You are on page 1of 56

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article
Dopamine D3/D2 Receptor Antagonist PF-4363467 Attenuates
Opioid Drug-Seeking Behavior without Concomitant D2 Side Effects
Travis T Wager, Thomas A. Chappie, David Horton, Ramalakshmi Y. Chandrasekaran, Brian
M. Samas, Elizabeth Dunn-Sims, Cathleen Hsu, Nawshaba Nawreen, Michelle A Vanase-
Frawley, Rebecca E O'Connor, Christopher Schmidt, Keith Dlugolenski, Nancy C. Stratman,
Mark J. Majchrzak, Bethany L. Kormos, David Nguyen, Aarti Sawant-Basak, and Andy N. Mead
ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.6b00297 • Publication Date (Web): 07 Oct 2016
Downloaded from http://pubs.acs.org on October 12, 2016

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Neuroscience is published by the American Chemical Society. 1155


Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 55 ACS Chemical Neuroscience

1
2
3
4
5
Dopamine D3/D2 Receptor Antagonist PF-4363467
6
7
Attenuates Opioid Drug-Seeking Behavior without
8
9 Concomitant D2 Side Effects
10
11
12
13 Travis T. Wager, Thomas Chappie, David Horton, Ramalakshmi Y. Chandrasekaran, Brian
14
15
16
Samas, Elizabeth R. Dunn-Sims, Cathleen Hsu, Nawshaba Nawreen, Michelle A. Vanase-
17
18 Frawley, Rebecca E. O'Connor, Christopher J. Schmidt, Keith Dlugolenski, Nancy C. Stratman,
19
20 Mark J. Majchrzak, Bethany L. Kormos, David Nguyen, Aarti Sawant-Basak, Andy N. Mead
21
22
23
24 Pfizer Worldwide Research and Development; Neuroscience Medicinal Chemistry and
25
26 Neuroscience Research Unit, 610 Main Street, Cambridge, MA 02139 (TTW, TC, CJS, KD,
27
28 NCS, BLK); Pharmacokinetics, Dynamics, and Metabolism, 610 Main Street, Cambridge, MA
29
30
31 02139 (DN, AS-B); Chemistry and Biology, Eastern Point Road, Groton, CT 06340 (RYC, BS
32
33 DH, ERD-S, CH, NN, MJM ANM, MAV-F, REO).
34
35
36
37
ABSTRACT: Dopamine receptor antagonism is a compelling molecular target for the
38
39 treatment of a range of psychiatric disorders, including substance use disorders. From our
40
41 corporate compound file we identified a structurally unique D3 receptor (D3R) antagonist
42
43
44 scaffold, 1. Through a hybrid approach, we merged key pharmacophore elements from 1 and D3
45
46 agonist 2 to yield the novel D3R/D2R antagonist PF-4363467 (3). Compound 3 was designed to
47
48 possess CNS drug-like properties as defined by its CNS MPO desirability score (≥4/6). In
49
50
51 addition to good physicochemical properties, 3 exhibited low nanomolar affinity for the D3R
52
53 (D3 Ki = 3.1 nM), good subtype selectivity over D2R (D2 Ki = 692 nM), and high selectivity for
54
55
D3R versus other biogenic amine receptors. In vivo, 3 dose-dependently attenuated opioid self-
56
57
58
59
60 1
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 2 of 55

1
2
3
administration and opioid drug-seeking behavior in a rat operant reinstatement model using
4
5
6 animals trained to self-administer fentanyl. Further, traditional extrapyramidal symptoms (EPS),
7
8 adverse side effects arising from D2R antagonism, were not observed despite high D2 receptor
9
10
11
occupancy (RO) in rodents, suggesting that compound 3 has a unique in vivo profile.
12
13 Collectively, our data support further investigation of dual D3R and D2R antagonists for the
14
15 treatment of drug addiction.
16
17
18
19
20
21 KEYWORDS: Substance use dependence, drug addiction, CNS MPO, dopamine D3 antagonist,
22
23 dopamine D2 antagonist, opioid, cessation, drug-seeking behavior, reinstatement, extrapyramidal
24
25
26 symptoms, locomotor.
27
28
29 a
Abbreviations: ADME, absorption, distribution, metabolism and excretion; CK1, casein kinase
30
31
32
1; AUC, area under the curve; Cave, average concentration; Ceff, efficacious concentration; Cmax,
33
34 maximum concentration; CNS, central nervous system; CPP, conditioned place preference; DA,
35
36 dopamine; D2R, dopamine 2 receptor; D3R, dopamine 3 receptor; EPS, extrapyramidal
37
38
39 symptoms; HBD, hydrogen bond donor; HLM, human liver microsomes; ICSS, intracranial self-
40
41 stimulation; MDR, multi-drug resistance; MPO, multi-parameter optimization; P-gp, P-
42
43 glycoprotein; SAR, structure-activity relationship; SBDD, structure-based drug design.
44
45
46
47 Introduction
48
49 Drug addiction is characterized as compulsive, out-of-control drug use despite negative
50
51
52
consequences including significant morbidity and mortality. Substance abuse is a global health
53
54 concern with major socio-economic ramifications, which has reached epidemic proportions. For
55
56 example, opioid dependency increased 74% from 1990 through 2010 and the economic cost
57
58
59
60 2
ACS Paragon Plus Environment
Page 3 of 55 ACS Chemical Neuroscience

1
2
3
attributable to illicit drug use in the United States alone is estimated to be over $500 billion
4
5
6 annually.1, 2 Opioid dependency is particularly troubling in that it accounted for nearly half the
7
8 estimated 20 million disability-adjusted life years worldwide for illicit drug use in 2010 and
9
10
11
more than half of the 78,000 deaths in 2010 due to illicit drug use.2 Furthermore, recent trends
12
13 suggest that opioid addiction afflicts young people (ages 15–29), a population that will likely
14
15 carry this burden of drug dependency throughout life. Given the continued rise in opioid
16
17
18 dependency, there is an urgent need to develop new medicines that can be used both to treat
19
20 addiction and to prevent relapse.
21
22 We recently disclosed that casein kinase 1 inhibitors (CK1i) attenuate opioid-seeking
23
24
25 behavior in a rodent reinstatement model of fentanyl self-administration.3 While CK1 inhibition
26
27 is an attractive target for the treatment of drug addiction, dopamine (DA) receptor antagonism is
28
29
also a compelling molecular target, given the rich preclinical and clinical literature data on this
30
31
32 approach. In particular, the dopamine D3 receptor (D3R) has been pursued for more than a
33
34 decade as a potential target for the treatment of substance use disorders,4, 5 with numerous groups
35
36
37 developing selective ligands targeting D3R.6-8 Several lines of evidence provide a rationale to
38
39 target D3R and D2 receptors (D2R) for the treatment of substance abuse disorders. D3R and
40
41 D2R are distributed in key brain regions that regulate the rewarding, motoric, cognitive, and
42
43
44 emotional effects of opioids. D2R expression is widespread in the brain, with the highest
45
46 concentration found in the striatum, followed by the nucleus accumbens, globus pallidus,
47
48 subtantia nigra and ventral tegmental area.9 Conversely, D3R is most abundant in the globus
49
50
51 pallidus, ventral pallidum/substantia innominata, and substantia nigra, which have all been
52
53 associated with motivational processing and are likely part of the neurocircuitry of addiction and
54
55
56
relapse.9-11 In contrast to D2R, D3R is virtually absent in the ventral tegmental area, which is
57
58
59
60 3
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 4 of 55

1
2
3
associated with a rewarding response to low-dose opioid injections.12 Blockade of D2R in the
4
5
6 ventral tegmental area may help reduce opioid use. Further, there is evidence that abnormal D2R
7
8 expression and function can provide a pre-existing vulnerability, in addition to arising as a
9
10
11
consequence of substance abuse disorder.13 Co-expression of D2R and D3R in many brain
12
13 regions suggests a functional convergence of the signals mediated by these receptors.9 The brain
14
15 regions that express D2R and D3R have been shown to be important for motivation and reward,
16
17
18 as well as drug-mediated and cue-induced relapse. Thus, the localization of these receptors in
19
20 these brain regions suggests a potential role for modulating addictive behaviors.
21
22 Repeated exposure to illicit drugs produces long-term molecular and neurochemical
23
24
25 changes in the brain in both humans and animals.14 Postmortem evaluation of the expression of
26
27 D3R in the brains of chronic cocaine users revealed an increased receptor density when
28
29
compared with age-matched and drug-free control subjects.15-17 This observation of increased
30
31
32 receptor density has been extended to other drugs of abuse, including methamphetamine and
33
34 alcohol.18, 19
The up-regulation of D3R in the addicted state suggests a role for D3R in the
35
36
37 neuroadaptive responses to addictive drugs, and implies that blocking D3R may reduce craving
38
39 and relapse rates.
40
41 Recent literature suggests that D3R/D2R ligands have potential efficacy in blocking
42
43
44 drug-seeking behavior in both preclinical and clinical models of addiction.6, 20-22
Early work
45
46 examining the effect of D3R preferential agonists on cocaine self-administration in rats revealed
47
48 the importance of D3Rs in the reinforcing effects of cocaine.23 Further research has shown that
49
50
51 BP 897, a D3R partial agonist, dose-dependently reduced cocaine seeking;20 additional studies,
52
53 however, have shown that this compound exhibits antagonism at D3R.24 More recently, Román
54
55
56
et al. reported that cariprazine, a 10-fold selective D3R/D2R partial agonist, was able to reduce
57
58
59
60 4
ACS Paragon Plus Environment
Page 5 of 55 ACS Chemical Neuroscience

1
2
3
the rewarding effect of cocaine and to attenuate relapse to cocaine-seeking in rats.21 Further work
4
5
6 with antagonists has shown that SB-277011-A, a D3R/D2R antagonist, inhibits both cocaine-
7
8 conditioned place preference (CPP) and reinstatement of cocaine-seeking behavior, as well as
9
10
11
cocaine-induced increases in brain stimulation reward.25 NGB 2904, a selective D3R antagonist,
12
13 was reported to inhibit cocaine self-administration under a progressive ratio schedule,
14
15 reinstatement of cocaine-seeking behavior, and enhancement of brain stimulation reward induced
16
17
18 by cocaine, methamphetamine, heroin, and nicotine.26 Others have disclosed the discovery of
19
20 novel selective D3R antagonists and reported the potential utility of D3R antagonist in treating
21
22 opioid addiction.27-30 Finally, the selective D3R antagonist GSK598809 dose-dependently
23
24
25 attenuated nicotine-induced CPP in rats; this effect translated into the clinic, where GSK598809
26
27 transiently alleviated craving in smokers at sub-maximal receptor occupancy (72–89%).31 This
28
29
investigation provided the first clinical evidence supporting the hypothesis that D3R antagonism
30
31
32 would be effective in treating substance use disorders.
33
34 Given the growing preclinical and clinical data arising from subtype-preferring D3R/D2R
35
36
37 antagonists and selective D3R antagonists, antagonism of these receptors might be a promising
38
39 approach to treat substance abuse disorders and addiction. High DA receptor subtype selective
40
41 D3R antagonists are ideal for preclinical study of the specific role of D3R in addiction; however,
42
43
44 they may fall short of desired outcomes in the clinic due to the complex nature of addiction and
45
46 dopamine signaling. For this reason we pursued compounds that had the potential to engage
47
48 D3R at high occupancy levels, while simultaneously interacting with D2R at a wide range of
49
50
51 occupancies, given the potentially significant roles that both D3R and D2R play in addiction.
52
53 Based on the receptor occupancy achieved in the GSK598809 clinical study, and our knowledge
54
55
56
of D2R-related EPS effects, our initial target was a compound that could achieve high D3
57
58
59
60 5
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 6 of 55

1
2
3
receptor occupancy (>90%) with low D2 receptor occupancy (<10%). A second objective was to
4
5
6 establish whether it was possible to titrate in higher D2R occupancy and thereby achieve
7
8 improved efficacy in rodent models of substance use and relapse disorders. We hypothesized
9
10
11
that this dual-pharmacology strategy would improve efficacy by broadening the modulatory
12
13 influence over cue-reward processing pathways.
14
15 Results and Discussion
16
17
18 Our re-entry into identifying subtype-selective D3R antagonists began with a review of
19
20 our corporate file, which at the time included compounds from legacy Pfizer and Pharmacia &
21
22 Upjohn D3R and D2R antagonist and agonist programs; these analogues had been assayed for
23
24
25 binding and functional activity. Of the compounds that were in the file, PNU-177864 (1)
26
27 represented a structurally unique D3R antagonist scaffold (Table 1).32 Structurally, 1 had an aryl
28
29
sulphonamide, distinguishing it from many of the published selective D3R antagonists, which
30
31
32 often contained an aryl-piperidine moiety.33 Compound 1 exhibited moderate affinity for D3R
33
34 and good selectivity over D2R (Table 1) and was shown to be a functional antagonist, as
35
36
37 measured by a human D3 SH-SY5Y GTP-γ-S functional assay. The known in vivo
38
39 phospholipidosis safety liability of 1 prompted us to identify an amine replacement.32
40
41 Specifically, we wanted to reduce the pKa of the basic amine and lower the lipophilicity of the
42
43
44 compound. This approach has been successfully applied in a histamine H3 receptor program to
45
46 reduce phospholipidosis liability, and resulted in the identification of two clinical candidates.34
47
48 At this point we revisited our corporate file, seeking other D3R ligands with predicted pKa < 9
49
50
51 for the most basic center. PF-592379 (2) was identified as a potential lead that retained affinity
52
53 and selectivity for D3R over D2R, although functionally it was a full agonist (Table 1).35 The
54
55
56
physicochemical properties of 2 were significantly better than those of 1 (Table 2), thus allowing
57
58
59
60 6
ACS Paragon Plus Environment
Page 7 of 55 ACS Chemical Neuroscience

