Structure-activity relationships of privileged structures lead to the discovery of novel biased ligands at the dopamine D2 receptor

Article

pubs.acs.org/jmc

Structure−Activity Relationships of Privileged Structures Lead to the

Discovery of Novel Biased Ligands at the Dopamine D₂Receptor

Monika Szabo,^†,‡Carmen Klein Herenbrink,^‡Arthur Christopoulos,^‡J. Robert Lane,*^,‡

and Ben Capuano*^,†

^†Medicinal Chemistry and ^‡Drug Discovery Biology, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville 3052,

Victoria, Australia

S

* Supporting Information

ABSTRACT: Biased agonism at GPCRs highlights the

potential for the discovery and design of pathway-selective

ligands and may confer therapeutic advantages to ligands

targeting the dopamine D₂receptor (D₂R). We investigated

the determinants of eﬃcacy, aﬃnity, and bias for three

privileged structures for the D₂R, exploring changes to linker

length and incorporation of a heterocyclic unit. Proﬁling the

compounds in two signaling assays (cAMP and pERK1/2)

allowed us to identify and quantify determinants of biased agonism at the D₂R. Substitution on the phenylpiperazine privileged

structures (2-methoxy vs 2,3-dichloro) inﬂuenced bias when the thienopyridine heterocycle was absent. Upon inclusion of the

thienopyridine unit, the substitution pattern (4,6-dimethyl vs 5-chloro-6-methoxy-4-methyl) had a signiﬁcant eﬀect on bias that

overruled the eﬀect of the phenylpiperazine substitution pattern. This latter observation could be reconciled with an extended

binding mode for these compounds, whereby the interaction of the heterocycle with a secondary binding pocket may engender

bias.

Biased agonism (sometimes termed “functional selectivity” or

“stimulus bias”) refers to the phenomenon by which diﬀerent

agonists acting at the same receptor can stabilize distinct

receptor conformations linked to diﬀerent functional re-

sponses.¹⁰⁻¹²Of interest, aripiprazole has been identiﬁed as a

ligand that displays biased agonism, and evidence has been

provided that this bias may contribute to its antipsychotic

eﬃcacy.¹³Previous SAR studies, focused on aripiprazole,

revealed that small structural changes to the phenylpiperazine

core, the linker region, and the bicyclic heterocycle resulted in

changes in biased agonism. These studies identiﬁed novel D₂R

ligands that displayed bias toward the recruitment of β-

arrestin.^14,15Further examples of biased agonists targeting the

D₂R have been provided by a number of groups, including

Mailman et al. with dihydrexidine,^16,17Tschammer et al. with

1,4-disubstituted phenylpiperazines,¹⁸and Shonberg et al. with

tetrahydroisoquinoline derivatives.¹⁹

Privileged structures are often versatile scaﬀolds that can be

functionalized to produce ligands that may be selective for a

class of receptors and, more ideally, a single receptor

target.²⁰⁻²³Privileged structures can be “built in” to a molecule

by replacing certain functional groups or can be used as the

starting scaﬀold, where one can “build out” to create an

optimized ligand. The aim of the current study was to

investigate the structure−activity relationships underlying

aﬃnity and eﬃcacy of distinct privileged structures for the

INTRODUCTION

■

Dopamine receptors (DRs) belong to the G-protein-coupled

receptor (GPCR) superfamily that is characterized by seven

transmembrane (TM) domains. There are ﬁve receptor

subtypes (D₁−D₅) expressed both in the central nervous

system (CNS) and in the periphery.¹In the CNS the dopamine

D₂receptor (D₂R) is a primary target for the treatment of

disease including Parkinson’s and schizophrenia.²Antipsy-

chotics have evolved from ﬁrst generation antipsychotics

(FGAs), which act as D₂antagonists, to second generation

antipsychotics (SGAs), which display robust polypharmacology

with high aﬃnities for GPCRs in addition to the D₂R.³⁻⁵

Aripiprazole (1, Figure 1) is classiﬁed as a third generation

antipsychotic (TGA). This comes from its unique pharmaco-

logical proﬁle, as it is a potent partial agonist at both

presynaptic and postsynaptic D₂receptors while also displaying

partial agonist activity at the serotonin 5-HT_1Areceptor.⁶⁻⁸The

exploration of D₂R partial agonists for treating schizophrenia

was driven by the hypothesis that they would stabilize levels of

dopamine in the CNS but avoid the extrapyramidal side eﬀects

associated with complete D₂R blockade (for a detailed review of

the use of partial agonists in the treatment of schizophrenia and

other related psychoactive disorders, see Tamminga et al.⁹).

Since the discovery of aripiprazole, other partial agonists have

emerged, such as cariprazine (2), brexpiprazole (3), and

bifeprunox (4) (Figure 1), with 2 currently awaiting FDA

approval and 3 in phase III clinical trials for the treatment of

schizophrenia.

Received: March 24, 2014

A

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 1. Partial agonists: aripiprazole (1), cariprazine (2), brexpiprazole (3), and bifeprunox (4).

Figure 2. Privileged structures: 2-methoxyphenylpiperazine (5), 2,3-dichlorophenylpiperazine (6), and 4,4-chlorophenyl-4-hydroxypiperidine (7).

Antipsychotics and/or potent D₂R antagonists: haloperidol (8), L-741,626 (9), risperidone (10), and ziprasidone (11).

Figure 3. Model pharmacophore for the development of novel D₂R antagonists and/or partial agonists.

D₂R. Many ligands targeting the D₂R incorporate a substituted

phenylpiperazine; therefore, we selected 2-methoxyphenylpi-

perazine (5) and 2,3-dichlorophenylpiperazine (6) to use as

privileged structures in our study (Figure 2). We also

investigated 4,4-chlorophenyl-4-hydroxypiperidine (7), as it

represents a common structural feature of two very potent D₂

antagonists, haloperidol (8) and L-741,626 (9)²⁴(Figure 2).

Another structural attribute of many D₂R targeting antagonists

and partial agonists is the incorporation of a linker to the

ionizable nitrogen, usually from two to ﬁve atoms in length

followed by a heterocyclic group. For example, the two

antipsychotics risperidone (10) and ziprasidone (11), while

devoid of a substituted phenylpiperazine moiety, retain either a

piperazine or a piperidine system to bear the ionizable nitrogen,

a spacer, and a heterobicyclic group (Figure 2). Such scaﬀolds

have been shown to confer subtype selectivity across the D₂-like

receptor subfamily.²⁵However, of relevance to this study,

Newman and co-workers revealed that the presence of a

heterocyclic group, or even the linker alone, could modulate the

eﬃcacy of compounds based on 6 in an assay measuring

activation of a Gα_o1G protein. In contrast, an equivalent

compound based on 5 displayed no agonism at this assay end

point.²⁶These observations beg the question whether such

structural determinants of eﬃcacy are consistent across

diﬀerent signaling end points or whether they diﬀer and thus

engender biased agonism.

In our study, we have synthesized and characterized a

focused library of novel D₂R antagonists and partial agonists

based on a previously described structural model²⁷(Figure 3)

incorporating the aforementioned three privileged structures,

B

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

a

Scheme 1. Synthesis of Various Linker Lengths Incorporating Privileged Structures

a

Reagents and conditions: (a) Boc anhydride, Et₃N, DCM, rt, 1−1.5 h, 28−96%; (b) methanesulfonyl chloride, Et₃N, DCM, 0 °C → rt; (c) 5, 6, or

7, CH₃CN, reﬂux, 24 h, 28−64%; (d) 5 or 7, NaI, DIPEA, CH₃CN, reﬂux, 24 h, 34−93%.

a

Scheme 2. Synthesis of Key Thienopyridine Scaﬀolds

a

Reagents and conditions: (a) P₄S₁₀, EtOAc, reﬂux 1.5−4 h, 30%; (b) acetylacetone, KOH, MeOH, reﬂux, 4 h, 67%; (c) ethyl 2-chloroacetate, Et₃N,

0 °C → rt, 3.5 h, 86%; (d) 1 M KOH, DMF, rt, 15 min, 93%; (e) 2 M NaOH, EtOH, reﬂux, 4 h, 83%; (f) methyl acetoacetate, morpholine, EtOH,

reﬂux 8 h, 51%; (g) ethyl 2-chloroacetate, Et₃N, DMF, 0 °C → rt, 5 h, 81%; (h) (i) MeI, K₂CO₃, DMF, 4 h, rt; (ii) 1 M KOH, 15 min, 58%; (i)

phthalic anhydride, AcOH, reﬂux, 20 h, 55%; (j) N-chlorosuccinimide, conc HCl, EtOH, reﬂux, 1.5 h, 93%; (k) (NH₂)₂·H₂O, EtOH, reﬂux, 3 h,

84%; (l) 2 M NaOH, EtOH, reﬂux, 1.5 h, 84%.

diﬀerent spacer lengths (ranging from two to six carbon atoms),

and novel heterocyclic units identiﬁed from our in-house

muscarinic GPCR drug discovery program. These versatile

heterocyclic motifs were selected, as they exhibit structural

similarities to that of the tetrahydroquinolinone moiety of

aripiprazole. Additionally, the use of a thiophene has previously

been shown to be advantageous in increasing the aﬃnity for D₂-

like receptor subtypes, relating to its electron-rich system and

its ability to form hydrophobic interactions within a binding

pocket, essentially acting as an isostere of benzene.²⁸As such,

we envisioned that our thienopyridine scaﬀolds may also exhibit

these properties. The ligands generated were evaluated using in

vitro assays to measure their ability to displace [³H]spiperone

binding at the D₂R and to stimulate ERK1/2 phosphorylation

C

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

through activation of the D₂R. For selected compounds, we

extended our characterization to a second functional assay,

inhibition of forskolin-induced cAMP production, to allow

identiﬁcation and quantiﬁcation of biased agonism.²⁹

chlorine atom at the 5′ position to give 30 in excellent yield.

Subsequent removal of the phthalimide protecting group using

hydrazine monohydrate aﬀorded 31 in excellent yield. Finally,

the base-catalyzed ester hydrolysis conditions mentioned earlier

were commissioned that, following acidic workup, aﬀorded the

key carboxylic acid intermediate compound, 32.

The syntheses of the target compounds incorporating

privileged structures 5, 6, and 7 and the thienopyridine

scaﬀolds are summarized in Scheme 3. We initially combined

RESULTS AND DISCUSSION

■

Chemistry. Derivatives of the previously described priv-

ileged structures (5−7) were synthesized utilizing a single

linkage point, namely, the aliphatic nitrogen of the piperazine

or piperidine, and included chain lengths of two to six carbon

atoms (Scheme 1). To furnish derivatives with linker lengths of

two to three carbon atoms, we commenced with bromoalkan-

amines (15a,b) and subsequently Boc-protected the amines to

aﬀord the intermediates 16a,b. These compounds were then

reacted with 5 or 7 to give products 17a,b and 18a,b in yields

ranging from 34% to 93%. We utilized a modiﬁed pathway for

generating the required analogues 17c−e and 18c−e, since

linker lengths greater than four carbon atoms were not directly

available as the bromoalkanamines. Aminoalkanols (12c−e)

with carbon atom spacers of four to six were Boc-protected to

give intermediates 13c−e in excellent yields (83−96%). The

primary alcohol functionality was activated with methanesul-

fonyl chloride to generate the mesylated intermediates 14c−e.

Because of the reactivity and/or stability of the mesylated

compounds, the formation of products was conﬁrmed by the

disappearance of starting material via TLC and then reacted

immediately without further puriﬁcation in the next reaction.

Subsequent reaction of 5 or 7 with the mesylates 14c−e

produced analogues 17c−e and 18c−e. To furnish the ﬁve

carbon atom spaced Boc protected 2,3-dichlorophenylpiper-

azine (19), the mesylate 14d was reacted with 6 under the same

conditions as described above.

Scheme 3. Synthesis of Full Length Structures Including

Privileged Structures, Linker, and Thienopyridine Scaﬀold

a

Synthesis of the thienopyridine scaﬀold is represented in

Scheme 2 and commences with reacting 2-cyanoacetamide

(20) with phosphorus pentasulﬁde (P₄S₁₀) to aﬀord 2-

cyanothioacetamide (21) in approximately 30% after recrystal-

lization. To synthesize the 4,6-dimethylthienopyridine com-

pound, 21 was reacted with acetylacetone under basic

conditions to form the 4,6-dimethyl substituted pyridine core

(22) in 67% yield. Subsequent reaction of 22 with ethyl 2-

chloroacetate under basic conditions gave the monocyclic ethyl

ester (23). The monocyclic structure was successfully

converted to the bicyclic form (24) in rapid time (15 min)

in the presence of 1 M aqueous potassium hydroxide. Ester

hydrolysis of 24 using 2 M sodium hydroxide in ethanol,

followed by acidic workup, gave the ﬁrst key thienopyridine

core carboxylic acid (25). To furnish the 5-chloro-6-methoxy-4-

methylthienopyridine scaﬀold, compound 21 was reacted with

methyl acetoacetate and morpholine which initially aﬀorded the

product as the morpholinium salt. The desired thiol (26) was

isolated in the free form by acidiﬁcation with 1 M aqueous

hydrochloric acid. Further reaction with ethyl chloroacetate

gave the monocyclic ethyl ester (27) in good yield (81%). In

order to convert the pyridinone into the corresponding

pyridine alkyl ether, 27 was reacted with iodomethane under

basic conditions. One equivalent of a stronger base, such as 1 M

aqueous potassium hydroxide, was necessary to ensure

complete conversion to the thienopyridine scaﬀold (28). It

was imperative to protect the free amine of the bicycle prior to

chlorination at the 5′ position to avoid N-halogenation.

