Discovery of Novel Biased Opioid Receptor Ligands through Structure-Based Pharmacophore Virtual Screening and Experiment

Accepted Article

Title: Discovery of Novel Biased Opioid Receptor Ligands through

Structure-Based Pharmacophore Virtual Screening and

Experiment

Authors: Pyeonghwa Jeong, Soo-Kyung Kim, Quanjie Li, Su-jin Oh,

Seonil Son, Guangju Chen, Hongwei Tan, Siwon Kim, Jong-

Hyun Park, Ki Duk Park, Yeo Ok Kim, Myung Ha Yoon, Yong-

Chul Kim, and William Goddard III

This manuscript has been accepted after peer review and appears as an

Accepted Article online prior to editing, proofing, and formal publication

of the final Version of Record (VoR). This work is currently citable by

using the Digital Object Identifier (DOI) given below. The VoR will be

published online in Early View as soon as possible and may be different

to this Accepted Article as a result of editing. Readers should obtain

the VoR from the journal website shown below when it is published

to ensure accuracy of information. The authors are responsible for the

content of this Accepted Article.

To be cited as: ChemMedChem 10.1002/cmdc.201900418

Link to VoR: http://dx.doi.org/10.1002/cmdc.201900418

A Journal of

www.chemmedchem.org

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

Discovery of Novel Biased Opioid Receptor Ligands through Structure-

Based Pharmacophore Virtual Screening and Experiment

Pyeonghwa Jeong^#,4, Soo-Kyung Kim^#,1, Quanjie Li^1,2,3, Su-jin Oh⁵, Seonil Son⁵, Guangju Chen²,

Hongwei Tan², Siwon Kim^6,7, Jong-Hyun Park⁶, Ki Duk Park^6,7,8, Yeo Ok Kim⁹, Myung Ha Yoon⁹, Yong-

Chul Kim^4,5*, William A. Goddard III^1*

[a]

Prof. William A. Goddard III

¹Materials and Process Simulation Center (MC-139-74), California Institute of Technology, Pasadena, California 91125, US

E-mail: wag@caltech.edu ORCID:0000-0003-0097-5716;

Prof. Y.-C. Kim

⁵School of Life Sciences, ⁴Department of Biomedical Science & Engineering, Gwangju Institute of Science & Technology

123 Cheomdan-gwagiro, Buk-gu, 61005, South Korea

E-mail: yongchul@gist.ac.kr

[b]

PH. Jeong

⁴Department of Biomedical Science & Engineering, Gwangju Institute of Science & Technology, 123 Cheomdan-gwagiro, Buk-gu, 61005, South Korea

Dr. S-K. Kim

¹Materials and Process Simulation Center (MC-139-74), California Institute of Technology, Pasadena, California 91125, US

Dr. Q. Li

¹Material and Process Simulation Center (MC-139-14)

²College of Chemistry, Beijing Normal University

Beijing, 100875, People’s Republic of China

³Institue of Medicinal Biotechnology, Chinese Academy of Medical Science, Beijing, 100050, People’s Republic of China

S. Oh, S. Son

⁵School of Life Sciences, Gwangju Institute of Science & Technology

123 Cheomdan-gwagiro, Buk-gu, 61005, South Korea

Prof. G. Chen, Prof. H. Tan

²College of Chemistry, Beijing Normal University

Beijing, 100875, People’s Republic of China

S. Kim

⁶Convergence Research Center for Diagnosis, Treatment and Care system of Dementia, Korea Institute of Science and Technology (KIST), Seoul 02792,

South Korea

⁷Division of Bio-Medical Science & Technology, KIST School, Korea University of Science and Technology, Seoul 02792, South Korea

Dr. J-H. Park

⁶Convergence Research Center for Diagnosis, Treatment and Care system of Dementia, Korea Institute of Science and Technology (KIST), Seoul 02792,

South Korea

Dr. K.D. Park

⁶Convergence Research Center for Diagnosis, Treatment and Care system of Dementia, Korea Institute of Science and Technology (KIST), Seoul 02792,

South Korea

⁷Division of Bio-Medical Science & Technology, KIST School, Korea University of Science and Technology, Seoul 02792, South Korea

⁸KHU-KIST Department of Converging Science and Technology, Kyung Hee University, Seoul 02447, South Korea

Prof. M.H. Yoon, Dr. Y.O. Kim

⁹Department of Anesthesiology and Pain Medicine, Medical School, Chonnam National University, Gwangju 501-757, South Korea

^#These authors contributed equally to this work.

Abstract: ^G_i^{-protein biased agonists with minimal β-arrestin}

recruitment have shown opportunities for alternative safe pain

treatment to overcome the serious adverse effects of human mu

opioid receptor (µ-OR) agonists. In order to discover novel non-

morphine opioid receptor (OR) agonists, we applied hierarchical

virtual screening of our in-house database against a pharmacophore

based on modeling the active conformation of ORs. We discovered

Introduction

Human mu opioid receptor (µ-OR) agonists such as morphine

are clinically useful as effective analgesics for the treatment of

moderate to severe pain.^[1]In addition, synthetic morphine (1,

Figure 1.) analogues, e.g. buprenorphine^[2], and non-morphine

synthetic compounds, e.g. fentanyl^[3], have been developed in

recent decades to increase pharmacological activity and reduce

side effects. Nevertheless, current treatment for pain using

single administration of conventional opioids involves such risks

initial hit compound (4),

a

novel µ-OR agonist with

pyrazoloisoquinoline scaffold. We applied computational R-group

screening to compound 4 and synthesized 14 derivatives predicted to

be best. Of these, the new G_i-protein biased compound (19) shows

EC₅₀= 179 nM at µ-OR. This resulting in significant pain-relief effects

for mice at the phase II period in formalin tests. This study provides a

new strategy to identify diverse sets of promising compounds that

might prove useful for drug developments targeting other G protein-

coupled receptors (GPCRs).

as addiction, constipation, and respiratory depression^[4],[5]

.

Recent studies on biased G_i-protein activation with minimal β-

arrestin recruitment showed opportunities for alternative efficient

and more safe pain treatment to replace traditional narcotic

analgesics.^[6]Accordingly, a G_i-biased µ-OR-specific agonist,

TRV-130 (2, Oliceridine, Figure 1.),^[7]which showed significantly

less side effects compared with morphine, has been developed

and currently in the NDA stage of the US FDA ^[8]. More recently,

another biased agonist with highly selective pharmacological

1

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

profile to pain, PZM-21(3, Figure 1.)^[9]was discovered through

structure based approaches using the antagonist bound µ-OR

co-crystal structure for initial screening. However, recent studies

have shown some controversial results of β-arrestin knock-out

animal experiments, bringing forward a counterarguement

against the simple separation of therapeutic effects and side

effects due to G_i-protein dependent signalling versus β-arrestin

signalling.^[10]For example, although TRV-130 and PZM-21, the

representative G_i-protein biased agonists, are devoid of

addictive effects using conditioned place preference (CPP),^{[9, 11]}

the research with β-arrestin knock-out mice showed enhanced

morphine CPP, which provokes questions about the merits of G_i-

protein biased opioid agonists. The controversial results might

be explained by the lack of receptor desensitization of β-arrestin

deficient conditions.^[12]Nevertheless, the discovery of novel G_i-

protein biased µ-OR agonists and their precise pharmacological

and precilinical studies are required for the development of safe

opioid analgesics. Here, we report the discovery of non-

morphine structural µ-OR agonists by means of additional

refinements and optimization based on predicting the active G

protein-coupled receptor (GPCR) conformations, in-silico

screening, and side chain optimization. This led to the discovery

of pyrazoloisoquinoline derivatives that provide a novel human

µ-OR specific G_i-biased agonist chemotype, Compound 19, that

leads to nanomolar EC₅₀values with minimal β-arrestin

the energy of all (12)⁷=35 million packings in which the side chains

are optimized independently for each pair of interacting TMDs

(BiHelix). Then we select the top 1000, build the 7-TMD bundle,

reassign side each of the side chains, evaluate the energy, and

select 25 for further analysis (CombiHelix). We did this for 6

templates based on crystal structures of:

•

µ-OR activated structures: BU72 agonist bound mouse µ-OR

(PDB ID code 5C1M)^[15]

κ-OR inactivated structures: antagonist JDTic bound to

human κ-OR (PDB ID code 4DJH),^[16]

Antagonist morphinan bonded to mouse µ-OR (PDB ID code

4DKL),^[17]

Antagonist Naltrindole bonded to human δ-OR (PDB ID code

4N6H),^[18]

Peptide mimetic antagonist compound-24 (C-24) bonded

human nociceptin receptor (PDB ID code 4EA3),^[19]and

One partial agonist (EKC) bounded to κ-OR.^[20]

From these 6*25 = 150 packings, we selected the top 25 by

energy and used the superfamily-wide GPCR Sequence-

Structure (GRoSS) alignment of the TM for all 800 human GPCR

sequences (based on maximizing the 8 inter-helical contacts

characteristic of the activated GPCRs (the active hotspots) and 7

different contacts characteristic of inactive GPCRs (the GRoSS

inactive hotspots). In the Top 25 of µ-OR structures from BiHelix,

13 structures are µ-OR templates (2 from inactive and 11 from

active structures), 10 structures are from inactive δ-OR crystal

structure, and 2 structures are from active κ-OR predicted model.

