Structure-Aided Identification and Optimization of Tetrahydro-isoquinolines as Novel PDE4 Inhibitors Leading to Discovery of an Effective Antipsoriasis Agent

Article abstract of DOI:10.1021/acs.jmedchem.9b00518

Psoriasis is a common, chronic inflammatory disease characterized by abnormal skin plaques, and the effectiveness of phosphodiesterase 4 (PDE4) inhibitor to lessen the symptoms of psoriasis has been proved. Aiming to find a novel PDE4 inhibitor acting as an effective, safe, and convenient therapeutic agent, we constructed a library consisting of berberine analogues, and compound 2 with a tetrahydroisoquinoline scaffold was identified as a novel and potent hit. The structure-aided and cell-based structure-activity relationship studies on a series of tetrahydro-isoquinolines lead to efficient discovery of a qualified lead compound (16) with the high potency and selectivity, well-characterized binding mechanism, high cell permeability, good safety and pharmacokinetic profile, and impressive in vivo efficacy on antipsoriasis, in particular with a topical application. Thus, our study presents a prime example for efficient discovery of novel, potent lead compounds derived from natural products using a combination of medicinal chemistry, biochemical, biophysical, and pharmacological approaches.

Full text of DOI:10.1021/acs.jmedchem.9b00518

Subscriber access provided by University of Rochester | River Campus & Miner Libraries
Article
Structure-Aided Identification and Optimization of
Tetrahydro-isoquinolines as Novel PDE4 Inhibitors
Leading to Discovery of an Effective Anti-Psoriasis Agent
Xianglei Zhang, Guangyu Dong, Heng Li, Wuyan Chen, Jian Li, Chunlan Feng,
Zhanni Gu, Fenghua Zhu, Rui Zhang, Minjun Li, Wei Tang, Hong Liu, and Yechun Xu
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 17 May 2019
Downloaded from http://pubs.acs.org on May 17, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure-Aided Identification and Optimization of Tetrahydro-
isoquinolines as Novel PDE4 Inhibitors Leading to Discovery of an
Effective Anti-Psoriasis Agent
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Xianglei Zhang,Δ,§,† Guangyu Dong,§,‡,† Heng Li,Δ,§,† Wuyan Chen,Δ Jian Li,Δ Chunlan
Feng,Δ Zhanni Gu,Δ Fenghua Zhu,Δ Rui Zhang,Δ,§ Minjun Li,ǂ Wei Tang,Δ,§,* Hong
Liu,Δ,§,* and Yechun XuΔ,§,*
ΔCAS Key Laboratory of Receptor Research, State Key Laboratory of Drug
Research,
Drug
Discovery
and
Design
Center,
Laboratory
of
Immunopharmacology, Shanghai Institute of Materia Medica, Chinese Academy of
Sciences, Shanghai 201203, China
§University of Chinese Academy of Sciences, Beijing 100049
‡School of Life Science and Technology, ShanghaiTech University, Shanghai
201210, China
ǂShanghai Synchrotron Radiation Facility, Shanghai Advanced Research Institute,
Chinese Academy of Sciences, Shanghai 201210, China
ABSTRACT
Psoriasis is a common, chronic inflammatory disease characterized by abnormal skin
plaques, and the effectiveness of PDE4 inhibitor to lessen the symptoms of psoriasis
has been proved. Aiming to find a novel PDE4 inhibitor acting as an effective, safe and
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 66
1
2
3
4
5
6
7
8
9
convenient therapeutic agent, we constructed a library consisting of berberine analogs
and compound 2 with a tetrahydro-isoquinoline scaffold was identified as a novel and
potent hit. The structure-aided and cell-based SAR studies on a series of tetrahydro-
isoquinolines lead to efficient discovery of a qualified lead compound (16) with the
high potency and selectivity, well-characterized binding mechanism, high cell-
permeability, good safety and pharmacokinetic profile, and impressive in vivo efficacy
on anti-psoriasis in particular with a topical application. Thus, our study presents a
prime example for efficient discovery of novel, potent lead compounds derived from
natural products using a combination of medicinal chemistry, biochemical, biophysical,
and pharmacological approaches.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
Cyclic nucleotide phosphodiesterases (PDEs) catalyse the hydrolysis of cAMP and
cGMP, regulating the cAMP- and cGMP-associated signalling pathways as well as the
resulting physiological and pathophysiological responses, and they are proving to be
fruitful targets for drug discovery.1-3 PDE4, one of the 11 known human PDEs,
specifically catalyzes the hydrolysis of cAMP which in turn interferes with the function
of transcription factors like CREB and NF-kB, altering the expression of inflammatory
mediators such as TNF-α, IFN-γ, IL-12, and IL-10.4,5 Moreover, PDE4 isoforms
(PDE4A/4B/4C/4D) have a relatively high level of expression in cells that regulate
inflammatory responses.6 Accordingly, therapeutic benefits of PDE4 inhibitors have
2
ACS Paragon Plus Environment

