M Bakare1, Jia Zhan1, Ashwini K Banala1, Caitlin Burzynski1, Elie Pommier1, Thomas M Keck1, Prashant Donthamsetti2, Jonathan A Javitch2, Rana Rais3, Barbara S Slusher3, Zheng-Xiong Xi1, Amy Hauck Newman1. 1. †Molecular Targets and Medications Discovery Branch, National Institute on Drug Abuse- Intramural Research Program, National Institutes of Health, 333 Cassell Drive, Baltimore, Maryland 21224, United States. 2. ∥Departments of Psychiatry and Pharmacology, Columbia University College of Physicians and Surgeons, and Division of Molecular Therapeutics, New York State Psychiatric Institute, New York, New York 10032, United States. 3. §Department of Neurology, Brain Science Institute, The Johns Hopkins University School of Medicine, 855 North Wolfe Street, Baltimore, Maryland 21205, United States.
Abstract
The dopamine D3 receptor (D3R) is a promising target for the development of pharmacotherapeutics to treat substance use disorders. Several D3R-selective antagonists are effective in animal models of drug abuse, especially in models of relapse. Nevertheless, poor bioavailability, metabolic instability, and/or predicted toxicity have impeded success in translating these drug candidates to clinical use. Herein, we report a series of D3R-selective 4-phenylpiperazines with improved metabolic stability. A subset of these compounds was evaluated for D3R functional efficacy and off-target binding at selected 5-HT receptor subtypes, where significant overlap in SAR with D3R has been observed. Several high affinity D3R antagonists, including compounds 16 (Ki = 0.12 nM) and 32 (Ki = 0.35 nM), showed improved metabolic stability compared to the parent compound, PG648 (6). Notably, 16 and the classic D3R antagonist SB277011A (2) were effective in reducing self-administration of heroin in wild-type but not D3R knockout mice.
The dopamine D3 receptor (D3R) is a promising target for the development of pharmacotherapeutics to treat substance use disorders. Several D3R-selective antagonists are effective in animal models of drug abuse, especially in models of relapse. Nevertheless, poor bioavailability, metabolic instability, and/or predicted toxicity have impeded success in translating these drug candidates to clinical use. Herein, we report a series of D3R-selective 4-phenylpiperazines with improved metabolic stability. A subset of these compounds was evaluated for D3R functional efficacy and off-target binding at selected 5-HT receptor subtypes, where significant overlap in SAR with D3R has been observed. Several high affinity D3R antagonists, including compounds 16 (Ki = 0.12 nM) and 32 (Ki = 0.35 nM), showed improved metabolic stability compared to the parent compound, PG648 (6). Notably, 16 and the classic D3R antagonist SB277011A (2) were effective in reducing self-administration of heroin in wild-type but not D3R knockout mice.
The dopamine D2-like family of receptors, comprising
the D2, D3, and D4 receptor subtypes
(D2R, D3R, and D4R, respectively),
regulate physiological functions such as movement, emotion, and cognition.
Numerous medications for the treatment of neuropsychiatric disorders,
such as schizophrenia, target this important family of receptors.[1] The D3R subtype is localized in key
neurocircuits that underlie motivation and cognition, and in contrast
to D2R, do not appear to play a major role in movement.[2,3] Hence, the D3R has been proposed as a promising target
for development of psychostimulant addiction and relapse pharmacotherapy.[2,4−9] Additionally, seminal reports using [11C]raclopride for
PET imaging in both humans[10] and nonhuman
primates[11] have demonstrated that overall
D2-like receptor availability in the basal ganglia is significantly
reduced after chronic exposure to cocaine. Interestingly, the converse
appears to be true for the D3R subtype: recent humanPET
studies with the D3R-preferential PET ligand [11C]PHNO[12,13] confirmed earlier reports in post-mortem
brains of cocaineoverdose victims[14−16] that D3R
availability is actually increased upon chronic exposure to cocaine
and methamphetamine.[17−19] Inspired by early preclinical studies using lead
compounds BP897 (1)[20,21] and SB277011A
(2),[22−24] intensive efforts toward discovering D3R-selective antagonists and partial agonists as potential therapeutics
for substance use disorders have been underway for nearly two decades.Although early D3R-selective partial agonists, such
as 1, and antagonists such as 2, NGB2904
(3), and PG01037 (4), (Chart ) have served as highly useful
preclinical tools, none of these compounds have progressed to the
clinic due to poor bioavailability and the advancement of newer generation
analogues.[4−7,25−27] One example
of a D3R-selective compound tested in humans is GSK598,809
(5) (Chart ), which was first reported in 2007.[25] Recently, [11C]PHNOPET imaging studies demonstrated
a correlation between this compound’s D3R occupancy
and efficacy toward smoking cessation.[28,29] Although clinical
studies have been conducted with 5 for smoking cessation,[28−31] it is unlikely that this compound will be developed as a medication
to treat other substance use disorders, as GlaxoSmithKline terminated
this research and development program.
Chart 1
Selected Lead D3R-Selective Antagonists and Partial Agonists
One of our lead compounds, R-PG648 (-6) binds with
∼500-fold selectivity
for D3R (Ki = 0.53 nm) over
D2R (Ki = 295 nm) in vitro[26,27,32,33] and is an antagonist/weak partial agonist.[34] Using molecular simulations based on the D3R X-ray crystal
structure,[35] we demonstrated that the 4-phenylpiperazine
moiety binds to the structurally similar D2R/D3R orthosteric binding site (OBS) in which dopamine and the D2R/D3R antagonist eticlopride bind, whereas the
arylamide terminus binds in a secondary binding pocket (SBP) that
is topographically divergent between D2R and D3R.[34−36] Despite early preclinical promise, -6 and its racemate (6) proved to be unsuitable for translation to clinical study.[33]The development of D3R antagonists
with sufficient metabolic
stability and efficacy as medications to treat substance use disorders
remains a major challenge. Hence the purpose of the present study
was to extend D3RSAR by exploring modifications to all
three segments (Chart ) of the 4-phenylpiperazine class of molecules to identify ligands
that demonstrate high D3R binding affinity and subtype
selectivity, with improved bioavailability and metabolic stability.
Specifically, we replaced the 2,3-diCl-substitution of the 4-phenylpiperazine
in lead compound 6 with a 2-OMe,3-Cl or a 2,3-naphthyl-substituent.
Additionally, we explored bioisosteric replacement of the indolylamide
of 6 with various heteroarylamides as well as further
investigating the 3-hydroxy (3-OH) substituent in the 4-carbon linker.
After assessing in vitro binding and functional efficacy profiles,
and metabolic stability in mouse microsomes, two lead candidates were
selected for behavioral evaluation in murine models of heroin self-administration
and compared to the classic D3R antagonist, 2.Although D3R antagonists have largely been developed
toward nicotine, cocaine, and methamphetamine abuse, they have shown
efficacy in other preclinical models of drug and alcohol abuse.[37−39] For example, the D3R antagonist 2 has been
shown to block the acquisition and expression of heroin conditioned
place preference in rats.[40] Nevertheless,
to our knowledge, D3R antagonists have not been evaluated
for their effects on heroin self-administration. As D3R
antagonists are typically not effective in attenuating cocaine or
methamphetamine self-administration under low fixed-ratio (FR) schedules
(e.g., FR1, FR2),[41,42] we hypothesized that in the presence
of these psychostimulants, high extracellular dopamine (DA) levels
may compete at D3R and thus render these D3R-selective
antagonists ineffective. Indeed, most reports show D3R
antagonists as being most effective in models of relapse when extracellular
DA levels would be substantially lower than during active exposure
to psychostimulants.[2,33] In comparison to psychostimulant
drugs, heroin induces relatively moderate increases in extracellular
DA within the accumbens during active drug-taking,[43−46] thus we posited that D3R antagonists may be more effective in reducing heroin self-administration
than psychostimulant use. If the data support our hypothesis, this
would represent an alternative strategy to treating opioid abuse and
addiction. Herein, we evaluate our lead compounds in both wild-type
(WT) and D3R knockout (KO) mice to not only assess effectiveness
in attenuating heroin self-administration but also to determine the
role of D3R in these behaviors.
Chemistry
The
synthesis of 24 final compounds and their common intermediates
are outlined in Schemes and 2. In Scheme , 7a(47) was synthesized using 3-chloro-2-methoxyaniline and bis(2-chloroethyl)amine
HCl, in diglyme under reflux conditions. 1-(Naphthalene-1-yl)piperazine, 7b, was synthesized as previously reported[48] via a nucleophilic substitution reaction with naphthalen-1-amine.