These structures are tabulated in Table S1 of the SI.

recruitment. The computational strategies used in this study

should be useful for applications to other GPCRs.

The next step in GEnSeMBLE is the SuperBihelix optimization of

tilts and rotations, allowing each tilt angle (θ) to change by 0 or

±10°, each azumuthal projection (φ) to change by 0 or ±15°, and

each rotation (η) to change by 0 or ±15°. We selected 3 following

templates form BiHelix based on energy and diversity (see Table

S1 of the SI):

•

the all-zero structure from the active µ-OR (BHtop1)

the all-zero structure from the inactive δ-OR (BHtop2), and

the all-zero structure from µ-OR (BHtop3).

For each of these 3, we optimized the side chains and predicted

the energies for all (3*5*5)⁷= 13 trillion combinations of tilts and

rotations. We selected the best 2000 to build the full 7-TMD

bundle, repredicted the side chains and selected the best 25 by

energy. From these 75 bundles we selected the best 25 based

on energy for futher analysis. Out of 25, 9 of the SuperBiHelix

top25 (#2, 3, 4, 5, 8, 9, 10, 12,13) results are from the active

crystal structure of µ-OR, the other 16 are from the inactive

structure (Table S2).

Figure 1. Structures of mu opioid receptor agonists.

Results and Discussion

The number of Class A conserved contacts, and the number and

energy of interhelical hydrogen bonding are improved after

SuperBiHelix sampling. The all-zero structure with the same

rotation and tilting as the X-ray crystal structure of active µ-OR is

ranked as topA8. The Active hotspot analysis using currently

available X-ray of Class A GPCRs ^[21]shows that some 10 of the

top25 SuperBiHelix ensemble contain mainly active contacts

(Table S3). The X-ray template of the active µ-OR bound to

agonist has the most active hotspot contacts of 7, while the X-ray

template the inactive µ-OR bound to antagonist has the most

inactive hot spot contacts of 7. The X-ray template for the inactive

δ-OR and µ-OR has the most Class A contacts of 39. The highest

hydrogen bond (HB) number increased through SuperBihelix (20

to 27 for Inter HB, 39 to 46 for Tot HB). The best HB Energy was

also improved (-55.25 to -62.43 for Inter HB E, -138.97 to -146.30

Generating the ensemble of Opioid Receptors structures for

ligand docking

In order to find optimum ligand-GPCR structures for novel ligands,

we develop an ensemble of 25 GPCR conformations (with various

tilts and rotations of the 7 transmembrane domains (TMD)

expected to be thermally accessible. Here we use the GPCR

Ensemble of Structures in Membrane BiLayer Environment

(GEnSeMBLE) complete sampling method ^[13], which has been

applied successfully for adenosine A2A, Olfactory OR1G1,

glucagon-like peptide GLP1, bitter taste TAS2R38, Chemokine

CCR5, and cannabinoid CB1 receptors.^[14]Starting from

templates having high homology to the µ-OR and κ-OR targets,

we first fix the tilts and optimize simultaneous rotations about each

of the 7 TMD. Considering rotation increments of 30°, we evaluate

2

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

Table 1. Binding energy, % activation, and EC₅₀of 14 compounds at 10 µM at hOPRM for Top 10 structures from the unified cavity energy (UCav) and the binding

energy (BE) with ligand solvation (BE-Solv) rank from R-group screening. The table was ranked by % activation. The experiments show 19 to be the most active.

Method

VS^[a]

Compound

4

R1

R2

R3

UCav^[b]

-35.87

BE-Solv ^[c]

-55.44

% activation^[d]

EC₅₀(nM)^[d]

50 ± 9.37

602 ± 33.35

De novo

13

-31.74

-43.17

62.20 ± 11.81

UCav1

UCav2

14

15

m-Cl

o-Cl

p-Cl

m-Cl

o-Cl

p-Cl

o-Cl

R1

p-NO₂-Bn

m-MeO-Bn

H

-43.57

-42.26

-42.17

-41.85

-40.85

-40.61

-40.07

-40.06

-40.00

-39.57

UCav

-38.70

-41.96

-38.54

-41.58

-41.76

-41.36

-40.47

-36.33

-39.75

-40.19

BE-Solv

-39.31

-36.77

-35.01

-36.22

-34.17

-34.04

-28.16

-35.80

-32.20

-26.56

N.S^[d]

94.07 ± 1.33

46.07 ± 2.36

N.S

Dimethyl

H

235 ± 32.62

179 ± 21.83

UCav3

16

naphthalen-2-ylmethyl

m-MeO-Bn

UCav4

17

Dimethyl

H

UCav5

18

m-MeO-Bn

N.S

UCav6

19

p-MeSO₂-Bn

100.01 ± 13.65

64.95 ± 7.04

N.S

UCav7

20

naphthalen-2-ylmethyl

R2

UCav8

21

UCav9

22

H

46.81 ± 6.18

N.S

UCav10

Method

23

Dimethyl

R3

Compound

24

% activation

N.S

EC₅₀(nM)

BE-Solv1

BE-Solv2

BE-Solv3

BE-Solv4

BE-Solv5

BE-Solv6

BE-Solv7

BE-Solv8

BE-Solv9

BE-Solv10

m-Cl

p-Cl

o-Cl

m-Cl

p-Cl

p-NO₂-phenethyl

p-MeO- phenethyl

CH₂CH₂CN

Dimethyl

-52.08

-46.68

-46.39

-46.36

-46.00

-45.94

-45.89

-45.65

-45.58

25

48.52 ± 0.14

55.88 ± 5.92

57.84 ± 10.82

N.S

26

27

28

29

CH₂CH₂CN

87.00 ± 8.94

44.11 ± 9.39

42.15 ± 6.07

N.S

1,450 ± 360

940 ± 43.37

30

isobutyl

31

2-(N-methyl pyrrolidin-2-yl)ethyl

CH₂CH₂Cl

32

33

isobutyl

79.17 ± 0.76

[a] VS: Virtual screening, [b] UCav: unified cavity E is the nonbond energy between the ligand and the union of all protein residues within a constant cutoff distance

of the ligand in all the complexes, [c] BE: vertical binding energy, snap binding E= complex E-protein E-ligand E. [d] All biological data were normalized compared

to positive control DAMGO (3.7 µM) [e] N.S: not synthesized. All biological data were obtained by triplet testing and presented as the mean ± standard error of the

mean (SEM).

for Sum HB E). The Class A contact number also increased from

39 to 40. Thus, the Top25 models exhibit a mixture of active and

inactive ensembles. The predicted µ-OR structure exhibit the

conserved TM hydrogen bond networks common to Class A

GPCRs.^[22]The polar hydrogen bonding between L261 (5.65)

backbone oxygen and R278 (6.31) is observed for 8 out of 25

structures. Consistent with traditional models of Class A GPCR

activation, the active ensemble structures of µ-OR show TM5-6

interactions which are shown in Class A GPCRs. Using the same

methods, we predicted the ensemble of 25 structures for κ-OR

(See Table S4-S5 in supporting information, SI).

Molecular Docking Studies

The in-house GIST database was developed to contain diverse

heterocyclic core structures. To generate and visualize 3D

conformers of molecules in the GIST data set, we used the design

libraries tools in Discovery Studio 3.5. All generated ligands were

energy minimized using the CHARMm (Chemistry at HARvard

Macromolecular Mechanics) force field. These databases were

3

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

used for further structure-based virtual screening with our

predicted active µ-OR conformation. To predict the optimum

ligand binding site to each of the 25 protein conformations from

the GEnSeMBLE procedure, we used the DarwinDock method to

calculate the docking energy values of UCavE and SnapBE,

which previously used to predict ligand-protein structures for

fluorescence-based assays (Figure S1). However, for κ-OR we

observed no agonist activity (-1% at 10 µM) but good antagonist

activity (84% at 30 µM). The detailed synthetic procedures of

compound 4 are described in the Scheme S1.

To understand the agonism of µ-OR and antagonism of κ-OR, we

analysed the predicted binding pose of Compound 4. (Figure 2)

The common interactions are a salt bridge (SB) interactions from

D3.32 residue. However, the orientation and the conformation of

the ligands are different. The π-π interaction of 4 with the H6.52

residue in µ-OR is replaced with Y3.33 residue in κ-OR. There is

an additional HB of the phenolic hydroxyl group to the unique

W7.35 residue in µ-OR (Y7.35 residue in κ-OR or L7.35 residue

in δ-OR). However, the K5.39 residue forms an HB with this

hydroxyl group in κ-OR. Because of the different orientation, the

A2.53 residue and Q2.60 residue interactions are observed only

in µ-OR. Thus, various residues in the binding cavity and different

conformations of µ-OR leads to a favourable binding pose for

Compound 4. Compound 4 in µ-OR prefers to bind to active

structures #13 from SuperBihelix, while it binds to inactive

structure #14 in κ-OR. Thus, these docking studies to subtypes of

the OR explain fully why Compound 4 is an antagonist for κ-OR

and an agonist for µ-OR. The best docking pose of µ-OR with 4

has UCavE of -35.87 and SnapBE of -69.76 kcal/ mol. Compared

to the inactive κ-OR (UCavE: -45.40, SnapBE: -63.10 kcal/ mol),

µ-OR shows better SnapBE. In summary, Compound 4 prefers to

bind to the active structure in µ-OR, while it binds to the inactive

structure in κ-OR.