Page 3 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
been used in the treatment of inflammatory diseases including chronic obstructive
pulmonary disease (roflumilast), psoriasis (apremilast) and atopic dermatitis
(crisaborole) (Figure 1).7-10 Given the effectiveness of the target for antiinflammation,
there is considerable interest in the discovery of novel PDE4 inhibitors.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Structures and inhibitory activities of berberine and representative PDE4
inhibitors including three approved drugs (roflumilast, apremilast and crisaborole).
Psoriasis is a common, chronic skin disease with a prevalence of ~2% worldwide.11
It has been recognized as an immune-mediated genetic disease or a systemic
inflammatory disease manifested as raised, well-demarcated, erythematous oval
plaques with adherent silvery scales.12 Conventional therapies include topical therapies
and biologics. Glucocorticosteroids are common topical agents for psoriasis treatment,
however, the side effect of skin atrophy hinders its long-term use. Although several
biologics including etanercept, adalimumab and infliximab have been approved for the
treatment of psoriasis and psoriatic arthritis, they are expensive and require either
subcutaneous or intravenous administration.13 There is a substantial unmet need for the
treatment of psoriasis with cheaper oral drugs or safer topical agents. Apremilast, an
oral small-molecule PDE4 inhibitor, was approved by the FDA in 2014 for the
treatment of moderate-to-severe plaque psoriasis.14 Crisaborole, another PDE4
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 66
1
2
3
4
5
6
7
8
9
inhibitor, used as a topical medication for the treatment of mild-to-moderate atopic
dermatitis was also approved by the FDA in 2016.10,15 It prompts us to seek a new
therapeutic agent for psoriasis targeting PDE4.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Traditional Chinese medicine (TCM) has evolved over thousands of years and is
an invaluable source of inspiration for drug discovery. The therapeutic benefits of TCM
led us to extract its active components, the vast majority of which are natural products,
as lead compounds for drug development.16 Berberine (BBR), the major active
ingredient of Coptis chinensis, is a small-molecule alkaloid and is broadly used as an
antimicrobial drug.17,18 Mounting evidence also suggests that BBR has a promising
pharmacological efficacy on inflammation.19-21 It has been reported that BBR processes
anti-inflammatory activities and suppresses pro-inflammatory responses by inhibition
of mitogen-activated protein kinase (MAPK) signalling and cellular reactive oxygen
species (ROS) production, although the full range of biological target(s) of BBR
remains unclear.22 In addition, BBR possesses a typical catechol group, which is a
common pharmacophore in known PDE4 inhibitors (Figure 1). Therefore, we built a
chemical library consisting of precursors, intermediates and various derivatives of
berberine in order to discover a new scaffold PDE4 inhibitor for the treatment of
psoriasis (Figure 2).
RESULTS AND DISCUSSION
4
ACS Paragon Plus Environment

Page 5 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Hit Discovery and Chirality Identification. At the outset of our project, we used
virtual screening to efficiently dig out hit compounds from our in-house BBR analogues
library. The top 30 compounds ranked by scoring values were subsequently evaluated
for inhibitory activity on PDE4D by a scintillation proximity assay (SPA).23 Among
these compounds, we found a tetrahydro-isoquinoline-based compound (compound 1)
as a novel scaffold inhibitor of PDE4. It turned out to be a racemic compound with an
IC50 of 0.51 μM. We synthesized and soaked the racemic entity with apo crystals of the
catalytic domain of PDE4D, determining the complex structure of PDE4D-1 (Scheme
1 and Table S1, PDB code: 6INM). It revealed that the (S)-enantiomer occupied the
active site of PDE4D (Figure S1). A further separation and purification of two
enantiomers followed by the SPA-based activity assessment indicated that (S)-
enantiomer (compound 2) is more potent than (R)-enantiomer (compound 3), with an
IC50 of 0.27 μM and 3.71 μM, respectively (Scheme 1). We further solved the crystal
structure of PDE4D soaked with the (S)-enantiomer which results in a binding
conformation of the compound nearly identical to that in the PDE4D-1 complex, but
we were unable to obtain the crystals of PDE4D bound with the (R)-enantiomer (Figure
S1 and Table S1, PDB code: 6INK). As confirmed by the SPA assay together with the
proof for the binding conformation of the ligand in the crystal structures, compound 2
was identified as a good hit for further optimization.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 66
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Structure-aided and cell-based discovery and design of the lead compound
(16) based on an in-house BBR analogues library.
Scheme 1. Synthesis Route of Compounds 1-3.
O
OH
H
O
O
NH2
NH
a
b
O
O
N
+
O
N
H
1a
3a
2a
O
O
O
O
d
d
c
O
NH
N
5a
6a
1
HN
HN
O
O
O
O
O
N
.
HCl
O
e
.
HCl
O
NH
N
HN
4a
2
HN
HN
O
O
O
O
f
d
.
HCl
O
NH
N
7a HN
HN
3
Reagents and conditions: (a) EDCI, HOBt, TEA, rt, DCM, 80-90%; (b) POCl3, CH3CN
reflux, 90-95%; (c) NaBH4, MeOH, 82%; (d) HCOOEt, TEA reflux, 72-85%; (e)
RuCl[(R,R)-Tsdpen](p-cymene), AgSbF6, La(OTf)3, HCOONa, H2O/MeOH= 2:1, Ar,
6
ACS Paragon Plus Environment