The synthesis of the 4-(4-arylpiperazine)butylamine intermediates 10a–10d and 13a–13d, with or without a 3-OH substituent, respectively, was
achieved starting with 7a, 7b, or commercially
available 1-(2-methoxyphenyl)piperazine and 1-(2,3-dichlorophenyl)piperazine.[49] Specifically, an epoxide ring-opening reaction
of 8 with the corresponding 4-arylpiperazines under reflux
conditions yielded the 3-OH-phthalimides 9a–9d, which were subjected to phthalimide deprotection using
hydrazine to give the 3-OH-substituted butylamine intermediates 10a–10d. The synthesis of intermediate
compounds 9c, 9d, 10c, and 10d was previously reported[49] using
microwave conditions; however, the methods reported herein use conventional
(oil bath) heating methods. To generate the 4-(4-arylpiperazine)butylamine
intermediates 13a–13d, commercially
available N-(4-bromobutyl)phthalimide 11 was reacted with the appropriate 4-arylpiperazine to give phthalimides 12a–12d, which were then subjected to
deprotection with hydrazine to give the desired 4-(4-arylpiperazine)butylamines13a–13d. While intermediates 12b–12d and 13a–13d have been previously reported,[34,50] our synthetic
methods either differ completely from or improve upon these methods.
The final ligands 14–37 (Scheme ) featuring the arylpiperazine,
four-carbon linker chain (with or without a 3-OH substituent), and
arylamide moieties were synthesized via a general amidation procedure
(method A),[50,51] in which various arylcarboxylic
acids were coupled to corresponding primary amines 10 or 13 via an in situ N,N′-carbonyldiimidazole coupling reaction. All 3-OH-substituted
compounds were racemic mixtures. As considerable enantioselectivity
has so far not been demonstrated with either the 3-OH or 3-F substitution
in this class of compounds,[26,27] no attempt to separate
enantiomers was made.
Scheme 1
Synthesis of 4-(4-Arylpiperazine)butylamine
Intermediates
Reagents and conditions: (a)
3-chloro-2-methoxyaniline or naphthalen-1-amine, K2CO3, bis(2-chloroethyl)amine HCl, diglyme, reflux, 48 h; (b)
appropriate 4-arylpiperazine, 2-propanol, reflux, overnight; (c) hydrazine
(anhydrous), EtOH, reflux, 2–3 h.
Scheme 2
Synthesis
of Target Compounds 14–37
Reagents and conditions: (a)
(i) 1,1′-carbonyldiimidazole (CDI), THF, room temperature,
2 h; (ii) appropriate 4-arylpiperazine amine, THF, 0 °C to room
temperature, overnight.
Synthesis of 4-(4-Arylpiperazine)butylamine
Intermediates
Reagents and conditions: (a)
3-chloro-2-methoxyaniline or naphthalen-1-amine, K2CO3, bis(2-chloroethyl)amine HCl, diglyme, reflux, 48 h; (b)
appropriate 4-arylpiperazine, 2-propanol, reflux, overnight; (c) hydrazine
(anhydrous), EtOH, reflux, 2–3 h.
Synthesis
of Target Compounds 14–37
Reagents and conditions: (a)
(i) 1,1′-carbonyldiimidazole (CDI), THF, room temperature,
2 h; (ii) appropriate 4-arylpiperazine amine, THF, 0 °C to room
temperature, overnight.
Pharmacological Results
and Discussion
SAR at D2R and D3R
The objective
of this study was to design metabolically stable ligands with high
D3R binding affinity and subtype selectivity. We extended
our previous SAR studies by exploring modifications to all three segments
of the lead compound 6 (Chart : arylamide moiety, 4-carbon linker, and
4-phenylpiperazine moiety) in an attempt to overcome bioavailabilty
and metabolic shortcomings of previously reported preclinical candidates.[33]We recently evaluated the binding affinities
and selectivities at D3R and D2R of a library
of synthons along with their full-length ligands.[34] Computational modeling studies demonstrated that the 3-Cl
of the 2,3-diCl-phenylpiperazine moiety makes polar contacts with
transmembrane domain 5 (TM5) serine (Ser) amino acid residues (particularly
Ser1935.43), which correlates with observed partial agonist
actions in vitro.[34] However, when the 2,3-diCl
substituent was replaced with a 2-OMe, there were no polar contacts
with TM5, which corresponds to the much weaker partial agonist effects
observed with the 2-OMe-phenylpiperazine-substituted ligands.[34] In view of these data, we explored the effects
of incorporating a hybrid 2-OMe,3-Cl-phenylpiperazine substituent
as the primary pharmacophore. This 2-OMe,3-Cl- substitution pattern
is present in the nonselective high affinity D2R/D3R receptor antagonist/inverse agonist, eticlopride. Thus,
we hypothesized that this substitution would lead to an increase in
affinity at D3R compared to the 2,3-diCl-phenylpiperazine.We also replaced the substituted phenyl ring on the piperazine
with a naphthalene moiety, a bioisosteric replacement for the 2,3-diCl-phenyl
ring, and predicted that this substituent might also result in compounds
with high-affinity for D3R. The naphthylpiperazine moiety
is a well-described pharmacophore for serotonin receptors and specifically
for 5-HT1A agonists, potentially reducing D3R selectivity of these analogues.[48,52] Notably, buspirone,
a 5-HT1A partial agonist that also antagonizes D3R,[53] was recently reported to reduce cocaine
and nicotine self-administration (alone or in combination) in nonhuman
primates,[53−55] Although a recent clinical trial testing buspirone
for cocaine dependence treatment proved unsuccessful,[56] developing novel D3R antagonists/5-HT1A partial agonists might also be an interesting combination for compounds
with therapeutic potential.To evaluate the binding affinities
of this series of compounds,
we performed competition binding studies with [3H]N-methylspiperone using membranes prepared from HEK293 cells
expressing either the humanD2R or D3R (Tables –3.) In addition, cLogP values are provided as a relative
measure of lipophilicity. Compounds 14–21 displayed moderate to high binding affinities for D2R
(Ki = 8–330 nM; Table ). Moreover, all of these analogues, with the exception of 21 (Ki = 1.8 nM), displayed subnanomolar
affinities at D3R (Ki = 0.1–0.9
nM), demonstrating that the 2-OMe,3-Cl- and the 2,3-naphthyl-piperazines,
as predicted, were well tolerated as the primary pharmacophores in
this series. For example, the 2-OMe,3-Cl-phenylpiperazine (e.g., 14; Ki = 0.39 nM) and the 2,3-naphthyl-piperazine
(e.g., 16; Ki = 0.12 nM)
indolylamide analogues generally showed higher D3R binding
affinities compared to lead compound 6 (Ki = 1.9 nM), although neither was as D3R selective.
We also explored the bioisosteric replacement of the indole moiety
(as the secondary pharmacophore) with benzofuran and benzothiophene
moieties, substitutions that have previously demonstrated high D3R affinities, in related series.[7,34,51,57,58]
Table 1
Human Dopamine D2-Like
Receptor Binding Data in HEK Cells for Ligands with Varying Arylpiperazine
and Arylamide Moietiesa
Ki values
determined by competitive inhibition of [3H]N-methylspiperone binding in membranes harvested from HEK 293 cells
stably expressing hD2R or hD3R.
Table 3
Human Dopamine D2-Like
Receptor Binding Data in HEK Cells for Quinoline-3-carboxamide Ligands
with Varying Arylpiperazines and 4-Carbon Linker Chainsa
Ki values
determined by competitive inhibition of [3H]N-methylspiperone binding in membranes harvested from HEK 293 cells
stably expressing hD2R or hD3R.