Summarizing, we discovered that compound 4 is a novel partial

agonist for µ-OR and an antagonist for κ-OR. Next, we used de

novo design with our predicted structure for compound 4 to

simplify the synthetic process while increasing overall yield. We

replaced the para-hydroxyl benzyl by a phenethyl group and we

replaced the hydroxyl group at 7 position by a methoxy group in

the isoquinoline ring. This modified compound 13 still shows 50%

activation for µ-OR and 10% activation for κ-OR at 10 µM

concentration (Table 1). Fig. S2 shows the docking result of

compound 13. The common interactions are a SB interaction from

D3.32 residue. There is an additional HB of the Nitrogen in the

ring to Q2.60 residue in µ-OR. The π-π interaction of 13 was

shown at the H6.52 residue in µ-OR.

To improve the activity and selectivity of compound 13, we carried

out R-group screening as follows:

1) At the R1 position (phenyl ring), we considered Cl at three

positions: ortho, meta, and para)

2) At the R2 position, we considered 56 commercially available

groups as summarized in Figure S3)

3) For the R3 position, we considered three cases: methyl and

benzyl substitution or no substituent.

For lead optimization we calculated the binding energy for all

Figure 2. Predicted ligand-protein structures of compound 4 by molecular

docking studies. (A) the superimposition of kappa (yellow) and mu (green)

opioid receptor bound with compound 4, (B) the binding site of kappa opioid

receptor, and (C) the binding site of mu opioid receptor. The H-bond is

represented by the arrows between the donor and the acceptor. Residues within

4 Å of ligand are shown on the corresponding 2D ligand interaction diagram

displayed on the right. For clarity, only polar hydrogen atoms were shown. Here

the color code is that dark blue is charged, light blue is polar amino acid, green

is nonpolar hydrophobic amino acid, and orange is glycine.

3x56x3=504 cases to human µ-OR (topA-02). In each case we

applied SCREAM (Side Chain Rotamer Excitation Analysis

[23]

Method)

to optimize the conformations for all residues in the

binding site for all cases. The UCavE and the BE with solvation

(BE-Solv) of the resulting ligand-protein complexes were

calculated after 100 steps of energy minimization. Then we

selected five compounds from the Top 10 of UCavE ranking and

seven compounds from the Top 10 of BE-Solv ranking in the bold

face in Table 1.

many GPCRs.^[14]After this initial screening, we selected

compound 4 to be synthesized. We used 12 steps overall with

poor yield. We found that compound 4 showed 50% activation of

µ-OR at 10 µM compared to the DAMGO positive control, using

Chemistry

4

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

Based on the predictions from computer-aided molecular design,

we synthesized candidate compounds as shown in Scheme 1

and S1. The detailed synthetic procedures of initial hit compound

4 and common intermediates 5 and 6 are described in Scheme

S1. The synthesis of 13 derivatives selected from de novo design

and R-group screening results at µ-OR is depicted in Scheme 1.

Briefly, cyclization of the pyrazole intermediate, 6, with various

benzaldehydes using microwave irradiation gave 7a-c. After O-

demethylation using reactions using thionyl chloride and 2M

ammonia solution in THF, respectively to provide 9a-c. The final

dimethyl amino derivatives, 15, 19, 20, 25-27, 29-31 and 33 were

obtained by reductive amination reactions with formaldehyde

(12a-c) and N-alkylations of pyrazole moieties with various R₂

groups. For the synthesis of final primary amine derivatives, 13,

16 and 22, the primary amines of 9a-c were first protected with

Boc group (10a-b), alkylation reactions (11a-c) and de-protection

of the Boc group were accomplished.

19, 29, and 33) were further tested for the concentration-

dependency to calculate EC₅₀values (Figure 3B). Among them,

O

N

O

6

N

NH

O

CN

c

a

b

NH₂

N

5

R₁

7a-c

d

N

HO

N

H₂N

N

NH

N

NH

N

NH

O

R₂

O

e

g

O

f

N

O

R₁

15, 19, 20,

25-27, 29-31, 33

8a-c

12a-c

9a-c

h

Figure 3. (A) Relative agonist activity of human µ opioid receptor (µ-OR) in

calcium flux FLIPR assay at the concentration of 10 µM of compounds

suggested by R-group screening, (B) The dose-response activation of the

selected compounds at human µ-OR. Agonist activity was normalized to that of

3.7 µM DAMGO. All data were obtained by triplet testing and the error bars are

the mean (SEM)

Boc

N

H

H₂N

N

NH

N

H

N

O

R₂

O

R₂

j

i

N

O

R₁

11d-e

R₁

13, 16, 22

10a-b

a: = m-Cl

R₁

=

d: R₁

e:

R

p-Cl,

naphthalen-2-ylmethyl

₁=m-Cl, R²₂

=

b: R₂

p-Cl

c: = o-C

R₃

l

Scheme 1. Reagents and conditions : (a) Ethyl methoxy acetate, potassium

tert-butoxide, THF, rt, 2 h; (ii) Hydrazine hydrate, Acetic acid, MeOH, reflux, 4

h; (c) Various benzaldehyde, TFA, Microwave irradiation, 140℃, 2 h; (d) BBr₃,

DCM, 0℃, 4 h; (e) Thionyl chloride, reflux, 2 h then 2M NH₃solution, THF 1 h;

(f) 37% formaldehyde in water, sodium cyanoborohydride, acetic acid, DCE, rt,

3 h; (g) Cesium carbonate, various halide, DMF, rt; (h) Boc-anhydride, TEA,

DCM, rt, 1 h; (i) 2-bromomethyl naphthalene, cesium carbonate, DMF, rt, 4 h;

(j) TFA, DCM, 0℃, 1 h.

Biological Evaluation of Synthesized Compounds

The predicted compounds from R-group screening were tested

for agonist activity using a calcium flux FLIPR assay. In this assay,

all 14 synthesized compounds showed activation of human µ-OR

signal at the concentration of 10 µM. (Table 1 and Figure 3A).

Particularly, compounds 15 and 19 promoted as much activity as

DAMGO (94% and 100% of 3.7 µM DAMGO activity, respectively).

The remaining 12 compounds showed partial agonism of µ-OR in

calcium flux FLIPR assay. The four most effective compounds (15,

Figure 4. Agonist activity in β-arrestin recruitment assay at the concentration of

10 µM. Agonist activity was normalized to that of 3.7 µM of DAMGO. All data

were obtained by triplet testing and the error bars are standard deviations.

5

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

Compound 19 displayed the best efficacy with an EC₅₀value of

179 nM. In addition, the best µ-OR agonists, Compound 15 and

19, showed no κ-OR or δ-OR agonism (See Supporting

information). All 14 compounds that displayed µ-OR agonism

were tested at concentration of 10 µM for β-arrestin recruitment

assay. All 14 showed minimal β-arrestin recruitment compared to

DAMGO (Figure 4). This result indicates that the new series of

pyrazoloisoquinioline derivatives are G_i-biased agonists for µ-OR.

in Figure 4. The additional binding site at the upper TM2 seems

to be critical to exhibit G_i-protein biased agonism. This is plausible

since binding or stabilizing the upper TM2 might affect the stability

of TM1-2-7 network connecting to helix 8 and the C-terminal end

that is an anchoring site of the β-arrestin.

In Vivo Experiment of Compound 15 and 19

We investigated the in vivo analgesic effects of newly developed

µ-OR biased agonists, compound 15 and 19, for the initial proof

of concept in formalin induced pain animal models. Subcutaneous

injection of 50 µL of 5% formalin solution was performed into the

plantar surface of the hind paw to induce nociception 10 mins after

pre-treatment with 10 µg of compound 19 and 30 µg of compound

15, respectively. The morphine (3 µg) was used as a positive

control. The behavioural responses were measured by counting

the number of rat flinches for 1 min periods from 1 to 2 min, 5 to

6 min and every 5 min from 10 to 60 min. Both compounds

displayed significant analgesic effects at the phase II period of

inflammatory pain with more than 80 % reduction (Figure 5).

Figure 6. Binding site and comparisons of compound 19 with other µ-OR

agonists. (A) the binding site for µ-OR, (B) 2D ligand interaction diagram, and

superimposition with (C) the conventional µ-OR agonist, BU72, in green and (D)

the biased µ-OR agonist, PZM-21, in yellow. The H-bond is represented by the

arrows between the donor and the acceptor. Residues within 4 Å of ligand are

shown on the 2D ligand interaction diagram displayed on the right. For clarity,

only polar hydrogen atoms are shown.