Phthalic anhydride was employed under acidic conditions

which resulted in the phthalimide-protected product (29) in

good yield. N-Chlorosuccinimide (NCS) was used to install the

a

Reagents and conditions: Compounds 17a−e, 18a−e, and 19 were

all deprotected prior to performing the coupling reaction. (a) BOP

reagent, DIPEA, DMF, rt, 1−12 h, 29−70%.

the 4,4-chlorophenyl-4-hydroxypiperidine analogues (17a−e),

which were deprotected prior to use, with the 4,6-

dimethylthienopyridinecarboxylic acid (25). This was carried

out using a BOP-mediated coupling reaction to give analogues

33a−e in 31−56% yield. Similarly the 2-methoxyphenylpiper-

azine analogues 18a−e following Boc deprotection were

coupled to the carboxylic acid 32 to furnish the target

compounds 34a−e in respectable yield. Compound 19,

following removal of the protecting group, was then

D

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

immediately reacted with 32 to yield the analogue 35, while

compound 25 furnished analogue 36. Reaction of the

carboxylic acid precursor 25 with 18d (Boc deprotected) in

the presence of BOP also aﬀorded compound 37 in respectable

yield.

Table 2. Binding Data at the Dopamine D₂R for 2-

Methoxyphenylpiperazine and 4,4-Chlorophenyl-4-

hydroxypiperidine Full Length Derivatives

a

Pharmacology. Binding Characterization of Privileged

Structures and Linkers. Both 5 and 6 privileged structures have

been evaluated in terms of their ability to bind the D₂R in a

previous study.²⁶We therefore evaluated the binding aﬃnities

of 7, incorporating linker lengths of two to six carbon atoms

and conserving the Boc group for stability purposes and to add

lipophilicity to the end of the linker, with equivalent 5

structures as comparators. As expected, both privileged

structures (5 and 7) demonstrated poor binding aﬃnity (∼10

μM, Table 1), most likely because of their small fragment-like

a

Compounds are tested against [³H]spiperone through competition

binding studies using D_2LCHO cell membranes. Data represent the

mean SEM of three separate experiments performed in duplicate.

Table 1. Binding Data at the Dopamine D₂R for 2-

Methoxyphenylpiperazine and 4,4-Chlorophenyl-4-

a

hydroxypiperidine Linker Derivatives

1). A similar analogue using privileged structure 7 in

combination with the heterocycle present in aripiprazole

(dihydroquinolinone) showed poor binding aﬃnity at the

D₂R;³⁰therefore, this is consistent with our results. Structures

incorporating 5 retained their aﬃnity, and in the cases of the

two and six carbon atom spacers (34a and 34e), their binding

aﬃnities were improved in comparison to their linker

precursors (18a and 18e). On this occasion, a deﬁnite trend

of increasing binding aﬃnity with carbon linker length was

evident, with the six carbon atom linker derivative (34e) as the

standout displaying subnanomolar aﬃnity (K_i= 0.6 nM). This

notable enhancement in aﬃnity is supported by literature that

describes a secondary hydrophobic pocket in the D₂R that is

only reached via longer chain lengths. Interaction with this

secondary pocket has been demonstrated to confer both

increases in aﬃnity and subtype selectivity.^18,26,31While we

acknowledge that the heterocycles added to these two distinct

privileged structures are subtly diﬀerent, the lack of aﬃnity gain

observed for the 4,6-dimethyl substituted thienopyridine

derivatives (33a−e) is unlikely to be attributed to this

diﬀerence.

SAR of 2-Methoxyphenylpiperazine Linkers and Full

Structures: ERK1/2 Phosphorylation Assays. To expand on

our initial results with the 5 and 7 series linkers and full

structures (Tables 1 and 2), we tested these compounds in

functional ERK1/2 phosphorylation (pERK1/2) assays. Ana-

logues 5, 7, 17a−e, 18a−e, 33a−c, and 34a−e at 10 μM were

initially tested in time-course assays to identify potential

agonists and determine the appropriate peak stimulation time

for subsequent construction of concentration−response curves.

Compounds derived from 7 displayed no agonism (data not

shown). As such, while also considering their poor binding

aﬃnities, we did not pursue these compounds for further

characterization. Ligands 5, 18a, and 18e were determined to

be antagonists (based on time-course data and binding data).

They were further analyzed in a pERK1/2 assay for their ability

to antagonize a 10 nM concentration of dopamine, to obtain

values of inhibitory potency (pIC₅₀) for both compounds. We

observed a striking increase in inhibitory potency from the

parent molecule 5 (>1000 nM) to compounds containing a

linker (18a, 93.3 nM; 18e, 3.98 nM). The remainder of

compounds displayed agonism in our time-course experiments

(18b−d). These analogues exhibited partial agonism, consistent

with the observations of Newman et al. using similar linked

a

Compounds were tested against [³H]spiperone through competition

binding studies using D_2LCHO cell membranes. Data represent the

mean SEM of three separate experiments performed in duplicate.

size that confers minimal points for contact with residues in the

D₂R orthosteric pocket. Modest increases in aﬃnity were

observed upon addition of the linkers to 7 (17a−e) with the six

carbon atom linker (17e) conferring a 10-fold increase in

aﬃnity. Comparatively, a more pronounced increase in binding

aﬃnity was evident for the 2-methoxyphenylpiperazine

derivatives (Table 1). Analogues containing a three, four, or

ﬁve carbon atom linker (18b−d) demonstrated enhanced

aﬃnity compared to the two carbon atom linker (18a), with

these analogues displaying a 200-fold increase in aﬃnity

compared to 5 alone. However, there seems to be no clear

linker length dependence for aﬃnity within this set of

compounds 18b−e (n = 3−6).

Binding Characterization of Full-Length Structures. We

then investigated the eﬀect of the addition of a heterocyclic

group to each of the linkers as indicated in our model

pharmacophore (Figure 3). While the substitution pattern on

each of the heterocycles listed in Table 2 is diﬀerent (4,6-

dimethyl vs 5-chloro-6-methoxy-4-methyl), the general core of

the heterocycle (thienopyridine) is the same. The subtle

diﬀerences in these substitutions and their consequences on

pharmacology at the D₂R will be explored at a later stage in our

SAR study. The results in Table 2 showed that analogues based

on privileged structure 7 (33a−e) bind the D₂R with only

moderate aﬃnity (micromolar range) with no signiﬁcant

increase compared to the equivalent linker precursors (Table

E

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 3. Functional ERK1/2 Phosphorylation Assays at the Dopamine D₂Receptor for 2-Methoxyphenylpiperazine Linker

a

Derivatives

b

c

d

compd

n

pIC₅₀SEM (IC₅₀, nM)

pEC₅₀SEM (EC₅₀, nM)

E_maxSEM (% DA)

5

na

2

<6 (>1000)

18a

18b

18c

18d

18e

7.03 0.54 (93.3)

3

9.41 0.15 (0.39)

9.22 0.25 (0.60)

8.84 0.16 (1.43)

31

13

19

1

4

5

6

8.40 0.29 (3.98)

a

b

Data represent the mean SEM of three to four separate experiments performed in duplicate. Values are obtained via interacting ligands with a 10

c

d

nM concentration of dopamine. Values are obtained via concentration−response assays at the appropriate stimulation time. E_maxdata are

represented as a % of the maximal eﬀect of dopamine (DA).

Table 4. Functional ERK1/2 Phosphorylation Assays at the Dopamine D₂R for Compounds Containing the 2-

a

Methoxyphenylpiperazine Privileged Structure

b

c

d

compd

n

pK_BSEM (K_B, nM)

pEC₅₀SEM (EC₅₀, nM)

E_maxSEM (% DA)

5

na

2

6.32 0.18 (479)

6.45 0.17 (355)

6.13 0.16 (741)

6.31 0.14 (490)

34a

34b

34c

34d

34e

3

4

5

8.57 0.66 (2.69)

8.26 0.31 (5.49)

17

14

4

2

6

a

b

Data are the mean of four to ﬁve experiments SEM. Data are based on interaction studies with varying concentrations of dopamine. Data are ﬁt

c

to the Gaddum−Schild model of competitive antagonism using Graph Pad Prism (version 6) with Schild slopes constrained to 1. Values are

d

attained from concentration−response curves at the appropriate concentration and stimulation time. E_maxdata are represented as a % of the

maximal eﬀect of dopamine (DA).

derivatives of 5²⁶(Table 3). The four-carbon atom spacer 18c

has a lower E_max(the maximum stimulation achieved as a

percentage of the maximal eﬀect of dopamine, E_max= 13 1)

compared to the ﬁve-carbon atom spacer 18d (E_max= 19 1).

The three-carbon atom spacer analogue (18b) displayed a

signiﬁcantly higher E_maxof 31 1 (P < 0.05). An enhancement

of aﬃnity at the D₂R was observed for a similar series of

compounds upon addition of alkyl linkers to 5.²⁶We also

observed an enhancement of aﬃnity (Table 1) consistent with

the hypothesis of Newman et al. whereby this increase in

aﬃnity is conferred by additional hydrophobic interactions

between residues within the orthosteric pocket of the D₂R and

the linker. However, we also observe a gain in agonist eﬃcacy

that is dependent on linker length (Table 3). This pattern is

distinct from the lack of eﬀect of the addition of a linker to the

scaﬀold of 5 observed by Newman et al. This discrepancy could

be due to the subtle diﬀerences between the two sets of

compounds, the diﬀerent cell backgrounds used in these

experiments, or the diﬀerent assay end point used to measure

agonism (pERK1/2 of our study vs Gα_o1G protein coupling of

Newman et al).

the functional data for compounds 34a−e at the D₂R. We

identiﬁed compounds 34a−c (linker lengths from two to four

carbon atoms) as antagonists and compounds 34d−e (with

linker lengths of ﬁve and six carbon atoms, respectively) as

partial agonists. This pattern is distinct from the linker only

compounds (18a−e) in which a linker length of three carbon

atoms was optimal in terms of the maximal eﬀect of the agonist.

The partial agonists 34d and 34e displayed the same maximum

response at the D₂R (E_max= 17 4 and 14 2, respectively).

Moreover, the partial agonism observed in the ﬁve carbon atom

2-methoxyphenylpiperazine linker (18d, E_max= 19

1) was

maintained upon incorporation of the heterocyclic group to the

linker (34d, E_max= 17 4). Compounds 34a−c displayed no

gain in functional aﬃnity (pK_B) compared to the parent

molecule 5.

In combination, these data highlight that subtle changes to

the linker length appear to be important in switching between

antagonism and partial agonism. Of note, the pattern is

diﬀerent for linker analogues as opposed to the extended

structures with a heterocycle group suggesting a diﬀerent

binding mode at the D₂R.

Next we investigated if such linker length dependencies in

eﬃcacy are also observed for those compounds with a

heterocyclic group attached to the linker. Table 4 represents

Biased Agonism at the D₂R. The partial agonists 34d and

34e displayed similar binding aﬃnity (Table 2) and functional

activity (Table 4). The 2,3-dichlorophenylpiperazine moiety

F

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 5. Binding Data for Compounds at the D₂R That Incorporate 2,3-Dichlorophenylpiperazine and Have Variations to the

a

Thienopyridine Unit

a

b

Compounds are tested using [³H]spiperone through competition binding studies using D_2LCHO cell membranes. Data represent the mean

SEM of three separate experiments performed in duplicate.

Figure 4. (A) ERK1/2 phosphorylation concentration−response assays. (B) Inhibition of FSK-induced cAMP production. (C) Bias plot

representing the bias factor (ΔΔlog(τ/K_A)) of ligands between pERK1/2 and cAMP signaling pathways. Values are represented in Table 6

and the aromatic bicycle of aripiprazole are separated by a ﬁve-

atom spacer. This is analogous to the ﬁve carbon atom spacer

separating the substituted thienopyridine and the 2-methox-

yphenylpiperazine motif of compound 34d. Therefore, we

decided to further investigate 34d based upon its linker length

and structural similarity to aripiprazole. In particular, given that

a number of studies have described aripiprazole as a biased

agonist,^8,13we extended our characterization of a limited

number of compounds to activity at two assay end points,

ERK1/2 phosphorylation and inhibition of forskolin stimulated

cAMP production. In order to gain some preliminary SAR

around this compound, we synthesized a small number of

compounds related to 34d. We retained the ﬁve-carbon atom

linker for all derivatives and instead focused on subtle structural

G

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Table 6. Calculated Bias Factors and Functional Aﬃnities for Selected Full Agonists and Partial Agonists at the Dopamine D_2L

Receptor

a

b

Data are the mean of n = 4 experiments SEM. ND: no agonist activity detected. Bias is deﬁned as the fold bias relative to the reference agonist

dopamine: (∗) P < 0.05, (∗∗∗) P < 0.001, signiﬁcantly diﬀerent from the reference agonist dopamine determined by a one-way ANOVA, Tukey post

hoc test.

changes to the phenylpiperazine scaﬀold and the aromatic

heterocyclic unit (thienopyridine scaﬀold). We utilized the 2,3-

dichlorophenylpiperazine moiety present in aripiprazole and

attached a ﬁve carbon atom spacer with the Boc appendage to

generate ligand 19. This compound was used as a comparator

to compound 18d which contains the same spacer but with the

2-methoxyphenylpiperazine moiety. We subsequently replaced

the 2-methoxyphenylpiperazine moiety of 34d with the 2,3-

dichlorophenylpiperazine moiety to generate the target

compound 35. We also synthesized a further two analogues

incorporating the simpliﬁed 4,6-dimethylthienopyridine to

explore whether substitution around the thienopyridine core

engendered an inﬂuence on bias. These structural motifs were

combined with 6 and 5 to give 36 and 37, respectively.