Ki values
determined by competitive inhibition of [3H]N-methylspiperone binding in membranes harvested from HEK 293 cells
stably expressing hD2R or hD3R.As shown in Table , all these benzothienyl and benzofuranyl
analogues displayed high
D3R affinities, with Ki values
between 0.13 and 0.91 nM and good selectivity for D3R over
D2R (61–80-fold). The presence of an electron-donating
OMe group on the indole ring, in compound 17, was well-tolerated
and resulted in high affinity at D3R and 80-fold selectivity
for D3R over D2R. The presence of an electron-withdrawing
F group on the indole ring in compound 20(50) also resulted in high D3R affinity
(Ki = 0.43 nM), and D3R selectivity
was improved with the addition of a 3-OH in the linking chain (21). A small reduction in selectivity for D3R over
D2R in comparison with 6 was observed with
compound 21 due to a ∼2-fold improvement in D2R affinity.In summary, for the ligands reported in Table , compounds without
the 3-OH substituent
in the 4-carbon linker chain possessed higher cLogP values (i.e.,
were more lipophilic/hydrophobic) and displayed higher affinities
at both D2R and D3R but lower selectivity for
D3R over D2R (e.g., compound 20 vs 21 in Table ). Bioisosteric
replacement of the indole moiety with other [5,6]-heteroaromatic moieties,
such as benzofuran and benzothiophene, did not significantly influence
D3R binding affinity or selectivity.To improve chemical
and metabolic stability, we opted to further
explore other heteroaromatic groups such as quinoline that does not
have an acidic hydrogen or hydrogen-bond donor group like the N–H
group of the indole. To this end, compounds 22(59)–31 were prepared that contained
the 2-OMe-phenylpiperazine moiety along with either electron withdrawing
or donating substituents on the quinoline moiety. We explored whether
the position of the quinolinenitrogen (N), that is 2-quinolinyl (compounds 22–25), 3-quinolinyl (compounds 26–30), and 7-quinolinyl (compound 31) positions, had any effect on binding affinity and/or selectivity.
We also compared quinolinylamide analogues with or without a 3-OH
substituent in the 4-carbon linker between the arylamide terminus
and the 4-phenylpiperazine moiety.Binding affinities for the
2-OMe-phenylpiperazine quinolinylamide
analogues 22–31 are reported in Table . Compared to the indolylamide, benzofuran, and benzothiophene
analogues (Table ),
the quinoline analogues generally displayed lower affinities at D2R and D3R as well as decreased selectivity for
the D3R over D2R. Moreover, quinolinylamide
analogues with a 3-OH substituent in the 4-carbon linker chain displayed
reduced D3R binding affinities and in contrast to previous
reports, reduced D3R-selectivity over D2R as
compared to the unsubstituted analogues.[26,49] Specifically, compounds 22, 24, and 27, without a 3-OH substituent in the linker chain, displayed
low nanomolar affinities at D3R (Ki = 3.5, 2.5, and 2.1 nM, respectively) and showed equal or
higher selectivity for D3R compared to their corresponding
3-OH substituted analogues 23, 25, and 28 (Ki = 97.9, 53.1 and 44.8 nM,
respectively). The position of the quinolinyl N did not appear to
strongly influence binding affinity and/or selectivity at D3R, although there appeared to be a slight preference for 3- and 7-quinolinyl
positions over the 2-quinolinyl position (e.g., compounds 26 and 31 compared to 23). While affinities
at D2R remained similarly low for compounds 23, 26, and 31, there was a 3-fold decrease
in affinity at D3R and selectivity for D3R over
D2R for compound 23 compared to compounds 26 and 31. Hence, D3R affinity and
selectivity for the 2-, 3-, and 7-quinolinyl analogues follow the
order 2- < 3- = 7-quinoline. Of note, within the 3-quinolineseries
(compounds 26–30), the presence of
an electron-withdrawing or donating substituent on the quinoline moiety
did not substantially affect affinity or selectivity at D3R, although it appeared that electron-withdrawing substituents (e.g.,
F) induced less pronounced loss of affinity compared to electron-donating
substituents (e.g., OMe). In general, with the 2-OMe-phenylpiperazine
moiety, the quinolinylamide analogues were less potent and selective
for D3R over D2R. On the basis of relative SAR
within the quinoline–carboxamide analogues, we decided to explore
the 3-quinoline moiety as a bioisosteric replacement for indole in
the arylamide terminus in combination with the 2-OMe,3-Cl-phenylpiperazine,
or 2,3-naphthyl-piperazine. In addition, we prepared the 2,3-diCl
analogues for comparison.
Table 2
Human Dopamine D2-Like
Receptor Binding Data in HEK Cells for N-(4-(4-(2-Methoxyphenyl)piperazin-1-yl)butyl)quinoline
Carboxamide Ligandsa
Ki values
determined by competitive inhibition of [3H]N-methylspiperone binding in membranes harvested from HEK 293 cells
stably expressing hD2R or hD3R.
Ki values
determined by competitive inhibition of [3H]N-methylspiperone binding in membranes harvested from HEK 293 cells
stably expressing hD2R or hD3R.The D2R and D3R binding affinities of ligands
that incorporate the quinoline-3-carboxamide moiety with either 2-OMe,3-Cl-phenylpiperazine,
2,3-naphthyl-piperazine, or 2,3-diCl-phenylpiperazine (compounds 32–37) are shown in Table . Comparison of these compounds and the quinoline-3-carboxamide
analogues in Table revealed improvement in binding affinities at the D3R.
While the 2-OMe,3-Cl-phenylpiperazine analogue 35 (Ki = 31.6 nM; D2R/D3Rratio
= 28) is essentially equipotent at D2R and D3R as compound 26 (Table ; Ki = 33.8 nM; D2R/D3Rratio = 27), the equivalent 2,3-naphthyl-piperazine
analogue, 33, displayed ∼10-fold increase in affinity
(Ki = 3.63 nM) and ∼2-fold more
D3R selective (D2R/D3Rratio = 71)
for D3R over D2R than 26. Compound 32 had the highest D3R affinity (Ki = 0.35 nM) of the 2,3-naphpthyl or 2-OMe,3-Cl-analogues.
Moreover, the 2,3-diCl-phenylpiperazine analogues with a 6-F-quinoline
(D2R/D3Rratio = 69 and 103 for 36 and 37, respectively) were slightly more selective
than the unsubstituted quinolines in this series. These data suggest
that among the quinoline-3-carboxamides, both the 2,3-diCl-phenylpiperazine
and the 2,3-naphthyl-piperazine are good replacements for the 2-OMe-phenylpiperazine
moiety. Addition of the 3-OH group in the linker in the 6-F-quinoline
compound, 37, showed somewhat lower affinity and selectivity
as compared to its 5-F-indole analogue, 21. Among the
quinoline-3-carboxamide analogues, compound 32 with the
2,3-naphthyl-piperazine and 36 with 2,3-diCl-phenylpiperazine
showed the highest D3R affinity and their 3-OH-analogues, 33 and 37, respectively, displayed highest D3R-selectivity.Ki values
determined by competitive inhibition of [3H]N-methylspiperone binding in membranes harvested from HEK 293 cells
stably expressing hD2R or hD3R.
Functional Data and Serotonergic Actions
Six analogues
were selected to compare with the parent compound 6 for
their ability to antagonize quinpirole-induced activation of hD2R or hD3R in a cell-based mitogenesis assay (Table .) All of the analogues were moderately potent antagonists
in the D3R mitogenesis assay (IC50 = 7.4–72
nM), with compounds 16 and 32 being the
most potent. Of note these compounds, as well as 26, 33, and 34, showed weak partial agonist profiles
at higher concentrations. In general, functional potencies of these
analogues were lower than their binding affinities. Most of the analogues
were more potent at D3R than at D2R in this
assay, although selectivities varied.
Table 4
Functional
Data for Selected Compounds
Using Stimulation or Inhibition of Quinpirole-Stimulated Mitogenesis
in CHO Cells with Human Dopamine D2R or D3Ra
D2R mitogenesis assay
D3R mitogenesis assay
compd
agonist EC50 ± SEM (nM)
% stimulation
antagonist IC50 ± SEM (nM)
agonist EC50 ± SEM (nM)
% stimulation
antagonist IC50 ± SEM (nM)
6
170 ± 61b
22b
4400 ± 60
>10000
NDc
30 ± 5.8b
16
76 ± 14
4.5
87 ± 20
140 ± 51
25
8.0 ± 1.9
20
>10000
NDc
564.40 ± 109
>10000d
NDc
52 ± 1.2d
26
NDc
NDc
NDc
0.95 ± 0.39
11
47 ± 11
32
170 ± 52
18
29 ± 8.7
20 ± 4.6
31
7.4 ± 2.5
33
200 ± 28
8.2
720 ± 190
13 ± 1.9
36
72 ± 12
34
>10000
<7
63 ± 13
16 ± 6.4
29
59 ± 24
Data were obtained
through the NIDA
Addiction Treatment Discovery Program contract (ADA151001) with Oregon
Health & Science University.
Previously published data.[26]
ND = not determined. Functional
assays were not conducted if the Ki value
for the binding assay was >500 nM.