Conclusions

In summary, we used ensemble docking based on the predicted

ensemble of 25 predicted best GPCR packings (from

GEnSeMBLE) to predict active conformation of µ-OR for in silico

lead discovery. This was followed by experimental testing of the

best cases to find new lead compound 4 with

a novel

Figure 5. Reduced pain behavior in formalin injection test. Animals were pre-

treated with vehicle (white, n = 4), compound 15 at 30 µg/10 µL (red, n=4) and

19 at 10 µg/10 µL (blue, n=4). Results are presented as the mean ± standard

error of the mean (SEM).

pyrazoloqiunoline scaffold exhibiting has EC₅₀= 602 nM with a

maximum 50% activity for µ-OR. Then we applied R-group

computational optimization with the predicted binding site for 504

combinations. This was followed by experimental screening of 14

cases to discover compound 19: a fully active agonist with an

EC₅₀of 179 nM at µ-OR and biased activity for all 14. This led to

our discovery of the new pyrazolo[3,4-c]isoquinoline derivatives

as G_i-protein biased full agonists at µ-OR, constituting new

candidates for µ-OR pharmacology, distinct from such

conventional nonbiased agonists as morphine. We further

validated the in vivo analgesic effects of two new G_i-biased µ-OR

agonists using the formalin induced nociception model studies in

rats. We found significant analgesic effects of both compounds.

This combined computational in silico optimization with

experimental synthesis and characterization procedure

successfully developed novel selective biased agonists for µ-OR

provides an improved strategy to search for diverse new

promising compounds. We expect that this procedure will be

useful for other GPCR targeting drug developments.

Binding Pose Analysis of Compound 19

Figure 6A and B shows the predicted binding pose of 19 in µ-OR.

The major interactions are a SB between the protonated amine of

the ligand and the D149 (3.32) residue. There is also a π- π

interaction of the diphenyl ring of 19 with the Y150 (3.33) residue

in µ-OR. However, the orientation of phenyl ring is different from

compound 4. The methane sulfonyl benzyl moiety of compound

19 is directed toward TM2 rather than TM7. Superimposing

compound 19 into the binding site of the conventional µ-OR

agonist, BU72, in green (Figure 6C) and the biased µ-OR agonist,

PZM-21, in yellow (Figure 6D), we see that this methyl sulfoxide

phenyl ring overlaps the thiophene ring of PZM-21 (Figure 6D).

Based on the similarity to the PZM-21 binding site, we expected

compound 19 to be a G_i-protein biased µ-OR agonist like PZM-

21. Indeed, this was validated by the β-arrestin recruitment assay

6

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

To a microwave tube charged with chloro-substituted benzaldehyde (100

mg, 0.710 mmol) and compound 6 (150 mg, 0.592 mmol) in

trifluoroaceticacid (2 ml) was added. The reaction was irradiated at

140°C for 2 h in the microwave apparatus, cooled to room temperature.

The reaction mixture was concentrated in vacuo and dried to afford (7a-

c) compound as a pale-yellow solid.

Experimental Section

DarwinDock

For each of the 25 protein conformations and each of the 10-20 ligand

conformations obtained for conformational search, we generated

iteratively ~50,000 poses spanning the putative binding regions of the

bulky residue-alanized protein. These poses were generated in

increments of 5,000 and clustered into Voronoi families based on RMSD

until <2% new families are generated. The family heads were scored

using the Dreiding force field^[24]and the top10% by total energy were

selected. Then we calculated the binding energy of all the members of

these top 10% of families and selected the lowest 100 poses for further

optimization. At this point the protein side chains were de-alanized using

SCREAM^[23]to find the optimum side chains for each of the 100 poses.

Then we neutralized the protein and the ligand by transferring protons

appropriately within SB and protonating or deprotonating exterior ligands,

followed by further full geometry minimization. The final docked structure

with the best binding energy was selected.

5-(3-chlorophenyl)-7,8-dimethoxy-1-(methoxymethyl)-3H-

pyrazolo[3,4-c]isoquinoline (7a)

Compound 6 was reacted with 3-chlorobenzaldehyde according to the

general procedure I, to yield the compound 7a as a pale yellow solid. ¹H

NMR (400 MHz, CDCl₃) δ 8.00 (s, 1 H), 7.75 - 7.73 (m, 1 H), 7.62 (ddd,

J=6.30, 2.40, 1.60 Hz, 1 H), 7.54 - 7.50 (m, 2 H), 7.42 (s, 1 H), 5.04 (s, 2

H), 4.14 (s, 3 H), 3.87 (s, 3 H), 3.50 (s, 3 H).

5-(4-chlorophenyl)-7,8-dimethoxy-1-(methoxymethyl)-3H-

pyrazolo[3,4-c]isoquinoline (7b)

Compound 6 was reacted with 4-chlorobenzaldehyde according to the

general procedure I, to yield the compound 7b as a pale yellow solid. ¹H

NMR (400 MHz, CDCl₃) δ 8.00 (s, 1 H), 7.71 - 7.66 (m, 2 H), 7.58 - 7.53

(m, 2 H), 7.42 (s, 1 H), 5.03 (s, 2 H), 4.13 (s, 3 H), 3.87 (s, 3 H), 3.49 (s,

3 H).

Materials

Starting materials, reagents, and solvents were purchased from

commercial suppliers and used as supplied without further purification.

Thin layer chromatography (TLC) was performed on fluorescent silica gel

plates (60 F253 from Merck) and visualized with either short wave ultra

violet light at 254nm or 365nm. Microwave irradiation and conventional

heating systems were achieved using Anton-paar Monowave 400 and

Monowave 50, respectively. Chromatography purifications were achieved

using kieselgel 60 (Merck) 0.040-0.0636 mm column chromatography.

1H NMR was recorded on a JEOL JNM-ECX 400P spectrometer at 400

MHz and spectra were taken in CDCl₃and DMSO-d₆. All NMR data were

processed by ACD NMR Processor Academic version. Chemical shifts

are expressed as ppm and coupling constants in Hz. Data were reported

as follows: chemical shift, multiplicity (s, singlet; d, doublet; t, triplet; dd,

doublet of doublet; ddd, doublet of doublet of doublet; m, multiplet),

integration, and coupling constant.

5-(2-chlorophenyl)-7,8-dimethoxy-1-(methoxymethyl)-3H-

pyrazolo[3,4-c]isoquinoline (7c)

Compound 6 was reacted with 2-chlorobenzaldehyde according to the

general procedure I, to yield the compound 7c as a pale yellow solid. ¹H

NMR (400 MHz, CDCl₃) δ 8.00 (s, 1 H), 7.63 - 7.57 (m, 1 H), 7.55 - 7.45

(m, 3 H), 7.00 (s, 1 H), 5.05 (d, J=1.60 Hz, 2 H), 4.14 (s, 3 H), 3.81 (s,

3H), 3.52 (s, 3 H).

General Procedure II for the preparation of compounds-

demethylation

4-(3,4-dimethoxyphenyl)-3-(methoxymethyl)-1H-pyrazol-5-amine (6)

(5-(3-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-1-

yl)methanol (8a)

The compound 5 (500 mg, 2 mmol) was fully dissolved in methanol with

3% acetic acid and then hydrazine hydrate (2 ml) was added dropwise

with stirring. The reaction mixture was refluxed at 150°C for 4 h then

cooled gradually to ambient temperature. The solvent was removed

under reduced pressure and purified by silica gel column

chromatography with MeOH:CHCl₃(5%) to yield 6 as a yellow solid. ¹H

NMR (400 MHz, CDCl₃) δ 6.96 (d, J=1.83Hz, 1 H), 6. 95 - 6.92 (m, 1 H),

6.91 - 6.87 (m, 1 H), 4.45 (s, 2 H), 3.90 (s, 3 H), 3.91 (s, 3 H), 3.40 (s, 3

H).

The compound 7a (30 mg, 0.078 mmol) was dissolved in

dichloromethane (3 mL) under argon atmosphere then 1 M boron

tribromide in dichloromethane solution was added dropwise at 0 °C. The

resulting mixture was kept in ice bath for 4 h and quenched with

saturated aqueous ammonium chloride and extracted with

dichloromethane 3 times. The organic layers were combined and dried

with anhydrous Na₂SO₄. After Na₂SO₄was removed, the organic layer

was concentrated under reduced pressure and purified by silica gel

column chromatography with MeOH:CHCl₃(5%) to afford 8a as a yellow

solid. ¹H NMR (400 MHz, DMSO-d₆) δ 8.06 (s, 1 H), 7.80 - 7.74 (m, 2 H),

General Procedure I for the preparation of compounds (7a-c)

7

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

7.66 - 7.62 (m, 2 H), 7.37 (s, 1 H), 4.95 - 5.01 (m, 2 H), 4.01(s, 3H), 3.75

(s, 3 H).