Page 7 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
40℃, 65-73%; (f) RuCl[(S,S)-Tsdpen](p-cymene), AgSbF6, La(OTf)3, HCOONa,
H2O/MeOH = 2:1, Ar, 40℃, 65-73%.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Structure-Aided SAR Studies on Substituents of the Tetrahydro-isoquinoline
Ring. In order to efficiently carry out the hit-to-lead optimization, it is of utmost
importance to understand the underlying structure-activity relationship (SAR) of
tetrahydro-isoquinolines binding with PDE4D. In addition, compound 2 possesses an
excellent potency with a submicromolar IC50. It is thus interesting to evaluate
contribution of each functional group of the scaffold to binding and inhibition of
PDE4D. To this end, the flexible “tail” (the indole ring attached with the ethylidene)
was removed from 2 to yield a fragment only containing the tetrahydro-isoquinoline
core (compound 4) (Figure 2). The crystal structure determination revealed that
compound 4 could bind into the pocket of PDE4D with two orientations associated with
hydrogen bond (H-bond) formation or not between the carbonyl group and H160
(Figure 3A and Table S1, PDB codes: 6IMB and 6IMD). This is consistent with the
thermodynamic profile of 4 determined by isothermal titration calorimetry (ITC) that it
has a relatively weak binding enthalpy while its entropic contribution is favourable
(Table S2). It is not difficult to understand that the lack of the “tail” and the existence
of two identical methoxyl groups allow the tetrahydro-isoquinoline ring to adopt two
different bound conformations in the enzyme. However, the complex structures
together with the thermodynamic data suggest that interaction of the carbonyl group
with H160 is not sufficient to fix the binding conformation of 4. To further validate and
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 66
1
2
3
4
5
6
7
8
9
understand it, we designed and synthesized compounds 5 and 6 by replacement of each
methoxyl group of 4 with an ethyoxyl one (Figure 2 and Scheme 2). The solved crystal
structures show that the ethyoxyl group of 5 or 6 always inserts into a hydrophobic sub-
pocket, named the Q-pocket, formed by the sidechains of N321, Y329, T333, and I336,
playing a key role in determination of the binding conformation of the compounds
(Figure 3B,C and Table S1, PDB codes: 6IM6 and 6IMI). Similar to what was found
with two bound conformations of 4, the carbonyl group of 5 makes the H-bond to H160
with a distance of 2.7 Å while such a H-bond is not established in the case of 6 (Figure
S2). In addition, superimposition of the structures of 4, 5 and 6 suggests that a small
shift of the tetrahydro-isoquinoline ring of 5 towards the solvent region results in
stronger π-π stacking interactions with F372 (Figure S2). These underlying interactions
lay a solid foundation for the fact that inhibition of 5 and 6 to PDE4D is much improved
compared to 4 and the activity of 5 is higher than that of 6 (Figure 2). Moreover, the
hydrophobic interactions of the methoxyl or ethyoxyl group with the Q-pocket and the
H-bond of the carbonyl group to H160 both contribute to binding of the tetrahydro-
isoquinoline core to PDE4D, although the former is more significant than the latter. As
a result, compound 5 with a best inhibitory activity among three fragments and a similar
binding conformation to the tetrahydro-isoquinoline of compound 2 provides a good
candidate for further optimization.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Synthesis Route of Compounds 4-6.
8
ACS Paragon Plus Environment

Page 9 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
R1
O
R1
O
NH2
1.HCOOH,(CHO)X
2.HCOOEt, Et3N
N
H
O
R2
O
R2
O
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1a: R1 = Me; R2 = Me
1b: R1 = Me; R2 = Et
1c: R1 = Et; R2 = Me
4: R1 = Me; R2 = Me
5: R1 = Me; R2 = Et
6: R1 = Et; R2 = Me
Figure 3. Crystal structures of the catalytic domain of PDE4D in complex with
compounds 4 (cyan and green, PDB codes: 6IMB and 6IMD), 5 (magenta, PDB code:
6IM6) and 6 (yellow, PDB code: 6IMI). (A) Two opposite orientations of 4 were found
in the binding pocket of PDE4D. (B,C) The ethyoxyl groups of 5 and 6 both insert into
a hydrophobic sub-pocket (the Q-pocket) formed by the sidechains of N321, Y329,
T333, and I336 (grey surface). Dash lines indicate H-bonds between compounds and
residues.
Structure-Aided SAR Studies on the “Tail”. The IC50 of 4 is 58.5 μM, which is >200-
fold drop in activity relative to compound 2, suggesting that the “tail” which includes
the indole ring and the ethylidene group exerts a significant influence on inhibition of
2 to PDE4D (Figure 2). Analysis of the crystal structure of PDE4D-2 shows the terminal
indole ring is not well fixed and it forms few hydrophobic interactions with residues
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 66
1
2
3
4
5
6
7
8
9
M273 and M357 (Figures 4A and S1). Moreover, occupation of the solvent accessible
region by the “tail” leads to the release of water molecules from the binding pocket
(Figure S3). These binding features result a dramatic entropic contribution to the
binding affinity of 2 to PDE4D (Table S2). These data make us to speculate that the
length of the linker between two rings may affect flexibility of the compound and the
binding of the indole ring with surrounding residues. In view of this, we changed the
linker between the indole ring and the tetrahydro-isoquinoline ring, i.e. the ethylidene
in 2 replaced by a methylene and propylidene in compounds 7 and 8, respectively. The
IC50 values of 7 and 8 are 0.65 μM and 4.3 μM, respectively, demonstrating that
shortening or lengthening of the linker does not improve the inhibitory activity.
Analysis of crystal structures of PDE4D in complex with 2, 7 and 8 shows the indole
ring of 7 adopts a similar orientation to that of 2 but in the case of 8 it shifts into a well-
defined hydrophobic sub-pocket formed by M273, H276, H315, and I376 (Figures 4
and S4, and Table S1; PDB codes: 6IMO and 6IMR). A short linker in 7 or the
additional hydrophobic interactions to 8 both reduces the mobility of the indole ring,
which is in line with the reduced entropic contribution to the binding affinities of 7 and
8 to PDE4D (Table S2). Additionally, the carbonyl group of the tetrahydro-isoquinoline
ring of 7 and 8 forms either no, or a weak H-bond to H160, respectively (Figures 4B,C
and S4). Taken together, these results suggest that the “tail” group has a profound
influence on the binding affinity of 2 and an optimal linker between two rings is the
ethylidene for keeping potency of the compound.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
ACS Paragon Plus Environment