The new structures (19 and 35−37) were ﬁrst tested in a

binding assay to measure their biochemical aﬃnity for the D₂R

(Table 5). Compound 6 displayed a micromolar binding

aﬃnity for the D₂R similar to that of the other two privileged

structures (5 and 7). A small but signiﬁcant (3-fold) increase in

aﬃnity was observed upon addition of a heterocyclic unit

(compounds 35 and 36) to the linker derivative of 6,

compound 19 (P < 0.05). There is, however, no diﬀerence in

the binding aﬃnities of compounds 35−37.

pathways.¹¹This transduction coeﬃcient is a descriptor of the

agonist eﬀect of a single pathway that takes into account the

intrinsic eﬃcacy (τ) of the compound and its aﬃnity (K_A) for

the receptor coupled to that signaling pathway. The results

from performing these analyses are summarized in Table 6 and

displayed in Figure 4C. For values of potency (pEC₅₀) and

maximal stimulation (E_max) for all compounds, see Supporting

Information Table 1. We conﬁrmed the biased action of

aripiprazole, which displayed a 400-fold bias toward cAMP

(ΔΔlog(τ/K_A) = 2.65 0.32) when compared to our reference

agonist dopamine. Ligands 34d and 35 both displayed no

signiﬁcant bias to either pathway; however, the linker derivative

of 6 (19) showed a signiﬁcant bias toward cAMP (ΔΔlog(τ/

K_A) = 1.73 0.40). Conversely, the linker derivative of 5 (18d)

did not display signiﬁcantly diﬀerent bias compared to

dopamine. Compound 36 displayed biased agonism toward

cAMP (ΔΔlog(τ/K_A) = 2.04

0.67), a bias proﬁle that is

statistically indistinguishable from that of aripiprazole and

equates to a 100-fold bias toward the cAMP pathway.

Compound 37 did not produce a response in pERK1/2

signaling assays but acted as a partial agonist in the cAMP assay.

As such, we were unable to quantify the bias of this compound

compared to dopamine. If we compare the transduction

coeﬃcient determined in the cAMP assay for 37 (log(τ/K_A)

= 8.45 0.36) with those determined for 36 (changes in the

phenyl substitution), 34d (changes in the thienopyridine

substitution pattern), and 18d (no heterocyclic group), then

compound 37 has a higher log(τ/K_A) value in comparison to

34d and 18d and is similar to 36. This means that compound

37 is at least as eﬃcacious in the cAMP assay as compounds

34d, 18d, and 36 and suggests that compound 37 may also be a

biased ligand. However, it is important to note that of all the

Given that the new set of analogues all exhibited noteworthy

binding aﬃnities (42.6−9.1 nM) for the D₂R, we extended our

characterization to functional assays. Compounds 18d, 19, 34d,

35−37, the reference agonist dopamine, and control ligand

aripiprazole were tested in pERK1/2 and cAMP (inhibition of

forskolin-induced cAMP production) signaling pathways

(Figure 4A and Figure 4B). We utilized a derivation of the

operational model of agonism³²to quantify bias by determining

a transduction coeﬃcient, log(τ/K_A), for all ligands at both

H

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 5. Progression of functionalizing privileged structure 6 with a linker (19) and substituted thienopyridine heterocycles (35, 36).

compounds, 37 displays the lowest maximal eﬀect (E_max= 33

5) in the cAMP assay (Supporting Information Table 1). It is

therefore possible that compound 37 is not a biased ligand

relative to dopamine but that its action in the pERK1/2 assay

was too weak to be detected.

substitution of the thienopyridine core has an important role in

determining the bias proﬁle of such compounds. The inﬂuence

of the addition of a heterocycle upon both ligand aﬃnity and

eﬃcacy suggests that this heterocycle makes an interaction with

the D₂R. Newman et al. provided evidence of a secondary

binding pocket that was exploited by extended phenyl-

piperazine ligands to gain subtype selectivity and modulate

ligand eﬃcacy.²⁶It is tempting to speculate that our novel

extended phenylpiperazine derivatives interact with an

analogous secondary binding pocket and the nature of this

interaction inﬂuences biased agonism.

Table 6 illustrates that the 2,3-dichlorophenylpiperazine

derivative containing the tert-butylcarbamate ﬁve-carbon linker

(19) is biased toward the cAMP pathway. Altering the

substitution pattern on the phenyl ring from 2,3-dichloro to

2-methoxy gives compound 18d that exhibits no bias for either

the cAMP or pERK1/2 end points, thus highlighting that the

substitution pattern on the phenylpiperazine moiety can itself

confer diﬀerences in bias. Substitution around the phenyl ring

on the phenylpiperazine moieties also seems to be important

for bias for compounds that incorporate a tail heterocycle (34d

versus 36). Diﬀerences between 2-methoxyphenylpiperazine-

and 2,3-dichlorophenylpiperazine-substituted analogues could

be attributed to the electronic eﬀects of these substituents, i.e.,

weakly inductively electron-withdrawing (-Cl) compared to

moderately electron-donating (-OCH₃). Tschammer et al. also

provided evidence that diﬀerential substitution of the phenyl

moiety of phenylpiperazine ligands can modulate biased

agonism at the D₂R.¹⁸Such observations are an extension of

those made by Newman et al. whereby measuring G protein

activation by the D₂R revealed that addition of linkers of

diﬀerent length and/or tail heterocyclic groups modulated

agonist eﬃcacy for derivatives of 6 whereas derivatives of 5

displayed no agonism.²⁶However, replacing the 2-methox-

yphenylpiperazine moiety on 34d with 6 to give 35 does not

produce signiﬁcant biased agonism when compared to the

reference agonist dopamine. If we simplify the thienopyridine

core by altering the 5-chloro-6-methoxy-4-methyl substitution

pattern with that of the 4,6-dimethyl substitution to give

compound 36, the biased agonism proﬁle is recapitulated. This

suggests that (1) the signaling pathway bias does not simply

arise from the 2,3-dichlorophenylpiperazine moiety and (2) the

Taking advantage of the partial agonism displayed by the

compounds in this study, we were able to determine values of

functional aﬃnity (K_A) and eﬃcacy (τ) as separate parameters

rather than the composite transduction coeﬃcient (log τ/K_A)

(Table 6). By doing this, we can observe which of these

parameters might be important drivers for signal bias at the

D₂R. First, if we compare our partial agonists in Table 6, there

are no signiﬁcant diﬀerences between the ligands in terms of

functional aﬃnity (pK_A) in the pERK1/2 signaling pathway or

eﬃcacy (τ) at both pathways. However, in the cAMP signaling

pathway, compound 36 exhibits a functional aﬃnity that is

signiﬁcantly diﬀerent from analogues 18d, 34d, and 35 (P <

0.05). As such, the distinct bias proﬁle displayed by compound

36 is largely driven through a change in functional aﬃnity at the

cAMP assay. This is consistent with the ﬁndings of an SAR

exploration of biased agonsim at the D₂R focused on

cariprazine.¹⁹It is also of interest to compare the diﬀerent

parameters of compound 36 with that of aripiprazole. Both

have very similar scaﬀolds and display a bias proﬁle that is not

statistically diﬀerent. Although both have a similar transduction

coeﬃcient in the cAMP assay, compound 36 has a 10-fold

higher functional aﬃnity but a 10-fold lower eﬃcacy at this

pathway compared to aripiprazole. Similarly, in the pERK1/2

end point, compound 36 displays a 10-fold higher functional

aﬃnity than aripirazole but a smaller value of eﬃcacy (τ).

I

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

As previously mentioned in the Introduction, aripiprazole has

a high aﬃnity at serotonin receptors, particularly 5-HT_1A.⁷

Therefore, in light of the structural similarities of aripiprazole

with some of our analogues, particularly the biased analogue

(36), it is worth mentioning that these ligands may also have

high aﬃnities at serotonin receptors. As such, the action of

these ligands may have an unprecedented complexity of action

that should be considered in the characterization of bias not

only at the D₂R but also at serotonin receptors as potential dual

dopamine−serotonin biased ligands.

We have revealed that the SAR of biased agonism at the D₂R

is complex and, depending on the functional scaﬀold explored,

cannot be deﬁned by a single functional group or trans-

formation.¹⁹The diﬀerential bias observed for compounds 18d

and 19 is governed only by the nature of the privileged

structure, i.e., 5 or 6, with the latter conferring bias toward the

cAMP assay. For the extended structures, it appears that the 4-

methyl-5-chloro-6-methoxy substitution pattern on the thieno-

pyridine scaﬀold (34d and 35) engenders no bias irrespective

of the privileged structure used. However, the 4,6-dimethyl

substitution pattern on the thienopyridine motif appears to

confer bias regardless of the privileged structure implemented

(36) (Figure 5). Therefore, for these compounds, this suggests

that the bias is driven primarily by the thienopyridine

heterocycle and more speciﬁcally the substitution pattern

upon this heterocycle whereby the inﬂuence of the privileged

structure is overridden.

EXPERIMENTAL SECTION

■

Chemistry. All solvents and chemicals were purchased from

standard suppliers and were used without any further puriﬁcation. H

1

NMR and ¹³C NMR spectra were acquired at 400.13 and 100.62 MHz,

respectively, on a Bruker Advance III 400 MHz Ultrashield Plus NMR

spectrometer using TOPSPIN 2.1 software. Chemical shifts (δ) for all

¹H spectra are reported in parts per million (ppm) using

tetramethylsilane (TMS, 0 ppm) as the reference. The data for all

spectra are reported in the following format: chemical shift (δ),

(multiplicity, coupling constants J (Hz), integral), where the

multiplicity is deﬁned as s = singlet, d = doublet, t = triplet, q =

quartet, p = pentet, and m = multiplet. For ¹³C NMR spectra C =

quaternary carbon, CH = methine carbon, CH₂= methylene carbon,

and CH₃= methyl carbon.

The purity and retention time of ﬁnal products were determined on

an analytical reverse-phase HPLC system ﬁtted with a Luna C8 (2)

100 Å column (50 mm × 4.60 mm, 5 μm) using a binary solvent

system: solvent A of 0.1% TFA/H₂O; solvent B of 0.1% TFA/80%

MeOH/20% H₂O. Gradient elution was achieved using 100% A for 10

min, 20% A and 80% B over 2 min, and 100% A over 10 min at a ﬂow

rate of 1 mL/min monitored at 214 nm using a Waters 996

photodiode array detector.

Thin layer chromatography (TLC) was carried out routinely on

silica gel 60F₂₅₄precoated plates (0.25 mm, Merck). Flash column

chromatography was carried out using Merck silica gel 60, 230−400

mesh ASTM. Melting points were determined using an electronic

MP50 melting point system by Mettler Toledo analytical 2009 and are

uncorrected.

Synthesis of Linkers. General Procedure A. The required

aminoalkanol 12c−e (1 equiv) was added to 10−30 mL of

dichloromethane. Di-tert-butyl dicarbonate (1 equiv) was then added

followed by dropwise addition of triethylamine (1.1 equiv) at room

temperature. The reaction mixture was stirred for 1 h. After this time,

the solvent was removed in vacuo and the resulting oil puriﬁed via

column chromatography (ethyl acetate).

tert-Butyl (4-Hydroxybutyl)carbamate (13c). 12c (0.41 mL,

4.49 mmol) and di-tert-butyl dicarbonate (979 mg, 4.49 mmol) were

added to dichloromethane (10 mL). Triethylamine (0.69 mL, 4.94

mmol) was added dropwise and the mixture stirred for 1 h at room

temperature and worked up according to general procedure A.