1
2
3
for considerable structural modification opportunities with which to switch the in vitro functional
4
5
6 activity from agonist to antagonist.
7
8 Hybridization of 1 and 2, along with application of traditional structure activity
9
10
11
relationships (SAR), led to the identification of PF-4363467 (3) and its enantiomer PF-4363476
12
13 (not shown) (Table 1). Compound 3 exhibited single digit nM affinity for D3R (Ki = 3.1 nM)
14
15 and greater than 100-fold selectivity over D2R (Ki = 692 nM; Table 1). Further, 3 was an
16
17
18 antagonist in the human D3 SH-SY5Y GTP-γ-S functional assay and in the human D2 SH-SY5Y
19
20 cAMP stimulation assay (Table 1). Functional D2/D3 selectivity was in line with binding
21
22 selectivity, thus achieving our initial goal of a novel D3-selective antagonist.
23
24
25 Table 1. Corporate file mining leads and hybrid compound 3.
26
27
28
29
30
31
32
33
34
35
36
37 D3 Ki D2 Ki D2/ D3 Functional D2 Functional Functional
38 ID
(nM)a (nM)a D3 (nM)b (nM)b Activity
39
40 1 38.2 4480 117 Ki = 22 Ki > 33c Antagonist
41 (23.6 – 61.9) (3290 – 6090) (10.1 - 48.9)c
42 2 322 >4160 >12 EC50 = 21d EC50 > 9,000d Agonist
43
(57.9 – 1790)
44
45 3 3.1 692 223 Ki = 464
Ki = 1.1 Antagonist
c c
46 (2.47 – 3.99) (539 – 890) (0.61 - 2.1) (188 – 1147)
a
47 Human dopamine D3 and D2 binding data reported as the geometric mean of at least 3
48
49
determinations. bHuman dopamine D3 and D2 functional data reported as the geometric mean of
50
51 c d
52 at least 3 determinations. Data reported as mean antagonist activity with (95% CI). Data
53
54 reported by Attkins et al.35 as agonist activity.
55
56
57
58
59
60 7
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 8 of 55

1
2
3
The overall physicochemical properties of 3 are in line with marketed CNS drugs, with a
4
5
6 CNS multi-parameter optimization (CNS MPO) desirability score of 4.0/6.0 (Table 2).36-38
7
8 Improvements in ClogP, HBD, and pKa were achieved relative to compound 1; most notable is
9
10
11
the change in pKa, which was lowered from 10.6 to 7.5. ClogD7.4 is still relatively high for 3
12
13 (ClogD7.4 = 3.96, which may be a contributing factor in its high human liver microsome (HLM)
14
15 clearance (Table 3). Future work around this chemotype should focus on further lowering ClogP
16
17
18 and ClogD7.4, with a goal of improving CNS MPO desirability and aligning all ADME
19
20 attributes to desirable values.
21
22 Table 2. Summary of physicochemical properties and CNS MPO desirability
23
24
25 Compound 1 Compound 2 Compound 3
26
27 Physicochemical Component Component Component
28 Property Value Score Value Score Value Score
29
30
31
MW 402 0.70 235 1.00 403 0.70
32 ClogP 4.46 0.27 1.48 1.00 4.06 0.47
33 TPSA 67.4 1.00 51.4 1.00 58.6 1.00
34 ClogD7.4 1.06 1.00 0.76 1.00 3.96 0.02
35
36 HBD 2 0.50 2 0.50 1 0.83
37 pKa 10.6 0.00 7.5 1.00 7.5 1.00
38
39 CNS MPO 3.5/ 6.0 5.5/ 6.0 4.0/ 6.0
40 a
Calculated CNS MPO desirability scores were obtained using the published algorithm.37 Color
41
42
43 code for component score (x): red, x ≤ 0.33; yellow, 0.33 < x ≤ 0.66; green, x > 0.66.
44
45 In addition to good physicochemical properties and potency, compound 3 exhibited good
46
47
48
passive permeability, with a Papp value of 8.4 × 10-6 cm/sec and predicted low-to-moderate P-gp
49
50 liability (Table 3). Brain availability (Cb,u/Cp,u) in rats was 1.0 and free brain exposure of 3
51
52 (Cmax,b,u) at a 10 mg/kg s.c. dose was determined to be 23.7 nM. At a 10 mg/kg s.c. dose, 3 had
53
54
55 sufficient free drug exposure to cover the D3R Ki (∼7 fold). Cmax,b,u was further used to estimate
56
57 target occupancy (%) as:
58
59
60 8
ACS Paragon Plus Environment
Page 9 of 55 ACS Chemical Neuroscience

1
2
3
4
5 (1)
6
7
8 where the dissociation constant Kd represents in vitro target binding potency. Using the binding
9
10
11 Ki as a surrogate for Kd, compound 3 was initially estimated to have maximum CNS target
12
13 occupancy of 88% for D3R and 5% for D2, after a 10 mg/kg s.c. dose.
14
15
16 Table 3. Summary of ADME properties for 3.
17
18
19 Rat Cmax,b,u Rat AUC0-∞ Rat AUC0-∞
20 Compound HLMa Pappb P-gpc
[nM]d Cb/Cp Cb,u/Cp,u
21
22 3 126 8.4 2.3 23.7 2.4 1.01
23 a
24 Human liver microsomal clearance (µL/min/mg). Passive permeability (Papp AB x 10-6 cm/sec).
b

25
26 c d
MDR1 Efflux Ratio (BA/AB). Rat Cmax,b,u values are reported as maximum free drug
27
28
29
concentration in brain at a 10 mg/kg dose, s.c.
30
31
32 Compound 3 was further evaluated in a broad selectivity panel, where it demonstrated
33
34 high selectivity for D3R over receptors for other biogenic amines, ion channels, and enzymes
35
36
37 (Figure 1, see supplementary material, Table 1s for details). In this 129-target selectivity panel,
38
39 3 inhibited 22 targets at >50% inhibition at 10 µM drug concentration. Of the off-target
40
41 activities tested in IC50 format, 3 had the highest affinity for CYP2C19 (110 nM, 34-fold higher
42
43
44 than the D3R binding IC50), followed by 5HT6 (990 nM, 309-fold higher than the D3R IC50) and
45
46 5HT2A (1.4 µM, 437-fold higher than the D3R IC50). Overall, compound 3 possessed an
47
48
49
excellent selectivity profile and was therefore progressed into ex vivo brain target occupancy
50
51 studies.
52
53
54
55
56
57
58
59
60 9
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 10 of 55

1
2
3
10% at 10 uM 100% at 10 uM
4
5
6
7
8
9
10
11
12 7600 nM 4 nM
13
14
15
16
17
18
19

D2S
D1

D2

D3
CHT1

Ca Chan

Na Chan

H3
M5

M1

M4
5HT2B

5HT6
5HT1A

5HT2A

5HT2A
CYP3A4
CYP3A4b
SIGMA
ALPHA2B

CYP2C9

CYP2C19
ALPHA2A

20
LTB4 BLT1

ALPHA2C
M3

21
22
23
24 Figure 1. A) Percent inhibition of a 129-target promiscuity panel at 10 µM of compound 3.
25
26
27 Heat map colors for this continuum: green, 0% inhibition; yellow, 50% inhibition; red, 100%
28
29 inhibition. B) IC50 determination for 26 targets hitting in the promiscuity panel with inhibition
30
31 >50% for compound 3. Heat map colors continuum: green, 7,600 nM; yellow, 100 nM and red,
32
33
34 4 nM. IC50 determinations for the 10 highest affinity targets for compound 3 were: D3R, 4 nM;
35
36 CYP2C19, 110 nM; D2R, 420 nM; D2S, 540 nM; 5HT6, 990 nM; 5HT2A, 1400 nM; CYP3A4b,
37
38
2300 nM; H3, 2400 nM; D1R, 2500 nM.
39
40
41
42 Using an ex vivo autoradiography technique with rat brain slices, compound 3 was
43
44 evaluated for both D3R and D2R occupancy (Table 4, Figure 2A, 2B). The 0.1 mg/kg dose
45
46
47 resulted in moderate D3R occupancy (33.6%) and low D2R occupancy (10.8%). High D3R
48
49 occupancy was achieved at the 3.2 mg/kg dose with near maximal occupancy (>95%) and
50
51 moderate D2R occupancy (34.7%). It was not until 32 mg/kg that D2R occupancy was over
52
53
54 75%. The ex vivo receptor occupancy EC50 was determined to be 1.5 nM for D3R and 46.6 nM
55
56 for D2R. The D3R occupancy EC50 is within 3-fold of the binding Ki; however, the D2R
57
58
59
60 10
ACS Paragon Plus Environment
Page 11 of 55 ACS Chemical Neuroscience

1
2
3
occupancy EC50 is 10-fold left-shifted when compared to the in vitro binding Ki. This was the
4
5
6 first piece of data that suggested this compound does not behave like a traditional D3/D2
7
8 antagonist (i.e. eticlopride and sulpiride). The ex vivo D3R and D2R occupancy characteristics
9
10
11
of 3 represented a reasonable profile to test two hypotheses with one molecule, simply by
12
13 varying the dose. First, can a selective D3R antagonist (low dose) attenuate opioid drug-seeking
14
15 behavior or, second, is dual pharmacology required (high dose) to yield a robust in vivo
16
17
18 biological response?
19
20 Table 4. D3R and D2R brain receptor occupancy from autoradiography techniques in rats
21
22
23 Treatment Cmax,b,u D3 EVRO % RO D2 EVRO %
Compound
24 (mg/kg) (nM) Observeda RO Observedb
25
26 32 207.1 --- 85.9 ± 1.7
27 10 43.9 98.8 ± 2.4 50.6 ± 3.5
28
29 3.2 13.6 99.9 ± 0.5 34.7 ± 1.5
30 3
0.1 0.7 33.6 ± 4.5 10.8 ± 4.1
31
32 0.032 0.2 6.3 ± 10.1 ---
33
34 EC50 1.5 ± 0.4 nM 46.6 ± 9.1 nM
35 a b
Brain region evaluated: rat cerebellum (mean ± SEM, n = 3–4). Brain region evaluated: rat
36
37
38
striatum (mean ± SEM, n = 4). Ex vivo Receptor Occupancy (EVRO).
39
40
41 A) B)
100
100 96
42
Ex-Vivo, D3 Receptor Occupancy
Ex-Vivo, D2 Receptor Occupancy

43 80 80
44
45 60 60

46
40 40
47
48
20 15
49 7
50 0 0
51
Dose = 0.1 3.2 10 32 mg/kg Dose = 0.01 0.032 0.1 3.2 10 mg/kg
52 -12 -10 -8 -6
-13 -12 -11 -10 -9 -8 -7 10 10 10 10
53 10 10 10 10 10 10 10
Free Brain Concentration (nM)
Free Brain Concentration (nM)
54
55 Figure 2. A) D2 (striatum) and B) D3 (cerebellum) brain receptor occupancy curves from the ex
56
57 vivo autoradiography protocol in rat. The gray and black data points represent actual measured
58
59
60 11
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 12 of 55

1
2
3
RO and brain PK from individual animals. The solid line is the data-fitted RO model. The D3R
4
5
6 binding was measured in cerebellum, while the D2R binding was measured in striatum.
7
8
9 Given the higher-than-expected D2R occupancy in the ex vivo receptor occupancy assay,
10
11
12 we were concerned that significant D2R side effects would be observed at low doses (≤10
13
14 mg/kg). Haloperidol and other D2R antagonists provoke prolactin secretion and elicit catalepsy
15
16 in preclinical and clinical studies.39-42 The induction of catalepsy in rat preclinical studies is
17
18
19 believed to be a predictor of extrapyramidal syndrome (EPS) in man. Further, it has been
20
21 suggested that D2R antagonism may decrease locomotor activity.43 Prolactin serves as a marker
22
23 for D2R antagonism and the magnitude of response is roughly correlated with the occupancy of
24
25
26 D2 receptors in the pituitary.44, 45
As expected, 3 (0.1–10 mg/kg) dose-dependently increases
27
28 prolactin release in rats (Figure 3A). At the highest dose tested in rats (10 mg/kg), 3 increases
29
30
31
plasma prolactin 9.3-fold (p < 0.05). Unexpectedly, compound 3 had no significant effects on
32
33 spontaneous locomotor activity (Figure 3B) at the highest dose tested (32 mg/kg), where D2R
34
35 occupancy is 85.9%. Further, 3 did not produce a statistically significant cataleptic effect in rats
36
37
38 at any dose tested up to a dose of 32 mg/kg (Figure 3C). While this was surprising, there have
39
40 been reports of selective D3R antagonists attenuating catalepsy induced by haloperidol.46 The
41
42 lack of significant locomotor and catalepsy side effects after administration of 3 make this a
43
44
45 unique compound, and we undertook further evaluations to determine whether there are
46
47 advantages of dosing 3 to both high D3R and D2R occupancy in models of addiction.
48
49
50
51
52
53
54
55
56
57
58
59
60 12
ACS Paragon Plus Environment
Page 13 of 55 ACS Chemical Neuroscience

1
2
3 A) B) Vehicle C) Vehicle
** 9.3 fold ↑
4 Haldol, 1 mg/kg
3, 1.0 mg/kg
Haldol, 1 mg/kg
3, 1.0 mg/kg
**
5 3, 3.2 mg/kg
3, 10 mg/kg
3, 3.2 mg/kg
3, 10 mg/kg **
6 3, 32 mg/kg 3, 32 mg/kg **

7
8
9 1.5 fold ↑
1.8 fold ↑
10
11 e
ic
l 0.
1
1.
0 10
12 V
eh
3, Dose (mg/kg, SC)
13
14
15 Figure 3. A) Effect of 3 on prolactin. Data represent mean ± SEM, n = 5/group. B) Effects of 3
16
17 on total activity in a rat model of locomotor activity. Data represent mean ± SEM locomotor
18
19
20
activity over 30 minute time bins, n = 8/group. C) Effects of 3 in a rat model of catalepsy. Data
21
22 represents mean ± SEM seconds on bar demonstrating the effect of 3 (1.0–32.0 mg/kg, s.c.), n =
23
24 8/group. ** indicates p < 0.01 compared to vehicle group.
25
26
27
28 To evaluate the utility of D3R/D2R antagonism in substance use disorder treatment, the
29
30 effect of 3 was assessed in both a cessation model of drug-taking (Figure 4A) and a relapse
31
32 model of drug-seeking behavior (Figure 4B). With respect to cessation, administration of 3
33
34
35 reduced self-administration of fentanyl (Figure 4A), as indicated by a significant main effect of
36
37 dose (F(4, 312) = 23.93, p < 0.0001), a significant main effect of session (F(6,312) = 8.14, p < 0.0001)
38
39
and a significant dose by session interaction (F(24,312) = 2.64, p < 0.0001). Follow-up analysis of
40
41
42 infusions for each test session by one-way ANOVA revealed a significant main effect of dose
43
44 across sessions 1, 2, and 3 (F(4,37) = 7.47, p<0.001; F(4,37) = 6.89, p < 0.001; F(4,37) = 11.2, p <
45
46
47
0.0001, respectively). Post hoc comparisons at each test session revealed a significant decrease
48
49 in infusions following pretreatment with 32 mg/kg of 3 for all three sessions (p < 0.001) and
50
51 when fentanyl was replaced with saline under the saline extinction condition for sessions 1 and 3
52
53
54 (p < 0.05 and p < 0.01, respectively). Analysis of the average infusions obtained over three
55
56 consecutive test days in the fentanyl cessation model following treatment with 3 revealed a
57
58
59
60 13
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 14 of 55