Page 11 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. The effect of different linkers on orientations of the indole ring. Surface
representation of the PDE4-2 (cyan, A, PDB code: 6INK), PDE4-7 (magenta, B, PDB
code: 6IMO), PDE4-8 (green, C, PDB code: 6IMR), and PDE4-16 (yellow, D, PDB
code: 6IND) complexes. Dash lines indicate H-bonds between compounds and residues.
Hit-to-Lead Optimization. After fully understanding the binding signature of each
part of compound 2, we decided to keep the ethylidene linker between the tetrahydro-
isoquinoline ring and the indole ring, and only carry out minor modifications on two
rings to improve the enzymatic as well as the cell-based activities (Tables 1 and 2). It
has been well demonstrated that elevated levels of cAMP by inhibition of PDE4 would
reduce the release of TNF-α, a key cytokine in the inflammatory cascade, from activated
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 66
1
2
3
4
5
6
7
8
9
monocytes and human peripheral blood mononuclear cells (hPBMC).24 The TNF-α
production in lipopolysaccharide (LPS) stimulated hPBMC was thereby measured
using an enzyme linked immunosorbent assay to evaluate inhibition of compounds to
PDE4 in cells.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The multiple crystal structures described above show that the carbonyl group of
our compounds has two orientations with or without the H-bond to H160 and it seems
that the contribution of this group to the binding of compounds to PDE4D is not
efficient. Given that this group faces to and is close to two metal ions in the binding
pocket, it is possible to introduce a larger group to form interactions with the metal ions,
like the binding of roflumilast to PDE4D (Figures 1 and S5). We thereby synthesized
compounds 9-14 by introducing the isonicotinoyl (9), tetrahydro-2H-pyran-4-carbonyl
(10), (tetrahydro-2H-pyran-4-yl)methyl (11), 2,2,2-trifluoroethyl (12), 2-methoxy-2-
oxoethyl (13), or 2-methoxyethyl (14) to replace the carbonyl group of compound 1
(Table 1). Unexpectedly, all compounds showed a significant decreased potency to
PDE4D compared to 1. Accordingly, the substitution at R1 position was not favourable
for potency improvement and we keep the carbonyl group in the next optimization steps.
Table 1. Chemical Structures of Compounds 9-14 and Their Inhibitory Activities
against PDE4D and TNF-α Release from hPBMC Stimulated with LPS.
12
ACS Paragon Plus Environment

Page 13 of 66
Journal of Medicinal Chemistry
1
2
3
O
4
5
6
N
O
R1
7
8
9
HN
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PDE4D inhibition
TNF-α inhibition
(1 μM)
Compds
9
R1
(1 μM)
N
O
O
30.7%
39.8%
O
O
10
11
16.4%
33.3%
34.6%@ 10 μM
36.7%@ 10 μM
F
F
12
19.5%
39.1%
19.2%
N/Aa
21.7%
F
O
13
14
O
O
25.7%@ 10 μM
aNot available.
Scheme 3. Synthesis Route of Compounds 9-14.
O
O
O
OH
N
.
HCl
H
O
O
NH2
NH
a
b
O
O
N
+
O
N
H
HN
F
4a
1a
3a
2a
N
F
O
O
O
12: R1 =
9: R1 =
N
F
NH
d
R1
c
O
O
O
O
10: R1 =
11: R1 =
13: R1 =
14: R1 =
O
O
O
HN
O
HN
5a
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 66
1
2
3
4
5
6
7
8
9
Reagents and conditions: (a) EDCI, HOBt, TEA, rt, DCM, 80-90%; (b) POCl3, CH3CN
reflux, 90-95%; (c) NaBH4, MeOH, 82%; (d) (e) R’X /R’OTf, KI, K2CO3, CH3CN,
reflux or R’’COX,TEA, DCM or R’’COOH, EDCI, DMAP, DCM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As mentioned above, the indole ring of compound 2 oriented towards a
hydrophobic region but had few interactions with surrounding residues (Figures 4A, S1
and S6). To increase hydrophobic interactions of the indole ring with residues, we
added a fluorine (15) or a methyl group (16) to the indole ring first (Scheme 4).
Determination of complex structures of 15 and 16 with PDE4D shows that the indole
rings of two compounds formed more hydrophobic interactions with residues compared
to 2 (Figure S6, PDB codes: 6IMT and 6IND), which is in line with a higher enthalpic
contribution of 16 over 2 to the binding affinity of the compound with the enzyme
(Table S2). The enzymatic IC50 of 15 and 16 is 0.25 and 0.24 μM, respectively, which
is slightly better than that of 2 (0.27 μM). Surprisingly, compound 16 has a significant
inhibition on the TNF-α release in hPBMC, yielding an IC50 of 0.65 μM. It is more
potent than 2 or 15 (Table 2). Given that PDE4 is an intracellular target, the enzymatic
and cell-based activities of 2, 15 and 16 indicate that the intrinsic physicochemical
property of compound 16 may enable it to be more permeable, which is further
validated by a cell permeability experiment. The data in Table 3 show that both
compound 16 and apremilast possess a perfect cell-penetrating ability.
14
ACS Paragon Plus Environment