CONCLUSION

■

We explored the determinants of eﬃcacy and aﬃnity for three

distinct privileged structures for the D₂R. The privileged

structure 7 showed little gain in aﬃnity in both the precursors

containing the linkers (17a−e) and also the extended structures

incorporating an additional thienopyridine heterocycle (33a−

e). Focusing on the privileged structure 5, we conﬁrm that the

addition of linker groups resulted in a gain in aﬃnity (18a−e,

Table 1) and in eﬃcacy (Table 3). In particular, we observed a

linker length relationship with eﬃcacy with three to ﬁve carbon

atoms engendering partial agonism (18b−d). Inclusion of a

heterocyclic structure (34a−e) resulted in a diﬀerent pattern of

linker length dependent eﬃcacy, as only extended compounds

containing linker lengths of ﬁve and six carbon atoms (34d−e)

exhibited partial agonism. It is therefore likely that compounds

that possess only the linker component bind the D₂R in a

diﬀerent orientation to the extended structures incorporating a

thienopyridine motif. Extending our studies around the novel

partial agonist 34d, we generated a focused set of compounds

that explored the SAR around the two privileged structures 5

and 6. Of note is that not only can the type of substitution on

the phenylpiperazine inﬂuence bias, the substitution on the

thienopyridine scaﬀold can have dramatic eﬀects on bias to the

point whereby the type of privileged structure used is no longer

the principal driver of bias. This observation is consistent with

the interaction of the thienopyridine scaﬀold with a secondary

pocket previously identiﬁed to interact with other extended

D₂R ligands. Our ﬁndings suggest that the nature of this

interaction confers bias at the D₂R receptor. Our novel D₂R

partial agonists 35 and 36 display a similar aﬃnity for the D₂R

as aripiprazole but exhibit distinct bias proﬁles. As such, they

may represent useful tools to explore the contribution of biased

agonism to antipsychotic eﬃcacy.

1

Puriﬁcation gave the product as a pale yellow oil (809 mg, 95%). H

NMR (CDCl₃): δ 1.44 (s, 9H), 1.53−1.62 (m, 4H), 2.52 (br s, 1H,

OH), 3.14 (m, 2H), 3.65 (m, 2H), 4.78 (br s, 1H, NH). ¹³C NMR

(CDCl₃): δ 26.7 (CH₂), 28.5 (CH₃), 29.8 (CH₂), 40.4 (CH₂), 62.3

(CH₂), 79.2 (C), 156.3 (C).

General Procedure B. The required bromoalkanamine 15a,b (1.1

equiv) was added to 20−30 mL of dichloromethane. Di-tert-butyl

dicarbonate (1 equiv) was then added followed by dropwise addition

of triethylamine (1.1 equiv) at room temperature. The reaction

mixture was stirred for 1−1.5 h. After this time, an additional 20 mL of

dichloromethane was added and the reaction mixture was washed with

3 × 50 mL portions of saturated potassium hydrogen sulfate solution.

The organic layer was then dried over anhydrous sodium sulfate,

ﬁltered and the solvent removed to aﬀord the crude product. Further

puriﬁcation of the product was achieved via column chromatography

(9:1 petroleum spirits/ethyl acetate).

tert-Butyl (2-Bromoethyl)carbamate (16a). 15a (1.50 g, 7.32

mmol) and di-tert-butyl dicarbonate (1.45 g, 6.66 mmol) were added

to dichloromethane (20 mL). Triethylamine (1.02 mL, 7.32 mmol)

was added dropwise and the mixture stirred for 1.5 h at room

temperature and worked up according to general procedure B to give

1

the product as a colorless oil (413 mg, 28%). H NMR (CDCl₃): δ

1.45 (s, 9H), 3.45 (m, 2H), 3.53 (m, 2H), 5.01 (br s, 1H, NH). ¹³C

NMR (CDCl₃): δ 28.5 (CH₃), 32.9 (CH₂), 42.5 (CH₂), 80.0 (C),

155.7 (C).

Coupling of Linkers with Privileged Structures 5, 6, and 7.

General Procedure C. To a solution of 5 or 7 (1 equiv) in

acetonitrile (30 mL) at room temperature under N₂were added

sodium iodide (1 equiv) and N,N-diisopropylethylamine (2 equiv).

J

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

The required Boc protected amines 16a,b (1.1−1.5 equiv) were

dissolved in 1−2 mL of acetonitrile and added slowly to the reaction

mixture which was then heated at reﬂux for 24 h. The reaction was

then concentrated in vacuo and the resulting residue partitioned

between ethyl acetate (50 mL) and 1 M potassium carbonate (50 mL).

The organic layer was removed, and the aqueous layer was further

extracted with 2 × 50 mL portions of ethyl acetate. The organic

fractions were pooled, washed with water (50 mL), brine (50 mL),

dried over anhydrous sodium sulfate, and ﬁltered and the solvent was

removed to aﬀord the crude product. Puriﬁcation is achieved by

column chromatography as indicated.

General Procedure D. The Boc protected aminoalkanol 13c−e (1

equiv) was added to dichloromethane (5 mL). Triethylamine (3

equiv) was added and the reaction mixture cooled to 0 °C via an ice

bath. Methanesulfonyl chloride (2 equiv) was then added slowly, and

stirring continued for 30 min at 0 °C. After this time the temperature

was warmed to room temperature and stirring continued for 1−12 h.

The reaction mixture is then diluted with dichloromethane (10 mL),

washed with water (3 × 15 mL), dried over anhydrous sodium sulfate,

ﬁltered and the solvent removed in vacuo. The mesylates were then

reacted immediately with 5, 6, or 7 (2−2.5 equiv) in acetonitrile (5

mL) and heated at reﬂux for 8−24 h. The solvent was then removed in

vacuo and the resulting residue redissolved in dichloromethane and

washed with saturated sodium bicarbonate (3 × 20 mL). The organic

layer was then dried over anhydrous sodium sulfate, ﬁltered, and

evaporated to dryness to give the crude product. Further puriﬁcation

was achieved by column chromatography as indicated.

4H), 3.22 (m, 2H), 3.86 (s, 3H), 5.49 (br s, 1H, NH), 6.85−7.02 (m,

4H). ¹³C NMR (CDCl₃): δ 26.5 (CH₂), 28.6 (CH₃), 40.2 (CH₂), 50.8

(CH₂), 53.6 (CH₂), 55.5 (CH₃), 57.1 (CH₂), 78.9 (C), 111.3 (CH),

118.3 (CH), 121.1 (CH), 123.0 (CH), 141.4 (C), 152.4 (C), 156.2

(C). HPLC purity (λ = 214 nm): 99%, t_R= 7.74 min. HRMS (ESI)

TOF (m/z): [M + H]⁺350.2444 calcd for C₁₉H₃₁N₃O₃; found [M +

H]⁺350.2455.

tert-Butyl (5-(4-(2-Methoxyphenyl)piperazin-1-yl)pentyl)-

carbamate (18d). 5 (683 mg, 2.99 mmol) was dissolved in

acetonitrile (10 mL) and triethylamine (0.21 mL, 1.49 mmol) and

stirred for 5 min. 14d (420 mg, 1.49 mmol) was then added dropwise

to the reaction mixture which was reﬂuxed for 24 h. Workup

proceeded as described in general procedure D and the crude product

was puriﬁed by column chromatography (10% chloroform/methanol−

1

100% acetonitrile) to give a yellow/orange oil (334 mg, 59%). H

NMR (CDCl₃): δ 1.36 (m, 2H), 1.44 (s, 9H), 1.47−1.59 (m, 4H),

2.43 (m, 2H), 2.68 (m, 4H), 3.09−3.15 (m, 6H), 3.86 (s, 3H), 4.63

(br s, 1H, NH), 6.84−7.01 (m, 4H). ¹³C NMR (CDCl₃): δ 24.8

(CH₂), 26.4 (CH₂), 28.5 (CH₃), 40.6 (CH₂), 50.5 (CH₂), 53.4 (CH₂),

55.4 (CH₃), 58.6 (CH₂), 79.1 (C), 111.2 (CH), 118.3 (CH), 121.1

(CH), 123.0 (CH), 141.3 (C), 152.3 (C), 156.1 (C). HPLC purity (λ

= 214 nm): 98%, t_R= 8.18 min. HRMS (ESI) TOF (m/z): [M + H]⁺

378.2757 calcd for C₂₁H₃₅N₃O₃; found [M + H]⁺378.2768.

tert-Butyl (5-(4-(2,3-Dichlorophenyl)piperazin-1-yl)pentyl)-

carbamate (19). 6 (1.09 g, 4.07 mmol) was added to acetonitrile

(15 mL) followed by 14d (572 mg, 2.03 mmol) and triethylamine

(0.57 mL, 4.07 mmol) and the mixture reﬂuxed for 19 h. Workup

proceeded as described in general procedure D and the crude product

was puriﬁed by column chromatography (chloroform/methanol 99:1)

tert-Butyl (2-(4-(4-Chlorophenyl)-4-hydroxypiperidin-1-yl)-

ethyl)carbamate (17a). 7 (300 mg, 1.21 mmol), sodium iodide

(181 mg, 1.21 mmol), and N,N-diisopropylethylamine (0.21 mL, 1.21

mmol) were added to acetonitrile (30 mL) at room temperature under

N₂. 16a (350 mg, 1.56 mmol) was then combined with acetonitrile (2

mL) and added slowly to the reaction mixture (full amount added after

1.5 h). After 6 h of reﬂux, the reaction mixture was worked up as

described in general procedure C to give the product as a white foam

1

to give a yellow oil (431 mg, 51%). H NMR (CDCl₃): δ 1.36 (m,

2H), 1.44 (s, 9H), 1.47−1.59 (m, 4H), 2.41 (m, 2H), 2.63 (m, 4H),

3.07 (m, 4H), 3.12 (m, 2H), 4.62 (br s, 1H, NH), 6.96 (m, 1H), 7.12−

7.17 (m, 2H). ¹³C NMR (CDCl₃): δ 24.8 (CH₂), 26.6 (CH₂), 28.5

(CH₃), 30.1 (CH₂), 40.6 (CH₂), 51.4 (CH₂), 53.4 (CH₂), 58.6 (CH₂),

79.1 (C), 118.7 (CH), 124.6 (CH), 127.5 (CH), 127.6 (C), 134.1

1

(C), 151.4 (C), 156.1 (C). HPLC purity (λ = 214 nm): 98%, t_R

=

(348 mg, 81%). Mp: 129.8−131.1 °C. H NMR (CDCl₃): δ 1.46 (s,

10.43 min. HRMS (ESI) TOF (m/z): [M + H]⁺416.1872 calcd for

C₂₀H₃₁Cl₂N₃O₂; found [M + H]⁺416.1870.

9H), 1.71−1.75 (m, 2H), 2.11 (td, J 13.2, 4.3 Hz, 2H), 2.46−2.55 (m,

4H), 2.79 (m, 2H), 3.26 (m, 2H), 5.02 (br s, 1H, NH), 7.31 (m, 2H),

7.44 (m, 2H). ¹³C NMR (CDCl₃): δ 28.6 (CH₃), 38.5 (CH₂), 40.7

(CH₂), 49.4 (CH₂), 57.4 (CH₂), 71.1 (C), 79.4 (C), 126.3 (CH),

128.6 (CH), 133.0 (C), 147.0 (C), 156.1 (C). HPLC purity (λ = 214

nm): 97%, t_R= 7.68 min. HRMS (ESI) TOF (m/z): [M + H]⁺

355.1788 calcd for C₁₈H₂₇ClN₂O₃; found [M + H]⁺355.1796.

tert-Butyl (6-(4-(4-Chlorophenyl)-4-hydroxypiperidin-1-yl)-

hexyl)carbamate (17e). 7 (292 mg, 1.38 mmol) was added to

acetonitrile (5 mL) followed by dropwise addition of 14e (204 mg,

0.691 mmol). Reﬂux occurred overnight, and then the reaction

mixture was worked up as described in general procedure D to give the

product as a white foam (183 mg, 64%). Mp: 64.4−65.7 °C. ¹H NMR

(CDCl₃): δ 1.32−1.35 (m, 4H), 1.44 (s, 9H), 1.48 (m, 2H), 1.54 (m,

2H), 1.73 (m, 2H), 2.15 (m, 2H), 2.39−2.45 (m, 4H), 2.84 (m, 2H),

3.1 (m, 2H), 4.53 (br s, 1H, NH), 7.31 (m, 2H), 7.45 (m, 2H). ¹³C

NMR (CDCl₃): δ 26.8 (CH₂), 27.0 (CH₂), 27.4 (CH₂), 28.6 (CH₃),

30.2 (CH₂), 38.5 (CH₂), 40.7 (CH₂), 49.6 (CH₂), 58.9 (CH₂), 71.2

(C), 79.2 (C), 126.3 (CH), 128.6 (CH), 132.9 (C), 147 (C), 156.2

(C). HPLC purity (λ = 214 nm): 95%, t_R= 8.77 min. HRMS (ESI)

TOF (m/z): [M + H]⁺411.2414 calcd for C₂₂H₃₅ClN₂O₃; found [M

+ H]⁺411.2429.

Synthesis of Substituted Thienopyridine Scaﬀolds. 2-

Cyanothioacetamide (21). P₄S₁₀(1.00 g, 4.52 mmol) and 2-

cyanoacetamide (20, 1.00 g, 11.89 mmol) were dissolved in ethyl

acetate (12 mL). The mixture was heated at reﬂux for 1.5−4 h. The

reaction mixture was then cooled to room temperature and the orange

solid ﬁltered oﬀ and washed with ethyl acetate (100 mL). The solid

was recrystallized from toluene to give sharp yellow needles (389 mg,

∼30%). ¹H NMR (DMSO-d₆): δ 3.98 (s, 2H), 9.48 (s, 1H, NHa), 9.83

(s, 1H, NHb). ¹³C NMR (DMSO-d₆): δ 33.9 (CH₂), 116.5 (C), 194.5

(C).