1
2
3
significant main effect of dose (F4,37 = 10.8, p < 0.0001) and post hoc comparisons indicated a
4
5
6 significant decrease in average infusions following pretreatment of 32 mg/kg and when fentanyl
7
8 was replaced with saline (saline extinction) compared to when fentanyl was available for self-
9
10
11
administration (vehicle; Figure 1s, supporting information). A trend for a reduction in fentanyl
12
13 self-administration was observed following 10 mg/kg of 3, but the effect did not reach statistical
14
15 significance (p = 0.08).
16
17
18 To further determine the effect of 3 on fentanyl-seeking behavior, the ability of 3 to
19
20 attenuate reinstatement of fentanyl seeking was assessed, following exposure to both a fentanyl
21
22 drug prime and fentanyl-associated cues. Analysis of results revealed a main effect of dose
23
24
25 (F(3,21) = 3.28; p < 0.05), a main effect of test (F(1,7) = 7.66; p < 0.05) and a significant dose by
26
27 test interaction (F(3,21) = 3.32; p < 0.05; Figure 4B). Post hoc comparisons revealed a significant
28
29
reinstatement effect following administration of 3.2 (t(7) = −2.52; p < 0.05) and 10 mg/kg (t(7) =
30
31
32 −2.94; p < 0.05) of 3 compared to baseline, with a strong trend for an increase following vehicle
33
34 pretreatment (p = 0.058). While the effect of vehicle on reinstatement was just outside
35
36
37 significance, normalization of the active lever responses to fentanyl intake obtained during the
38
39 last two sessions of fentanyl self-administration training (Figure 2s, supporting information),
40
41 revealed a main effect of dose (F(3,21) = 4.30; p < 0.05), a main effect of test (F(1,7) = 24.4; p <
42
43
44 0.01) and a significant dose by test interaction (F(3,21) = 4.70; p < 0.05; Figure 2s). Post hoc
45
46 analysis revealed a significant reinstatement effect following pretreatment with vehicle, 10
47
48 mg/kg of 3, or 3.2 mg/kg of 3 (p < 0.05). With respect to the effect of 3 on reinstatement of
49
50
51 fentanyl seeking, post hoc analysis revealed that 3 decreased active lever responses under
52
53 reinstatement test conditions at 32 mg/kg compared to vehicle. Analysis of inactive lever
54
55
56
responses failed to reveal a main effect of test, main effect of dose, or dose by interaction effect.
57
58
59
60 14
ACS Paragon Plus Environment
Page 15 of 55 ACS Chemical Neuroscience

1
2
3
At this dose it is expected that both the D3R and D2R receptors will be highly occupied, so dual
4
5
6 antagonism seems to be responsible for the observed activity. These results make a compelling
7
8 case for D3R/D2R antagonism in both the cessation and relapse phase of opioid addiction
9
10
11
treatment.
12
13 A) B)
14
15 **
16
17
18
19
20 Extinction
21 Vehicle
3, 3.2 mg/kg
22 3, 10 mg/kg
23 3, 32 mg/kg
24
25
26
27 Treatment 3, Dose (mg/kg, SC)
28
29
30 Figure 4. A) Effects of vehicle or 3 pretreatment on fentanyl self-administration over 3 test days
31
32 in the rat model of cessation. Data represent the average infusions of fentanyl across baseline
33
34 responding (fentanyl) before and after test sessions as well as three consecutive test sessions with
35
36
37 3 (Days 1–3). Saline extinction tests (Extinction) are also shown, in which saline was substituted
38
39 for fentanyl during test sessions. n = 9–11; *** p < 0.001. B) Effects of 3 on combined cue and
40
41
42
fentanyl prime-induced reinstatement of fentanyl-seeking behavior. Data represent mean ± SEM
43
44 active lever responses, with white bars representing baseline active lever responses and blue bars
45
46 representing active lever response on reinstatement test conditions. n = 8; ** p < 0.01
47
48
49 Conclusion
50
51 Via examination of compounds with D3R and D2R binding and functional data in our
52
53 corporate compound file, we identified the unique structural scaffold 1. Through a hybrid
54
55
56 approach, we merged key pharmacophore elements from D3 antagonist 1 and D3 agonist 2 to
57
58
59
60 15
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 16 of 55

1
2
3
yield the novel D3R/D2R antagonist 3. Compound 3 possessed CNS drug-like properties as
4
5
6 defined by ADME attributes and CNS MPO desirability score (≥4/6).36, 37 In addition to good
7
8 physicochemical properties, 3 exhibited low nanomolar affinity for the D3R (D3 Ki = 3.1 nM)
9
10
11
and good subtype selectivity over D2R (D2 Ki = 692 nM). The D3R functional potency of 3 was
12
13 in line with its binding affinity, establishing 3 as an antagonist with a D3R functional Ki of 1.1
14
15 nM. Compound 3 exhibited excellent brain penetration in rats, with equivalent free drug levels
16
17
18 in plasma and brain compartments (Cb,u/Cp,u = 1). Given the higher-than-expected D2R
19
20 occupancy in the ex vivo receptor occupancy assay, we were concerned that significant D2R side
21
22 effects would be observed at low doses. When we examined 3 in models that evaluate D2R-
23
24
25 associated liabilities, we were surprised to discover that high D2R occupancy (86% RO) did not
26
27 result in either the locomotor or catalepsy side effects characteristic of traditional D2R
28
29
antagonists such as haloperidol and eticlopride which induce catalepsy in rodents at D2R
30
31
32 occupancies of >50%. While we do not fully understand why 3 was devoid of these side effects,
33
34 one potential explanation is that very high D3R occupancy counters these responses to moderate
35
36
37 D2 receptor blockade in specific neural pathways. An alternative explanation could be that 3 has
38
39 a unique binding mode in either D3R or D2R that differentially modulates in vivo functional
40
41 signaling, compared to traditional D2R antagonists. Recent reports47, 48
suggest that D3/D2
42
43
44 dimerization is critical to the functional effects of D2R, allowing for the hypothesis that 3 is
45
46 binding in a unique way to this heteromer complex and thus changing the downstream signaling
47
48 events. Lastly, moderate affinity of 3 to 5HT2a could help explain the lack of locomotor or
49
50
51 catalepsy side effects, however, other compounds from this series devoid of 5HT2a activity have
52
53 similar in vivo profiles to 3. Ongoing studies, including the generation of a crystal structure of 3
54
55
56
interacting with either D3R or D2R, may provide insight into this potential unique binding pose,
57
58
59
60 16
ACS Paragon Plus Environment
Page 17 of 55 ACS Chemical Neuroscience

1
2
3
and single molecule fluorescence experiments could be used to determine if binding of 3 alters or
4
5
6 prevents dimerization of dopamine receptors, thereby differentially modulating in vivo
7
8 functional effects. Induced-fit docking49 of 3 into the published D3R X-ray crystal structure
9
10
11
(PDB ID: 3PBL)50 suggests the possibility of two binding modes. Neither of these binding
12
13 modes is fully predictive of the SAR and selectivity in this series, however, nor is it clear what
14
15 could potentially disrupt D3/D2 dimerization from either of these binding poses; the true binding
16
17
18 mode of 3 to D3R or D2R therefore remains unclear.
19
20 Compound 3 is well tolerated in animals at high D3R and D2R occupancy, allowing for
21
22 the full examination of a range of both D3R and D2R occupancy of 3 on cessation of drug-taking
23
24
25 and relapse to drug-seeking. The current study demonstrated a connection between D3R/D2R
26
27 antagonism and attenuation of drug-seeking behavior, with compound 3 decreasing fentanyl self-
28
29
administration (cessation) and robustly attenuating the reinstatement of fentanyl-seeking
30
31
32 behavior under combined cue and prime conditions. Drug-seeking behavior can be characterized
33
34 by the acquisition and maintenance of drug-taking behavior and the relapse to drug-seeking
35
36
37 behavior, consisting of forced abstinence from drug-taking and a reinstatement/relapse event
38
39 triggered by a drug prime or drug cue presentation (Figure 5). In the current model, the proposed
40
41 efficacy of a D3/D2 receptor antagonist is highlighted by a potential benefit in both cessation of
42
43
44 a drug-taking (as evidenced by the decrease in fentanyl self-administration) as well as a
45
46 prevention of relapse (as evidenced by a decrease in fentanyl reinstatement) following
47
48 pretreatment with compound 3. Considering the increased efficacy observed following
49
50
51 administration of 32 mg/kg relative to 10 mg/kg, the current results suggest that addition of D2R
52
53 antagonism is beneficial in driving the efficacy of reducing drug-seeking behavior compared to
54
55
56
D3R antagonism alone. A potential alternative interpretation of these results is that compound 3
57
58
59
60 17
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 18 of 55

1
2
3
is inducing a non-specific effect on motor function or general motivation, resulting in decreased
4
5
6 fentanyl self-administration and reinstatement, rather than specifically attenuating fentanyl-
7
8 seeking. However, results from the current study demonstrate that compound 3 does not alter
9
10
11
locomotor activity up to 32 mg/kg. Furthermore, the effect of compound 3 in an intracranial self-
12
13 stimulation (ICSS) model designed to understand potential anhedonic effects of compounds
14
15 revealed no change in ICSS behavior at efficacious doses, which suggests a specific effect on
16
17
18 fentanyl-associated motivation.. Thus, given that the rat reinstatement model of drug relapse
19
20 possesses high predictive validity for efficacy in substance use disorders,51 these data suggest
21
22 that dual D3/D2 receptor antagonists warrant further investigation in clinical studies examining
23
24
25 both the drug cessation and relapse phases of substance use disorder treatment.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46 Figure 5. General depiction of the experiment. Selective D3/D2 antagonist 3 attenuates
47
48
fentanyl-seeking behavior in models of both cessation of fentanyl self-administration and
49
50
51 reinstatement (as a model of relapse).
52
53 In summary, compound 3 is an excellent preclinical tool to functionally probe the
54
55
56
biology of D3R and D2R in vivo. While it does not meet our guidelines [Ceff ≤ 250 nM (total
57
58
59
60 18
ACS Paragon Plus Environment
Page 19 of 55 ACS Chemical Neuroscience

1
2
3
drug)] to move into preclinical in vivo safety testing,52 the discovery of 3 is a significant
4
5
6 advancement in the development of D3R/D2R antagonists for the treatment of drug addiction.27
7
8 The unique in vivo profile of 3 makes it a promising candidate for preclinical testing of D3/D2
9
10
11
antagonism for multiple substance use disorders.
12
13 METHODS
14
15 Physicochemical Properties and Data Analysis. For the work herein, calculated CNS MPO
16
17
18 desirability scores were obtained using the published algorithm,37 and calculated
19
20 physicochemical properties were obtained using standard commercial packages: Biobyte for
21
22 ClogP calculations; for calculation of TPSA, see Ertl.53 Statistical analyses were carried out
23
24
25 using SigmaPlot version 11.0, from Systat Software, Inc., San Jose California USA,
26
27 www.sigmaplot.com .
28
29
30 ADME Data. Data on the following in vitro ADME properties were obtained from the company
31
32
33 database. All assays were performed via reported methods as described previously for: (a)
34
35 passive apparent permeability, Papp, assay;54 (b) P-glycoprotein (P-gp) efflux liability assay;54 (c)
36
37 metabolic stability, expressed as unbound intrinsic clearance (CLint,u);55, 56 and (d) plasma protein
38
39
40 binding, Fu mouse and human.
41
42 Chemistry
43
44
45
The synthesis of 3 begins with commercially available starting materials, as illustrated in Scheme
46
47 1. Conversion of 3-methyl-4-nitrobenzoic acid (4) to the corresponding acid chloride with
48
49 thionyl chloride and catalytic DMF gave 3-methyl-4-nitrobenzoyl chloride (5) in 91% crude
50
51
52 yield (500 g). Treatment of crude 5 in acetonitrile/ THF with (trimethylsilyl)diazomethane
53
54 generated the α-diazoketone in situ, which by addition of 40% aq. HBr was converted to 2-
55
56 bromo-1-(3-methyl-4-nitrophenyl)ethanone (6) in 93% isolated yield (600 g). Morpholin-2-ol 7
57
58
59
60 19
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 20 of 55

1
2
3
was prepared in 52% isolated yield (320 g) via bromide displacement of α-bromoketone 6 with
4
5
6 2-(ethylamino)ethanol and subsequent cyclization to the lactol. Hydrodehydroxylation of 7 with
7
8 TFA/triethylsilane in CH2Cl2 yielded 4-ethyl-2-(3-methyl-4-nitrophenyl)morpholine (8) (200 g)
9
10
11
in 66% isolated yield. Reduction of the nitro group in 8 to the corresponding amine was
12
13 accomplished via Raney nickel hydrogenation in MeOH to give 4-(4-ethylmorpholin-2-yl)-2-
14
15 methylaniline (9) in 63% isolated yield (110 g). Completing the sequence, racemic sulfonamide
16
17
18 10 was prepared in 99% isolated yield (20 g) by treatment of aniline 9 with commercially
19
20 available 4-isopropylbenzene-1-sulfonyl chloride in pyridine. Preparative chiral chromatography
21
22 of this sulfonamide, followed by HCl salt formation, afforded the enantiomerically pure title
23
24
25 compound (-)-N-{4-[(2S)-4-ethylmorpholin-2-yl]-2-methylphenyl}-4-(propan-2-
26
27 yl)benzenesulfonamide (3).
28
29
Scheme 1: Synthesis of D3R/D2R antagonist 3.
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 20
ACS Paragon Plus Environment
Page 21 of 55 ACS Chemical Neuroscience

1
2
3 O O O
4
a b Br
5 OH Cl
6
7 O2N O2N O2N
8
9 4 5 6
10
11
12
13 N N N
14 HO
c d e
15
16 O O O
17
O2N O2N H2N
18
19
20 7 8 9
21
22
23 N N
24
25
26 O O O O O O
27 f S g S
28 N N
H H HCl
29
30
31 10 3 (PF-4363467)
32
33
34
Reagents and conditions: a) thionyl chloride, DMF; b) (trimethylsilyl)diazomethane, ACN/THF,
35
36
37 then 40% HBr; c) 2-(ethylamino)ethanol, Et3N, CH2Cl2; d) triethylsilane, TFA, CH2Cl2; e) H2,
38
39 Raney Ni, MeOH; f) 4-isopropylbenzene-1-sulfonyl chloride, pyridine; g) chiral
40
41
42
chromatographic separation, then HCl/EtOAc/Et2O.
43
44 Chemistry Experimental Section. General Information. All solvents and reagents were
45
46 obtained from commercial sources and were used as received. All reactions were followed by
47
48
49 TLC (TLC plates F254, Merck) or LCMS (liquid chromatography-mass spectrometry) analysis.
50
13
51 Proton and C NMR spectra were obtained using deuterated solvents on a Varian 400 MHz
52
53 instrument. All proton shifts are reported in δ units (ppm) downfield of TMS (tetramethylsilane)
54
55
56 and were measured relative to the signals for chloroform (7.27 ppm) and methanol (3.31 ppm).
57
58
59
60 21
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 22 of 55