Previously published data.[50]
Data were obtained
through the NIDA
Addiction Treatment Discovery Program contract (ADA151001) with Oregon
Health & Science University.Previously published data.[26]ND = not determined. Functional
assays were not conducted if the Ki value
for the binding assay was >500 nM.Previously published data.[50]These analogues were also
tested for binding at 5-HT1A, 5-HT2A, and 5-HT2C receptors as well as for
functional activity at 5-HT1A (Table ). Although all the analogues had affinity toward these 5-HT
receptor subtypes, compounds 32–34 exhibited the highest affinities, especially at 5-HT1A and 5-HT2A. All these analogues were full agonists at
5-HT1A receptors, which was not surprising based on previously
reported SAR for 5-HT1A.[48,52] However, with
the exception of 32, all analogues were still significantly
D3R-selective. Interestingly, 32 showed relatively
high affinity for all the 5-HT receptor subtypes tested, providing
a unique profile among this series.
Table 5
Additional in Vitro
Binding and Functional
Data for Selected Compounds at 5-HT1A, 5-HT2A, and 5-HT2C Receptorsa
5-HT1A [35S]GTPγS binding
compd
5-HT1A [3H]-8-OH-DPATKi ± SEM (nM)
5-HT2A [125I]DOIKi ± SEM (nM)
5-HT2C [125I]DOIKi ± SEM (nM)
agonist EC50 ± SEM (nM)
% stimulation
6
100 ± 24b
48 ± 13
86 ± 0.61
NDc
NDc
16
84 ± 7.1
55 ± 12
73 ± 16
250 ± 67
108
20
540 ± 48
>10000
990 ± 310
NDc
NDc
26
68 ± 9.8
350 ± 50
1,800 ± 130
340 ± 91
94
32
2.5 ± 0.35
0.33 ± 0.12
3.8 ± 1.4
9.0 ± 3.0
108
33
9.0 ± 3.2
2.4 ± 0.77
44 ± 5.0
17 ± 5.7
104
34
21 ± 5.4
1.7 ± 0.60
14 ± 2.6
43 ± 11
106
Data were obtained through the NIDA
Addiction Treatment Discovery Program contract (ADA151001) with Oregon
Health & Science University.
Previously published data.[26]
ND = Not determined. Functional
assays were not conducted if the Ki value
for the binding assay was >250 nM.
Data were obtained through the NIDA
Addiction Treatment Discovery Program contract (ADA151001) with Oregon
Health & Science University.Previously published data.[26]ND = Not determined. Functional
assays were not conducted if the Ki value
for the binding assay was >250 nM.
Microsomal Metabolism Results
Metabolic stability assays
were conducted in mouse liver microsomes to predict in vivo stability
after oral administration. The analogues were tested for phase I metabolism
in mouse liver microsomal incubations in the presence of NADPH and
were compared to parent compound 6 following procedures
previously described.[60] Compound stability
over a 1 h incubation is presented in Figure . In general, the compounds with the unsubstituted
linker, regardless of the primary or secondary pharmacophores, were
more metabolically stable (see Table S3 in Supporting
Information). Of the quinoline analogues, 32 was
the most stable, with calculated in vitro rate constant (k) of −0.017 and in vitro half-life (t1/2) of 41.8 min, whereas 16 with the k value of −0.010 and in vitro t1/2 of 67.5 min was the most stable of the indoles and
significantly more stable than the parent compound, 6, with k of −0.037 and t1/2 of 18.8 min.
Figure 1
(a) Phase I metabolism data for all compounds
with an unsubstituted
linker chain. (b) Phase I metabolism data for all compounds with a
3-OH substituted linker chain. (c) Phase I metabolism data for all
quinoline compounds. (d) Phase I metabolism data for all indole compounds.
(a) Phase I metabolism data for all compounds
with an unsubstituted
linker chain. (b) Phase I metabolism data for all compounds with a
3-OH substituted linker chain. (c) Phase I metabolism data for all
quinoline compounds. (d) Phase I metabolism data for all indole compounds.
Heroin Self-Administration
Studies in WT and D3KO
Mice
On the basis of their pharmacological profile and microsomal
stability, compounds 16 and 32 were chosen
for behavioral studies in WT and D3KO mice. Of note, compound 16 was D3R selective over the 5-HT receptor subtypes
in contrast to compound 32. Moreover, these two analogues
were the most metabolically stable in this series. Mice were trained
on an FR1 schedule of reinforcement, where every lever press results
in an injection, initially with 0.1 mg/kg (iv) heroin, followed by
0.05 mg/kg and finally 0.025 mg/kg, to obtain more robust lever pressing
behavior via a compensatory response to a reduced reinforcer (e.g.,
a lower dose of heroin). Both WT and D3KO mice acquired
heroin self-administration after 1–2 weeks of training. Overall,
D3KO mice display higher total number of heroin infusions
taken over 3 h than WT mice although the difference was not statistically
significant (Figures A, 3A, 4A). After stable
heroin self-administration was achieved, we evaluated the effects
of the classic D3R antagonist 2(22−24) in both WT and D3KO mice. Figure A shows that pretreatment with 2 (25, 50, 100 mg/kg, i.p., 30 min prior to test) dose-dependently
decreased the total number of infusions of heroin self-administered
by the WT mice, with 100 mg/kg (ip) as the most effective dose.
Figure 2
Effects of
compound 2 on heroin self-administration
in WT and D3KO mice. All the data are expressed as mean
± SEM. (A) Total numbers of heroin infusions after injection
of each dose of compound 2. (B) Normalized data showing
drug-induced % change in heroin self-administration over new basal
levels immediately before each test day. One-way ANOVA with repeated
measures over drug dose revealed a significant drug treatment main
effect in WT mice ((A) F3,24 = 10.93, p < 0.001; (B) F3,24 = 13.06, p < 0.001) but not in D3KO mice ((A) F3,12 = 2.21, p > 0.05; (B) F3,12 = 1.46, p > 0.05).
*p < 0.05, ***p < 0.001, compared
to vehicle.
Figure 3
Effects of compound 16 on heroin self-administration
in WT and D3R-KO mice. All the data are expressed as mean
± SEM. (A) Total numbers of heroin infusions after injection
of each dose of compound 16. (B) Normalized data showing
drug-induced % change in heroin self-administration over new baseline
immediately before each test day. One-way ANOVA with repeated measures
over drug dose revealed a statistically significant drug treatment
main effect in WT mice ((A) F3,18 = 2.13, p > 0.05; (B) F3,18 = 9.09, p < 0.001) but not in D3KO mice ((A) F3,18 = 0.63, p > 0.05; (B) D3KO, F3,18 = 1.78, p >
0.05). We
note that one-way ANOVA did not reveal a significant treatment main
effect for the data shown in (A) WT mice. However, the direct two
group comparison between the vehicle and 10 mg/kg 16 groups
revealed a statistically significant reduction ((A) WT, paired t test, q = 5.07, p <
0.05). These may be related to the relatively smaller group size and/or
the relatively variable basal levels of heroin self-administration
in different subjects. Therefore, the renormalized data (% change
over baseline) are provided ((B) in this figure; the same as in Figures ) in this study. *p < 0.05, ***p < 0.001, compared to vehicle.
Figure 4
Effects of compound 32 on heroin self-administration
in WT and D3KO mice. All the data are expressed as mean
± SEM. (A) Total numbers of heroin infusions after injection
of each dose of compound 32. (B) Normalized data showing
drug-induced % change in heroin self-administration over new baseline
immediately before each test day. One-way ANOVA with repeated measures
over drug dose revealed a statistically significant drug treatment
main effect in WT mice ((A) F3,18 = 0.83, p > 0.05; (B) F3,18 = 3.34, p < 0.05) and D3KO mice ((A) F3,15 = 3.63, p < 0.05; (B) F3,15 = 4.39, p < 0.05) *p < 0.05, compared to vehicle.
Effects of
compound 2 on heroin self-administration
in WT and D3KO mice. All the data are expressed as mean
± SEM. (A) Total numbers of heroin infusions after injection
of each dose of compound 2. (B) Normalized data showing
drug-induced % change in heroin self-administration over new basal
levels immediately before each test day. One-way ANOVA with repeated
measures over drug dose revealed a significant drug treatment main
effect in WT mice ((A) F3,24 = 10.93, p < 0.001; (B) F3,24 = 13.06, p < 0.001) but not in D3KO mice ((A) F3,12 = 2.21, p > 0.05; (B) F3,12 = 1.46, p > 0.05).