Followed by the procedure III, compound 9c was obtained from

compound 8c. ¹H NMR (400 MHz, CDCl₃) δ 7.84 (s, 1 H), 7.42 - 7.64 (m,

4 H), 6.98 (s, 1 H), 4.56 (s, 2 H), 4.13 (s, 3 H), 3.79 (s, 3 H).

(5-(4-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-1-

yl)methanol (8b)

tert-butyl ((5-(3-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-

c]isoquinolin-1-yl)methyl)carbamate (10a)

Followed by the procedure II, compounds 8b were obtained from

compounds 7b. ¹H NMR (400 MHz, DMSO-d₆) δ 8.06 (s, 1 H), 7.80 - 7.74

(m, 2 H), 7.66 - 7.62 (m, 2 H), 7.37 (s, 1 H), 4.95 - 5.01 (m, 2 H), 4.01(s,

3H), 3.75 (s, 3 H).

To a solution of compound 9a (90 mg, 0.244 mmol) in dichloromethane

(5 mL) was treated with di-tert-butyl dicarbonate (106 mg, 0.488 mmol)

and triethylamine (49 mg, 0.488 mmol). The reaction mixture was stirred

at room temperature for 1 h. After the reaction was completed, the

residue was diluted with saturated aqueous ammonium chloride and

extracted with dichloromethane 3 times. The organic layer was dried over

anhydrous MgSO₄, filtered and concentrated under reduced pressure.

The crude materials were treated with diethyl ether and the mixture was

filtrated to obtain compound 10a as a pale-yellow solid. ¹H NMR (400

MHz, DMSO-d₆) δ 7.92 - 7.86 (m, 2 H), 7.82 - 7.75 (m, 2 H), 7.73 - 7.68

(m, 1 H), 7.67 - 7.59 (m, 2 H), 4.83 - 4.70 (m, 2 H), 4.06 (s, 3 H), 3.75 (s,

3H), 1.43 - 1.38 (m, 9 H).

(5-(2-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-1-

yl)methanol (8c)

Followed by the procedure II, compounds 8c were obtained from

compounds 7c. ¹H NMR (400 MHz, DMSO-d₆) δ 8.06 (s, 1 H), 7.71 - 7.66

(m, 1 H), 7.63 - 7.54 (m, 3 H), 6.83 (s, 1 H), 5.00 (br. s., 2 H), 4.01 (s,

3H), 3.63 (s, 3H).

tert-butyl ((5-(4-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-

c]isoquinolin-1-yl)methyl)carbamate (10b)

General procedure III – amination followed by chlorination

To a solution of compound 9b (90 mg, 0.244 mmol) in dichloromethane

(5 mL) was treated with di-tert-butyl dicarbonate (106 mg, 0.488 mmol)

and triethylamine (49 mg, 0.488 mmol). The reaction mixture was stirred

at room temperature for 1 h. After the reaction was completed, the

residue was diluted with saturated aqueous ammonium chloride and

extracted with dichloromethane 3 times. The organic layer was dried over

anhydrous MgSO₄, filtered and concentrated under reduced pressure.

The crude materials were treated with diethyl ether and the mixture was

filtrated to obtain compound 10b as a pale-yellow solid. ¹H NMR (400

MHz, CDCl₃) δ 7.76 (s, 1 H), 7.72 - 7.63 (m, 2 H), 7.60 - 7.52 (m, 2 H),

7.41 (s, 1H), 5.06 – 4.95 (m, 2 H), 4.17 (s, 3 H), 3.85 (s, 3 H), 1.52 - 1.46

(m, 9 H).

(5-(3-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-1-

yl)methanamine (9a)

The compound 8a (20 mg, 0.054 mmol) was added in SOCl₂(2 mL) and

refluxed at 80°C for 2 h. Excess SOCl₂was removed under reduced

pressure. The residual dark-brown crystalline was dissolved in 3 mL of

tetrahydrofuran (THF), and 2 M ammonia solution in THF(1 mL) was

added. After being stirred at room temperature for 1 h, the reaction

mixture was diluted with ethyl acetate and quenched with aqueous 1N

HCl. The organic layer was separated and washed with brine twice and

dried with anhydrous Na₂SO₄. After Na₂SO₄was filtered, solvent was

removed in vacuum. The residue was purified by silica gel column

chromatography with MeOH:CHCl₃(2%) to afford 9a as an off-white

liquid. ¹H NMR (400 MHz, CDCl₃) δ 7.96 (s, 1 H), 7.75 - 7.73 (m, 1 H),

7.63 - 7.59 (m, 1 H), 7.56 - 7.50 (m, 2 H), 7.43 (s, 1 H), 5.27 (s, 2 H),

4.15 (s, 3 H), 3.88 (s, 3 H).

tert-butyl ((5-(2-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-

c]isoquinolin-1-yl)methyl)carbamate (10c)

To a solution of compound 9c (90 mg, 0.244 mmol) in dichloromethane

(5 mL) was treated with di-tert-butyl dicarbonate (106 mg, 0.488 mmol)

and triethylamine (49 mg, 0.488 mmol). The reaction mixture was stirred

at room temperature for 1 h. After the reaction was completed, the

residue was diluted with saturated aqueous ammonium chloride and

extracted with dichloromethane 3 times. The organic layer was dried over

anhydrous MgSO₄, filtered and concentrated under reduced pressure.

The crude materials were treated with diethyl ether and the mixture was

filtrated to obtain compound 10c as a white solid. ¹H NMR (400 MHz,

CDCl₃) δ 7.78 (s, 1 H), 7.62 - 7.57 (m, 1 H), 7.53 - 7.45 (m, 3 H), 7.00 (s,

1 H), 5.04 - 4.95 (m, 2 H), 4.17 (s, 3 H), 3.80 (s, 3H), 1.53 - 1.45 (m, 9 H).

(5-(4-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-1-

yl)methanamine (9b)

Followed by the procedure III, compound 9b was obtained from

compound 8b. ¹H NMR (400 MHz, DMSO-d₆) δ 8.08 (s, 1 H), 7.81 - 7.73

(m,2 H), 7.68 - 7.61 (m, 2 H), 7.37 (s, 1 H), 4.33 (s, 2 H), 4.04 (s, 3 H),

3.76 (s, 3 H).

(5-(2-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-1-

yl)methanamine (9c)

8

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

Tert-butyl ((5-(3-chlorophenyl)-7,8-dimethoxy-3-(naphthalen-2-

ylmethyl)-3H-pyrazolo[3,4-c]isoquinolin-1-yl)methyl)carbamate (11d)

Following the general procedure IV, compound 12b was obtained from

compound 8b. ¹H NMR (400 MHz, CDCl₃) δ 8.43 (s, 1 H), 7.73 - 7.67 (m,

2 H), 7.59 - 7.53 (m, 2 H), 7.41 (s, 1 H), 4.15 (s, 3 H), 3.99 (s, 2 H), 3.87

(s, 3 H), 2.42 (s, 6 H).

To a solution of compound 10b (40 mg, 0.085 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 2-

(bromomethyl)naphthalene (37.5 mg, 0.170 mmol). After the mixture was

stirred at room temperature for 3 h, the solvent was removed in vacuo.

The residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford 11d as a pale-yellow oily liquid. ¹H NMR (400 MHz, CDCl₃) δ

7.87 - 7.73 (m, 6 H), 7.64 - 7.61 (m, 1 H), 7.53 - 7.41 (m, 6 H), 5.89 (s, 2

H), 5.10 (br. s., 1 H), 4.95 (d, J=5.50 Hz, 2 H), 4.15 (s, 3 H), 3.86 (s, 3 H),

1.45 (s, 9 H).

1-(5-(2-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-

1-yl)-N,N-dimethylmethanamine (12c)

Following the general procedure IV, compound 12c was obtained from

compound 8c. ¹H NMR (400 MHz, CDCl₃) δ 8.40 (s, 1 H), 7.57 - 7.61 (m,

1 H), 7.46 - 7.53 (m, 3 H), 6.98 (s, 1 H), 4.14 (s, 3 H), 3.99 (d, J=4.27 Hz,

2 H), 3.80 (s, 3 H), 2.43 (s, 6 H).

(5-(4-chlorophenyl)-7,8-dimethoxy-3-phenethyl-3H-pyrazolo[3,4-

c]isoquinolin-1-yl)methanamine (13)

Tert-butyl ((5-(4-chlorophenyl)-7,8-dimethoxy-3-(naphthalen-2-

ylmethyl)-3H-pyrazolo[3,4-c]isoquinolin-1-yl)methyl)carbamate (11e)

A solution of compound 11c (17 mg, 0.029 mmol) in dichloromethane (5

mL) was cooled to 0°C, and trifluoroaceticacid (1 mL) was slowly added.

Upon completion of reaction, the mixture was concentrated under

reduced pressure. The residue was taken up in solution of saturated

aqueous sodium bicarbonate and extracted dichloromethane 3 times.