1
2
3
All 13C shifts are reported in δ units (ppm) relative to the signals for chloroform (77.0 ppm) and
4
5
6 methanol (49.1 ppm) with 1H-decoupled observation. NMR abbreviations are as follows: br,
7
8 broadened; s, singlet; d, doublet; t, triplet; q, quartet; p, pentuplet; m, multiplet; dd, doublet of
9
10
11
doublets; dddd, doublet of doublet of doublet of doublets; sept, septuplet; tt, triplet of triplets.
12
13 Analytical analyses by UPLC were performed on a Waters Acquity system with PDA detection
14
15 (UV 210 nm) at 45 ºC, flow rate 0.5 mL/min, with a 95/5 buffer/acetonitrile (0 to 7.55 min),
16
17
18 10/90 buffer/acetonitrile (7.55 to 7.85 min) using the following columns and buffers: Waters
19
20 BEH C8 column (2.1 x 100 mm, 1.7 µm) with 50 mM sodium perchlorate/0.1% phosphoric acid
21
22 or 10 mM ammonium bicarbonate as buffer; Waters BEH RP C18 column (2.1 x 100 mm, 1.7
23
24
25 µm) or Waters HSS T3 (2.1 x 100 mm, 1.8 µm) column with 0.1% methanesulfonic acid buffer.
26
27 All melting points are uncorrected. Elemental analyses were performed by QTI Development.
28
29
Mass spectra were recorded on a Micromass ADM atmospheric pressure chemical ionization
30
31
32 instrument (MS, APCI). High resolution mass spectra were obtained on an Agilent LC-MS TOF
33
34 equipped with a Zorbax Eclipse column (50 mm x 4.6 mm, 1.8 µm XDB-C18) using 0.1%
35
36
37 aqueous formic acid as mobile phase A1 and acetonitrile containing 0.1% formic acid as mobile
38
39 phase B1. Column chromatography was carried out on silica gel 60 (32–60 mesh, 60 Ǻ) or on
40
41 pre-packed Biotage™ columns. The purities of final compound 3 and 4–10 as measured by
42
43
44 UPLC were found to be above 95%.
45
46
47
48
49
50
51
52
53 Preparation of 2-bromo-1-(3-methyl-4-nitrophenyl)ethanone (6). A mixture of 3-methyl-4-
54
55 nitrobenzoic acid (4) (500.0 g, 2.8 mol), DMF (10 mL) and thionyl chloride (2.0 L, 26.8 mol)
56
57 was refluxed for 10 h. The thionyl chloride was removed in vacuo and the residue was re-
58
59
60 22
ACS Paragon Plus Environment
Page 23 of 55 ACS Chemical Neuroscience

1
2
3
concentrated from heptane three times to yield 3-methyl-4-nitrobenzoyl chloride (5) (500.0 g,
4
5
6 91%) as a brown oil which was used without purification. For analytical purposes, a portion of
7
8 this material was recrystallized from heptane to give a light yellow solid: 1H NMR (400 MHz,
9
10
11 CDCl3) δ ppm 8.05 - 8.08 (m, 2H), 7.98 (d, J = 8.8 Hz, 1H), 2.63 (s, 3H). The crude acid
12
13 chloride (500.0 g, 2.5 mol) was dissolved in CH3CN/THF (1:1, 8.0 L) and cooled to 0 ºC. A
14
15
solution of 2 M (trimethylsilyl)diazomethane in hexanes (1.5 L, 3.0 mol) was added drop-wise.
16
17
18 The resulting mixture was warmed to rt and stirred for 1 h. The reaction was re-cooled to 0 ºC
19
20 and 40% aq. HBr (1.0 kg, 5.0 mol) was added drop-wise. After stirring at rt for 30 min, the
21
22
23
reaction was concentrated in vacuo to yield 2-bromo-1-(3-methyl-4-nitrophenyl)ethanone (6)
24
25 (600.0 g, 93.0%) as a brown solid. 1H NMR (400 MHz, CDCl3) δ ppm 7.89 - 8.05 (m, 3H), 4.45
26
27 13
28
(s, 2H), 2.66 (s, 3H); C NMR (100 MHz, CDCl3) δ ppm 190.2, 152.3, 137.0, 134.2, 133.5,
29
30 127.6, 125.2, 30.5, 20.3; HRMS m/z (+TOF) [M + H]+ calcd. for C9H8NO3S, 257.9760; found,
31
32 257.9753.
33
34
35
36
37
38
39
40
41
42
43
Preparation of 4-ethyl-2-(3-methyl-4-nitrophenyl)morpholin-2-ol (7). 2-Bromo-1-(3-methyl-
44
45
46 4-nitrophenyl)ethanone (6) (600.0 g, 2.3 mol), 2-(ethylamino)ethanol (228 mL, 2.6 mol) and
47
48 Et3N (259 g, 2.6 mol) in dichloromethane (3.5 L) were stirred at rt for 16 h. The reaction mixture
49
50
51
was washed with brine (1.5 L), and the aqueous layer was extracted with ethyl acetate (3×3 L).
52
53 The combined organic layers were dried over MgSO4 and concentrated. The residue was purified
54
55 by silica gel chromatography (pet. ether/EtOAc 10:1) to give 4-ethyl-2-(3-methyl-4-
56
57
58 nitrophenyl)morpholin-2-ol (7) (320 g, 52.0%) as a brown oil. 1H NMR (400 MHz, CDCl3) δ
59
60 23
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 24 of 55

1
2
3
ppm 7.96 (d, J = 8.4 Hz, 1H), 7.57 - 7.64 (m, 2H), 4.95 (br s, 1H), 4.18 (ddd, J = 11.9, 11.9, 2.9
4
5
6 Hz, 1H), 3.86 (dd, J = 11.6, 3.6 Hz, 1H), 2.76 - 2.85 (m, 2H), 2.61 (s, 3H), 2.41 - 2.53 (m, 2H),
7
8 2.32 (ddd, J = 11.7, 11.7, 3.8 Hz, 1H), 2.19 (d, J = 10.9 Hz, 1H), 1.09 (t, J = 7.2 Hz, 3H); 13C
9
10
11
NMR (100 MHz, CDCl3) δ ppm 149.1, 147.0, 133.6, 130.6, 124.8, 94.4, 63.4, 60.9, 52.3, 52.0,
12
13 20.8, 11.9; HRMS m/z (+TOF) [M + H]+ calcd. for C13H18N2O4, 267.1339; found, 267.1331.
14
15
16
17
18
19
20
21
22
23
24 Preparation of 4-ethyl-2-(3-methyl-4-nitrophenyl)morpholine (8). A mixture of 4-ethyl-2-(3-
25
26
methyl-4-nitrophenyl)morpholin-2-ol (7) (320 g, 1.2 mol), triethylsilane (720 mL, 6.0 mol) and
27
28
29 TFA (3.2 L) in CH2Cl2 (6.0 L) was stirred at rt for 72 h. Saturated aq. K2CO3 was added until the
30
31 pH was 10. The organic layer was separated and the aqueous layer was extracted with ethyl
32
33
34
acetate (3x1.5 L). The combined organic layers were dried over MgSO4 and concentrated, and
35
36 the residue was purified by silica gel chromatography (petroleum ether) to yield 4-ethyl-2-(3-
37
38 methyl-4-nitrophenyl)morpholine (8) (200 g, 66.0%) as a brown oil. This material was suitable
39
40
41 for use without further purification. For analytical purposes, a portion was converted to the
42
43 hydrochloride salt and recrystallized from CHCl3/ EtOAc, yielding a light orange-tinged solid:
44
45 1
H NMR (400 MHz, CDCl3) δ ppm 13.54 (br s, 1H), 7.99 (d, J = 8.6 Hz, 1H), 7.36 - 7.42 (m,
46
47
48 2H), 5.44 (d, J = 9.4 Hz, 1H), 4.53 - 4.61 (m, 1H), 4.23 (dd, J = 13.1, 3.1 Hz, 1H), 3.57 (t, J =
49
50 10.9 Hz, 2H), 3.10 - 3.18 (m, 2H), 2.86 - 2.98 (m, 1H), 2.56 - 2.65 (m, 4H), 1.54 (t, J = 7.2 Hz,
51
52 13
53
3H); C NMR (100 MHz, CDCl3) δ ppm 149.2, 141.8, 134.3, 130.3, 125.3, 124.3, 73.8, 63.6,
54
55 56.2, 52.9, 50.4, 20.4, 8.9; HRMS m/z (+TOF) [M + H]+ calcd. for C13H18N2O3, 251.1390;
56
57 found, 251.1387.
58
59
60 24
ACS Paragon Plus Environment
Page 25 of 55 ACS Chemical Neuroscience

1
2
3
4
5
6
7
8
9
10
11
12
Preparation of 4-(4-ethylmorpholin-2-yl)-2-methylaniline (9). A mixture of 4-ethyl-2-(3-
13
14
15 methyl-4-nitrophenyl)morpholine (8) (200 g, 0.80 mol) and Raney nickel (20 g) in MeOH (5.0
16
17 L) was stirred under 1 atm of H2 at rt for 24 h. The mixture was filtered and washed with MeOH
18
19
20 (3x100 mL). The filtrate was concentrated and the residue was purified by chromatography on
21
22 silica gel (petroleum ether/EtOAc 5:1) to give 4-(4-ethylmorpholin-2-yl)-2-methylaniline (9)
23
24 (110 g, 62%) as a brown oil: 1H NMR (400 MHz, CDCl3) δ ppm 7.07 (s, 1H), 7.03 (dd, J = 8.1,
25
26
27 1.9 Hz, 1H), 6.64 (d, J = 8.0 Hz, 1H), 4.45 (dd, J = 10.4, 2.3 Hz, 1H), 4.02 (ddd, J = 11.4, 3.4,
28
29 1.4 Hz, 1H), 3.83 (ddd, J = 11.5, 11.5, 2.3 Hz, 1H), 3.59 (br s, 2H), 2.91 (ddd, J = 11.5, 2.0, 2.0
30
31 Hz, 1H), 2.81 (dd, J = 11.3, 1.8 Hz, 1H), 2.45 (q, J = 7.1 Hz, 2H), 2.15 - 2.23 (m, 4H), 2.05 (dd,
32
33 13
34 J = 11.4, 10.4 Hz, 1H), 1.11 (t, J = 7.2 Hz, 3H); C NMR (100 MHz, CDCl3) δ ppm 144.5,
35
36 130.8, 128.7, 125.3, 122.4, 114.9, 78.5, 67.4, 60.5, 52.9, 52.8, 17.6, 12.0; HRMS m/z (+TOF) [M
37
38
39
+ H]+ calcd for C13H20N2O; 221.1648, found, 221.1646.
40
41
42 N O O
43 N N
44 O O O O O
f S g S
45 N N
46 H2N H H HCl
47
48 9
10 3 (PF-4363467)
49
50
51
52
Preparation of (-)-N-{4-[(2S)-4-ethylmorpholin-2-yl]-2-methylphenyl}-4-(propan-2-
53
54 yl)benzenesulfonamide (3). 4-Isopropylbenzene-1-sulfonyl chloride (8.9 mL, 49.9 mmol) was
55
56 added to an ice cooled solution of 4-(4-ethylmorpholin-2-yl)-2-methylaniline (9) (11.0 g, 49.9
57
58
59
60 25
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 26 of 55

1
2
3
mmol) in pyridine (106 mL) and the resulting reaction mixture was slowly warmed to rt. After
4
5
6 stirring overnight, the reaction was concentrated and the residual oil was dissolved in CH2Cl2
7
8 and washed with 1 N NaOH, dried over MgSO4, and concentrated to yield 28 g of crude N-[4-(4-
9
10
11
ethylmorpholin-2-yl)-2-methylphenyl]-4-(propan-2-yl)benzenesulfonamide. Chromatography on
12
13 a 350 g ISCO cartridge using EtOAc and 5% EtOH / EtOAc gave 20 g (>99%) of purified
14
15 material as an amorphous glass: 1H NMR (CDCl3) δ 7.66 (d, J = 8.5 Hz, 2H), 7.28 - 7.23 (m,
16
17
18 3H), 7.12 - 7.08 (m, 2H), 4.49 (dd, J = 10.3, 1.9 Hz, 1H), 4.5 - 3.95 (m, 1H), 3.82 (ddd, J = 11.6,
19
20 11.2, 2.5 Hz, 1H), 2.94 - 2.78 (m, 3H), 2.50 - 2.42 (m, 2H), 2.24 - 2.15 (m, 1H), 2.02 - 1.96 (m,
21
22 1H), 1.97 (s, 3H), 1.26 - 1.20 (m, 6H), 1.05 (t, J = 9.5 Hz, 3H); 13
C NMR (CDCl3) δ 154.7,
23
24
25 138.4, 134.2, 131.6, 129.2, 128.7, 127.5, 127.3, 125.0, 124.4, 67.3, 60.2, 52.8, 52.7, 34.5, 34.3,
26
27 23.8, 17.8, 11.8; LRMS m/z calcd for C22H30N2O3S, 402.56; found, 403.3 [M + H]+, (APCI); Rf
28
29
= 0.4 (10% EtOH / EtOAc). The enantiomers of N-[4-(4-ethylmorpholin-2-yl)-2-methylphenyl]-
30
31
32 4-(propan-2-yl)benzenesulfonamide were separated by preparative chiral chromatography (10
33
34 cm x 50 cm Chiralpak AD column using 75/25 heptane/EtOH as mobile phase with a flow rate
35
36
37 of 500 mL/ min) to yield 7.7 g of enantiomer #1 (3, r.t. = 8.49 min; [α]D = −3.025, c = 1.20 in
38
39 MeOH) and 7.7 g of enantiomer #2 (PF-04363476; r.t. 13.43 min). The HCl salt of enantiomer
40
41 #1 (3) was prepared in EtOAc (150 mL) by adding 1 eq. of 2 N HCl/ether followed by ether (150
42
43
44 mL). The resulting slurry was stirred overnight, filtered, dried under N2 and evacuated overnight
45
46 to yield 7.7 g of the mono HCl salt of 3 as a light yellow-tinted solid: mp = 218–220 ºC; 1H
47
48 NMR (400 MHz, CDCl3) δ ppm 13.05 (br s, 1H), 7.67 (d, J = 8.4 Hz, 2H), 7.28 - 7.30 (m, 3H),
49
50
51 7.10 - 7.12 (m, 2H), 7.06 (s, 1H), 5.20 (dd, J = 10.7, 1.6 Hz, 1H), 4.45 - 4.54 (m, 1H), 4.15 (dd, J
52
53 = 13.0, 3.4 Hz, 1H), 3.48 - 3.56 (m, 2H), 3.09 - 3.17 (m, 2H), 2.83 - 2.99 (m, 2H), 2.60 - 2.68
54
55
56
(m, 1H), 2.03 (s, 3H), 1.50 (t, J = 7.3 Hz, 3H), 1.24 (d, J = 6.9 Hz, 6H); 13C NMR (100 MHz,
57
58
59
60 26
ACS Paragon Plus Environment
Page 27 of 55 ACS Chemical Neuroscience