Page 15 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Encouraged by the good potency of 16 in cells, we carried out further modification
on this compound. The aforementioned structure-aided SAR studies on substituents of
the tetrahydro-isoquinoline ring have suggested that the ethyl group at R2 position is
better than the methyl one to fit into the Q-pocket and thus improves the inhibitory
activity of compound 5 as compared to 4. In addition, in the representative PDE4
inhibitors shown in Figure 1 imply that a group larger than the ethyl group, such as a
cyclopropylmethyl or a cyclopentyl, is also possible to be used as substituents of R2.
Therefore, an ethyl group, a cyclopropylmethyl, and a cyclopentyl were introduced at
the R2 position of compounds 17, 18 and 19, respectively. As anticipated, the use of
the ethyl group increased the inhibitory activity of 17 on PDE4D at the enzymatic level.
Its IC50 is 0.098 µM which is about 3-fold potent of 16. The cyclopropylmethyl in 18
and the cyclopentyl group in 19, in particular the latter, seem too large for the Q-pocket
as their enzymatic activities reduced one by one in comparison with 17 or 16. One
possibility is that these large hydrophobic groups may compete with the phenyl ring of
the indole for the same space, and one group pushes the other out of alignment and so
binding is lost. Unfortunately, the potency of three compounds in hPBMC is all worse
than compound 16. Furthermore, we designed and synthesized compounds 20 and 21
in which an ethyl group and a cyclopropylmethyl were installed at R2, respectively, and
a hydrogen was used at R3. The enzymatic IC50 of 20 is better than 16 but is worse than
17 and meanwhile 21 shows a decreased potency to PDE4D compared to 18, which are
consistent with the above SAR studies. However, the potency of 20 and 21 on the TNF-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 66
1
2
3
4
5
6
7
8
9
α release in hPBMC is also worse than 16, demonstrating again the importance of the
methyl group at the indole ring for 16 to maintain its potency in cells. Eventually,
compound 16 was selected for further evaluation of selectivity, pharmacokinetics and
in vivo efficacy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Impact of R2- and R3-Substitutents on Enzymatic (PDE4D) Activity and
Inhibition of TNF-α Release from hPBMC Stimulated with LPS.
O
R2
N
O
O
HN
R3
PDE4D
TNF-α
Compds
R2
R3
IC50 (μM)
IC50 (μM)
apremilast
-
-
-
0.074a
0.49b
0.11a
0.54b
crisaborole
-
2
Me
Me
Me
Et
H
F
0.27 ± 0.01
2.73 ± 0.17
5.83 ± 1.72
0.65 ± 0.12
36.5%@ 1 μM
15
16
17
0.25 ± 0.003
0.24 ± 0.02
Me
Me
0.098 ± 0.004
18
Me
0.70 ± 0.01
2.16 ± 0.04
19
20
Me
H
24%@ 1 μM
12.1%@ 1 μM
Et
0.13 ± 0.01
2.71 ± 0.70
21
H
0.82 ± 0.17
2.58 ± 0.72
a,bThe inhibitory activity data of apremilast and crisaborole are cited from
references.14,25
Scheme 4. Synthesis Route of Compounds 7-8 and 15-21.
16
ACS Paragon Plus Environment