4,6-Dimethyl-2-thioxo-1,2-dihydropyridine-3-carbonitrile

(22). 2-Cyanothioacetamide (21, 1.02 g, 10.2 mmol) and

acetylacetone (1.04 mL, 10.2 mmol) were suspended in methanol

(10 mL) and warmed until dissolved. Potassium hydroxide (0.75 g)

dissolved in methanol (50 mL) was added to the solution, which was

stirred for 2 h and then heated at reﬂux for a further 2 h. The mixture

was cooled to room temperature and acidiﬁed with 6 M aqueous

hydrochloric acid to pH 2−3, and a precipitate resulted. The

precipitous mixture was then cooled overnight, ﬁltered, and washed

with water. The product was recrystallized from methanol to aﬀord the

title compound as yellow needles (1.13 g, 67%). Mp: 266.9−270 °C.

¹H NMR (DMSO-d₆): δ 2.34 (s, 3H), 2.35 (s, 3H), 6.69 (s, 1H),

tert-Butyl (3-(4-(2-Methoxyphenyl)piperazin-1-yl)propyl)-

carbamate (18b). 5 (200 mg, 0.87 mmol) and N,N-diisopropylethyl-

amine (0.15 mL, 1.87 mmol) were added to acetonitrile (5 mL) and

stirred at room temperature under N₂for 2 min to generate the free

amine. Sodium iodide (131 mg, 0.87 mmol) and additional N,N-

diisopropylethylamine (0.15 mL, 1.87 mmol) were then also added.

16b (280 mg, 1.18 mmol) was then combined with acetonitrile (2

mL) and added slowly to the reaction mixture. After heating at reﬂux

for 24 h, the reaction mixture was worked up as described in general

procedure C and puriﬁed via a silica plug (ethanol) to give the product

13.83 (br s, 1H). ¹³C NMR (DMSO-d₆): δ 18.8 (CH₃), 20.8 (CH₃),

113.4 (C), 114.9 (CH), 116.3 (C), 152.6 (C), 156.9 (C), 177.5 (C).

HPLC purity (λ = 214 nm): 99%, t_R= 5.56 min.

Ethyl 2-((3-Cyano-4,6-dimethylpyridin-2-yl)thio)acetate

(23). Compound 22 (1.00 g, 6.09 mmol) and triethylamine (0.93

mL, 6.70 mmol) were added to N,N-dimethylformamide (15 mL) and

cooled to 0 °C. Ethyl-2-chloroacetate (0.65 mL, 6.09 mmol) was

added to 1 mL of N,N-dimethylformamide and then added dropwise

to the reaction mixture. The mixture was then warmed to room

temperature for 3.5 h, and a yellow precipitate resulted. The solvent

1

as a colorless oil (286 mg, 93%). H NMR (CDCl₃): δ 1.44 (s, 9H),

1.70 (p, J 6.6 Hz, 2H), 2.49 (t, J 6.8 Hz, 2H), 2.65 (m, 4H), 3.10 (m,

K

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

was then removed and the resulting residue partitioned between

dichloromethane (50 mL) and water (50 mL). The organic layer was

removed, and the aqueous layer was then washed with further 3 × 50

mL portions of dichloromethane. The organic fractions were

combined, dried over anhydrous sodium sulfate, and ﬁltered, and

the solvent was removed to leave a yellow solid. Further puriﬁcation

using a silica plug aﬀorded the ﬁnal compound as a white solid (1.30 g,

Ethyl 3-Amino-6-methoxy-4-methylthieno[2,3-b]pyridine-2-

carboxylate (28). Compound 27 (1.52 g, 6.02 mmol) was added to

N,N-dimethylformamide (20 mL) followed by potassium carbonate

(1.25 g, 9.04 mmol) and methyl iodide (0.41 mL, 6.63 mmol). The

reaction mixture was then stirred at room temperature for 2 h before

another 0.5−1 equiv of the methyl iodide was added. After 4 h, 1 M

aqueous potassium hydroxide is added and the reaction mixture stirred

for 15 min. Water (15 mL) is then added and the resulting white

precipitate is ﬁltered and dried overnight over high vacuum (0.91 g,

1

86%). H NMR (DMSO-d₆): δ 1.17−1.21 (t, J 7.1 Hz, 3H), 2.41 (s,

3H), 2.43 (s, 3H), 4.07 (s, 2H), 4.09- 4.14 (q, J 7.1 Hz, 2H), 7.12 (s,

1H). ¹³C NMR (DMSO-d₆): δ 14.1 (CH₃), 19.6 (CH₃), 24.2 (CH₃),

32.2 (CH₂), 61.0 (CH₂), 103.6 (C), 114.6 (C), 120.6 (CH), 152.6

(C), 159.7 (C), 161.4 (C), 168.5 (C).

1

58%). Mp: 149.5−150.2 °C. H NMR (CDCl₃): δ 1.38 (t, J 7.1 Hz,

3H), 2.68 (d, J 0.9 Hz, 3H), 3.98 (s, 3H), 4.33 (q, J 7.1 Hz, 2H), 6.11

(br s, 2H, NH₂), 6.43 (m, 1H). ¹³C NMR (CDCl₃): δ 14.7 (CH₃),

20.4 (CH₃), 54.0 (CH₃), 60.4 (CH₂), 95.2 (C), 110.0 (CH), 119.4

(C), 145.9 (C), 149.5 (C), 160.6 (C),164.6 (C), 165.9 (C).

Ethyl 3-Amino-4,6-dimethylthieno[2,3-b]pyridine-2-carbox-

ylate (24). 23 (1.29 g, 5.15 mmol) was dissolved in 4−5 mL of N,N-

dimethylformamide followed by 1 M potassium hydroxide (5.15 mL,

5.15 mmol). The reaction mixture was stirred for 15 min at room

temperature. Quenching with water (15 mL) caused precipitation of

the product that was then ﬁltered to leave a yellow crystalline solid

Ethyl 3-(1,3-Dioxoisoindolin-2-yl)-6-methoxy-4-

methylthieno[2,3-b]pyridine-2-carboxylate (29). To a solution

of 28 (65 mg, 0.244 mmol) in acetic acid (5 mL), a total of 4 equiv of

phthalic anhydride (150 mg, 1.01 mmol) was added over the 20 h

period. The reaction was stopped by cooling the mixture to room

temperature and then over an ice bath to initiate precipitation of the

ﬁnal product as a white solid (53 mg, 55%). ¹H NMR (CDCl₃): δ 1.11

(t, J 7.1 Hz, 3H), 2.37 (s, 3H), 4.02 (s, 3H), 4.20 (q, J 7.1 Hz, 2H),

6.59 (s, 1H), 7.83 (m, 2H), 8.00 (m, 2H). ¹³C NMR (CDCl₃): δ 13.9

(CH₃), 18.5 (CH₃), 54.2 (CH₃), 61.7 (CH₂), 112.2 (CH), 124.2

(CH), 124.7 (C), 125.4 (C), 127.6 (C), 132.4 (C), 134.7 (CH), 145.9

(C), 159.1 (C), 161.1 (C), 164.5 (C), 167.6 (C).

1

(1.20 g, 93%). Mp: 154.7−155.6 °C. H NMR (CDCl₃): δ 1.39 (t, J

7.1 Hz, 3H), 2.58 (s, 3H), 2.72 (s, 3H), 4.34 (q, J 7.1 Hz, 2H), 6.14

(br s, 2H, NH₂), 6.84 (s, 1H). ¹³C NMR (CDCl₃): 14.6 (CH₃), 20.3

(CH₃), 24.5 (CH₃), 60.6 (CH₂), 91.2 (C), 121.9 (CH), 122.5 (C),

143.7 (C), 149.1 (C), 159.8 (C), 161.5 (C), 165.9 (C). HPLC purity

(λ = 214 nm): 99%, t_R= 8.38 min.

3-Amino-4,6-dimethylthieno[2,3-b]pyridine-2-carboxylic

Acid (25). To a solution of 24 (1.16 g, 4.62 mmol) in ethanol (15

mL) was added an aqueous solution of sodium hydroxide (2 M, 8

mL), and the mixture was reﬂuxed for 4 h. The solvent was removed in

vacuo and the resulting residue dissolved in water (20 mL). Excess

aqueous hydrochloric acid (1 M, 10 mL) was added which caused

precipitation of the desired compound. Filtration of the precipitate

gave the title compound as a yellow solid (856 mg, 83%). Mp: 151.9−

152.9 °C. ¹H NMR (DMSO-d₆): δ 2.50 (s, 3H), 2.72 (s, 3H), 7.05 (s,

1H). ¹³C NMR (DMSO-d₆): δ 19.9 (CH₃), 23.4 (CH₃), 95.2 (C),

121.9 (CH), 122.9 (C), 145.8 (C), 149.3 (C), 158.8 (C), 159.0 (C),

166.5 (C).

2-Mercapto-4-methyl-6-oxo-1,6-dihydropyridine-3-carboni-

trile (26). A mixture of 21 (8.00 g, 79.89 mmol) and morpholine

(6.99 mL, 79.9 mmol) in ethanol (50 mL) was warmed until all

components dissolved. Methyl acetoacetate (8.59 mL, 79.9 mmol) was

then added and the mixture heated at reﬂux for 8 h. After cooling to

room temperature, the mixture was ﬁltered and the residue washed

with cold dichloromethane to give morpholinium 3-cyano-4-methyl-6-

oxo-1,6-dihydropyridine-2-thiolate. The desired thiol was isolated by

dissolving the thiolate in a minimum amount of water and then

acidifying with excess 1 M aqueous hydrochloric acid, which caused

precipitation of the molecule. The mixture was refrigerated overnight

and then ﬁltered to give the title compound as a pale yellow powder

(6.83 g, 51%). Mp: 267−268.9 °C. LRMS (ESI) m/z: 165.1 [M − H]⁺

(100%).

Ethyl 5-Chloro-3-(1,3-dioxoisoindolin-2-yl)-6-methoxy-4-

methylthieno[2,3-b]pyridine-2-carboxylate (30). Compound 29

(500 mg, 1.26 mmol), N-chlorosuccinimide (337 mg, 2.52 mmol), and

3 drops of concentrated HCl were reﬂuxed in acetonitrile (10 mL) for

1.5 h. The solvent was then removed in vacuo and the resulting residue

redissolved in chloroform (50 mL) and washed with water (3 × 20

mL) and brine (50 mL). The organic layer was then dried over

anhydrous sodium sulfate, ﬁltered, and evaporated to dryness to give

1

the compound as a white solid (506 mg, 93%). H NMR (CDCl₃): δ

1.12 (t, J 7.1 Hz, 3H), 2.49 (s, 3H), 4.13 (s, 3H), 4.21 (q, J 7.1 Hz,

2H), 7.85 (m, 2H), 8.01 (m, 2H). ¹³C NMR (CDCl₃): δ 13.9 (CH₃),

14.7 (CH₃), 55.4 (CH₃), 61.9 (CH₂), 118.8 (C), 124.3 (CH), 125.1

(C), 127.0 (C), 127.4 (C), 132.4 (C), 134.9 (CH), 143.0 (C), 155.9

(C), 159.5 (C), 160.9 (C), 167.6 (C).

Ethyl 3-Amino-5-chloro-6-methoxy-4-methylthieno[2,3-b]-

pyridine-2-carboxylate (31). A mixture of 30 (144 mg, 0.33

mmol) and hydrazine monohydrate (65 μL, 1.34 mmol) in ethanol

(10 mL) was reﬂuxed for 3 h. Upon cooling to room temperature, the

white precipitate was ﬁltered oﬀ and washed with cold chloroform (5

mL). The ﬁltrate was then collected and diluted with a further 20 mL

of chloroform and washed with water (3 × 20 mL), dried over

anhydrous sodium sulfate, and ﬁltered and the solvent removed in

1

vacuo to give the product as a pale yellow solid (85 mg, 84%). H

NMR (CDCl₃): δ 1.38 (t, J 7.1 Hz, 3H), 2.83 (s, 3H), 4.08 (s, 3H),

4.33 (q, J 7.1 Hz, 2H), 6.14 (br s, 2H, NH₂). ¹³C NMR (CDCl₃): δ

14.6 (CH₃), 16.4 (CH₃), 55.2 (CH₃), 60.6 (CH₂), 116.3 (C), 120.1

(C), 142.9 (C), 149.1 (C), 157.3 (C), 159.5 (C), 165.7 (C), 168.4

(C).

Ethyl 2-((3-Cyano-4-methyl-6-oxo-1,6-dihydropyridin-2-yl)-

thio)acetate (27). Compound 26 (200 mg, 1.2 mmol) and

triethylamine (0.18 mL, 1.32 mmol) were added to N,N-

dimethylformamide (8 mL) and cooled to 0 °C. Ethyl 2-chloroacetate

(0.13 mL, 1.2 mmol) was added to 1 mL of N,N-dimethylformamide

and then added dropwise to the reaction mixture. The mixture was

then warmed to room temperature and stirred for 5 h. The solvent was

then removed and the resulting residue dissolved in dichloromethane

(50 mL) and partitioned between water (50 mL). The aqueous layer

was then washed with further 3 × 50 mL portions of dichloromethane,

and the organic layers were combined, dried over anhydrous sodium

sulfate, and ﬁltered. The solvent was removed to leave a yellow solid.