1
2
3
CDCl3) δ 154.7, 137.0, 135.3, 134.3, 132.1, 128.7, 127.2, 127.1, 124.5, 124.4, 74.4, 63.6, 56.3,
4
5
6 52.8, 50.5, 34.1, 23.6, 17.7, 8.8; Anal. calcd. for C22H30N2O3S•HCl: C, 60.19; H, 7.12; N, 6.38;
7
8 Found: C, 60.32; H, 7.37, N, 6.26. The absolute stereochemistry was determined using small
9
10
11
molecule crystallography. Pertinent crystal, data collection and refinement information is
12
13 summarized in the small molecule report, supplementary material.
14
15 In vitro Materials.
16
17
18 Binding Assay.
19
20 Binding assays were performed using over-expressing human D2 and D3 CHO cell lines. To
21
22 determine basic assay parameters, ligand concentrations were determined from saturation
23
24
25 binding studies, where the Kd using [3H]-spiperone (PerkinElmer NET1187250UC) for D2
26
27 binding was determined to be 1.6 nM and a Kd of 1.4 nM was determined for hD3 using [3H]-7-
28
29
OH-DPAT (PerkinElmer NET1169250UC). From tissue concentration curve studies, the
30
31
32 optimal amount of tissue was determined to be 4 mg/mL for hD2 and 7 mg/mL for hD3 per 96-
33
34 well plate. These ligand and tissue concentrations were used in time-course studies to determine
35
36
37 linearity and equilibrium conditions for binding. It was found that binding was at equilibrium
38
39 with the specified amount of tissue in 20 minutes at 37 °C for both receptors. From these
40
41 parameters, Ki values were determined by homogenizing the specified amount of tissue for each
42
43
44 receptor in 50 mM Tris (pH 7.4 at 4 °C) containing 2.0 mM MgCl2, and spinning in a centrifuge
45
46 at 40,000 x g for 10 minutes. The pellet was resuspended in either the D2 assay buffer [50 mM
47
48
49 Tris (pH 7.4 @ 37 °C), 100 mM NaCl and 1 mM MgCl2] or the D3 assay buffer [50 mM Tris
50
51 (pH 7.4 at 37 °C), 120 mM NaCl, 5 mM MgCl2, 5 mM KCl and 2 mM CaCl2]. Incubations were
52
53
initiated by the addition of 200 µL of tissue to 96-well plates containing test drugs (2.5 µL) and
54
55
56 1.6 nM [3H]-spiperone for D2 or 1.5 nM of [3H]-7-OH-DPAT for D3 (both 50 µL) for a final
57
58
59
60 27
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 28 of 55

1
2
3
4
assay volume of 250 µL. Non-specific binding was determined by radioligand binding in the
5
6 presence of a saturating concentration of haloperidol. After the 20 minute incubation period at
7
8 37 °C, assay samples were rapidly filtered through Unifilter-96 GF/B PEI-coated filter plates and
9
10
11 rinsed with ice-cold 50 mM Tris buffer (pH 7.4 at 4 °C). Membrane radioligand levels were
12
13 determined by liquid scintillation counting of the filter plates in 50 µL EcoLume. The IC50 value
14
15
16 (concentration at which 50% inhibition of specific binding occurs) was calculated by linear
17
18 regression of the concentration-response data in Activity Base (IDBS). Ki values were calculated
19
20 according to the Cheng-Prusoff equation.
21
22
23 D2 Functional Assay. cAMP Accumulation Assay in SH-SY5Y hD2L Cells:
24
25 Human D2L-transfected SH-SY5Y cells were grown at 37 °C and 5% CO2 in DMEM/F12
26
27
medium supplemented with 10% dialyzed FBS and 100 µg/mL G418. The cells were frozen as
28
29
30 assay-ready vials and thawed immediately prior to the assay. On the day of the experiment,
31
32 compounds dissolved in DMSO were diluted in pH 7.4 assay buffer containing PBS, 5 µM
33
34
35
HEPES, and 100 µM IBMX (final concentrations). The vial of frozen cells was rapidly thawed
36
37 in a 37 °C water bath. Cells were washed with PBS and centrifuged to obtain a pellet. The
38
39 pellet was resuspended in warm PBS and the cells were allowed to sit for 5 minutes, prior to
40
41
42 counting and dilution. The reaction was initiated by addition of the cells into 384-well plates
43
44 containing test compounds; the final number of cells used in the assay was 10000–15000 cells
45
46 per well. Cells and compounds were pre-incubated together for 10 minutes at 37 °C. NKH477
47
48
49 was then added to the plates at a final concentration of 3 µM to activate adenylyl cyclase and
50
51 increase basal levels of cAMP. For testing compounds in agonist mode, assay buffer was added
52
53
to the plate; in antagonist mode, a D2 agonist was added to produce an 80% agonist effect
54
55
56 (quinelorane, 32 nM). Plates were then returned to the 37 °C incubator and incubated for 15
57
58
59
60 28
ACS Paragon Plus Environment
Page 29 of 55 ACS Chemical Neuroscience

1
2
3
minutes at 37 °C. To stop the reaction, Cisbio cAMP Dynamic 2 screening kit reagents (cat#
4
5
6 62AM4PEB) were added to the plate. Homogenous time-resolved fluorescence-based cAMP
7
8 (Schering) detection was determined according to the manufacturer's instruction. A Wallac
9
10
11
EnVision was used to measure HTRF (excitation 320 nm, emission 665 nm/620 nm, delay time
12
13 50 µs, window time 400 µs). Data was analyzed based on the ratio of fluorescence intensity of
14
15 each well at 620 nm and 665 nm followed by cAMP quantification using a cAMP standard
16
17
18 curve. Agonist activity of a compound was measured by its ability to decrease the cAMP
19
20 production elicited by NKH477. The agonist effect was normalized to the % effect of 1 µM
21
22 quinelorane. Antagonist activity was measured by a compound’s ability to reverse the agonist-
23
24
25 induced decrease of cAMP (in the presence of NHK477). The antagonist effect was normalized
26
27 to the % effect of NHK477 alone.
28
29
30
D3 Functional Assay. GTP-γγ-S Binding Assay in SH-SY5Y hD3 Cells: SH-SY5Y hD3 cells
31
32 were grown at 37 °C and 5% CO2 in DMEM/F12 supplemented with 10% dialyzed FBS and 100
33
34 µg/mL G418. Cell paste was frozen in a pellet and stored at −80 °C until ready for use. The day
35
36
37 of the assay, cell paste was homogenized using a glass-Teflon homogenizer in buffer containing
38
39 50 mM HEPES, 1 mM EDTA and protease inhibitors (pH 7.4, 4 °C). The membranes were
40
41
centrifuged at 40000 x g for 10 minutes. The pellet was resuspended in pH 7.4 assay buffer (20
42
43
44 mM HEPES, 100 mM NaCl, 30 mM MgCl2). Saponin and guanosine diphosphate (44 µg/mL
45
46 and 30 µM final assay concentrations, respectively) were added to the membrane solution. The
47
48
49
membranes were incubated at 4 °C for 1 hour to promote GDP binding. Membranes were then
50
51 added to test compounds (in DMSO) in the assay plate and incubated for 30 minutes at 30 °C.
52
53 [35S] GTP-γ-S was added to the assay plate (final assay concentration of 1 nM) and incubated
54
55
56 another 30 minutes at 30 °C. To stop the reaction, assay samples were rapidly filtered through
57
58
59
60 29
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 30 of 55

1
2
3
GF/B fired UniFilter plates (PerkinElmer) and rinsed with ice-cold 50 mM Tris buffer (pH 7.4).
4
5
6 Filter plates were allowed to dry overnight. The next day, membrane-bound [35S] GTP-γ-S
7
8 levels were determined by liquid scintillation counting of the filter plates in EcoLume
9
10
11 scintillation fluid. A background control of 10 µM cold guanosine triphosphate was subtracted
12
13 from raw assay data prior to data analysis. Percent effect in the agonist assay was calculated by
14
15
normalizing compound concentration data to 1 µM quinelorane (full agonist). Activity in the
16
17
18 antagonist assay was determined by a compound’s ability to reverse an 80% agonist effect (10
19
20 nM quinelorane). Percent effect in the antagonist assay was calculated by normalizing data back
21
22
23 to basal levels of [35S] GTP-γ-S binding. IC50 values (concentration at which 50% inhibition of
24
25 specific binding occurs) were calculated by curve-fitting of the concentration-response data.
26
27
Ex Vivo Methods and Materials; Receptor Occupancy.
28
29
Frozen −80 oC male Sprague Dawley rat brain slices (20 µm) on gelatin-coated slides were
30
31
32 thawed in a humidified chamber at 37 oC for 15 min and subsequently incubated for 15 min at
33
34 room temperature with 0.3 nM [125I]iodosulpiride in buffer (50 mM Tris-HCl, 120 mM NaCl, 5
35
36
37
mM KCl, 2 mM CaCl2, 1 mM MgCl2, 1 mM L-ascorbic acid pH 7.4). The reaction was stopped
38
39 by 2 min rinses (2x) in cold buffer followed by a 30 second rinse in 4 oC deionized water. Slices
40
41 were air-dried for 15 to 20 min under a stream of cool air and further dried for 30 min at room
42
43
44 temperature under vacuum in a desiccant-containing vacuum bell. Dried slices were placed in x-
45
46 ray cassettes and exposed to hyperfilm (Kodak BioMax MR) for 5 days with cerebellum slices
47
48 for D3 quantitation or 36 hours with dorsal striatum slices for D2 quantitation. Specific binding
49
50
51 was determined by calculation of total binding minus non-specific binding observed in the
52
53 presence of 1 µM haloperidol. Images were semi-quantitatively evaluated using standard curves
54
55
generated with [14C]microscales.
56
57
58
59
60 30
ACS Paragon Plus Environment
Page 31 of 55 ACS Chemical Neuroscience

1
2
3
In Vivo Methods and Materials: Effects of Administration of Compound 3 on Locomotor
4
5
6 Activity and Catalepsy.
7
8
9 General Experimental. This study took place over 6 days with 8 animals tested each day.
10
11 Locomotor activity was monitored for 3 hours, with catalepsy measurements taken at the
12
13
14 beginning of each experiment and then at 30 minutes, 1 hour, 2 hours and 3 hours.
15
16 Animals and Conditions. Male Sprague Dawley rats (220–260 g) from Charles River were
17
18
19
pair-housed with free access to water and food. Standard light/dark (0600 h/1800 h), temperature
20
21 & humidity conditions were controlled and monitored.
22
23 Drugs. Compound 3 was dissolved in 10% 2-hydroxypropyl-beta-cyclodextrin at concentrations
24
25
26 of 1.0, 3.2, 10.0 and 32.0 mg/mL, and dosed at 1.0, 3.2, 10.0 and 32.0 mg/kg. Haloperidol
27
28 (Sigma) was dissolved in 1% acetic acid. The vehicle group received the same 10% 2-
29
30 hydroxypropyl-beta-cyclodextrin that was used for dissolution of 3. The route of administration
31
32
33 was s.c., with an injection volume of 1.0 mL/kg; the pretreatment time was zero. All drug
34
35 concentrations and doses refer to the active moiety.
36
37 Spontaneous Locomotor Activity Procedure/Chamber Apparatus. Procedure: Animals
38
39
40 were brought in their home cages into the procedural room at least 30 minutes prior to testing
41
42 and allowed to acclimate to the test room. Animals were assigned to treatment groups and
43
44
45
locomotor boxes. On days 1–6, animals (1–2 per treatment group) were given a baseline
46
47 catalepsy assessment, then given a s.c. injection of vehicle or drug and placed in Amlogger
48
49 locomotor monitoring equipment for three hours of activity monitoring. Dependent measures
50
51
52 included total activity and total distance travelled.
53
54 Inside the Amlogger equipment the rat was placed in a rat-sized housing box with a
55
56 microisolator top located within a Linton Instrumentation (Norfolk, England) AM548 / AM524
57
58
59
60 31
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 32 of 55

1
2
3
IR Locomotor Activity Monitor. Three hours of activity were monitored by 24 infrared beams
4
5
6 arranged in an 8 x 16, 1” (25.4 mm) pitched grid. Data were reduced by AMON Control and
7
8 Data Logging Windows Software. The AMON software runs on Windows PCs, and was used to
9
10
11
configure the experimental protocol and record activity data via a standard RS232 serial port.
12
13 The AMON software discriminates between different types and speeds of movement. As such,
14
15 the data that were logged provide a detailed breakdown of the various categories of activity. This
16
17
18 software has the ability to start cages independently of one another and to review files during and
19
20 after data acquisition. Data were output to an Excel spreadsheet. The software includes a self-
21
22 test utility which verifies that all IR beams are functioning correctly.
23
24
25 Catalepsy Testing. Apparatus and procedure: Rats from the spontaneous locomotor activity
26
27 procedure were used. At 0 (pre-injection) 30, 60, 120, 180 min after injection, the rats were
28
29
removed from the Amlogger equipment and tested for catalepsy. Testing was accomplished by
30
31
32 placing each rat in an upright position with its forepaws resting on a horizontal bar (9 mm in
33
34 diameter) suspended at a height of 10 cm; the latency to removal of forepaws and reversion to a
35
36
37 “normal” position was recorded. The maximum time limit was set at 90 seconds. After
38
39 assessment of catalepsy, the rats were returned to the Amlogger equipment and activity was
40
41 monitored until the next catalepsy measurement, over a total of 3 hours of testing.
42
43
44 Data Analysis. Data from 6 groups of 8 rats were collected and analyzed by calculating the total
45
46 activity for each parameter, in selected time periods of testing. Locomotor comparisons were
47
48 made through multiple unpaired t-tests with a significance level set at 95%, and corrected for
49
50
51 multiple comparisons using the Holm-Sidak method for each dose against vehicle at each time
52
53 block. Catalepsy comparisons were made through a repeated measures ANOVA test with a
54
55
56
significance level set at 95% and followed by a Dunnett’s post-test.
57
58
59
60 32
ACS Paragon Plus Environment
Page 33 of 55 ACS Chemical Neuroscience