Page 17 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
R3
H
O
NH2
n
O
N
OH
a
b
n
R2
+
R2
O
NH
O
O
R3
N
H
3b: R2 = Me, R3 = H, n = 1
3c: R2 = Me, R3 = H, n = 3
3d: R2 = Me, R3 = F, n = 2
3e: R2 = Me, R3 = Me, n = 2
2a: R3 = H, n = 2
2b: R3 = H, n = 1
2c: R3 = H, n = 3
2d: R3 = Me, n = 2
2e: R3 = F, n = 2
3b-j
1a: R2 = Me
1b: R2 = Et
1d: R2 = cyclopropylmethyl
1e: R2 = cyclopentyl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3f: R2 = Et, R3 = Me, n = 2
3g: R2 = cyclopropylmethyl, R3 = Me, n = 2
3h: R2 = cyclopentyl, R3 = Me, n = 2
3i: R2 = Et, R3 = H, n = 2
3j: R2 = cyclopropylmethyl, R3 = H, n = 2
O
7: R2 = Me, R3 = H, n = 1
8: R2 = Me, R3 = H, n = 3
15: R2 = Me, R3 = F, n = 2
16: R2 = Me, R3 = Me, n = 2
O
O
R2
O
N
R2
.
NH HCl
N
R2
O
O
.
HCl
NH
O
n
n
c
n
d
NH
NH
17: R2 = Et, R3 = Me, n = 2
18: R2 = cyclopropylmethyl, R3 = Me, n = 2
19: R2 = cyclopentyl, R3 = Me, n = 2
20: R2 = Et, R3 = H, n = 2
R3
R3
R3
4b-j
6b-j
7-8 & 15-21
21: R2 = cyclopropylmethyl, R3 = H, n = 2
Reagents and conditions: (a) EDCI, HOBt, TEA, rt, DCM, 80-90%; (b) POCl3, CH3CN
reflux, 90-95%; (c) RuCl[(R,R)-Tsdpen](p-cymene), AgSbF6, La(OTf)3, HCOONa,
H2O/MeOH= 2:1, Ar, 40 ℃, 65-73%, (d) HCOOEt, TEA reflux, 72-85%.
Table 3. Cell Permeability Results of Compound 16 and Apremilast.
Mean
a
Flow
Mean
Mean Papp
(10-6cm/s)
Efflux Permeability
Ratioc
b
Compds
Fabs
Direction
Recovery (%)
classd
(%)
apremilast
apremilast
apical to basal
basal to apical
96.68 ± 4.81
92.12 ± 2.78
14.93 ± 1.27
29.02 ± 1.33
100
1.9
high
16
16
apical to basal
basal to apical
92.27 ± 2.51
96.84 ± 0.27
18.43 ± 0.70
19.87 ± 3.75
100
1.1
high
aThe Papp values in the A to B (A-B) or B to A (B-A) were calculated by the following
formula: Papp=(VA/(Area×Time))×([drug]acceptor/[drug]initial donor) where, VA=the volume
in the acceptor well (in this assay: 0.3 or 1.0 mL), Area=the surface area of the
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 66
1
2
3
4
5
6
membrane (in this assay: 0.33 cm2), Time=the total transport time in seconds (in this
assay: 5400 s), Papp values are expressed as 10-6 cm/s.
7
8
9
bHuman absorbed fraction (Fabs).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
cThe B-A/A-B ratio was calculated by dividing the Papp values from B to A by the
Papp values from A to B as follows. Efflux ratio (B-A/A-B ratio) = Papp (B to A) /Papp
(A to B)
dFabs % > 70%, high permeability; 30% < Fabs % < 70%, moderate permeability; Fabs
%
< 30%, poor permeability.
Selectivity and Safety Profiles of Compound 16. Prior to evaluate in vivo efficiency,
selectivity study of compound 16 against other PDE isoforms and dopamine receptor
D1/D2 which are potential targets of berberine were performed.26 The data in Table 4
reveal that 16 is highly selective for PDE4D vs PDE1, 2, 5-9, and PDE11 and exhibits
at least 4-fold selectivity over PDE3 and PDE10. As shown in Figure S7, compound 16
is neither an agonist nor an antagonist for dopamine receptor D1/D2. Moreover, we
tested hERG (the human Ether-à-go-go-Related Gene) inhibition effect of 16 using
QPatch of automated patch clamping measurement (Figure S8).27 It inhibits hERG with
an IC50 value over 40 µM, much weaker than that of a real hERG inhibitor (cisapride,
IC50: 0.07 µM). These data indicate that 16 is a PDE4 selective inhibitor without the
inhibition to dopamine receptors or hERG.
Table 4. Selectivity of Compound 16 to 11 PDE Enzymes
PDEs
IC50 (μM)
Selectivety Index
18
ACS Paragon Plus Environment

Page 19 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
PDE1Aa
PDE2Aa
PDE3Aa
PDE4Db
PDE5Ab
PDE6Ca
PDE7Ab
PDE8A1a
PDE9Ab
PDE10Ab
PDE11A4a
>50
>200
>200
-
>50
56.6%@ 1 µM
0.24 ± 0.01
>5
-
>20
13
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.3 ± 0.03
>5
>20
>200
>200
4
>50
>50
0.92 ± 0.07
>50
>200
aCommercial recombinant human full length PDEs (BPS Bioscience Inc. U.S.).
bRecombinant human PDE catalytic domain prepared in our laboratory.
Pharmacokinetic Profiles of 16. A preliminary pharmacokinetic assessment was
performed on 16 and the results were summarized in Table 5. Six male Sprague-Dawley
rats were randomly divided into two groups. After oral administration of a 25 mg/kg
dose of 16 to a group of three SD rats, it revealed a Tmax of 1.33 h, t1/2 of 0.94 h and a
good oral bioavailability of 68.3%. The same experiment with intravenous
administration of 16 at a dose of 10 mg/kg revealed a moderate plasma clearance (CL
= 1.70 L/h/kg), but a much smaller volume of distribution and a shorter t1/2.
Table 5. Pharmacokinetic Data of Compound 16 in Rats
Tmax
Cmax
AUC0-t
AUC0-∞
MRT
(h)
t1/2
CL
Vss
F
route
(h)
(ng/mL) (ng·h/mL) (ng·h/mL)
(h)
(L/h/kg) (L/kg) (%)
IV
/
/
6213±1750
8388±2201
6276±1769
8458±2232
0.92±0.05 0.60±0.07 1.70±0.46 1.56±0.51
/
(10 mg/kg)
PO
1.33±0.58 2447±703
2.50±0.10 0.94±0.20
/
/
68.3
(25 mg/kg)
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 66
1
2
3
4
5
6
7
8
9
In Vivo Efficacy of Compound 16. We further examined anti-inflammatory effects of
16 in animal models. Using apremilast as a positive control, the LPS-induced murine
acute inflammation model was tested at first.28 The compound was given 1.5 h prior to
LPS injection, and concentrations of TNF-α, IL-6, IL-10, and IL-12p40 in serum and
spleen were measured 2 h after the LPS treatment. LPS significantly increased the level
of inflammatory cytokines in the serum and spleen in vehicle mice, while oral
administration of 16 (1 mg/kg) prominently down-regulated the secretion of
inflammatory cytokines, such as TNF-α, IL-6 and IL-12p40 (Figure 5). The results
indicate that compound 16 is able to effectively attenuate inflammation and is
promising for the treatment of inflammatory diseases.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. (A) Histograms representation of inflammation cytokines concentrations in
serum of LPS-induced mice model after an oral administration of compound 16 (1
mg/kg) or apremilast (1 mg/kg) only once at 1.5 h before LPS application (n=6, female).
(B) Similar inflammation cytokines concentration profiles in spleen at the same dose
are displayed (n=6). Data are given as mean ± S.E.M. Statistical analysis is performed
20
ACS Paragon Plus Environment