The combined organic layer was dried with anhydrous Na₂SO₄. After

Na₂SO₄was filtered, solvent was removed in vacuum. The residue was

purified by silica gel column chromatography with MeOH: ammonia

saturated CHCl3 (1%) to afford 13 as a brown liquid. ¹H NMR (400 MHz,

MeOH-d₄) δ 7.66 (s, 1 H), 7.55 - 7.58 (m, 4 H), 7.33 (s, 1 H), 7.03 - 7.13

(m, 5 H), 4.72 - 4.77 (m, 2 H), 4.44 (m, 2 H), 4.08 (s, 3 H), 3.76 (s, 3 H),

3.20 (t, J=7.02 Hz, 2 H)

To a solution of compound 10a (40 mg, 0.085 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 2-

(bromomethyl)naphthalene (37.5 mg, 0.170 mmol). After the mixture was

stirred at room temperature for 4 h, the solvent was removed in vacuo.

The residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 11e as a transparent liquid. ¹H NMR (400 MHz,

CDCl₃) δ 7.81 - 7.74 (m, 5 H), 7.73 - 7.67 (m, 2 H), 7.58 - 7.52 (m, 2 H),

7.49 - 7.42 (m, 4 H), 5.88 (s, 2 H), 5.11 (br. s., 1 H), 4.94 (d, J=5.50 Hz, 2

H), 4.16 (s, 3 H), 3.86 (s, 3 H), 1.45 (s, 9 H).

1-(5-(3-chlorophenyl)-7,8-dimethoxy-3-(3-methoxybenzyl)-3H-

pyrazolo[3,4-c]isoquinolin-1-yl)-N,N-dimethylmethanamine (15)

General procedure IV – reductive amination

To a solution of compound 12a (40 mg, 0.085 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 3-

methoxybenzyl bromide (26.2 mg, 0.130 mmol). After the mixture was

stirred at room temperature for 1.5 h, the solvent was removed in vacuo.

The residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 15 as a white liquid (yield = 81%). ¹H NMR (500

MHz, CDCl₃) δ ppm 8.38 (s, 1 H), 7.74 - 7.76 (m, 1 H), 7.62 (ddd, J=6.42,

2.26, 1.59 Hz, 1 H), 7.49 - 7.51 (m, 2 H), 7.41 (s, 1 H), 7.20 (t, J=8.07 Hz,

1 H), 6.90 6.92 (m, 2 H), 6.77 - 6.79 (m, 1 H), 5.75 (s, 2 H), 4.13 (s, 4 H),

3.96 (s, 2 H), 3.87 (s, 3 H), 3.74 (s, 3 H), 2.40 (s, 6 H). ¹³C NMR (126

MHz, CDCl₃) δ ppm 157.87 (s, 1 C), 157.45 (s, 1 C), 153.63 (s, 1 C),

147.82 (s, 1 C), 143.25 (s, 1 C), 141.77 (s, 1 C), 139.09 (s, 1 C), 138.73

(s, 1 C), 134.50 (s, 1 C), 130.06 (s, 1 C), 129.58 (s, 1 C), 129.52 (s, 1 C),

129.46 (s, 1 C), 128.92 (s, 1 C), 128.78 (s, 1 C), 128.46 (s, 1 C), 128.03

(s, 1 C), 120.06 (s, 1 C), 113.72 (s, 1 C), 113.20 (s, 1 C), 105.49 (s, 1 C),

104.00 (s, 1 C), 56.18 (s, 1 C), 55.92 (s, 1 C), 55.92 (s, 1 C), 49.90 (s, 1

C), 44.53 (s, 1 C), 44.53 (s, 1 C), 40.99 (s, 1 C).

1-(5-(3-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-

1-yl)-N,N-dimethylmethanamine (12a)

To a solution of compound 9a (200 mg, 0.54 mmol) in 1,2-dichloroethane

with 5% acetic acid, sodium cyanoborohydride (340 mg, 1.62 mmol) was

added followed by 37% formaldehyde in water (300 µL, 3.8 mmol) and

stirred at room temperature for 3 h. The crude reaction mixture was

quenched with saturated aqueous sodium bicarbonate and extracted with

dichloromethane twice. The combined organic layer as dried with

anhydrous Na₂SO₄. After Na₂SO₄was filtered, solvent was removed in

vacuum. The residue was purified by silica gel column chromatography

with MeOH:CHCl₃(2%) to give 12a as a colourless liquid. ¹H NMR (400

MHz, CDCl₃) δ 8.43 (s, 1 H), 7.76 - 7.73 (m, 1 H), 7.64 - 7.59 (m, 1 H),

7.55 - 7.48 (m, 2 H), 7.41 (s, 1 H), 4.15 (s, 3 H), 3.98 (s, 2 H), 3.88 (s, 3

H), 2.42 (s, 6 H).

1-(5-(4-chlorophenyl)-7,8-dimethoxy-3H-pyrazolo[3,4-c]isoquinolin-

1-yl)-N,N-dimethylmethanamine (12b)

9

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

(5-(3-chlorophenyl)-7,8-dimethoxy-3-(naphthalen-2-ylmethyl)-3H-

pyrazolo[3,4-c]isoquinolin-1-yl)methanamine (16)

The residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 20 as a pale-yellow liquid. ¹H NMR (400 MHz,

CDCl₃) δ 8.37 (s, 1 H), 7.82 - 7.72 (m, 5 H), 7.65 - 7.60 (m, 1 H), 7.53 -

7.40 (m, 6 H), 5.93 (s, 2 H), 4.13 (s, 3 H), 3.96 (s, 2 H), 3.86 (s, 3 H),

2.40 (s, 6 H). ¹³C NMR (126 MHz, CDCl₃) δ ppm 153.29 (s, 1 C), 147.89

(s, 1 C), 147.57 (s, 1 C), 141.95 (s, 1 C), 135.02 (s, 1 C), 134.46 (s, 1 C),

133.30 (s, 1 C), 132.77 (s, 1 C), 131.94 (s, 1 C), 131.12 (s, 1 C), 130.10

(s, 1 C), 129.59 (s, 1 C), 129.59 (s, 1 C), 128.81 (s, 1 C), 128.21 (s, 1 C),

128.06 (s, 1 C), 127.90 (s, 1 C), 127.90 (s, 1 C), 127.60 (s, 1 C), 126.70

(s, 1 C), 126.05 (s, 1 C), 125.82 (s, 1 C), 124.85 (s, 1 C), 118.36 (s, 1 C),

108.02 (s, 1 C), 107.88 (s, 1 C), 56.07 (s, 1 C), 55.90 (s, 1 C), 55.90 (s, 1

C), 50.78 (s, 1 C), 45.33 (s, 1 C), 45.33 (s, 1 C).

A solution of compound 11b (20 mg, 0.035 mmol) in dichloromethane (5

mL) was cooled to 0°C, and trifluoroaceticacid (1 mL) was slowly added.

Upon completion of reaction, the mixture was concentrated under

reduced pressure. The residue was taken up in solution of saturated

aqueous sodium bicarbonate and extracted dichloromethane 3 times.

The combined organic layer was dried with anhydrous Na₂SO₄. After

Na₂SO₄was filtered, solvent was removed in vacuum. The residue was

purified by silica gel column chromatography with MeOH: ammonia

saturated CHCl₃(2%) to afford compound 16 as a yellow liquid. ¹H NMR

(400 MHz, CDCl₃) δ 7.71 (s, 1 H), 7.61 - 7.35 (m, 9 H), 7.15 - 7.04 (m, 2

H), 6.91 - 6.78 (m, 1 H), 5.81 (s, 2 H), 4.47 (s, 2 H), 3.97 (s, 3 H), 3.86 (s,

3 H). ¹³C NMR (126 MHz, CDCl₃) δ ppm 157.67 (s, 1 C), 153.53 (s, 1 C),

147.84 (s, 1 C), 147.59 (s, 1C), 145.15 (s, 1 C), 141.86 (s, 1 C), 134.89

(s, 1 C), 134.50 (s, 1 C), 133.29 (s, 1 C), 132.81 (s, 1 C), 130.10 (s, 1 C),

129.59 (s, 1 C), 128.88 (s, 1 C), 128.67 (s, 1 C), 128.28 (s, 1 C), 128.06

(s, 1 C), 127.92 (s, 1 C), 127.62 (s, 1 C), 126.91 (s, 1 C), 126.10 (s, 1 C),

125.99 (s, 1 C), 125.88 (s, 1 C), 118.29 (s, 1 C), 108.28 (s, 1 C), 106.64

(s, 1 C), 103.71 (s, 1 C), 77.22 (s, 1 C), 56.17 (s, 1 C), 55.92 (s, 1 C),

50.80 (s, 1 C), 41.33 (s, 1 C).

1-(5-(4-chlorophenyl)-7,8-dimethoxy-3-(naphthalen-2-ylmethyl)-3H-

pyrazolo[3,4-c]isoquinolin-1-yl)methanamine (22)

A solution of compound 11a (20 mg, 0.035 mmol) in dichloromethane (5

mL) was cooled to 0°C, and trifluoro acetic acid (1 mL) was slowly added.

Upon completion of reaction, the mixture was concentrated under

reduced pressure. The residue was taken up in solution of saturated

aqueous sodium bicarbonate and extracted dichloromethane 3 times.