*p < 0.05, ***p < 0.001, compared
to vehicle.Effects of compound 16 on heroin self-administration
in WT and D3R-KO mice. All the data are expressed as mean
± SEM. (A) Total numbers of heroin infusions after injection
of each dose of compound 16. (B) Normalized data showing
drug-induced % change in heroin self-administration over new baseline
immediately before each test day. One-way ANOVA with repeated measures
over drug dose revealed a statistically significant drug treatment
main effect in WT mice ((A) F3,18 = 2.13, p > 0.05; (B) F3,18 = 9.09, p < 0.001) but not in D3KO mice ((A) F3,18 = 0.63, p > 0.05; (B) D3KO, F3,18 = 1.78, p >
0.05). We
note that one-way ANOVA did not reveal a significant treatment main
effect for the data shown in (A) WT mice. However, the direct two
group comparison between the vehicle and 10 mg/kg 16 groups
revealed a statistically significant reduction ((A) WT, paired t test, q = 5.07, p <
0.05). These may be related to the relatively smaller group size and/or
the relatively variable basal levels of heroin self-administration
in different subjects. Therefore, the renormalized data (% change
over baseline) are provided ((B) in this figure; the same as in Figures ) in this study. *p < 0.05, ***p < 0.001, compared to vehicle.Effects of compound 32 on heroin self-administration
in WT and D3KO mice. All the data are expressed as mean
± SEM. (A) Total numbers of heroin infusions after injection
of each dose of compound 32. (B) Normalized data showing
drug-induced % change in heroin self-administration over new baseline
immediately before each test day. One-way ANOVA with repeated measures
over drug dose revealed a statistically significant drug treatment
main effect in WT mice ((A) F3,18 = 0.83, p > 0.05; (B) F3,18 = 3.34, p < 0.05) and D3KO mice ((A) F3,15 = 3.63, p < 0.05; (B) F3,15 = 4.39, p < 0.05) *p < 0.05, compared to vehicle.Given the observation that the basal levels of heroin self-administration
are variable in individual subjects, we further normalized drug-induced
percent changes in self-administration over new basal levels immediately
before each test day in each animal (Figure B) to more accurately reveal drug effects.
We found that pretreatment with 2 produced a dose-dependent
reduction in heroin self-administration in WT mice (Figure B). Of note, although the D3KO mice self-administered heroin similarly to the WT mice,
this behavior was not attenuated by 2, suggesting that
this D3R antagonist reduces heroin self-administration
through a D3R-mediated mechanism. Likewise, in Figure , compound 16, at the dose range of 1–10 mg/kg, also dose dependently
attenuated heroin self-administration in the WT but not in the D3KO mice. Of note, the effective dose of 16 was
5–10-fold lower than that of 2, suggesting that 16 is more potent than 2 in this model. In our
binding assays, compound 16 displayed ∼70-fold
higher affinity for D3R than 2 (Ki = 0.12 and 13.9 nM, respectively), which is consistent
with its higher potency in vivo.Compound 32 also
dose dependently attenuated the rewarding
effects of heroin, with 3 mg/kg being the most effective dose in both
WT and D3KO mice (Figure ), reflecting the lack of D3R selectivity
of this compound. It is enticing to speculate that the 5-HT1A, 5-HT2A, and/or 5-HT2C receptors may be playing
a role in the behavioral effects of this drug as these serotonergic
receptor subtypes have all been implicated in both drug reward and
impulsivity.[61−64] However, additional experiments must be conducted in order to confirm
this. Nevertheless, compound 32 effectively blocked self-administration
of heroin more potently than either 2 or 16, thus follow-up studies are underway.
Conclusion
In
summary, we have synthesized a novel series of analogues of 6 wherein the 2,3-diCl-phenyl piperazine was replaced with
a 2-OMe, a 2-OMe,3-Cl-, or a 2,3-naphthyl-substituent and a bioisosteric
replacement of the indole moiety. In addition, we also explored analogues
with either the unsubstituted or the 3-OH substituted 4-carbon linker
between the arylpiperazine and aryl amide to investigate its effects
on D3R affinity, selectivity, efficacy, and metabolic stability.
By varying the arylpiperazine substitution, we found several ligands
that displayed high affinity and selectivity profiles at the D3R versus D2R. Overall, we identified indolylamide
analogues, e.g., 16 (Ki =
0.12 nM), which displayed higher D3R binding affinity compared
to lead compound 6 (Ki =
1.9 nM) and >100-fold selectivity for D3R over D2R. Although in general, the 2-OMe-phenylpiperazine-quinolinylamides
had lower affinities at D3R, when substituted with the
2-OMe,3-Cl or the 2,3-naphthyl substituents, D3R affinity
was increased ∼10-fold. Also the 3-OH substituted analogues
showed an increase in selectivity for D3R over D2R, as well as a decrease in lipophilicity, as described for other
4-phenylpiperazine-butylarylamides.[26,49]On the
basis of the D2R/D3R binding profiles,
a subset of analogues was evaluated in a cell-based D2R
or D3R functional assay. These analogues were potent antagonists
in the D3R mitogenesis assay (e.g., 16 and 32; IC50 = 8.0 and 7.4 nM, respectively) but at
higher concentrations demonstrated weak partial agonist profiles.
They were also evaluated for activities at 5-HT1A, 5-HT2A, and 5-HT2C receptors. Among the selected ligands,
in addition to its high binding affinity to D3R (Ki = 0.35 nM), 32 displayed high
affinity for the 5-HT receptor subtypes (Ki = 2.46 nM for 5-HT1A, 0.33 nM for 5-HT2A,
3.80 nM for 5-HT2C) and was a potent 5-HT1A agonist.We then selected compounds based on their in vitro pharmacology
profiles and evaluated them for metabolic stability. Analogues with
or without the 3-OH substitution were compared; 16 and 32 were discovered to be the most metabolically stable in
the series − 37% and 54%, respectively, remained after 60 min
compared to the parent compound 6 (11% remaining after
60 min) in mouse microsomes. Of note, all the compounds with the 3-OH
substituent were metabolically less stable.D3R antagonists
have largely been developed as potential
therapeutic agents for the treatment of nicotine or psychostimulant
abuse. However, they are typically unable to block self-administration
under a low FR schedule of reinforcement and are primarily effective
in models of relapse.[2,4,33] We
hypothesized that D3R antagonists might be able to block
self-administration of addictive drugs, like heroin, that more modestly
stimulate dopamine release in the accumbens. Indeed, behavioral studies
in WT and D3KO mice that were trained to self-administer
heroin demonstrated that D3R-selective antagonists 2 and 16 significantly decreased drug self-administration
in WT but not D3KO mice, suggesting an effect that is mediated
through D3R. In addition, compound 16 was
5–10 times more potent than 2, a classic D3R antagonist, in its pharmacological effect in vivo. Compound 32 also attenuated heroin self-administration in the WT mice
but appeared to be less selective for D3R than either compounds 2 or 16, as at the most effective dose of 3 mg/kg,
self-administration behavior was reduced in both the WT and D3KO mice. Its binding profile suggests that actions at the
5-HT receptor subtypes might also play a role in the behavioral actions
of this compound. Additional behavioral assessment is naturally required,
however, these are the first studies to show the effectiveness of
D3R antagonists in heroin self-administration and suggest
an alternate medication strategy to methadone maintenance or buprenorphine,
the current pharmacotherapeutic treatments for heroin addiction.
Experimental Methods
Synthesis
Reaction
conditions and yields were not optimized.