1
2
3
In Vivo Methods and Materials: Effects of Administration of Compound 3 on Prolactin
4
5
6 Release.
7
8 Animals. Sprague Dawley rats from Charles River (225–250 grams at delivery) were pair-
9
10
11
housed in vent-rack cages and allowed free access to food and water. They were maintained on a
12
13 12 hour light/dark (0600 h/1800 h) cycle in environmentally controlled quarters. Rats were
14
15 given one week to acclimate to the conditions prior to study initiation.
16
17
18 Tissue Harvest. Rats were habituated to the room 60 minutes prior to dosing. Subjects were
19
20 dosed in 1 mL/kg volumes with 3 (0.1, 1.0, 10 mg/kg; s.c.), corrected for the weight of the salt or
21
22 vehicle one hour prior to trunk blood collection. Trunk blood was obtained via decapitation
23
24
25 without anesthesia and stored in EDTA-containing tubes from Becton, Dickinson and Company.
26
27 Blood was spun down at 4000 RPM for 5 minutes. Plasma was stored at −20 °C.
28
29
ELISA. ELISA kits were purchased from ALPCO (Product number 55-PRLRT-E01) and the kit
30
31
32 directions were precisely followed except that the standard curve was extended down to 0.625
33
34 ng/mL. Standards and individual subject plasma samples were run in duplicate. Plates were
35
36
37 washed on an EL 406 washer dispenser (BioTek) and read on a SpectraMax Plus (Molecular
38
39 Devices) using SoftMax Pro 4.8 software.
40
41 Data Analysis. Data were organized in Excel and the standard curve was calculated using
42
43
44 GraphPad ver. 6 software. Individual concentrations were interpolated from this curve using
45
46 GraphPad. Groups were compared and p-values were calculated within GraphPad using a one-
47
48 way ANOVA. The ANOVA revealed significantly different standard deviations; therefore the
49
50
51 data were logarithmically (base 10) transformed within GraphPad with the equation [y=log(y)].
52
53 Data were then compared using a one-way ANOVA followed by a Dunnett’s post-test with a
54
55
56
significance value of 0.05.
57
58
59
60 33
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 34 of 55

1
2
3
In Vivo Methods and Materials: Effects of Administration of (3) on cessation of fentanyl
4
5
6 self-administration and reinstatement of fentanyl seeking.
7
8 Drugs. Fentanyl citrate salt (Sigma-Aldrich, St. Louis, MO) was dissolved in 0.9% saline
9
10
11
solution. Compound 3 was formulated in 10% (2-hydroxypropyl)-beta-cyclodextrin (HPβCD).
12
13 All drug concentrations and doses refer to the active moiety.
14
15 Surgery. Sprague Dawley rats (Charles River Laboratories, North Carolina, USA) were
16
17
18 implanted with jugular vein catheters (JVC) with modifications to previously described
19
20 methods.57 Briefly, on the day of surgery, anti-inflammatory and antibiotic treatment was
21
22 provided (carprofen: 5 mg/kg and enrofloxacin: 2.25 mg/kg; s.c.). After the rats had been
23
24
25 anaesthetised with isoflurane (2.5–3.5% in 100% oxygen), a JVC (modified IVSAp40 catheter:
26
27 CamCaths, Cambridge, UK), previously sterilized by exposure to ethylene oxide gas, was
28
29
implanted. The proximal end was placed at the right atrium, entering at the right jugular vein,
30
31
32 while the distal end was passed over the right shoulder and exited dorsally, between the scapulae.
33
34 Carprofen (5 mg/kg) and enrofloxacin (2.25 mg/kg) were given subcutaneously for at least two
35
36
37 days post-operatively. A minimum of seven days of postsurgical recovery was allowed prior to
38
39 animal use in experiments. Catheter patency was maintained with daily iv infusions of 0.1 mL
40
41 heparinized (50 units/mL) sterile 0.9% saline. During training and cessation self-administration
42
43
44 sessions, animals were flushed daily with approximately 50 µL heparinized saline (50 U/mL)
45
46 containing gentamicin (0.1 mg/mL). When rats were not used for periods longer than 48 hr,
47
48
49 catheters were locked with approximately 50 µL of heparinized dextrose (100 U/mL) containing
50
51 gentamicin (0.1 mg/mL). Catheter patency was assessed by injection of 0.1 mL propofol. Rats
52
53
were food-restricted to ~22 grams of rodent diet per day, and fed at the end of each day after
54
55
56 operant sessions had completed.
57
58
59
60 34
ACS Paragon Plus Environment
Page 35 of 55 ACS Chemical Neuroscience

1
2
3
Apparatus. Self-administration training and testing occurred in operant chambers (ENV-008CT;
4
5
6 Med-Associates, Vermont, USA) individually located within sound-attenuating cubicles, which
7
8 were ventilated by an exhaust fan that also served to mask external noise. Two retractable
9
10
11
response levers (ENV-112CM, Med-Associates) were located on one wall of the operant
12
13 chamber, 12 cm apart and 6 cm from the grid floor. A cue light (ENV-221M, Med-Associates)
14
15 was located approximately 4 cm above each lever. On the wall opposite to the response levers
16
17
18 was positioned a house light (ENV-215M, Med-Associates) and response feedback clicker
19
20 module (ENV-135M). Drug was infused by a fixed-rate syringe pump (PHM100, Med-
21
22 Associates) located outside the sound-attenuating cubicle, via plastic tubing (polyethylene, SAI
23
24
25 Infusion Technologies, Illinois, USA) connected from the infusion syringe to a stainless steel
26
27 single-channel swivel (Model 375/22: Instech Laboratories, Pennsylvania, USA) mounted
28
29
directly above the operant chamber on a counterbalanced lever arm (PHM-110, Med-
30
31
32 Associates). A further length of tubing, shielded by a metal spring tether, connected the swivel to
33
34 the external guide cannula of the implanted JVC. Med-PC IV software (Med-Associates) was
35
36
37 used to control operant chambers and record data.
38
39 Self-administration Training Procedure. Self-administration, cessation and reinstatement
40
41 methods were adapted from.57 Rats were placed into the operant chambers and self-
42
43
44 administration sessions commenced with an automatic infusion designed to fill the JVC with
45
46 drug. The house light was then illuminated and both response levers were extended into the
47
48 operant chamber. Responding on one lever (the active lever) under a fixed ratio schedule resulted
49
50
51 in fentanyl delivery, followed by a timeout period of 20 seconds, during which time the cue
52
53 lights located above the levers were flashed at a rate of 0.5 Hz and a clicking sound was
54
55
56
presented at the same time as the light cue. Further responding during this timeout period on the
57
58
59
60 35
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 36 of 55

1
2
3
active lever was recorded but had no scheduled consequence. Responding on the alternative lever
4
5
6 (the inactive lever) was recorded throughout the experimental session but had no scheduled
7
8 outcome. Active and inactive levers were randomly assigned to each rat prior to the first session
9
10
11
and did not change throughout the study. The unit dose of drug available was determined by the
12
13 unit volume per infusion (as determined by the duration of the infusion) and adjusted for the
14
15 weight of the rat. The duration of infusion was maintained at between 1 and 3 seconds. All
16
17
18 experiments started between 8:00 AM and 3:00 PM, up to five days per week. All sessions were
19
20 1.5 hrs in duration.
21
22 Effects of Compound 3 on Cessation of Fentanyl Self-administration. Eleven rats were
23
24
25 trained to respond for fentanyl (1 µg/kg/infusion) under an FR5 schedule of reinforcement (i.e.,
26
27 five responses on the active lever resulted in a single infusion of fentanyl). When rats were
28
29
demonstrating stable self-administration of fentanyl under an FR5 schedule (as defined by ≤ 20%
30
31
32 variation in the number of infusions obtained during two consecutive sessions), fentanyl was
33
34 replaced with saline to extinguish responding. Following a minimum of three extinction sessions
35
36
37 and demonstration of a reduction of 50% of the number of infusions received during the last two
38
39 days of fentanyl self-administration behavior, rats were given the opportunity to reacquire
40
41 fentanyl self-administration (1 µg/kg/infusion) and once stability criteria were achieved, the
42
43
44 effect of 3 (0, 3.2, 10, and 32 mg/kg, s.c.) on fentanyl self-administration was assessed following
45
46 a 30 min pretreatment prior to three consecutive fentanyl self-administration sessions. Following
47
48 each three-session assessment of 3, rats were given the opportunity to self-administer fentanyl
49
50
51 for a minimum of two sessions until stability criteria were achieved before moving to the next
52
53 dose. Included in the dose response was an additional saline extinction group, where saline was
54
55
56
substituted for fentanyl during test sessions.
57
58
59
60 36
ACS Paragon Plus Environment
Page 37 of 55 ACS Chemical Neuroscience

1
2
3
Effects of Compound 3 on Reinstatement of Fentanyl-seeking Behavior. Eight rats were
4
5
6 trained to respond for fentanyl (1 µg/kg/infusion) under an FR5 schedule of reinforcement (i.e.,
7
8 five responses on the active lever resulted in a single infusion of fentanyl). When rats were
9
10
11
demonstrating stable self-administration of fentanyl under an FR5 schedule (as defined by ≤ 20%
12
13 variation in the number of infusions obtained during two consecutive sessions), responding was
14
15 extinguished by removing the contingency between the response and fentanyl delivery. In
16
17
18 addition, cues associated with the infusion were also removed, including the flashing cue lights,
19
20 clicking sound, and the sound of the syringe pump. Following a minimum of five extinction
21
22 sessions, and the demonstration of extinction (defined as a decrease in active lever responses to
23
24
25 <20% of the mean number of responses made during the final two sessions of fentanyl
26
27 responding), reinstatement sessions began. During a reinstatement test session, responses on the
28
29
active lever resulted in delivery of the fentanyl-associated cues for 6 seconds; however, no
30
31
32 fentanyl infusion was administered. Following each reinstatement session, rats were given
33
34 extinction sessions until a minimum of 2 sessions met stability criteria, prior to the next
35
36
37 reinstatement test session. To determine the effect of 3 on reinstatement of fentanyl-seeking
38
39 behavior, doses of 0, 3.2, 10, and 32 mg/kg were administered. Compound 3 was given 30
40
41 minutes prior to all reinstatement tests. The order dose presentation of 3 was randomized.
42
43
44 Analysis. For the cessation tests, the dependent variable for statistical analyses was the number
45
46 of drug infusions. The effects of 3 on fentanyl self-administration were analyzed using two-way
47
48 repeated measures analysis of variance (ANOVA), with dose and session as factors. To permit
49
50
51 direct comparison with baseline data, the baseline sessions preceding and following the test
52
53 sessions were included in the analysis alongside the three test sessions for each dose condition.
54
55
56
When a significant interaction was observed, comparisons on fentanyl self-administration were
57
58
59
60 37
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 38 of 55

1
2
3
performed between the 3 conditions and vehicle condition using one-way ANOVA followed by a
4
5
6 Dunnett’s post hoc test for each session. To understand the dose-response relationship, the
7
8 average numbers of infusions obtained over all three test sessions were compared using a one-
9
10
11
way ANOVA followed by a Dunnett’s post hoc test.
12
13 For reinstatement tests, the baseline was defined as the average response on the two days
14
15 preceding a reinstatement test, while the test was the number of responses on the reinstatement
16
17
18 test session. Data from animals that did not reinstate under vehicle conditions, (i.e., exhibited
19
20 fewer active lever presses on test compared to baseline) and outliers, (i.e., exhibited active lever
21
22 presses greater than three standard deviations of the mean on vehicle) were excluded for each
23
24
25 reinstatement condition.
26
27 For reinstatement tests, statistical analysis was performed using two-way repeated measures
28
29
ANOVA, with session (baseline vs. test) and dose as factors for comparing responses (active and
30
31
32 inactive lever presses) at baseline (i.e., the average of response rates during the sessions
33
34 immediately preceding reinstatement sessions) with those during test reinstatement sessions. If a
35
36
37 significant interaction or main-effect of dose was then observed, post hoc comparisons were
38
39 made between responses following each dose of drug and vehicle, using Dunnett’s test.
40
41
42 The significance level for all statistical tests was set at 0.05. All statistical analysis was
43
44
45
performed using Statistica version 12.0 (StatSoft, Inc. 2014; STATISTICA (data analysis
46
47 software system), version 12.0. www.statsoft.com.).
48
49 Computational Methods. Molecular docking of 3 into the D3R X-ray crystal structure, PDB
50
51
52 ID: 3PBL,50 was used to develop an understanding of its molecular mode of binding, SAR, and
53
54 selectivity. The structure was prepared using the Protein Preparation Wizard in Maestro (version
55
56 9.7, Schrödinger, LLC, New York, NY, 2014) using default options following deletion of the T4-
57
58
59
60 38
ACS Paragon Plus Environment
Page 39 of 55 ACS Chemical Neuroscience

1
2
3
lysozyme residues that replaced most of the third cytoplasmic loop (ICL3). Compound 3 was
4
5
6 docked in its protonated form using Induced Fit Docking (Induced Fit Docking protocol 2014-1,
7
8 Glide version 6.1, Prime version 3.4, Schrödinger, LLC, New York, NY, 2014)49, 58 using the
9
10
11
default options. The pose chosen for analysis was the lowest energy one that maintained a salt
12
13 bridge between the basic amine of the morpholine of compounds 3 and the conserved
14
15 Asp1103.32 of D3R, while also keeping compound 3 in hydrogen bonding proximity to
16
17
18 Tyr361.39.
19
20
21 Associated Content
22
23
24 Acknowledgements. The authors thank Katherine Brighty for her insightful comments on this
25
26
manuscript.
27
28
29
30 Supporting Information Available
31
32 This material is available free of charge via the Internet at http://pubs.acs.org.
33
34
35 Corresponding Author
36
37
38
*To whom correspondence should be addressed. Mailing address: Pfizer Worldwide Research
39
40
41 and Development, 610 Main Street, Cambridge, MA 02139; Tel: 857-225-2840; Fax 860-686-
42
43 6052; E-mail: travis.t.wager@pfizer.com
44
45
46 References
47
48 1. O’Connor, P. G., Sokol, R. J., and D’Onofrio, G. (2014) Addiction medicine: The birth of
49 a new discipline, JAMA Internal Medicine 174, 1717-1718.
50 2. Degenhardt, L., Whiteford, H. A., Ferrari, A. J., Baxter, A. J., Charlson, F. J., Hall, W.
51
52
D., Freedman, G., Burstein, R., Johns, N., Engell, R. E., Flaxman, A., Murray, C. J. L.,
53 and Vos, T. (2013) Global burden of disease attributable to illicit drug use and
54 dependence: findings from the Global Burden of Disease Study 2010, The Lancet 382,
55 1564-1574.
56 3. Wager, T. T., Chandrasekaran, R. Y., Bradley, J., Rubitski, D., Berke, H., Mente, S.,
57 Butler, T., Doran, A., Chang, C., Fisher, K., Knafels, J., Liu, S., Ohren, J., Marconi, M.,
58
59
60 39
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 40 of 55