Further puriﬁcation by column chromatography aﬀorded the ﬁnal

3-Amino-5-chloro-6-methoxy-4-methylthieno[2,3-b]-

pyridine-2-carboxylic Acid (32). Compound 31 (236 mg, 0.783

mmol) was was added to a 50:50 mix of ethanol and 2 M sodium

hydroxide. Reﬂux occurs for 1.5 h before the mixture was cooled to

room temperature. An excess of 2 M HCl is added to cause

precipitation of the product which is then ﬁltered and washed with a

small amount of cold water, giving the product as a beige-yellow solid

1

(179 mg, 84%). H NMR (DMSO-d₆): δ 2.77 (s, 3H), 3.96 (s, 3H),

6.41 (br s, 2H, NH₂). LCMS (ESI) m/z: 273.0 [M + H]⁺(90%),

275.0 (30%).

1

compound as a white solid (247 mg, 81%). Mp: 102.7−103.5 °C. H

General Procedure E for Amide Coupling Reactions Using

BOP Reagent. The required carboxylic acids (1 equiv) were added to

N,N-dimethylformamide (10 mL), followed by N,N-diisopropylethyl-

amine (1.05 equiv) and BOP (1.05 equiv) under N₂at room

temperature and stirred for 5−10 min. The amine (1 equiv) was

subsequently added and the reaction mixture stirred at room

NMR (CDCl₃): δ 1.28 (t, J 7.1 Hz, 3H), 2.38 (d, J 0.9 Hz, 3H), 3.93

(s, 2H), 4.23 (q, J 7.1 Hz, 2H), 6.33−6.35 (m, 1H), 11.03 (br s, 1H,

NH). ¹³C NMR (CDCl₃): δ 14.1 (CH₃), 20.9 (CH₃), 34.1 (CH₂),

62.8 (CH₂), 98.2 (C), 113.9 (CH), 114.9 (C), 153.4 (C), 154.3 (C),

163.6 (C), 169.6 (C).

L

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

temperature for 2−3 h. The solvent was then removed in vacuo and

the resulting residue partitioned between dichloromethane (30 mL)

and sodium bicarbonate (50 mL). The organic layer was removed, and

the aqueous phase was further extracted with 3 × 20 mL portions of

dichloromethane. The organic fractions were combined, washed with

water (50 mL), brine (50 mL), dried over anhydrous sodium sulfate,

ﬁltered, and concentrated to yield the crude product as an oily residue.

To remove excess HMPA, crude products are dissolved in ethyl

acetate and washed with 3 × 50 mL portions of 2 M brine. Puriﬁcation

of the product was performed by column chromatography and/or

recrystalisation as indicated.

3-Amino-N-(2-(4-(4-chlorophenyl)-4-hydroxypiperidin-1-yl)-

ethyl)-4,6-dimethylthieno[2,3-b]pyridine-2-carboxamide

(33a). Carboxylic acid 25 (50 mg, 0.225 mmol), N,N-diisopropylethyl-

amine (0.041 mL, 0.24 mmol), and BOP (104 mg 0.24 mmol) were

added to N,N-dimethylformamide (2 mL). 17a (92 mg, 0.28 mmol)

was dissolved in 2 mL of N,N-dimethylformamide and N,N-

diisopropylethylamine (0.078 mL, 0.450 mmol) and then added to

the reaction mixture. After the mixture was stirred for 16 h at room

temperature, the crude product was worked up as per general

procedure E to give an amber solid (32 mg, 31%). Mp: 159−163 °C.

¹H NMR (DMSO-d₆): δ 1.58 (d, J 12.1 Hz, 2H), 1.91 (td, J 12.9, 4.1

Hz, 2H), 2.44−2.53 (m, 7H), 2.71−2.72 (m, 5H), 3.36 (m, 2H), 4.9

(s, 1H, OH), 6.78 (s, 2H, NH₂), 7.02 (s, 1H), 7.36 (m, 2H), 7.50 (m,

2H), 7.55 (t, J 5.5 Hz, 1H, NH). ¹³C NMR (DMSO-d₆): δ 19.7

(CH₃), 23.8 (CH₃), 36.7 (CH₂), 37.9 (CH₂), 49.1 (CH₂), 57.2 (CH₂),

69.4 (C), 97.6 (C), 121.8 (CH), 123.2 (C), 126.8 (CH), 127.7 (CH),

130.8 (C), 144.5 (C), 147.5 (C), 149.1 (C), 158.4 (C), 158.5 (C),

165.2 (C). HPLC purity (λ = 214 nm): 98%, t_R= 8.06 min. HRMS

(ESI) TOF (m/z): [M + H]⁺459.1621 calcd for C₂₃H₂₇ClN₄O₂S;

found [M + H]⁺459.1634.

3-Amino-N-(6-(4-(4-chlorophenyl)-4-hydroxypiperidin-1-yl)-

hexyl)-4,6-dimethylthieno[2,3-b]pyridine-2-carboxamide

(33e). Carboxylic acid 25 (61 mg, 0.274 mmol), N,N-diisopropylethyl-

amine (0.146 mL, 0.837 mmol), and BOP (128 mg, 0.288 mmol) were

added to N,N-dimethylformamide (2 mL). 17e (116 mg, 0.302 mmol)

was dissolved in 2 mL of N,N-dimethylformamide and N,N-

diisopropylethylamine (0.096 mL, 0.550 mmol) and then added to

the reaction mixture. After the mixture was stirred for 5 h at room

temperature, the crude product was worked up as per general

procedure E to give the product as a pale yellow solid (49 mg, 35%).

Mp: 137.2−140.3 °C. ¹H NMR (CDCl₃): δ 1.39−1.42 (m, 4H),

1.57−1.65 (m, 4H), 1.72−7.75 (m, 3H), 2.19 (m, 2H), 2.43−2.5 (m,

4H), 2.58 (s, 3H), 2.74 (s, 1H), 2.87 (m, 2H), 3.42 (dd, J 13.1, 7 Hz,

2H), 5.55 (t, J 5.3 Hz, 1H, NH), 6.3 (s, 2H, NH₂), 6.87 (s, 1H), 7.29

(m, 2H), 7.44 (m, 2H). ¹³C NMR (CDCl₃): δ 20.2 (CH₃), 24.3

(CH₃), 26.8 (CH₂), 27.0 (CH₂), 27.4 (CH₂), 29.9 (CH₂), 38.4 (CH₂),

39.8 (CH₂), 49.5 (CH₂), 58.7 (CH₂), 71.1 (C), 98.7 (C), 122.3 (CH),

123.8 (C), 126.3 (CH), 128.5 (CH), 132.9 (C), 143.8 (C), 147.0 (C),

147.3 (C), 159.1 (C), 159.2 (C), 166 (C). HPLC purity (λ = 214

nm): 98%, t_R= 8.39 min. HRMS (ESI) TOF (m/z): [M + H]⁺

515.2247 calcd for C₂₇H₃₅ClN₄O₂S; found [M + H]⁺515.2254.

3-Amino-5-chloro-6-methoxy-N-(4-(4-(2-methoxyphenyl)-

piperazin-1-yl)butyl)-4-methylthieno[2,3-b]pyridine-2-carbox-

amide (34c). Carboxylic acid 32 (100 mg, 0.367 mmol), N,N-

diisopropylethylamine (0.067 mL, 0.385 mmol), and BOP (176 mg,

0.398 mmol) were added to N,N-dimethylformamide (10 mL). 18c

(121 mg, 0.403 mmol) was dissolved in 2 mL of N,N-

dimethylformamide, and N,N-diisopropylethylamine (0.077 mL,

0.403 mmol) was then added to convert to the free base, which was

added dropwise to the reaction mixture. The reaction mixture was

then worked up as described in general procedure E to give the

product as a tan crystalline solid (55 mg, 29%). Mp: 99.1−101.6 °C

(methanol/water). ¹H NMR (CDCl₃): δ 1.62−1.69 (m, 4H), 2.46 (m,

2H), 2.67 (m, 4H), 2.80 (s, 3H), 3.11 (m, 4H), 3.44 (q, J 6.5 Hz, 2H),

3.86 (s, 3H), 4.06 (s, 3H), 5.66 (t, J 5.6 Hz, 1H, NH), 6.28 (s, 2H,

NH₂), 6.84−7.01 (m, 4H). ¹³C NMR (CDCl₃): δ 16.4 (CH₃), 24.4

(CH₂), 27.0 (CH₂), 39.7 (CH₂), 50.8 (CH₂), 53.7 (CH₂), 55.2 (CH₃),

55.5 (CH₃), 58.3 (CH₂), 98.4 (C), 111.3 (CH), 116.6 (C), 118.4

(CH), 121.1 (CH), 121.3 (C), 123.1 (CH), 141.5 (C), 143.1 (C),

147.4 (C), 152.4 (C), 154.5 (C), 159.3 (C), 165.8 (C). HPLC purity

(λ = 214 nm): 97%, t_R= 9.30 min. HRMS (ESI) TOF (m/z): [M +

H]⁺518.1993 calcd for C₂₅H₃₂ClN₅O₃S; found [M + H]⁺518.2005.

3-Amino-5-chloro-6-methoxy-N-(5-(4-(2-methoxyphenyl)-

piperazin-1-yl)pentyl)-4-methylthieno[2,3-b]pyridine-2-car-

boxamide (34d). Carboxylic acid 32 (100 mg, 0.367 mmol), N,N-

diisopropylethylamine (0.067 mL, 0.385 mmol), and BOP (170 mg,

0.385 mmol) were added to N,N-dimethylformamide (10 mL). 18d

(127 mg, 0.403 mmol) was dissolved in 2 mL of N,N-

dimethylformamide, and N,N-diisopropylethylamine (0.11 mL, 0.403

mmol) was then added to convert to the free base, which was added

dropwise to the reaction mixture. The reaction mixture was then

worked up as described in general procedure E to give the product as

pale red microneedles (97 mg, 50%). Mp: 96−97.7 °C (methanol/

1

water). H NMR (CDCl₃): δ 1.43 (m, 2H), 1.56−1.69 (m, 4H), 2.43

(m, 2H), 2.66 (m, 4H), 2.83 (s, 3H), 3.10 (m, 4H), 3.42 (m, 2H), 3.86

(s, 3H), 4.07 (s, 3H), 5.44 (t, J 5.6 Hz, 1H, NH), 6.30 (br s, 2H, NH₂),

6.84−7.02 (m, 4H). ¹³C NMR (CDCl₃): δ 16.4 (CH₃), 25.1 (CH₂),

26.7 (CH₂), 29.9 (CH₂), 39.7 (CH₂), 50.8 (CH₂), 53.6 (CH₂), 55.2

(CH₃), 55.5 (CH₃), 58.7 (CH₂), 98.3 (C), 111.3 (CH), 116.6 (C),

118.3 (CH), 121.1 (CH), 121.3 (C), 123.0 (CH), 141.5 (C), 143.1

(C), 147.4 (C), 152.4 (C), 154.4 (C), 159.3 (C), 165.8 (C). HPLC

purity (λ = 214 nm): 99%, t_R= 9.84 min. HRMS (ESI) TOF (m/z):

[M + H]⁺532.2149 calcd for C₂₆H₃₄ClN₅O₃S; found [M + H]⁺

532.2166.

3-Amino-5-chloro-N-(5-(4-(2,3-dichlorophenyl)piperazin-1-

yl)pentyl)-6-methoxy-4-methylthieno[2,3-b]pyridine-2-car-

boxamide (35). Carboxylic acid 32 (80 mg, 0.293 mmol), N,N-

diisopropylethylamine (0.05 mL, 0.308 mmol), and BOP (136 mg,

0.308 mmol) were added together. Compound 19, which was Boc

deprotected and isolated as the HCl salt (114 mg, 0.298 mmol), was

dissolved in 2 mL of N,N-dimethylformamide, and N,N-diisopropy-

lethylamine (0.05 mL) was then added to convert to the free base,

which was added dropwise to the reaction mixture. The reaction

mixture was then worked up as described in general procedure E to

give the product as a beige solid (106 mg, 63%). Mp: 114.5−116.5 °C.

¹H NMR (CDCl₃): δ 1.41 (m, 2H), 1.57−1.67 (m, 4H), 2.44 (m, 2H),

2.64 (m, 4H), 2.81 (s, 3H), 3.07 (m, 4H), 3.42 (m, 2H), 4.06 (s, 3H),

5.48 (t, J 5.7 Hz, 1H, NH), 6.30 (br s, 2H, NH₂), 6.94 (dd, J 7.2, 2.4,

1H), 7.10−7.16 (m, 2H). ¹³C NMR (CDCl₃): δ 16.3 (CH₃), 24.9

(CH₂), 26.6 (CH₂), 29.9 (CH₂), 39.7 (CH₂), 51.3 (CH₂), 53.4 (CH₂),

55.1 (CH₃), 58.5 (CH₂), 98.2 (C), 116.5 (C), 118.7 (CH), 121.2 (C),

124.6 (CH), 127.5 (CH), 127.5 (C), 134.1 (C), 143.0 (C), 147.3 (C),

151.4 (C), 154.3 (C), 159.2 (C), 165.7 (C). HPLC purity (λ = 214

nm): 99%, t_R= 11.66 min. HRMS (ESI) TOF (m/z): [M + H]⁺

570.1264 calcd for C₂₅H₃₀Cl₃N₅O₂S; found [M + H]⁺570.1265.