Anhydrous solvents were purchased from Aldrich and were used without
further purification except for tetrahydrofuran, which was freshly
distilled from sodium-benzophenone ketyl. All other chemicals and
reagents were purchased from Sigma-Aldrich Co. LLC, Combi-Blocks,
TCI America, OChem Incorporation, Acros Organics, Maybridge, and Alfa
Aesar. All amine final products were converted into either the oxalate
or HCl salt. Spectroscopic data and yields refer to the free base
form of compounds. Teledyne ISCO CombiFlash Rf or glass flash column
chromatography were performed using silica gel (EMD Chemicals, Inc.;
230–400 mesh, 60 Å). 1H and 13C
NMR spectra were acquired using a Varian Mercury Plus 400 spectrometer
at 400 and 100 MHz, respectively. Chemical shifts are reported in
parts-per-million (ppm) and referenced according to deuterated solvent
for 1H spectra (CDCl3, 7.26, CD3OD,
3.31, or DMSO-d6, 2.50) and 13C spectra (CDCl3, 77.2, CD3OD, 49.0, or DMSO-d6, 39.5). Gas chromatography–mass spectrometry
(GC/MS) data were acquired (where obtainable) using an Agilent Technologies
(Santa Clara, CA) 6890N GC equipped with an HP-5MS column (cross-linked
5% PH ME siloxane, 30 m × 0.25 mm i.d. × 0.25 μm film
thickness) and a 5973 mass-selective ion detector in electron-impact
mode. Ultrapure grade helium was used as the carrier gas at a flow
rate of 1.2 mL/min. The injection port and transfer line temperatures
were 250 and 280 °C, respectively, and the oven temperature gradient
used was as follows: the initial temperature (100 °C) was held
for 3 min and then increased to 295 °C at 15 °C/min over
13 min and finally maintained at 295 °C for 10 min. Combustion
analysis was performed by Atlantic Microlab, Inc. (Norcross, GA),
and the results agree within ±0.4% of calculated values. cLogP
values were calculated using ChemBioDraw Ultra 4.0. Melting point
determination was conducted using a Thomas–Hoover melting point
apparatus and are uncorrected. On the basis of NMR and combustion
data, all final compounds are ≥95% pure.
1-(3-Chloro-2-methoxyphenyl)piperazine
(7a).[47]
K2CO3 (3.31 g,
23.98 mmol) was added to a solution of 3-chloro-2-methoxyaniline (3.15
g, 19.98 mmol) and bis(2-chloroethyl)amine HCl (4.30 g, 23.98 mmol)
in diglyme (20 mL). The resulting mixture was stirred at reflux for
48 h. The reaction mixture was cooled to room temperature, and diglyme
was removed under vacuum. The crude mixture was diluted with water
(100 mL) and EtOAc (150 mL) and then extracted with EtOAc (3 ×
100 mL). The combined organic layer was dried over MgSO4, filtered, and concentrated. The crude compound was purified by
flash column chromatography (90% CMA) to give the pure compound 7a (2.53 g, 56% yield). 1H NMR (CDCl3) δ 7.01 (dd, J = 8.0, 2.0 Hz, 1H), 6.95 (t, J = 8.0 Hz, 1H), 6.81 (dd, J = 8.0, 2.0
Hz, 1H), 3.88 (s, 3H), 3.10–3.03 (m, 8H), 1.89 (br s, 1H). 13C NMR (CDCl3) δ 148.9, 147.2, 128.9, 124.8,
123.5, 117.3, 59.2, 51.7, 46.6. GC-MS (EI) m/z 226 (M+).
1-(Naphthalen-1-yl)piperazine
(7b).[48]
Compound 7b was synthesized
as described for 7a, using naphthalen-1-amine (4.00 g,
27.94 mmol), bis(2-chloroethyl)amine HCl (5.98 g, 33.52 mmol), and
K2CO3 (4.63 g, 33.52 mmol) in diglyme (25 mL).
The product 7b (3.40 g, 57% yield) was converted to the
HCl salt in 2-propanol/acetone to form a gray solid; mp >250 °C
(HCl salt). 1H NMR (CDCl3) δ 8.24–8.21
(m, 1H), 7.84–7.81 (m, 1H), 7.55 (d, J = 8.4
Hz, 1H), 7.51–7.44 (m, 2H), 7.41 (t, J = 8.4
Hz, 1H), 7.09 (dd, J = 7.6, 1.2 Hz, 1H), 3.16 (t, J = 4.8 Hz, 4H), 3.09 (br s, 4H), 1.99 (br s, 1H). 13C NMR (CDCl3) δ 149.75, 134.4, 128.6, 128.0,
125.5, 125.4, 124.9, 123.2, 123.1, 114.3, 54.1, 46.3. GC-MS (EI) m/z 212 (M+).
2-(2-(Oxiran-2-yl)ethyl)isoindoline-1,3-dione
(8)
Phthalimide potassium salt (6.47 g, 34.9
mmol) was suspended
in DMF (60 mL) under an argon atmosphere and stirred at room temperature
for 10 min. A solution of 2-(2-bromoethyl)oxirane) (5.82 g, 38.50
mmol) in DMF (15 mL) was added dropwise, and the reaction mixture
was left to stir at room temperature for 20 h. The white precipitate
formed was filtered and the filtrate diluted with EtOAc (175 mL).
The white precipitate collected was then dissolved in water and extracted
with EtOAc (3 × 75 mL), and the extract was combined with the
original diluted filtrate. The combined EtOAc extracts were consecutively
washed with water (100 mL) and brine (100 mL), dried over Na2SO4, and filtered. The pure product, 8 (7.48
g, 98% yield), was obtained after solvent removal as an off-white
solid; mp 84–86 °C. 1H NMR (CDCl3) δ 7.87–7.85 (m, 2H), 7.73–7.71 (m, 2H), 3.96–3.83
(m, 2H), 3.02–2.98 (m, 1H), 2.72 (t, J = 5.0
Hz, 1H), 2.45 (dd, J = 4.8, 2.4 Hz, 1H), 2.04–1.96
(m, 1H), 1.90–1.83 (m, 1H). 13C NMR (CDCl3) δ 168.3, 133.9, 132.1, 123.3, 50.2, 46.4, 35.2, 31.6.
A solution
of 8 (2.09 g, 9.62 mmol) in 2-propanol (5 mL) was added
to a solution of 1-(2-methoxyphenyl)piperazine (1.85 g, 9.62 mmol)
in 2-propanol (40 mL). After refluxing the reaction mixture under
an argon atmosphere overnight, the solvent was reduced in vacuo to
give a clear, dark-orange solution. Ice-cold 2-propanol (30 mL) was
added to the orange solution at room temperature, and the mixture
was left to stir overnight at room temperature, resulting in precipitation
of the desired product. Compound 9c (2.87 g, 73% yield)
was recovered after filtration as an off-white solid and washed with
2-propanol; mp 108–109 °C. 1H NMR (CDCl3) δ 7.86–7.84 (m, 2H), 7.72–7.70 (m, 2H),
7.02–6.98 (m, 1H), 6.93–6.91 (m, 2H), 6.86 (d, J = 8.0 Hz, 1H), 3.94–3.83 (m, 2H), 3.85 (s, 3H),
3.80–3.76 (m, 1H), 3.57 (br s, 1H), 3.07 (br s, 4H), 2.86–2.82
(m, 2H), 2.61–2.58 (m, 2H), 2.45–2.35 (m, 2H), 1.79
(q, J = 7.2, 2H). 13C NMR (CDCl3) δ 168.5, 152.2, 141.2, 133.9, 132.2, 123.2, 122.99, 120.99,
118.2, 111.2, 64.5, 63.9, 55.4, 53.4, 50.7, 35.2, 33.6.
Anhydrous hydrazine (0.16 mL, 5.07 mmol)
was added to a suspension of 9a (0.45 g, 1.01 mmol) in
anhydrous ethanol (20 mL), and the mixture was stirred at reflux for
3 h under an argon atmosphere. Solvents were removed, and the recovered
residue was partitioned between chloroform (50 mL) and a 20% aqueous
K2CO3 (50 mL) solution. The layers were separated,
and the organic layer was washed with water (50 mL) and brine (50
mL). The combined organic layer was dried over MgSO4, filtered,
and the solvent removed in vacuo to give 10a (270 mg,
85% yield), which was used without further purification. 1H NMR (CDCl3) δ 7.01–6.92 (m, 2H), 6.80–6.77
(dd, J = 8, 2 Hz, 2H), 3.90–3.87 (m, 1H),
3.85 (s, 3H), 3.84–3.81 (m, 1H), 3.12 (br s, 5H), 3.07–2.93
(m, 2H), 2.78 (br m, 2H), 2.59 (br, 2H), 2.41–2.36 (m, 2H),
1.62–1.57 (m, 2H). 13C NMR (CDCl3) δ
148.6, 146.6, 128.7, 124.6, 123.3, 117.1, 66.3, 64.5, 59.0, 58.9,
53.8, 50.3, 39.5, 37.2. GC-MS (EI) m/z 313 (M+).
Anhydrous
hydrazine (0.28 mL, 9.00 mmol) was added to a solution of 12a (1.28 g, 3.00 mmol) in ethanol (30 mL). The solution was stirred
at reflux for 3 h. The reaction mixture was cooled and concentrated.