The combined organic layer was dried with anhydrous Na₂SO₄. After

Na₂SO₄was filtered, solvent was removed in vacuum. The residue was

purified by silica gel column chromatography with MeOH: ammonia

saturated CHCl₃(2%) to afford compound 22 as a brown liquid. ¹H NMR

(400 MHz, CDCl₃) δ 7.67 - 7.62 (m, 2 H), 7.59 - 7.51 (m, 3 H), 7.48 - 7.41

(m, 2 H), 7.41 - 7.35 (m, 2 H), 7.14 - 7.05 (m, 2 H), 6.83 (br. s., 1 H), 5.81

(s, 2 H), 4.45 (br. s., 2H), 3.99 (s, 3 H) 3.86 (s, 3 H). ¹³C NMR (126 MHz,

CDCl₃) δ ppm 157.05 (s, 1 C), 153.49 (s, 1 C), 147.90 (s, 1 C), 147.56 (s,

1 C), 145.19 (s, 1 C), 138.58 (s, 1 C), 134.96 (s, 1 C), 134.92 (s, 1 C),

133.29 (s, 1 C), 132.80 (s, 1 C), 131.26 (s, 1 C), 131.26 (s, 1 C), 128.67

(s, 1 C), 128.67 (s, 1 C), 128.27 (s, 1 C), 127.88 (s, 1 C), 127.88 (s, 1 C),

127.63 (s, 1 C), 126.82 (s, 1 C), 126.09 (s, 1 C), 125.94 (s, 1 C), 125.87

(s, 1 C), 118.33 (s, 1 C), 108.34 (s, 1 C), 106.54 (s, 1 C), 103.72 (s, 1 C),

56.17 (s, 1 C), 55.92 (s, 1 C), 50.80 (s, 1 C), 41.33 (s, 1 C).

1-(5-(3-chlorophenyl)-7,8-dimethoxy-3-(4-(methylsulfonyl)benzyl)-

3H-pyrazolo[3,4-c]isoquinolin-1-yl)-N,N-dimethylmethanamine (19)

To a solution of compound 12a (40 mg, 0.085 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 4-

(methylsulfonyl)benzyl bromide (37.1 mg, 0.140 mmol). After the mixture

was stirred at room temperature for 1.5 h, the solvent was removed in

vacuo. The residue was taken up in solution of saturated aqueous

ammonium chloride and extracted with ethyl acetate twice. The

combined extract was dried with anhydrous Na₂SO₄, filtered and

evaporated. The residue was purified by silica gel column

chromatography with MeOH:CHCl₃(1%) to afford compound 19 as a

white liquid (yield = 73%) ¹H NMR (400 MHz, CDCl₃-d) δ ppm 8.37 (s, 1

H), 7.83 - 7.86 (m, 2 H), 7.69 (dt, J=1.95, 1.09 Hz, 1 H), 7.58 (ddd,

J=6.30, 2.40, 1.60 Hz, 1 H), 7.47 - 7.51 (m, 2 H), 7.42 - 7.46 (m, 2 H),

7.39 (s, 1 H), 5.84 (s, 2 H), 4.12 (s, 3 H), 3.94 (s, 2 H), 3.85 (s, 3 H), 2.99

(s, 3 H), 2.38 (s, 6 H). ¹³C NMR (126 MHz, CDCl₃) δ ppm 157.99 (s, 1 C),

153.54 (s, 1 C), 147.92 (s, 1 C), 147.82 (s, 1 C), 143.76 (s, 1 C), 143.68

(s, 1 C), 141.68 (s, 1 C), 139.63 (s, 1 C), 134.51 (s, 1 C), 129.96 (s, 1 C),

129.96 (s, 1 C), 129.63 (s, 1 C), 129.63 (s, 1 C), 128.97 (s, 1 C), 128.46

(s, 1 C), 128.46 (s, 1 C), 128.46 (s, 1 C), 127.96 (s, 1 C), 127.70 (s, 1 C),

118.56 (s, 1 C), 108.12 (s, 1 C), 107.94 (s, 1 C), 56.18 (s, 1 C), 55.92 (s,

1 C), 55.92 (s, 1 C), 49.90 (s, 1 C), 44.53 (s, 1 C), 44.53 (s, 1 C), 40.99

(s, 1 C).

1-(5-(3-chlorophenyl)-7,8-dimethoxy-3-(4-methoxyphenethyl)-3H-

pyrazolo[3,4-c]isoquinolin-1-yl)-N,N-dimethylmethanamine (25)

To a solution of compound 12a (40 mg, 0.1 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 4-(2-

chloroethyl)anisole (14.5 mg, 0.170 mmol). After the mixture was stirred

at room temperature for 1 h, the solvent was removed in vacuo. The

residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 25 as a pale-brown liquid. ¹H NMR (500 MHz,

DMSO-d₆) δ ppm 8.20 (s, 1 H), 7.57 - 7.65 (m, 4 H), 7.29 (s, 1 H), 6.96

(m, J=8.80 Hz, 2 H), 6.68 (m, J= 8.80 Hz, 2 H), 4.65 (t, J=6.97 Hz, 2 H),

4.01 (s, 3 H), 3.86 (s, 2 H), 3.72 (s, 3 H), 3.62 (s, 3 H), 3.12 (t, J=6.97 Hz,

2 H), 2.27 (s, 6 H). ¹³C NMR (126 MHz, DMSO-d₆) δ ppm 158.18 (s, 1 C),

156.55 (s, 1 C), 153.39 (s, 1 C), 147.70 (s, 1 C), 147.52 (s, 1 C), 142.16

1-(5-(3-chlorophenyl)-7,8-dimethoxy-3-(naphthalen-2-ylmethyl)-3H-

pyrazolo[3,4-c]isoquinolin-1-yl)-N,N-dimethylmethanamine (20)

To a solution of compound 12a (40 mg, 0.1 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 2-

bromomethyl naphthalene (37.6 mg, 0.17 mmol). After the mixture was

stirred at room temperature for 2 h, the solvent was removed in vacuo.

10

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

(s, 1 C), 142.08 (s, 1 C), 133.71 (s, 1 C), 133.92 (s, 1 C), 130.84 (s, 1 C),

130.59 (s, 1 C), 130.16 (s, 1 C), 129.83 (s, 1 C), 129.09 (s, 1 C), 128.81

(s, 1 C), 128.42 (s, 1 C), 117.58 (s, 1 C), 114.29 (s, 1 C), 113.94 (s, 1 C),

107.85 (s, 1 C), 107.32 (s, 1 C), 105.88 (s, 1 C), 56.04 (s, 1 C), 55.89 (s,

1 C), 55.19 (s, 1 C), 55.19 (s, 1 C), 48.50 (s, 1 C), 45.26 (s, 1 C), 45.26

(s, 1 C), 35.50 (s, 1 C).

3-(5-(3-chlorophenyl)-1-((dimethylamino)methyl)-7,8-dimethoxy-3H-

pyrazolo[3,4-c]isoquinolin-3-yl)propanenitrile (29)

To a solution of compound 12a (40 mg, 0.085 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 3-

bromopropioinitrile (23.2 mg, 0.170 mmol). After the mixture was stirred

at room temperature for 1.5 h, the solvent was removed in vacuo. The

residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 29 as a white liquid (yield = 60%). ¹H NMR (400

MHz, CDCl₃) δ 8.35 (s, 1 H), 7.75 (dt, 1 H), 7.62 (ddd, 1 H), 7.55 - 7.50

(m, 2 H), 7.39 (s, 1H), 4.88 (t, 2 H), 4.14 (s, 3 H), 3.98 (s, 2 H), 3.87 (s, 3

H), 3.08 (t, 2 H), 2.43 (s, 6 H). ¹³C NMR (126 MHz, CDCl₃) δ ppm 157.30

(s, 1 C), 153.49 (s, 1 C), 147.80 (s, 1 C), 147.63 (s, 1 C), 141.63 (s, 1 C),

134.52 (s, 1 C), 129.94 (s, 1 C), 129.64 (s, 1 C), 128.98 (s, 1 C), 128.94

(s, 1 C), 128.94 (s, 1 C), 127.95 (s, 1 C), 118.60 (s, 1 C), 117.08 (s, 1 C),

108.27 (s, 1 C), 107.88 (s, 1 C), 105.51 (s, 1 C), 56.10 (s, 1 C), 55.89 (s,

1 C), 55.89 (s, 1 C), 45.31 (s, 1 C), 45.31 (s, 1 C), 42.37 (s, 1 C), 18.57

(s, 1 C).