3-Amino-N-(5-(4-(2,3-dichlorophenyl)piperazin-1-yl)pentyl)-

4,6-dimethylthieno[2,3-b]pyridine-2-carboxamide (36). Car-

boxylic acid 25 (73 mg, 0.32 mmol), N,N-diisopropylethylamine

(0.059 mL, 0.34 mmol), and BOP (150 mg, 0.34 mmol) were added

to N,N-dimethylformamide (2 mL). Compound 19, which was Boc

deprotected and isolated as the HCl salt (126 mg, 0.36 mmol), was

dissolved in 2 mL of N,N-dimethylformamide and N,N-diisopropyle-

thylamine (0.062 mL, 0.36 mmol) and then added to the reaction

mixture. The reaction mixture was then worked up as described in

general procedure E to give the product as a ﬂuﬀy beige solid (119 mg,

1

70%). Mp: 142.9−143.8 °C. H NMR (CDCl₃): δ 1.42 (m, 2H),

1.57−1.67 (m, 4H), 2.44 (m, 2H), 2.58 (s, 3H), 2.63 (m, 4H), 2.73 (s,

3H), 3.07 (m, 4H), 3.42 (m, 2H), 5.59 (t, J 5.6, 1H, NH), 6.31 (br s,

2H, NH₂), 6.86 (s, 1H), 6.94 (dd, J 7.0, 2.6 Hz, 1H), 7.10−7.15 (m,

2H). ¹³C NMR (CDCl₃): δ 20.3 (CH₃), 24.4 (CH₃), 24.9 (CH₂), 26.6

(CH₂), 29.9 (CH₂), 39.7 (CH₂), 51.4 (CH₂), 53.4 (CH₂), 58.5 (CH₂),

98.6 (C), 118.7 (CH), 122.3 (CH), 123.7 (C), 124.6 (CH), 127.5

(CH), 127.6 (C), 134.1 (C), 143.7 (C), 147.3 (C),151.4 (C), 159.0

(C), 159.1 (C), 165.9 (C). HPLC purity (λ = 214 nm): 99%, t_R

=

10.13 min. HRMS (ESI) TOF (m/z): [M + H]⁺520.1705 calcd for

C₂₅H₃₁Cl₂N₅OS; found [M + H]⁺520.1703.

3-Amino-N-(5-(4-(2-methoxyphenyl)piperazin-1-yl)pentyl)-

4,6-dimethylthieno[2,3-b]pyridine-2-carboxamide (37). Car-

boxylic acid 25 (80 mg, 0.36 mmol), N,N-diisopropylethylamine

M

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(0.066 mL, 0.38 mmol), and BOP (167 mg, 0.38 mmol) were added

to N,N-dimethylformamide (2 mL). 18d (124 mg, 0.40 mmol) was

dissolved in 2 mL of N,N-dimethylformamide and N,N-diisopropyle-

thylamine (0.069 mL, 0.40 mmol) and then added to the reaction

mixture. The reaction mixture was then worked up as described in

general procedure E to give the product as a yellow-orange solid (66

was added to the samples and incubated in the dark at 37 °C for 1.5 h.

Plates were read using a Fusion plate reader.

cAMP Accumulation Assays. The cells were grown and

incubated overnight and then preincubated for 45 min in 80 μL of

stimulation buﬀer (Hank’s buﬀered salt solution: 0.14 M NaCl, 5.4

mM KCl, 0.8 μM MgSO₄, 1.3 mM CaCl₂, 0.2 mM Na₂HPO₄, 0.44

mM KH₂PO₄, 5.6 mM D-glucose, 1 mg/mL BSA, 0.5 mM 3-isobutyl-1-

methylxanthine, and 5 mM HEPES, pH 7.4). The agonists (10 μL)

and 300 nM forskolin (10 μL) were added simultaneously to the cells

and incubated for 30 min at 37 °C. Stimulation was terminated via the

removal of the stimulation buﬀer and addition of 50 μL of ice cold

100% ethanol. The plates were then incubated at 37 °C to allow

evaporation of the ethanol. Then 50 μL of detection buﬀer (1 mg/mL

BSA, 0.3% Tween-20, and 5 mM HEPES, pH 7.4) was added and 5 μL

of each well transferred to a 384-well Optiplate (PerkinElmer,

Waltham, MA, U.S.). Anti-cAMP acceptor beads (0.2 units/μL)

diluted in stimulation buﬀer was added under green light for 30 min

before the addition of 15 μL of the donor beads/biotinylated cAMP

(0.07 units/μL) diluted in detection buﬀer. The plates were incubated

for 1 h at room temperature and read using a Fusion-α plate reader

using AlphaScreen presettings.

Inhibition Assays for Analogues 5, 18a, and 18e. The

required analogues were added to the cells at the appropriate dilutions

and incubated at 37 °C for 30 min before the addition of 10 nM

dopamine. After an additional 5 min stimulation time, termination of

the ERK1/2 phosphorylation was as stated above.

Concentration−Response Assays for Analogues 18b−d, 19,

34d, 35, and 36. The required analogues were added to the cells at

the appropriate dilutions and incubated at 37 °C for 5−10 min as

predetermined for each compound in the time-course assays.

Termination of the ERK1/2 phosphorylation was as stated above.

Interaction Studies with Dopamine for Compounds 34a−e.

The required analogues and dopamine were added to the cells

simulataneously at the appropriate dilutions and incubated at 37 °C for

5 min. Termination of the ERK1/2 phosphorylation was as stated

above.

1

mg, 38%). Mp: 127.6−128.2 °C. H NMR (CDCl₃): δ 1.42 (m, 2H),

1.57−1.67 (m, 4H), 2.53 (m, 2H), 2.58 (s, 3H), 2.72 (s, 3H), 2.77 (m,

4H), 3.14 (m, 4H), 3.40 (m, 2H), 3.85 (s, 3H), 5.64 (t, J 5.7, 1H,

NH), 6.30 (br s, 2H, NH₂), 6.84−7.02 (m, 5H). ¹³C NMR (CDCl₃): δ

20.2 (CH₃), 24.4 (CH₃), 24.8 (CH₂), 26.1 (CH₂), 29.8 (CH₂), 39.6

(CH₂), 50.2 (CH₂), 53.5 (CH₂), 55.5 (CH₃), 58.5 (CH₂), 98.7 (C),

111.3 (CH), 118.4 (CH), 121.1 (CH), 122.3 (CH), 123.2 (CH),

123.7 (C), 141.0 (C), 143.7 (C), 147.3 (C), 152.3 (C), 159.0 (C),

159.1 (C), 166.0 (C). HPLC purity (λ = 214 nm): 97%, t_R= 8.89 min.

HRMS (ESI) TOF (m/z): [M + H]⁺482.2590 calcd for

C₂₆H₃₅N₅O₂S; found [M + H]⁺482.2590.

Pharmacology. Cell Lines and Transfection. FlpIn CHO cells

(Invitrogen) were grown in Dulbecco’s modiﬁed Eagle medium

(DMEM) supplemented with 10% fetal bovine serum and maintained

at 37 °C in a humidiﬁed incubator containing 5% CO₂. The FlpIn

CHO cells were transfected with the pOG44 vector encoding Flp

recombinase and the pDEST vector encoding the wild-type long

isoform of the human D₂receptor (D_2LR) at a ratio of 9:1 using

polyethylenimine as the transfection reagent. At 24 h after transfection,

the cells were subcultured, and the medium was supplemented with

700 μg/mL HygroGold as a selection agent. Cells were grown and

maintained in DMEM containing 20 mM HEPES, 5% fetal bovine

serum, and 200 μg/mL hygromycin B. Cells were maintained at 37 °C

in a humidiﬁed incubator containing 5% CO₂and 95% O₂.

Preparation of FlpIN CHO Cell Membranes. When cells were

approximately 90% conﬂuent, they were harvested and centrifuged

(300g, 3 min). The resulting pellet was resuspended in assay buﬀer (20

mM HEPES, 100 mM NaCl, 6 mM MgCl₂, 1 mM EGTA, and 1 mM

EDTA, pH 7.4), and the centrifugation procedure was repeated. The

intact cell pellet was then resuspended in assay buﬀer and

homogenized using a Polytron homogenizer for three 10 s intervals

on the maximum setting, with 30 s periods on ice between each burst.

The homogenate volume was brought up to 30 mL. The sample was

centrifuged (1000g, 10 min, 25 °C). The pellet was discarded, and the

supernatant was recentrifuged at 30 000g for 1 h at 4 °C. The resulting

pellet was resuspended in 5 mL of assay buﬀer, and the protein

content was determined using the method of Bradford. The

homogenate was then separated into 0.5 mL aliquots and stored

frozen at −80 °C until it was required for binding assays.

Data Analysis. Computerized nonlinear regression was performed

using Prism 6.0 (GraphPad Software, San Diego, CA).

Agonist concentration−response curves were ﬁtted via nonlinear

regression to the three-parameter logistic function (eq 1):

E_max− basal

E = basal +

₅₀−log[A])

1 + 10^(−pEC

(1)

where E is response, E_maxand basal are the top and bottom asymptotes

of the curve, respectively, log[A] is the logarithm of the agonist

concentration, and pEC₅₀is the negative logarithm of the agonist

concentration that gives a response halfway between E_maxand basal.

To compare agonist proﬁles and quantify stimulus bias, agonist

concentration−response curves were ﬁtted to the following form of

the operational model of agonism (eq 2):

[³H]Spiperone Binding Assay. Cell membranes (D_2L-FlpIn

CHO, 3 μg) were incubated with varying concentrations of test

compound in binding buﬀer (20 mM HEPES, 100 mM NaCl, 6 mM

MgCl₂, 1 mM EGTA, and 1 mM EDTA, pH 7.4) containing 0.05 nM

of [³H]spiperone and 100 μM GppNHp to a ﬁnal volume of 1 mL and

were incubated at 37 °C for 3 h. Binding was terminated by fast-ﬂow

ﬁltration over GF/B membranes using a Brandel harvester followed by

three washes with ice-cold 0.9% NaCl. Bound radioactivity was

measured in a Tri-Carb 2900TR liquid scintillation counter

(PerkinElmer).

ERK1/2 Phosphorylation Assay. FlpIn CHO cells stably

expressing the D_2LR were seeded into 96-well plates at a density of

50 000 cells/well. After 5 h, cells were washed with phosphate buﬀered

saline (PBS) and incubated in serum-free DMEM overnight before

assaying. Initially, time-course experiments were conducted at least

twice for each ligand to determine the time required to maximally

promote ERK1/2 phosphorylation via the dopamine D_2LR. Dose−

response experiments were performed in the absence and presence of

increasing concentrations of each ligand at 37 °C. Stimulation of the

cells was terminated by removing the media followed by the addition

of 100 μL of SureFire lysis buﬀer (PerkinElmer) to each well. The

plate was shaken for 5 min at room temperature before transferring 5

μL of the lysates to a white 384-well Proxiplate (PerkinElmer). Then 8

μL of a 240:1440:7:7 mixture of Sureﬁre activation buﬀer/Sureﬁre

reaction buﬀer/Alphascreen acceptor beads/Alphascreen donor beads

(E_m− basal)

ⁿ[A]ⁿ

τ

A

(_K)

Y = basal +

n

τ

[A]

[A]ⁿ

+

1 +

(_K) (

)

K_A

(2)

A

where E_mis the maximal possible response of the system; basal is the

basal level of response; K_Adenotes the equilibrium dissociation

constant of the agonist (A); τ is an index of the signaling eﬃcacy of the

agonist and is deﬁned as R_T/K_E, where R_Tis the total number of

receptors and K_Eis the coupling eﬃciency of each agonist-occupied

receptor; and n is the slope of the transducer function that links

occupancy to response. The analysis assumes that the maximal system

responsiveness (E_m) and the transduction machinery utilized for a

given cellular pathway are the same for all agonists such that the E_m

and transducer slope (n) are shared between agonists. The ratio τ/K_A

(determined as a logarithm, i.e., log(τ/K_A)) is referred to herein as the

“transduction coeﬃcient”, as this composite parameter is suﬃcient to

describe agonism and bias for a given pathway; i.e., stimulus-biased

agonism can result from either a selective aﬃnity (K_A‑1) of an agonist

N

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

for a given receptor state(s) and/or a diﬀerential coupling eﬃcacy (τ)

toward certain pathways. To cancel the impact of cell-dependent

eﬀects on the observed agonism at each pathway, the log(τ/K_A) values

were then normalized to that determined for the endogenous agonist

dopamine at each pathway to yield a “normalized transduction

coeﬃcient”, Δlog(τ/K_A), i.e., Δlog(τ/K_A) = log(τ/K_A)_test− log(τ/

signal-regulated kinase; cAMP, cyclic adenosine monophos-

phate; FSK, forskolin; Boc, tert-butyloxycarbonyl; NMR,

nuclear magnetic resonance; BOP, (benzotriazol-1-yloxy)tris-

(dimethylamino)phosphonium hexaﬂuorophosphate

REFERENCES

K_A)_dopamine

.