The crude reaction mixture was partitioned between CHCl3 (50 mL) and 20% K2CO3 (20 mL) solution, and
the organic layer was collected and dried over MgSO4. The
organic layer was filtered and concentrated to give 13a (855 mg, 96% yield), which was used without further purification. 1H NMR (CDCl3) δ 7.01–6.92 (m, 2H),
6.81 (dd, J = 8.0, 1.2 Hz, 1H), 3.86 (s, 3H), 3.14
(br s, 4H), 2.73 (t, J = 6.8 Hz, 2H), 2.61 (br s,
4H), 2.41 (t, J = 7.2 Hz, 2H), 1.60–1.46 (m,
4H). 13C NMR (CDCl3) δ 148.7, 146.7, 128.8,
124.7, 123.3, 117.2, 59.0, 58.6, 53.8, 50.3, 42.0, 31.6, 24.3.
Compound 13d was synthesized as described for 13a using 12d (1.46 g, 3.37 mmol) and anhydrous hydrazine (0.32 mL, 10.19 mmol)
in ethanol (33 mL). The pure product, 13d (1.01 g, 99%
yield), was obtained as a light-yellow oil. 1H NMR (CDCl3) δ 7.16–7.11 (m, 2H), 6.96 (dd, J = 6.8, 3.2 Hz, 1H), 3.07 (br s, 4H), 2.73 (t, J = 6.6 Hz, 2H), 2.64 (br s, 4H), 2.43 (t, J = 7.6
Hz, 2H), 1.62–1.46 (m, 6H). 13C NMR (CDCl3) δ 151.3, 134.0, 127.5, 127.4, 124.5, 118.6, 58.5, 53.3, 51.3,
42.1, 31.7, 24.3.
General Amidation Procedure, Method A
1,1′-Carbonyldiimidazole
(1 equiv) was added to the solution of carboxylic acid (1 equiv) in
THF (10 mL/mmol). The resulting mixture was stirred at room temperature
for 2 h. The solution was cooled to 0 °C, and the amine substrate
(1 equiv) in THF (3 mL/mmol) was added dropwise. The reaction mixture
was allowed to warm to room temperature and stirred overnight. The
reaction mixture was concentrated, and the crude product was diluted
with CHCl3 (20 mL/mmol) and washed with saturated aqueous
NaHCO3 solution (2 × 10 mL). The organic layer was
dried over MgSO4, filtered, and concentrated. The crude
product was purified by flash column chromatography (EtOAc followed
by 8% MeOH:CHCl3), or as indicated.
Compound 37 was synthesized
according to general method A employing 10d (381 mg,
1.20 mmol) and commercially available 6-fluoroquinoline-3-carboxylic
acid (226 mg, 1.18 mmol). The crude product was purified by column
chromatography as described in general method A to give 37 (310 mg, 53% yield) as a cream-colored solid that was converted
to the oxalate salt in a 2-propanol/acetone solvent mixture; mp 201–203
°C (oxalate salt). 1H NMR (CDCl3) δ
9.25 (d, J = 2.0 Hz, 1H), 8.56 (d, J = 1.6 Hz, 1H), 8.15 (dd, J = 9.2, 5.2 Hz, 1H),
7.76 (br s, 1H), 7.57 (dd, J = 8.4, 2.8 Hz, 1H),
7.53 (dd, J = 7.4, 2.6 Hz, 1H), 7.20–7.14
(m, 2H), 6.96 (dd, J = 6.6, 2.6 Hz, 1H), 4.03–3.94
(m, 3H), 3.55–3.47 (m, 1H), 3.09 (br s, 4H), 2.92–2.88
(m, 2H), 2.64–2.62 (m, 2H), 2.53–2.43 (m, 2H), 1.91–1.84
(m, 1H), 1.71–1.60 (m, 1H). 13C NMR (CDCl3): δ 164.9, 160.8 (1JCF = 24.8 Hz), 150.9, 147.7 (4JCF = 3.0 Hz), 146.3, 134.8 (4JCF = 6.1 Hz), 134.1, 131.9 (3JCF = 9.1 Hz), 127.98, 127.8 (3JCF = 10.6 Hz), 127.6, 127.5, 124.8, 121.3 (2JCF = 25.8 Hz), 118.6, 111.6 (2JCF = 22.0 Hz), 66.9, 63.6, 53.2, 51.4, 39.0, 32.9. Anal.
(C24H25Cl2FN4O2·C2H2O4·H2O)
C, H, N.
Radioligand Binding Assays
Binding
at dopamine D2-like receptors was determined using previously
described
methods.[68] Membranes were prepared from
HEK293 cells expressing humanD2R, D3R, or D4R, grown in a 50:50 mix of DMEM and Ham’s F12 culture
media, supplemented with 20 mM HEPES, 2 mM l-glutamine, 0.1
mM nonessential amino acids, 1× antibiotic/antimycotic, 10% heat-inactivated
fetal bovineserum, and 200 μg/mL hygromycin (Life Technologies,
Grand Island, NY) and kept in an incubator at 37 °C and 5% CO2. Upon reaching 80–90% confluence, cells were harvested
using premixed Earle’s Balanced Salt Solution (EBSS) with 5
μM EDTA (Life Technologies) and centrifuged at 3000 rpm for
10 min at 21 °C. The supernatant was removed, and the pellet
was resuspended in 10 mL of hypotonic lysis buffer (5 mM MgCl2·6H2O, 5 mM Tris, pH 7.4 at 4 °C) and
centrifuged at 20000 rpm for 30 min at 4 °C. The pellet was then
resuspended in fresh EBSS buffer made from 8.7 g/L Earle’s
Balanced Salts without phenol red (US Biological, Salem, MA), 2.2
g/L sodium bicarbonate, pH to 7.4. A Bradford protein assay (Bio-Rad,
Hercules, CA) was used to determine the protein concentration and
membranes were diluted to 500 μg/mL and stored in a −80
°C freezer for later use.Radioligand competition binding
experiments were conducted using thawed membranes. Test compounds
were freshly dissolved in 30% DMSO and 70% H2O to a stock
concentration of 100 μM. To assist the solubilization of free-base
compounds, 10 μL of glacial acetic acid was added along with
the DMSO. Each test compound was then diluted into 13 half-log serial
dilutions using 30% DMSO vehicle; final test concentrations ranged
from 10 μM to 10 pM. Previously frozen membranes were diluted
in fresh EBSS to a 100 μg/mL (for hD2R or hD3R) or 200 μg/mL (hD4R) stock for binding.
Radioligand competition experiments were conducted in glass tubes
containing 300 μL of fresh EBSS buffer with 0.2 mM sodium metabisulfite,
50 μL of diluted test compound, 100 μL of membranes (10
μg total protein for hD2R or hD3R, 20
μg total protein for hD4R), and 50 μL of [3H]N-methylspiperone (0.4 nM final concentration;
PerkinElmer). Nonspecific binding was determined using 10 μM
butaclamol (Sigma-Aldrich, St. Louis, MO), and total binding was determined
with 30% DMSO vehicle. All compound dilutions were tested in triplicate
and the reaction incubated for 1 h at room temperature. The reaction
was terminated by filtration through Whatman GF/B filters, presoaked
for 1 h in 0.5% polyethylenimine, using a Brandel R48 filtering manifold
(Brandel Instruments, Gaithersburg, MD). The filters were washed 3
times with 3 mL of ice-cold EBSS buffer and transferred to scintillation
vials. Then 3 mL of CytoScint liquid scintillation cocktail (MP Biomedicals,
Solon, OH) was added and vials were counted using a PerkinElmer Tri-Carb
2910 TR liquid scintillation counter (Waltham, MA). IC50 values for each compound were determined from dose–response
curves, and Ki values were calculated
using the Cheng–Prusoff equation;[69] these analyses were performed using GraphPad Prism version 5.00
for Windows (GraphPad Software, San Diego, CA). Reported Ki values were determined from least three independent
experiments.
Mouse Microsomal Stability Assay
Phase I metabolic
stability assays were conducted in mouse liver microsomes as previously
described[60] with minor modifications. In
brief, the reaction was carried out with 100 mM potassium phosphate
buffer, pH 7.4, in the presence of NADPH regenerating system, (compound
final concentration was 1 μM; and 0.2 mg/mL microsomes). Negative
controls without cofactors were assessed to determine the non-CYP
mediated metabolism. Positive controls for phase I metabolism (testosterone)
were also evaluated. Compound disappearance was monitored over time
using a liquid chromatography and tandem mass spectrometry (LC/MS/MS)
method. All reactions were performed in triplicate.Chromatographic
analysis was performed using an Accela ultra high-performance system
consisting of an analytical pump and an autosampler coupled with TSQ
Vantage mass spectrometer (Thermo Fisher Scientific Inc., Waltham,
MA). Separation of the analyte from potentially interfering material
was achieved at ambient temperature using Agilent Eclipse Plus column
(100 mm × 2.1 mm i.d.) packed with a 1.8 μm C18 stationary
phase. The mobile phase used was composed of 0.1% formic acid in acetonitrile
and 0.1% formic acid in H2O with gradient elution, starting
with 10% (organic) linearly increasing to 99% up to 2.5 min, maintaining
at 99% (2.5–3.5 min), and reequlibrating to 10% by 4.5 min.