Page 21 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
by one-way ANOVA. *P < 0.05, **P < 0.01 and ***P < 0.001 compared with vehicle.
#P < 0.05, ##P < 0.01 and ###P < 0.001 compared with normal.
8
9
Owing to the effects of imiquimod (IMQ) on the development of psoriasis-like
skin inflammation, topical application of IMQ-containing cream to the skin of mice is
a typical model for mimicking the clinical symptoms of patients with psoriasis.29,30
Encouraged by a good suppressive effect of 16 in LPS-induced inflammation, we
further investigated efficacy of the compound in the IMQ-induced murine psoriasis
model. Daily application of IMQ-containing cream at a dose of 62.5 mg on BALB/c
mice for 7 days, severe lesions with thick scales and erythema appeared on the back
skin. Strikingly, as shown in Figure 6, an oral administration of 16 with a dose of 25
mg/kg significantly attenuates the clinical features of skin lesions. It even shows much
better therapeutic efficacy than apremilast at the same dose (25 mg/kg) both in
individual scores for scales, thickness and erythema, and in the cumulative score
(Figure 6). These results demonstrate that compound 16 is capable of effective
preventing the IMQ-induced psoriasis-like skin lesions in mice and provides a good
candidate for the development of a new anti-psoriasis drug.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 66
1
2
3
4
5
6
7
8
9
Figure 6. Efficacy of oral administration of 16 or apremilast at a dose of 25 mg/kg once
daily to the IMQ-induced psoriasis-like mice. (A) A representative phenotypic
manifestation of back skin in each group (n=6, female) after IMQ treatment for 7 days.
(B) The cumulative scores were monitored daily ranging from 0 to 12, including scales,
thickness and erythema. Data are given as mean ± SEM. *P < 0.05, **P < 0.01 and
***P < 0.001, compared with vehicle.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Efficacy of Topical Treatment with Compound 16. Local medication is more
convenient and usually less in side effects for psoriasis patients. Therefore, we further
tested therapeutic effects of a topical application of compound 16 in the IMQ-induced
psoriasis-like mice model. The same experiment setup was used to generate the IMQ-
applied mice but they were treated daily with ointment bases or a topical cream
containing 2% of compound 16 at the same dose as that of the IMQ cream (62.5 mg per
day). The topical application of compound 16 yielded a significant therapeutic efficacy
in the IMQ-induced skin inflammation, manifesting with scales, thickness, and
erythema, similar to those were resulted from the oral administration of 16 (Figures 6
and 7).
22
ACS Paragon Plus Environment

Page 23 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 7. The therapeutic efficacy of a topical application of compound 16 in the IMQ-
induced psoriasis-like mice. The cream with or without 2% of compound 16 at a dose
of 62.5 mg once daily were topically applied to the back skin of the mice. (A) A
representative phenotypic manifestation of back skin in each group (n=6, female)
during 7 days of treatment with or without 16. (B) The cumulative scores were
monitored daily ranging from 0 to 12, including scales, thickness and erythema. Data
are given as mean ± SEM. *P < 0.05, **P < 0.01 and ***P < 0.001, compared with
vehicle.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSIONS
Given the effectiveness of PDE4 as a pivotal target for numerous inflammatory
diseases, both in academia and pharmaceutical industry there is continuing interest in
the discovery of novel PDE4 inhibitors with minimal side effects. Here, we reported
the structure-aided identification of a hit compound with a tetrahydro-isoquinoline
scaffold as a novel inhibitor of PDE4 after the screening of a berberine-like analogue
library. The determination of multiple protein-ligand structures in conjunction with the
ligand binding thermodynamic signature leads us to rapidly gain sufficient knowledge
on the binding mode as well as SAR of the tetrahydro-isoquinolines acting on PDE4D.
Small modification on this scaffold results in the discovery of compound 16 as a potent
PDE4 inhibitor both at the enzymatic level and in inhibition of the TNF-α production
in hPBMC. Moreover, the compound has a perfect cell-penetrating ability, a good
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 66
1
2
3
4
5
6
7
8
9
selectivity to PDE4 over other PDEs in most cases, and a good safety reflected in
extremely low inhibition of hERG or dopamine receptor D1/D2 which are potential
targets of berberine. Most importantly, 16 exhibits a significant effect on preventing
LPS-induced acute inflammation and IMQ-induced psoriasis-like skin lesions in mice,
providing a preclinical proof for the topical application of tetrahydro-isoquinolines as
an effective treatment strategy in psoriasis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemistry. General Information. The reagents (chemicals) were purchased from
commercial chemical reagent company, and used without further purification unless
otherwise stated. Analytical thin-layer chromatography (TLC) is HSGF 254. All
products were characterized by their NMR and HRMS spectra. The purity of tested
compounds was determined by HPLC (Agilent Eclipse XBD-C18, 5 µm, 4.6*150 mm,
30 ℃, UV 214/230/238/254 nM, flow rate = 1.0 mL/min) with aqueous CH3CN or
CH3OH (50%-70%) for 15 min. All compounds used for crystal structures
determination and functional studies possess ≥ 95% purity (Table S4). Chemical shifts
were reported in parts per million (ppm, δ) downfield from tetramethylsilane. Proton
coupling patterns are described as singlet (s), doublet (d), triplet (t), quartet (q), multipet
(m), and broad (br).
24
ACS Paragon Plus Environment