1
2
3
DeMarco, G., Sneed, B., Walton, K., Horton, D., Rosado, A., and Mead, A. (2014)
4
5 Casein Kinase 1δ/ε Inhibitor PF-5006739 Attenuates Opioid Drug-Seeking Behavior,
6 ACS Chemical Neuroscience 5, 1253-1265.
7 4. Sokoloff, P., Giros, B., Martres, M.-P., Bouthenet, M.-L., and Schwartz, J.-C. (1990)
8 Molecular cloning and characterization of a novel dopamine receptor (D3) as a target for
9 neuroleptics, Nature 347, 146-151.
10
11
5. Newman, A. H., Grundt, P., and Nader, M. A. (2005) Dopamine D3 receptor partial
12 agonists and antagonists as potential drug abuse therapeutic agents, Journal of medicinal
13 chemistry 48, 3663-3679.
14 6. Heidbreder, C. A., and Newman, A. H. (2010) Current perspectives on selective
15 dopamine D3 receptor antagonists as pharmacotherapeutics for addictions and related
16
disorders, Annals of the New York Academy of Sciences 1187, 4-34.
17
18 7. Cho, D., Zheng, M., and Kim, K.-M. (2010) Current perspectives on the selective
19 regulation of dopamine D2 and D3 receptors, Archives of Pharmacal Research 33, 1521-
20 1538.
21 8. Keck, T. M., John, W. S., Czoty, P. W., Nader, M. A., and Newman, A. H. (2015)
22 Identifying Medication Targets for Psychostimulant Addiction: Unraveling the Dopamine
23
24
D3 Receptor Hypothesis, Journal of Medicinal Chemistry 58, 5361-5380.
25 9. Gurevich, E. V., and Joyce, J. N. (1999) Distribution of Dopamine D3 Receptor
26 Expressing Neurons in the Human Forebrain: Comparison with D2 Receptor Expressing
27 Neurons, Neuropsychopharmacology 20, 60-80.
28 10. Robinson, Mike J. F., and Berridge, Kent C. (2013) Instant Transformation of Learned
29
Repulsion into Motivational “Wanting”, Current Biology 23, 282-289.
30
31 11. Koob, G. F., and Volkow, N. D. (2009) Neurocircuitry of Addiction,
32 Neuropsychopharmacology 35, 217-238.
33 12. Ikemoto, S., and Wise, R. A. (2004) Mapping of chemical trigger zones for reward,
34 Neuropharmacology 47, Supplement 1, 190-201.
35 13. Kravitz, A. V., Tomasi, D., LeBlanc, K. H., Baler, R., Volkow, N. D., Bonci, A., and
36
37 Ferré, S. (2015) Cortico-striatal circuits: Novel therapeutic targets for substance use
38 disorders, Brain Research 1628, Part A, 186-198.
39 14. Koob, G. F. (2006) The neurobiology of addiction: a neuroadaptational view relevant for
40 diagnosis, Addiction 101, 23-30.
41 15. Staley, J. K., and Mash, D. C. (1996) Adaptive Increase in D3 Dopamine Receptors in the
42
43
Brain Reward Circuits of Human Cocaine Fatalities, The Journal of Neuroscience 16,
44 6100-6106.
45 16. Mash, D. C., and Staley, J. K. (1999) D3 Dopamine and Kappa Opioid Receptor
46 Alterations in Human Brain of Cocaine-overdose Victims, Annals of the New York
47 Academy of Sciences 877, 507-522.
48 17. Segal, D. M., Moraes, C. T., and Mash, D. C. (1997) Up-regulation of D3 dopamine
49
50 receptor mRNA in the nucleus accumbens of human cocaine fatalities, Molecular Brain
51 Research 45, 335-339.
52 18. Boileau, I., Payer, D., Houle, S., Behzadi, A., Rusjan, P. M., Tong, J., Wilkins, D., Selby,
53 P., George, T. P., Zack, M., Furukawa, Y., McCluskey, T., Wilson, A. A., and Kish, S. J.
54 (2012) Higher Binding of the Dopamine D3 Receptor-Preferring Ligand [11C]-(+)-
55
56
Propyl-Hexahydro-Naphtho-Oxazin in Methamphetamine Polydrug Users: A Positron
57 Emission Tomography Study, The Journal of Neuroscience 32, 1353-1359.
58
59
60 40
ACS Paragon Plus Environment
Page 41 of 55 ACS Chemical Neuroscience

1
2
3
19. Erritzoe, D., Tziortzi, A., Bargiela, D., Colasanti, A., Searle, G. E., Gunn, R. N., Beaver,
4
5 J. D., Waldman, A., Nutt, D. J., Bani, M., Merlo-Pich, E., Rabiner, E. A., and Lingford-
6 Hughes, A. (2014) In Vivo Imaging of Cerebral Dopamine D3 Receptors in Alcoholism,
7 Neuropsychopharmacology 39, 1703-1712.
8 20. Pilla, M., Perachon, S., Sautel, F., Garrido, F., Mann, A., Wermuth, C. G., Schwartz, J.-
9 C., Everitt, B. J., and Sokoloff, P. (1999) Selective inhibition of cocaine-seeking
10
11
behaviour by a partial dopamine D3 receptor agonist, Nature 400, 371-375.
12 21. Román, V., Gyertyán, I., Sághy, K., Kiss, B., and Szombathelyi, Z. (2013) Cariprazine
13 (RGH-188), a D3-preferring dopamine D3/D2 receptor partial agonist antipsychotic
14 candidate demonstrates anti-abuse potential in rats, Psychopharmacology 226, 285-293.
15 22. Le Foll, B., Goldberg, S. R., and Sokoloff, P. (2005) The dopamine D3 receptor and drug
16
dependence: Effects on reward or beyond?, Neuropharmacology 49, 525-541.
17
18 23. Caine, S. B., Koob, G. F., Parsons, L. H., Everitt, B. J., Schwartz, J.-C., and Sokoloff, P.
19 (1997) D3 receptor test in vitro predicts decreased cocaine self‐administration in rats,
20 NeuroReport 8, 2373-2377.
21 24. Wicke, K., and Garcia-Ladona, J. (2001) The dopamine D3 receptor partial agonist, BP
22 897, is an antagonist at human dopamine D3 receptors and at rat somatodendritic
23
24 dopamine D3 receptors, European Journal of Pharmacology 424, 85-90.
25 25. Vorel, S. R., Ashby, C. R., Paul, M., Liu, X., Hayes, R., Hagan, J. J., Middlemiss, D. N.,
26 Stemp, G., and Gardner, E. L. (2002) Dopamine D3 Receptor Antagonism Inhibits
27 Cocaine-Seeking and Cocaine-Enhanced Brain Reward in Rats, The Journal of
28 Neuroscience 22, 9595-9603.
29
30
26. Xi, Z.-X., and Gardner, E. L. (2007) Pharmacological Actions of NGB 2904, a Selective
31 Dopamine D3 Receptor Antagonist, in Animal Models of Drug Addiction, CNS Drug
32 Reviews 13, 240-259.
33 27. Hu, R., Song, R., Yang, R., Su, R., and Li, J. (2013) The dopamine D3 receptor
34 antagonist YQA14 that inhibits the expression and drug-primed reactivation of morphine-
35 induced conditioned place preference in rats, European Journal of Pharmacology 720,
36
37 212-217.
38 28. Boateng, C. A., Bakare, O. M., Zhan, J., Banala, A. K., Burzynski, C., Pommier, E.,
39 Keck, T. M., Donthamsetti, P., Javitch, J. A., Rais, R., Slusher, B. S., Xi, Z.-X., and
40 Newman, A. H. (2015) High Affinity Dopamine D3 Receptor (D3R)-Selective
41 Antagonists Attenuate Heroin Self-Administration in Wild-Type but not D3R Knockout
42
43
Mice, Journal of Medicinal Chemistry 58, 6195-6213.
44 29. Galaj, E., Manuszak, M., Babic, S., Ananthan, S., and Ranaldi, R. (2015) The selective
45 dopamine D3 receptor antagonist, SR 21502, reduces cue-induced reinstatement of heroin
46 seeking and heroin conditioned place preference in rats, Drug and Alcohol Dependence
47 156, 228-233.
48 30. Kumar, V., Bonifazi, A., Ellenberger, M. P., Keck, T. M., Pommier, E., Rais, R., Slusher,
49
50 B. S., Gardner, E., You, Z.-B., Xi, Z.-X., and Newman, A. H. (2016) Highly Selective
51 Dopamine D3 Receptor (D3R) Antagonists and Partial Agonists Based on Eticlopride
52 and the D3R Crystal Structure: New Leads for Opioid Dependence Treatment, Journal of
53 Medicinal Chemistry 59, 7634-7650.
54 31. Mugnaini, M., Iavarone, L., Cavallini, P., Griffante, C., Oliosi, B., Savoia, C., Beaver, J.,
55
56
Rabiner, E. A., Micheli, F., Heidbreder, C., Andorn, A., Merlo Pich, E., and Bani, M.
57
58
59
60 41
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 42 of 55

1
2
3
(2013) Occupancy of Brain Dopamine D3 Receptors and Drug Craving: A Translational
4
5 Approach, Neuropsychopharmacology 38, 302-312.
6 32. Rudmann, D. G., Mcnerney, M. E., Vandereide, S. L., Schemmer, J. K., Eversole, R. R.,
7 and Vonderfecht, S. L. (2004) Epididymal and Systemic Phospholipidosis in Rats and
8 Dogs Treated with the Dopamine D3 Selective Antagonist PNU-177864, Toxicologic
9 Pathology 32, 326-332.
10
11
33. Micheli, F., and Heidbreder, C. (2006) Selective Dopamine D3 Receptor Antagonist: A
12 Review 2001-2005, Recent Pat. CNS Drug Discovery 1, 271-288.
13 34. Wager, T. T., Pettersen, B. A., Schmidt, A. W., Spracklin, D. K., Mente, S., Butler, T.
14 W., Howard, H., Lettiere, D. J., Rubitski, D. M., Wong, D. F., Nedza, F. M., Nelson, F.
15 R., Rollema, H., Raggon, J. W., Aubrecht, J., Freeman, J. K., Marcek, J. M., Cianfrogna,
16
J., Cook, K. W., James, L. C., Chatman, L. A., Iredale, P. A., Banker, M. J., Homiski, M.
17
18 L., Munzner, J. B., and Chandrasekaran, R. Y. (2011) Discovery of Two Clinical
19 Histamine H3 Receptor Antagonists: trans-N-Ethyl-3-fluoro-3-[3-fluoro-4-
20 (pyrrolidinylmethyl)phenyl]cyclobutanecarboxamide (PF-03654746) and trans-3-Fluoro-
21 3-[3-fluoro-4-(pyrrolidin-1-ylmethyl)phenyl]-N-(2-
22 methylpropyl)cyclobutanecarboxamide (PF-03654764), Journal of Medicinal Chemistry
23
24
54, 7602-7620.
25 35. Attkins, N., Betts, A., Hepworth, D., and Heatherington, A. C. (2010) Pharmacokinetics
26 and elucidation of the rates and routes of N-glucuronidation of PF-592379, an oral
27 dopamine 3 agonist in rat, dog, and human, Xenobiotica 40, 730-742.
28 36. Wager, T. T., Chandrasekaran, R. Y., Hou, X., Troutman, M. D., Verhoest, P. R.,
29
Villalobos, A., and Will, Y. (2010) Defining Desirable Central Nervous System Drug
30
31 Space through the Alignment of Molecular Properties, in Vitro ADME, and Safety
32 Attributes, ACS Chem. Neurosci. 1, 420-434.
33 37. Wager, T. T., Hou, X., Verhoest, P. R., and Villalobos, A. (2010) Moving beyond rules:
34 The development of a central nervous system multiparameter optimization (CNS MPO)
35 approach to enable alignment of druglike properties, ACS Chem. Neurosci. 1, 435-449.
36
37 38. Wager, T. T., Hou, X., Verhoest, P. R., and Villalobos, A. (2016) Central Nervous
38 System Multiparameter Optimization Desirability: Application in Drug Discovery, ACS
39 Chemical Neuroscience 7, 767-775.
40 39. Ben-Jonathan, N., Arbogast, L. A., and Hyde, J. F. (1989) Neuroendrocrine regulation of
41 prolactin release, Progress in neurobiology 33, 399-447.
42
43
40. Rummel-Kluge, C., Komossa, K., Schwarz, S., Hunger, H., Schmid, F., Kissling, W.,
44 Davis, J. M., and Leucht, S. (2012) Second-Generation Antipsychotic Drugs and
45 Extrapyramidal Side Effects: A Systematic Review and Meta-analysis of Head-to-Head
46 Comparisons, Schizophrenia Bulletin 38, 167-177.
47 41. Coffin, V. L., Latranyi, M. B., and Chipkin, R. E. (1989) Acute extrapyramidal syndrome
48 in Cebus monkeys: development mediated by dopamine D2 but not D1 receptors, Journal
49
50 of Pharmacology and Experimental Therapeutics 249, 769-774.
51 42. Beart, P. (1988) Dopamine receptors (Vol. 8: Receptor biochemistry and methodology):
52 edited by I. Creese and CM Fraser, Alan R. Liss, 1987.(ix+ 256 pages) ISBN 0 8451
53 3707 7, Elsevier Current Trends.
54 43. Hartesveldt, C. V., Cottrell, G. A., Potter, T., and Meyer, M. E. (1992) Effects of
55
56
intracerebral quinpirole on locomotion in rats, European Journal of Pharmacology 214,
57 27-32.
58
59
60 42
ACS Paragon Plus Environment
Page 43 of 55 ACS Chemical Neuroscience