The total run time for each analyte was 4.5 min. The mass transitions
used for compounds for LC/MS/MS analysis are given in Table S1 (Supporting Information).The
calculations of in vitro half-lives were performed as described
previously.[70] Briefly, for the determination
of in vitro half-life (t1/2), the analyte/IS
peak area ratios were converted to percentage drug remaining, using
the T = 0 area ratio values as 100%. The slope of
the linear regression from log percentage remaining versus incubation
time relationships (−k) was used in the conversion
to in vitro t1/2, using the equation t1/2= −0.693/k.
Heroin
Self-Administration
Animals
Male WT and D3KO mice with a C57BL/6J
genetic background were bred at the National Institute on Drug Abuse
from three D3± breeding pairs purchased
from Jackson Laboratory. This strain of D3KO mice expresses
a truncated D3R, including the extracellular N-terminal,
the first intracellular loop, and part of the second intracellular
loop (a total of 148 residues) but lacks downstream sequences from
the second intracellular loop (from residue 149) (Song et al., 2011).
Genotyping was performed in our laboratory in accordance with a protocol
of Charles River Laboratories. All mice used in the present experiments
were matched for age (8–14 wk) and weight (25–35 g).
They were housed individually in a climate-controlled animal colony
room on a reversed light–dark cycle (lights on at 7:00 PM,
lights off at 7:00 AM) with free access to food and water. All experimental
procedures were conducted in accordance with the National Research
Council’s Guide for the Care and Use of Laboratory Animals
and were approved by the National Institute on Drug Abuse’s
Animal Care and Use Committee.
Heroin Self-Administration
Intravenous catheterization
surgery and the self-administration procedures were performed as described
previously (Xi et al., 2011; Song et al., 2012). Briefly, WT or D3KO mice were initially trained for oral sucrose (5% solution)
self-administration for 2–3 days to learn operant lever pressing
for sucrose reward. After lever pressing behavior was achieved, animals
were prepared for iv catheterization surgery. After 5–7 days
of recovery from surgery, each mouse was placed into a test chamber
(Med Associates, St. Albans, VT) and allowed to lever-press for iv
heroin, beginning with 0.1 mg/kg/infusion for a week, and then the
heroin dose was lowered to 0.05 and 0.025 mg/kg/infusion in the following
weeks. Each heroin dose was maintained for 5–7 days until stable
self-administration was achieved. A progressively decreased heroin
dose regimen was chosen based on the fact that total numbers of heroin
infusions at higher doses (0.1 mg/kg/infusion) are usually very low
(∼20). Lowering the heroin dose results in a significant increase
in the numbers of heroin infusions or active lever presses by a compensatory
mechanism that significantly increases “system sensitivity”
to detect the effects of experimental drugs on heroin self-administration
and prevent potential bottom effects. Each lever press led to a delivery
of 0.015 mL of the drug solution over 4.2 s under an FR1 reinforcement
schedule. Each session lasted 3 h or until the animal received the
maximally allowed 50 heroin infusions to prevent drug overdose. Daily
drug self-administration continued until stable day-to-day operant
behavior was established with a steady behavioral response pattern
for at least three consecutive days.
Effects of the Test Compounds
on Heroin Self-Administration
Animals were divided into two
groups to observe the effects of 16 (WT, n = 7; D3KO, n = 7) and 32 (WT, n = 7; D3KO, n = 6) on heroin self-administration, respectively.
After the completion of the tests with 16 and 32, the animals with stable heroin self-administration (WT, n = 9; D3KO, n = 5) were used
again to observed the effects of 2 on heroin self-administration.
On the test day, each dose of 2 (0, 25, 50, 100 mg/kg), 16 (0, 1, 3, or 10 mg/kg, ip), or 32 (0, 0.3,
1, or 3 mg/kg) was given 30 min prior to daily heroin self-administration.
After each test, animals continued until heroin self-administration
recovered to the basal levels before each drug test. The time intervals
between tests were 3–5 days. The sequence of tests for each
dose was counterbalanced.
Authors: Pradeep J Nathan; Barry V O'Neill; Karin Mogg; Brendan P Bradley; John Beaver; Massimo Bani; Emilio Merlo-Pich; Paul C Fletcher; Bridget Swirski; Annelize Koch; Chris M Dodds; Edward T Bullmore Journal: Int J Neuropsychopharmacol Date: 2011-07-12 Impact factor: 5.176
Authors: Michael A Nader; Drake Morgan; H Donald Gage; Susan H Nader; Tonya L Calhoun; Nancy Buchheimer; Richard Ehrenkaufer; Robert H Mach Journal: Nat Neurosci Date: 2006-07-09 Impact factor: 24.884
Authors: C Reavill; S G Taylor; M D Wood; T Ashmeade; N E Austin; K Y Avenell; I Boyfield; C L Branch; J Cilia; M C Coldwell; M S Hadley; A J Hunter; P Jeffrey; F Jewitt; C N Johnson; D N Jones; A D Medhurst; D N Middlemiss; D J Nash; G J Riley; C Routledge; G Stemp; K M Thewlis; B Trail; A K Vong; J J Hagan Journal: J Pharmacol Exp Ther Date: 2000-09 Impact factor: 4.030
Authors: Bernard Le Foll; Ginetta Collo; Eugenii A Rabiner; Isabelle Boileau; Emilio Merlo Pich; Pierre Sokoloff Journal: Prog Brain Res Date: 2014 Impact factor: 2.453
Authors: Fabrizio Micheli; Giorgio Bonanomi; Frank E Blaney; Simone Braggio; Anna Maria Capelli; Anna Checchia; Ornella Curcuruto; Federica Damiani; Romano Di Fabio; Daniele Donati; Gabriella Gentile; Andy Gribble; Dieter Hamprecht; Giovanna Tedesco; Silvia Terreni; Luca Tarsi; Andrew Lightfoot; Geoff Stemp; Gregor Macdonald; Alex Smith; Michela Pecoraro; Marcella Petrone; Ornella Perini; Jacqui Piner; Tino Rossi; Angela Worby; Maria Pilla; Enzo Valerio; Cristiana Griffante; Manolo Mugnaini; Martyn Wood; Claire Scott; Michela Andreoli; Laurent Lacroix; Adam Schwarz; Alessandro Gozzi; Angelo Bifone; Charles R Ashby; Jim J Hagan; Christian Heidbreder Journal: J Med Chem Date: 2007-09-15 Impact factor: 7.446
Authors: David V Herin; Marcy J Bubar; Patricia K Seitz; Mary L Thomas; Gilbert R Hillman; Yevgeniya I Tarasenko; Ping Wu; Kathryn A Cunningham Journal: Front Psychiatry Date: 2013-02-06 Impact factor: 4.157
Authors: Vivek Kumar; Amy E Moritz; Thomas M Keck; Alessandro Bonifazi; Michael P Ellenberger; Christopher D Sibley; R Benjamin Free; Lei Shi; J Robert Lane; David R Sibley; Amy Hauck Newman Journal: J Med Chem Date: 2017-02-10 Impact factor: 7.446
Authors: Chloe J Jordan; Bree A Humburg; Eric B Thorndike; Anver Basha Shaik; Zheng-Xiong Xi; Michael H Baumann; Amy Hauck Newman; Charles W Schindler Journal: J Pharmacol Exp Ther Date: 2019-09-27 Impact factor: 4.030
Authors: Amy E Moritz; R Benjamin Free; Warren S Weiner; Emmanuel O Akano; Disha Gandhi; Ara Abramyan; Thomas M Keck; Marc Ferrer; Xin Hu; Noel Southall; Joseph Steiner; Jeffrey Aubé; Lei Shi; Kevin J Frankowski; David R Sibley Journal: J Med Chem Date: 2020-05-12 Impact factor: 7.446
Chart 1
Scheme 1
Scheme 2
Figure 1
Figure 2
Figure 3
Figure 4