1-(5-(4-chlorophenyl)-3-isobutyl-7,8-dimethoxy-3H-pyrazolo[3,4-

c]isoquinolin-1-yl)-N,N-dimethylmethanamine (26)

To a solution of compound 12b (40 mg, 0.1 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 4-(2-

chloroethyl)anisole (14.5 mg, 0.170 mmol). After the mixture was stirred

at room temperature for 1 h, the solvent was removed in vacuo. The

residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 26 as a pale-brown liquid. ¹H NMR (400 MHz, CDCl₃)

δ 8.34 (br. s., 1 H), 7.62 (m, J=8.56 Hz, 2 H), 7.53 (m, J=8.31 Hz, 2 H),

7.37 (s, 1 H), 7.09 - 7.03 (m, 2 H), 6.76 - 6.71 (m, 2 H), 4.76 (br. S., 2 H),

4.17 - 4.09 (s, 3H), 3.95 (br. s., 2 H), 3.85 (s, 3 H), 3.73 (s, 3 H), 3.22 (t,

J=6.85 Hz, 2 H), 2.39 (br. s., 6 H). ¹³C NMR (126 MHz, CDCl₃) δ ppm

158.10 (s, 1 C), 153.13 (s, 1 C), 147.67 (s, 1 C), 147.42 (s, 1 C), 138.71

(s, 1 C), 134.75 (s, 1 C), 131.19 (s, 1 C), 131.19 (s, 1 C), 131.19 (s, 1 C),

130.65 (s, 1 C), 129.89 (s, 1 C), 129.89 (s, 1 C), 129.89 (s, 1 C), 128.57

(s, 1 C), 128.57 (s, 1 C), 128.57 (s, 1 C), 118.20 (s, 1 C), 113.71 (s, 1 C),

113.71 (s, 1 C), 113.71 (s, 1 C), 107.95 (s, 1 C), 103.62 (s, 1 C), 56.04

(s, 1 C), 55.89 (s, 1 C), 55.19 (s, 1 C), 55.19 (s, 1 C), 48.50 (s, 1 C),

45.26 (s, 1 C), 45.26 (s, 1 C), 35.50 (s, 1 C).

1-(5-(3-chlorophenyl)-3-isobutyl-7,8-dimethoxy-3H-pyrazolo[3,4-

c]isoquinolin-1-yl)-N,N-dimethylmethanamine (30)

To a solution of compound 12a (40 mg, 0.1 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 1-bromo-

2-methyl propane (23.2 mg, 0.170 mmol). After the mixture was stirred at

room temperature for 1 h, the solvent was removed in vacuo. The

residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 30 as a yellow liquid. ¹H NMR (400 MHz, CDCl₃) δ

8.34 (s, 1 H), 7.73 (t, J=1.83 Hz, 1 H), 7.61 (m, 1 H), 7.53 - 7.48 (m, 2 H),

7.40 - 7.37 (m, 1 H), 4.41 - 4.36 (m, 2 H), 4.13 (s, 3 H), 3.95 (s, 2 H),

3.86 (s, 3 H), 2.39 (s, 7 H), 0.96 - 0.91 (m, 6 H).¹³C NMR (126 MHz,

CDCl₃) δ ppm 157.58 (s, 1 C), 153.54 (s, 1 C), 148.00 (s, 1 C), 147.70 (s,

1 C), 142.42 (s, 1 C), 141.85 (s, 1 C), 134.47 (s, 1 C), 130.01 (s, 1 C),

129.58 (s, 1 C), 128.87 (s, 1 C), 128.40 (s, 1 C), 128.03 (s, 1 C), 118.24

(s, 1 C), 108.15 (s, 1 C), 106.83 (s, 1 C), 103.99 (s, 1 C), 60.12 (s, 1 C),

56.21 (s, 1 C), 56.21 (s, 1 C), 55.90 (s, 1 C), 54.20 (s, 1 C), 54.20 (s, 1

C), 29.40 (s, 1 C), 20.08 (s, 1 C), 20.08 (s, 1 C).

1-(5-(2-chlorophenyl)-7,8-dimethoxy-3-(4-methoxyphenethyl)-3H-

pyrazolo[3,4-c]isoquinolin-1-yl)-N,N-dimethylmethanamine (27)

To a solution of compound 12c (40 mg, 0.01 mmol) in dried DMF in the

presence of cesium carbonate (55 mg, 0.170 mmol) was added 4-(2-

chloroethyl)anisole (14.5 mg, 0.170 mmol). After the mixture was stirred

at room temperature for 1 h, the solvent was removed in vacuo. The

residue was taken up in solution of saturated aqueous ammonium

chloride and extracted with ethyl acetate twice. The combined extract

was dried with anhydrous Na₂SO₄, filtered and evaporated. The residue

was purified by silica gel column chromatography with MeOH:CHCl₃(1%)

to afford compound 27 as a transparent liquid. ¹H NMR (400 MHz,

CDCl₃) δ 8.31 (s, 1 H), 7.60 - 7.55 (m, 1 H), 7.49 - 7.44 (m, 2 H), 7.43 -

7.40 (m, 1 H), 7.04 (m, J=8.70 Hz, 2 H), 6.91 (s, 1 H), 6.74 - 6.68 (m, 2

H), 4.77 (t, J=7.56 Hz, 2 H), 4.12 (s, 3 H), 3.95 (s, 2 H), 3.77 (s, 3 H),

3.73 (s, 3 H), 3.22 (t, J=7.33 Hz, 2 H), 2.39 (s, 6 H). ¹³C NMR (126 MHz,

CDCl₃) δ ppm 158.05 (s, 1 C), 156.73 (s, 1 C), 153.29 (s, 1 C), 147.52 (s,

1 C), 138.85 (s, 1 C), 133.40 (s, 1 C), 131.42 (s, 1 C), 130.57 (s, 1 C),

129.88~129.86 (s, 5 C), 129.79 (s, 1 C), 129.78 (s, 1 C), 126.82 (s, 1 C),

118.86 (s, 1 C), 113.69 (s, 1 C), 113.69 (s, 1 C), 113.62 (s, 1 C), 108.03

(s, 1 C), 107.84 (s, 1 C), 56.05 (s, 1 C), 55.82 (s, 1 C), 55.17 (s, 1 C),

55.17 (s, 1 C), 48.51 (s, 1 C), 45.40 (s, 1 C), 45.25 (s, 1 C), 35.41 (s, 1

C).

1-(5-(3-chlorophenyl)-7,8-dimethoxy-3-(2-(1-methylpyrrolidin-2-

yl)ethyl)-3H-pyrazolo[3,4-c]isoquinolin-1-yl)-N,N-

dimethylmethanamine (31)

To a solution of compound 12a (10 mg, 0.021 mmol) in dried DMF in the

presence of cesium carbonate (14 mg, 0.045 mmol) was added 2-(2-

bromoethyl)-1-methylpyrrolidine (5.8 mg, 0.043 mmol). After the mixture

was stirred at room temperature for 1.5 h, the solvent was removed in

vacuo. The residue was taken up in solution of saturated aqueous

ammonium chloride and extracted with ethyl acetate twice. The

combined extract was dried with anhydrous Na₂SO₄, filtered and

11

10.1002/cmdc.201900418

ChemMedChem

FULL PAPER

evaporated. The residue was purified by silica gel column

chromatography with MeOH:CHCl₃(1%) to afford compound 31 as a

white liquid (yield = 62%) ¹H NMR (400 MHz, DMSO-d₆) δ 8.07 (s, 1 H),

7.76 (s, 1 H), 7.73 – 7.69 (m, 1 H), 7.65 – 7.60 (m, 2 H), 4.98 (t, 2 H),

4.01 (s, 3 H), 3.87 (s, 2 H), 3.75 (s, 3 H), 3.62 (s, 3 H), 3.15 (q, 2 H), 2.27

(s, 6 H), 2.18 – 1.63 (m, 6H).

The predicted compounds were tested for antagonist activity using an

intracellular calcium release assay. HEK/Ga15 stable cell line (Millipore)

expressing Ga15 protein to measure intracellular Ca²⁺ion release upon

receptor activation was transfected with opioid receptor genes (µ-OR1 or

κ-OR1). The compound solution was diluted with Ca²⁺buffer solution to

give 0.1% final concentration in cells. For antagonist assay, on the day

before the assay, frozen cell was re-suspended in fresh media and

seeded at 5*10^4/well, 100 µL/well in 96-well plate. Cells were cultured

overnight at 37°C with 5% CO₂. After 1 h the calcium dye loaded into the

wells, each compound was treated for 10 µM of working concentration.

The assay plates were incubated at 37°C for additional 1 h. Assay plates

were then placed onto the FLIPR tetra fluorescence plate reader for

treating 50 µL of the 0.08 mg/mL dynorphin solution for κ-OR 1 (or 0.2

mg/mL DAMGO for intracellular Ca²⁺ion release upon receptor activation

was transfected with opioid receptor genes (µ-OR 1) and measuring the

changes of intracellular calcium efflux response to the opioid receptor

activation. For each compound, relative fluorescence units (RFU) were

established by measuring the fluorescence every second from ten

seconds before treating agonist for 60 seconds and determining the peak

effect (maximum-minimum). %inhibition was calculated with mean value

of background treated (100%) and agonist only treated (0%).

1-(5-(4-chlorophenyl)-3-isobutyl-7,8-dimethoxy-3H-pyrazolo[3,4-

c]isoquinolin-1-yl)-N,N-dimethylmethanamine (33)