■

To determine the actual bias of each agonist for diﬀerent signaling

pathways, the Δlog(τ/K_A) values were evaluated statistically between

the pathways. The ligand bias of an agonist for one pathway, j1, over

another, j2 is given as (eq 3)

(1) Missale, C.; Nash, S. R.; Robinson, S. W.; Jaber, M.; Caron, M. G.

Dopamine receptors: from structure to function. Physiol. Rev. 1998, 78,

189−225.

(2) Seeman, P. Targeting the dopamine D2 receptor in

schizophrenia. Expert Opin. Ther. Targets 2006, 10, 515−531.

(3) Shin, J. K.; Malone, D. T.; Crosby, I. T.; Capuano, B.

Schizophrenia: a systematic review of the disease state, current

therapeutics and their molecular mechanisms of action. Curr. Med.

Chem. 2011, 18, 1380−1404.

(4) Tandon, R. Pharmacologic treatment of schizophrenia: current

status and future trends. Curr. Psychosis Ther. Rep. 2006, 4, 40−49.

(5) Rosen, W. G.; Mohs, R. C.; Johns, C. A.; Small, N. S.; Kendler, K.

S.; Horvath, T. B.; Davis, K. L. Positive and negative symptoms in

schizophrenia. Psychiatry Res. 1984, 13, 277−284.

(6) Inoue, T.; Domae, M.; Yamada, K.; Furukawa, T. Effects of the

novel antipsychotic agent 7-(4-[4-(2,3-dichlorophenyl)-1-piperazinyl]-

butyloxy)-3,4-dihydro-2(1H)-quinolinone (OPC-14597) on prolactin

release from the rat anterior pituitary gland. J. Pharmacol. Exp. Ther.

1996, 277, 137−143.

(7) Jordan, S.; Koprivica, V.; Chen, R.; Tottori, K.; Kikuchi, T.; Altar,

C. A. The antipsychotic aripiprazole is a potent, partial agonist at the

human 5-HT1A receptor. Eur. J. Pharmacol. 2002, 441, 137−140.

(8) Shapiro, D. A.; Renock, S.; Arrington, E.; Chiodo, L. A.; Liu, L.-

X.; Sibley, D. R.; Roth, B. L.; Mailman, R. Aripiprazole, a novel atypical

antipsychotic drug with a unique and robust pharmacology. Neuro-

psychopharmacology 2003, 28, 1400−1411.

(9) Tamminga, C. A. Partial dopamine agonists in the treatment of

psychosis. J. Neural Transm. 2002, 109, 411−420.

(10) Kenakin, T. Functional selectivity and biased receptor signaling.

J. Pharmacol. Exp. Ther. 2011, 336, 296−302.

⎛

⎞

⎛

⎞

⎛

⎞

τ

A

τ

A

τ

A

ΔΔlog

= Δlog

− Δlog

⎜

⎝

⎟

⎠

⎜

⎝

⎟

⎠

⎜

⎝

⎟

⎠

K

j1−j2

j1

j2

(3)

A lack of stimulus bias compared to a reference agonist (in this case

dopamine) will result in values of ΔΔlog(τ/K_A) not signiﬁcantly

diﬀerent from 0 between pathways. To account for the propagation of

error associated with the determination of composite parameters, eq 5

was used:

σ

n

SEM =

(4)

where σ is the standard deviation and n is the number of experiments,

pooled SEM = (SEM1)²+ (SEM2)²

(5)

All aﬃnity, potency, and transduction ratio parameters were estimated

as logarithms. All results are expressed as the mean SEM. Statistical

analyses were performed where appropriate using one-way ANOVA

with the Tukey’s post hoc test. Statistical signiﬁcance was taken as p <

0.05. For a more detailed explanation and analysis of using bias

calculations see van der Westhuizen et al.³³

ASSOCIATED CONTENT

* Supporting Information

■

S

Values of potency (pEC₅₀) and maximal stimulation (E_max) for

dopamine, aripiprazole, 18d, 19, 34d, and 35−37 for cAMP

and pERK1/2 signaling assays; full experimental and character-

ization of compounds 13d−e, 16b, 17b−d, 18a,c,e, 33b−d,

34a,b,e. This material is available free of charge via the Internet

at http://pubs.acs.org.

(11) Kenakin, T.; Christopoulos, A. Signalling bias in new drug

discovery: detection, quantification and therapeutic impact. Nat. Rev.

Drug Discovery 2013, 12, 205−216.

(12) Stallaert, W.; Christopoulos, A.; Bouvier, M. Ligand functional

selectivity and quantitative pharmacology at G protein-coupled

receptors. Expert Opin. Drug Discovery 2011, 6, 811−825.

(13) Urban, J. D.; Vargas, G. A.; von Zastrow, M.; Mailman, R. B.

Aripiprazole has functionally selective actions at dopamine D2

receptor-mediated signaling pathways. Neuropsychopharmacology

2006, 32, 67−77.

AUTHOR INFORMATION

Corresponding Authors

*J.R.L.: phone, +613 9903 9095; fax, +613 9903 9581; e-mail,

rob.lane@monash.edu.

■

(14) Allen, J. A.; Yost, J. M.; Setola, V.; Chen, X.; Sassano, M. F.;

Chen, M.; Peterson, S.; Yadav, P. N.; Huang, X.-p.; Feng, B.; Jensen,

N. H.; Che, X.; Bai, X.; Frye, S. V.; Wetsel, W. C.; Caron, M. G.;

Javitch, J. A.; Roth, B. L.; Jin, J. Discovery of β-arrestin-biased

dopamine D2 ligands for probing signal transduction pathways

essential for antipsychotic efficacy. Proc. Natl. Acad. Sci. U.S.A. 2011,

108, 18488−18493.

(15) Chen, X.; Sassano, M. F.; Zheng, L.; Setola, V.; Chen, M.; Bai,

X.; Frye, S. V.; Wetsel, W. C.; Roth, B. L.; Jin, J. Structure−functional

selectivity relationship studies of β-arrestin-biased dopamine D2

receptor agonists. J. Med. Chem. 2012, 55, 7141−7153.

(16) Gay, E. A.; Urban, J. D.; Nichols, D. E.; Oxford, G. S.; Mailman,

R. B. Functional selectivity of D2 receptor ligands in a Chinese

hamster ovary hD2L cell line: evidence for induction of ligand-specific

receptor states. Mol. Pharmacol. 2004, 66, 97−105.

(17) Mottola, D. M.; Kilts, J. D.; Lewis, M. M.; Connery, H. S.;

Walker, Q. D.; Jones, S. R.; Booth, R. G.; Hyslop, D. K.; Piercey, M.;

Wightman, R. M.; Lawler, C. P.; Nichols, D. E.; Mailman, R. B.

Functional selectivity of dopamine receptor agonists. I. Selective

activation of postsynaptic dopamine D2 receptors linked to adenylate

cyclase. J. Pharmacol. Exp. Ther. 2002, 301, 1166−1178.

*B.C.: phone, +61 3 9903 9556; fax, +61 3 9903 9581; e-mail,

ben.capuano@monash.edu.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

This research was supported by Project Grants 1049564 and

1011920 of the National Health and Medical Research Council

(NHMRC). A.C. is a Principle Research Fellow (NHMRC).

J.R.L. is a R. D. Wright Biomedical Career Development Fellow

(Grant 1052304, NHMRC) and a Larkin’s Fellow (Monash

University, Australia). M.S. acknowledges an Australian

postgraduate award. C.K.H. acknowledges a Monash graduate

scholarship. Special thanks are given to Dr. Jason Dang for

running HRMS on ﬁnal compounds.

ABBREVIATIONS USED

GPCR, G-protein-coupled receptor; CNS, central nervous

system; TLC, thin layer chromatography; ERK, extracellular

■

O

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

(18) Tschammer, N.; Bollinger, S.; Kenakin, T.; Gmeiner, P.

Histidine 6.55 is a major determinant of ligand-biased signaling in

dopamine D2L receptor. Mol. Pharmacol. 2011, 79, 575−585.

́

(19) Shonberg, J.; Herenbrink, C. K.; Lopez, L.; Christopoulos, A.;

Scammells, P. J.; Capuano, B.; Lane, J. R. A structure−activity analysis

of biased agonism at the dopamine D2 receptor. J. Med. Chem. 2013,

56, 9199−9221.

(20) DeSimone, R. W.; Currie, K. S.; Mitchell, S. A.; Darrow, J. W.;

Pippin, D. A. Privileged structures: applications in drug discovery.

Comb. Chem. High Throughput Screening 2004, 7, 473−493.

(21) Costantino, L.; Barlocco, D. Privileged structures as leads in

medicinal chemistry. Curr. Med. Chem. 2010, 5, 381−422.

(22) Duarte, C. D.; Barreiro, E. J.; Fraga, C. A. M. Privileged

structures: a useful concept for the rational design of new lead drug

candidates. Mini-Rev. Med. Chem. 2007, 7, 1108−1119.

(23) Bondensgaard, K.; Ankersen, M.; Thøgersen, H.; Hansen, B. S.;

Wulff, B. S.; Bywater, R. P. Recognition of privileged structures by G-

protein coupled receptors. J. Med. Chem. 2004, 47, 888−899.

(24) Kulagowski, J. J.; Broughton, H. B.; Curtis, N. R.; Mawer, I. M.;

Ridgill, M. P.; Baker, R.; Emms, F.; Freedman, S. B.; Marwood, R.;

Patel, S.; Patel, S.; Ragan, C. I.; Leeson, P. D. 3-[[4-(4-Chlorophenyl)-

piperazin-1-yl]methyl]-1H-pyrrolo[2,3-b]pyridine: an antagonist with

high affinity and selectivity for the human dopamine D4 receptor. J.

Med. Chem. 1996, 39, 1941−1942.

(25) Lober, S.; Hubner, H.; Tschammer, N.; Gmeiner, P. Recent

̈

advances in the search for D3- and D4-selective drugs: probes, models

and candidates. Trends Pharmacol. Sci. 2011, 32, 148−157.

(26) Newman, A. H.; Beuming, T.; Banala, A. K.; Donthamsetti, P.;

Pongetti, K.; LaBounty, A.; Levy, B.; Cao, J.; Michino, M.; Luedtke, R.

R.; Javitch, J. A.; Shi, L. Molecular determinants of selectivity and

efficacy at the dopamine D3 receptor. J. Med. Chem. 2012, 55, 6689−

6699.

(27) Capuano, B.; Crosby, I. T.; Lloyd, E. J.; Podloucka, A.; Taylor,

D. A. Synthesis and preliminary pharmacological evaluation of 4′-

arylalkyl analogues of clozapine. IV. The effects of aromaticity and

isosteric replacement. Aust. J. Chem. 2008, 61, 930−940.

(28) Abdel-Fattah, M. A. O.; Lehmann, J.; Abadi, A. H. An interactive

SAR approach to discover novel hybrid thieno probes as ligands for

D2-like receptors with affinities in the subnanomolar range. Chem.

Biodiversity 2013, 10, 2247−2266.

(29) Kenakin, T.; Watson, C.; Muniz-Medina, V.; Christopoulos, A.;

Novick, S. A simple method for quantifying functional selectivity and

agonist bias. ACS Chem. Neurosci. 2011, 3, 193−203.

(30) Vangveravong, S.; Zhang, Z.; Taylor, M.; Bearden, M.; Xu, J.;

Cui, J.; Wang, W.; Luedtke, R. R.; Mach, R. H. Synthesis and

characterization of selective dopamine D2 receptor ligands using

aripiprazole as the lead compound. Bioorg. Med. Chem. 2011, 19,

3502−3511.

(31) Tschammer, N.; Elsner, J.; Goetz, A.; Ehrlich, K.; Schuster, S.;

Ruberg, M.; Kuhhorn, J.; Thompson, D.; Whistler, J.; Hubner, H.;

̈

Gmeiner, P. Highly potent 5-aminotetrahydropyrazolopyridines:

enantioselective dopamine D3 receptor binding, functional selectivity,

and analysis of receptor−ligand interactions. J. Med. Chem. 2011, 54,

2477−2491.

(32) Black, J. W.; Leff, P. Operational models of pharmacological

agonism. Proc. R. Soc. London, Ser. B 1983, 220, 141−162.

(33) van der Westhuizen, E. T.; Breton, B.; Christopoulos, A.;

Bouvier, M. Quantification of ligand bias for clinically relevant β2-

adrenergic receptor ligands: implications for drug taxonomy. Mol.

Pharmacol. 2014, 85, 492−509.

P

dx.doi.org/10.1021/jm500457x | J. Med. Chem. XXXX, XXX, XXX−XXX

Article Doi

Structure-activity relationships of privileged structures lead to the discovery of novel biased ligands at the dopamine D2 receptor

DOI: 10.1021/jm500457x

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm500457x

Full text of DOI:10.1021/jm500457x

Products guided by the article

R&D Labs maybe for 1610591-73-5

Relevant to this article

Hot Product