1
2
3
44. Baron, J. C., Martinot, J. L., Cambon, H., Boulenger, J. P., Poirier, M. F., Caillard, V.,
4
5 Blin, J., Huret, J. D., Loc'h, C., and Maziere, B. (1989) Striatal dopamine receptor
6 occupancy during and following withdrawal from neuroleptic treatment: correlative
7 evaluation by positron emission tomography and plasma prolactin levels,
8 Psychopharmacology 99, 463-472.
9 45. Kapur, S., Langlois, X., Vinken, P., Megens, A. A. H. P., De Coster, R., and Andrews, J.
10
11
S. (2002) The Differential Effects of Atypical Antipsychotics on Prolactin Elevation Are
12 Explained by Their Differential Blood-Brain Disposition: A Pharmacological Analysis in
13 Rats, Journal of Pharmacology and Experimental Therapeutics 302, 1129-1134.
14 46. Gyertyán, I., and Sághy, K. (2007) The selective dopamine D3 receptor antagonists, SB
15 277011-A and S 33084 block haloperidol-induced catalepsy in rats, European Journal of
16
Pharmacology 572, 171-174.
17
18 47. Maggio, R., Aloisi, G., Silvano, E., Rossi, M., and Millan, M. J. (2009)
19 Heterodimerization of dopamine receptors: new insights into functional and therapeutic
20 significance, Parkinsonism & Related Disorders 15, Supplement 4, S2-S7.
21 48. Maggio, R., Scarselli, M., Capannolo, M., and Millan, M. J. (2015) Novel dimensions of
22 D3 receptor function: Focus on heterodimerisation, transactivation and allosteric
23
24
modulation, European Neuropsychopharmacology 25, 1470-1479.
25 49. Sherman, W., Day, T., Jacobson, M. P., Friesner, R. A., and Farid, R. (2006) Novel
26 Procedure for Modeling Ligand/Receptor Induced Fit Effects, Journal of Medicinal
27 Chemistry 49, 534-553.
28 50. Chien, E. Y. T., Liu, W., Zhao, Q., Katritch, V., Won Han, G., Hanson, M. A., Shi, L.,
29
Newman, A. H., Javitch, J. A., Cherezov, V., and Stevens, R. C. (2010) Structure of the
30
31 Human Dopamine D3 Receptor in Complex with a D2/D3 Selective Antagonist, Science
32 330, 1091-1095.
33 51. Marchant, N. J., Li, X., and Shaham, Y. (2013) Recent developments in animal models of
34 drug relapse, Current Opinion in Neurobiology 23, 675-683.
35 52. Wager, T. T., Kormos, B. L., Brady, J. T., Will, Y., Aleo, M. D., Stedman, D. B., Kuhn,
36
37 M., and Chandrasekaran, R. Y. (2013) Improving the Odds of Success in Drug
38 Discovery: Choosing the Best Compounds for in Vivo Toxicology Studies, Journal of
39 Medicinal Chemistry 56, 9771-9779.
40 53. Ertl, P. (2008) Polar surface area, Methods Princ. Med. Chem. 37, 111-126.
41 54. Feng, B., Mills Jessica, B., Davidson Ralph, E., Mireles Rouchelle, J., Janiszewski John,
42
43
S., Troutman Matthew, D., and de Morais Sonia, M. (2008) In vitro P-glycoprotein
44 assays to predict the in vivo interactions of P-glycoprotein with drugs in the central
45 nervous system, Drug Metab. Dispos. 36, 268-275.
46 55. Hosea, N. A., Collard, W. T., Cole, S., Maurer, T. S., Fang, R. X., Jones, H., Kakar, S.
47 M., Nakai, Y., Smith, B. J., Webster, R., and Beaumont, K. (2009) Prediction of human
48 pharmacokinetics from preclinical information: comparative accuracy of quantitative
49
50 prediction approaches, J. Clin. Pharmacol. 49, 513-533.
51 56. Gao, H., Yao, L., Mathieu, H. W., Zhang, Y., Maurer, T. S., Troutman, M. D., Scott, D.
52 O., Ruggeri, R. B., and Lin, J. (2008) In silico modeling of nonspecific binding to human
53 liver microsomes, Drug Metab. Dispos. 36, 2130-2135.
54 57. O’Connor, E. C., Parker, D., Rollema, H., and Mead, A. N. (2010) The α4β2 nicotinic
55
56
acetylcholine-receptor partial agonist varenicline inhibits both nicotine self-
57
58
59
60 43
ACS Paragon Plus Environment
ACS Chemical Neuroscience Page 44 of 55

1
2
3
administration following repeated dosing and reinstatement of nicotine seeking in rats,
4
5 Psychopharmacology 208, 365-376.
6 58. Sherman, W., Beard, H. S., and Farid, R. (2006) Use of an Induced Fit Receptor Structure
7 in Virtual Screening, Chemical Biology & Drug Design 67, 83-84.
8
9 Graphical Abstract
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 44
ACS Paragon Plus Environment
Page 45 of 55 ACS Chemical Neuroscience

1
2
3
4
5
6
7
Dopamine D3/D2 Receptor Antagonist PF-
8
9
10
04363467 Extinguishes and Blocks Opioid
11
12
13 Drug-Seeking Behavior without Concomitant
14
15
16
17
D2 Side Effects
18
19
20
21
22
23
Figures 1-6
24 Tables 1-4
25
26 Graphical Abstract
27
28
29
30
31
32
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42
ACS Chemical Neuroscience Page 46 of 55
Table 1. Corporate file mining leads and tool compound 3.

1
2
3
4
5
6
7
8
9
10
11
12
13 D3 Ki D2 Ki D2/D D2 Functional Functional
ID D3 Functional (nM)b
14 (nM)a (nM)a 3 (nM)b Activity
15 Ki > 33c
38.2 4480 Ki = 22
16 1 117 Antagonist
(23.6 – 61.9) (3290 – 6090) (10.1 - 48.9)c
17
18 322 EC50 > 9,000d
2 >4160 >12 EC50 = 21d Agonist
19 (57.9 – 1790)
20 3.1 692 Ki = 1.1 Ki = 464
21 3 223 Antagonist
22
(2.47 – 3.99) (539 – 890) (0.61 - 2.1)c (188 – 1147)c
23
24 aHuman dopamine D3 and D2 binding data reported as the geometric mean of at least 3 determinations.
25 bHuman
26 dopamine D3 and D2 functional data reported as the geometric mean of at least 3 determinations.
cData reported as mean antagonist activity with (95% CI). dData reported by Attkins et al.35 as agonist
27
28 activity.
29
30
31
32
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42
Page 47 of 55 ACS Chemical Neuroscience

1
2
3
4
5
6
7
Table 2. Summary of physicochemical properties and CNS MPO desirability
8 aCalculated CNS MPO desirability scores were obtained using the published algorithm.30
9
10
11 Compound 1 Compound 2 Compound 3
12
13 Physicochemical Component Component Component
14 Property Value Score Value Score Value Score
15
16 MW 402 0.70 235 1.00 403 0.70
17 ClogP 4.46 0.27 1.48 1.00 4.06 0.47
18
19 TPSA 67.4 1.00 51.4 1.00 58.6 1.00
20 ClogD7.4 1.06 1.00 0.76 1.00 3.96 0.02
21 HBD 2 0.50 2 0.50 1 0.83
22
23
pKa 10.6 0.00 7.5 1.00 7.5 1.00
24 CNS MPO 3.5/ 6.0 5.5/ 6.0 4.0/ 6.0
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42
ACS Chemical Neuroscience Page 48 of 55

1
2
3
4
5
6
7
8
9
10 Table 3. Summary of ADME properties for 3.
11
12 Rat Cmax,b,u Rat AUC0-∞ Rat AUC0-∞
13 Compound HLMa Pappb P-gpc
14
[nM]d Cb/Cp Cb,u/Cp,u
15 3 126 8.4 2.3 23.7 2.4 1.01
16
17 •aHuman liver microsomal clearance (mL/min/mg). bPassive permeability (Papp AB x
18 10-6 cm/sec). cMDR1 Efflux Ratio (BA/AB). dRat Cmax,b,u values are reported as
19 maximum free drug concentration in brain at a 10 mg/kg dose, s.c.
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42
Page 49 of 55 ACS Chemical Neuroscience

1
2
3
4 10% at 10 uM 100% at 10 uM
5
6
7
8
9
10
11
12
13
14
15 7600 nM 4 nM
16
17
18
19
20
21
22
23
24

D2S
D1

D2

D3
CHT1

Ca Chan

Na Chan

H3
M1
M5

M4
5HT2B

5HT6
5HT1A

5HT2A

5HT2A
CYP3A4
SIGMA

CYP3A4b
ALPHA2B

CYP2C9

CYP2C19
ALPHA2A

LTB4 BLT1

ALPHA2C
M3

25
26
27
28
29
30 Figure 1. A) Percent inhibition of a 129-target promiscuity panel at 10 µM of compound 3. Heat map colors for
31 this continuum: green, 0% inhibition; yellow, 50% inhibition; red, 100% inhibition. B) IC50 determination for 26
32 targets hitting in the promiscuity panel with inhibition >50% for compound 3. Heat map colors continuum: green,
33
34
7,600 nM; yellow, 100 nM and red, 4 nM. IC50 determinations for the 10 highest affinity targets for compound 3
35 were: D3R, 4 nM; CYP2C19, 110 nM; D2R, 420 nM; D2S, 540 nM; 5HT6, 990 nM; 5HT2A, 1400 nM; CYP3A4b,
36 2300 nM; H3, 2400 nM; D1R, 2500 nM.
37
38
39
40 ACS Paragon Plus Environment
41
42
ACS Chemical Neuroscience Page 50 of 55
Table 4. D3R and D2R brain receptor occupancy from autoradiography
techniques in rats
1
2 D3 EVRO % RO D2 EVRO % RO
Compound Treatment
3 Observeda Observedb
4
5 32 mg/kg --- 85.9 ± 1.7
6 10 mg/kg 98.8 ± 2.4 50.6 ± 3.5
7
8 3.2 mg/kg 99.9 ± 0.5 34.7 ± 1.5
9 3
0.1 mg/kg 33.6 ± 4.5 10.8 ± 4.1
10
11 0.03 mg/kg 6.3 ± 10.1 ---
12
13 EC50 1.5 ± 0.4 nM 46.6 ± 9.1 nM
14
15 aBrain region evaluated: rat cerebellum (mean ± SEM, n = 3–4). bBrain region
16 evaluated: rat striatum (mean ± SEM, n = 4). Ex vivo Receptor Occupancy (EVRO).
17
18
19 A) B)
100 100
20 96

Ex-Vivo, D3 Receptor Occupancy


Ex-Vivo, D2 Receptor Occupancy

21
22 80 80
23
24 60
60
25
26
40 40
27
28
29 20 15
30
7
31 0
0
32
33
Dose = 0.1 3.2 10 32 mg/kg Dose = 0.01 0.032 0.1 3.2 10 mg/kg
34 -12 -10 -8 -6
-13 -12 -11 -10 -9 -8 -7 10 10 10 10
35 10 10 10 10 10 10 10
36 Free Brain Concentration (nM) Free Brain Concentration (nM)
37
38 Figure 2. A) D2 (striatum) and B) D3 (cerebellum) brain receptor occupancy curves from the ex vivo autoradiography
39
40 protocol in rat. The gray and black data points
ACSrepresent
Paragon actual measured RO and brain PK from individual animals.
Plus Environment
41 The solid line is the data-fitted RO model. The D3R binding was measured in cerebellum, while the D2R binding was
42 measured in striatum.
Page 51 of 55 ACS Chemical Neuroscience

1
2
3
4
5
6
7
8
9
10
11
12 Vehicle Vehicle
13A) 8 0 ** 9.3 fold  B)
2000
Haldol, 1 mg/kg
C)
100
Haldol, 1 mg/kg
14 **
3, 1.0 mg/kg 3, 1.0 mg/kg
3, 3.2 mg/kg 3, 3.2 mg/kg **
15
3, 10 mg/kg 3, 10 mg/kg

M e a n T o t a l A c t iv it y
80 **

T im e O n B a r ( S e c )
16
P r o la c t in ( n g /m L )

6 0 1500 3, 32 mg/kg 3, 32 mg/kg **


17
60
18
40
19 1 0 0 0

20 40
**
21 1.8 fold  **
20 1.5 fold  500
22 20
23
24 0
0 0
le

10
1

25 0 -3 0 3 0 -6 0 6 0 -1 2 0 1 2 0 -1 8 0 0 30 60 120 180
0.

1.
ic
eh

T im e ( M in ) T im e ( M in )
V

26 3, Dose (mg/kg, SC)


27
28 Figure 3. A) Effect of 3 on prolactin. Data represent mean ± SEM, n = 5/group. B) Effects of 3 on total activity in a rat model of
29 locomotor activity. Data represent mean ± SEM locomotor activity over 30 minute time bins, n = 8/group. C) Effects of 3 in a rat model
30
of catalepsy. Data represents mean ± SEM seconds on bar demonstrating the effect of 3 (1.0–32.0 mg/kg, s.c.), n = 8/group. ** indicates
31
32 p < 0.01 compared to vehicle group.
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42
ACS Chemical Neuroscience Page 52 of 55

1
2
3
4 A) B)
5 300

M e a n A c t iv e L e v e r R e s p o n s e s
6 30
**
# I n f u s io n s ( M e a n + S E M )

7 250
8 25

(M e a n + S E M )
9
10 200
20
11
12 150
13 15
14 Extinction
15 10 Vehicle 100
16 3, 3.2 mg/kg
17
18 5 *** 3, 10 mg/kg 50
*** 3, 32 mg/kg
19 ***
20 0 0
21
yl

yl

yl

yl
1

le

10

32
2

2
ay

ay
n

n
22

3.
ay

ic
ta

ta

ta

ta
D

eh
D
en

en

en

en
23
3, Dose (mg/kg, SC)

V
F

24 Treatment
25
26 Figure 4. A) Effects of vehicle or 3 pretreatment on fentanyl self-administration over 3 test days in the rat model of cessation. Data
27 represent the average infusions of fentanyl across baseline responding (fentanyl) before and after test sessions as well as three
28 consecutive test sessions with 3 (Days 1–3). Saline extinction tests (Extinction) are also shown, in which saline was substituted for
29
30 fentanyl during test sessions. n = 9–11; *** p < 0.001. B) Effects of 3 on combined cue and fentanyl prime-induced reinstatement of
31 fentanyl-seeking behavior. Data represent mean ± SEM active lever responses, with white bars representing baseline active lever
32 responses and blue bars representing active lever response on reinstatement test conditions. n = 8; ** p < 0.01
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42
Page 53 of 55 ACS Chemical Neuroscience

1
2
3
4
5
6
7 Drug-seeking Responding
8
9 Drug-taking Relapse
10
“fentanyl self-administration”

11 Selective dual DA
Drug Priming and
12 D3/D2 antagonist
Cue Presentation
Responding

13 Stable
14 Acquisition maintenance
15 Reinstatement/
16 Forced Abstinence relapse
17
18
19 Drug-assisted
20 Drug-assisted
cessation relapse prevention
21
22
23
24
25
Days
26
27 Figure 5. General depiction of the experiment. Selective D3/D2 antagonist 3 attenuates fentanyl-
28 seeking behavior in models of both cessation of fentanyl self-administration and reinstatement (as a
29
model of relapse).
30
31
32
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42
ACS Chemical Neuroscience Page 54 of 55

1
2
3
Scheme 1: Synthesis of D3R/D2R antagonist 3.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35 Reagents and conditions: a) thionyl chloride, DMF; b) (trimethylsilyl)diazomethane, ACN/THF, then 40% HBr; c) 2-
36 (ethylamino)ethanol, Et3N, CH2Cl2; d) triethylsilane, TFA, CH2Cl2; e) H2, Raney Ni, MeOH; f) 4-isopropylbenzene-1-sulfonyl
37 chloride, pyridine; g) chiral chromatographic separation, then HCl/EtOAc/Et2O.
38
39
40 ACS Paragon Plus Environment
41
42
Page 55 of 55 ACS Chemical Neuroscience

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15 Graphical Abstract
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40 ACS Paragon Plus Environment
41
42