Page 25 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The preparation of compounds 1-21 is described below while the chemistry
experimental procedure of key intermediates (3a-7a) is provided in Supporting
Information.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(±)-1-(2-(1H-indol-3-yl)ethyl)-6,7-dimethoxy-3,4-dihydroisoquinoline-2(1H)-
methanal (1). To a solution of 5a (1.98 g, 5.325 mmol) in ethyl formate (20 mL) was
added trimethylamine (catalytic amount, 0.054 g). The mixture was stirred at 50 °C for
12 h. The reaction mixture was extracted with ethyl acetate. The organic layer was
washed with brine, dried over Na2SO4, filtered and concentrated under vacuum. The
residue was purified by the column chromatography (petroleum ether/ethyl acetate=1:1)
to afford the compound 1 (1.55 g, 80%) as a white solid. 1H NMR (500 MHz, CDCl3):
δ 8.32, 8.24 (2 × brs, 1H), 8.14, 8.07 (2 × brs, 1H), 7.63-7.56 (m, 1H), 7.43-7.34 (m,
1H), 7.25 – 7.02 (m, 3H), 6.59, 6.57 (2 × s, 1H), 6.43, 6.42 (2 × s, 1H), 5.51-5.45, 4.60
– 4.53 (2 × m, 1H), 4.50 – 4.42, 3.80-3.70 (2 × m, 1H), 3.87, 3.86(2 × s, 3H), 3.76, 3.70
(2 × s, 3H), 370-3.55, 3.19-3.11 (2 × m, 1H), 3.01 – 2.68 (m, 4H), 2.36-2.20 (m, 2H).
13C NMR (125 MHz, CDCl3): δ 161.98, 161.83, 148.07, 147.86, 147.67, 136.69,
136.46, 128.79, 128.77, 127.29, 127.08, 125.66, 124.57, 122.26, 121.98, 121.91,
121.84, 119.45, 119.18, 118.83, 118.75, 115.49, 114.45, 111.60, 111.42, 111.34,
111.18, 110.03, 109.60, 56.16, 56.01, 55.91, 55.88, 50.47, 40.35, 36.95, 36.25, 34.30,
+
29.41, 27.56, 22.29, 22.13. HRMS (ESI): exact mass calculated for C22H25N2O3
[M+H]+:365.1787, found.365.1857.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 66
1
2
3
4
5
6
7
(S)-1-(2-(1H-indol-3-yl)ethyl)-6,7-dimethoxy-3,4-dihydroisoquinoline-2(1H)-
methanal (2). This compound (a white solid) was prepared by replacement of 5a with
8
9
1
6a using a similar synthetic procedure of product 1. H NMR (400 MHz, CDCl3): δ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8.29, 8.21 (2 × s, 1H), 8.13, 8.06 (2 × brs, 1H), 7.67-7.55 (m, 1H), 7.41-7.34 (m, 1H),
7.24 – 7.06 (m, 3H), 6.57, 6.55 (2 × s, 1H), 6.41, 6.39 (2 × s, 1H), 5.50-5.43, 4.59-4.51
(2 × m, 1H), 4.47-4.40, 3.78-3.72 (2 × m, 1H), 3.84, 3.83 (2 × s, 3H), 3.74, 3.68 (2 × s,
3H), 3.67-3.58, 3.18-3.07 (2 × m, 1H), 3.02 – 2.63 (m, 4H), 2.36 – 2.17 (m, 2H). 13C
NMR (125 MHz, CDCl3): δ 161.81, 161.68, 148.06, 147.83, 147.66, 136.68, 136.46,
128.81, 127.29, 127.08, 125.68, 124.58, 122.26, 121.95, 121.91, 121.79, 119.45,
119.18, 118.83, 118.74, 115.51, 114.47, 111.59, 111.41, 111.34, 111.18, 110.02,
109.56, 56.11, 55.99, 55.91, 55.87, 50.44, 40.27, 36.95, 36.26, 34.26, 29.39, 27.56,
+
22.28, 22.10. HRMS (ESI): exact mass calculated for C22H25N2O3 [M+H]+: 365.1860,
found: 365.1861.
(R)-1-(2-(1H-indol-3-yl)ethyl)-6,7-dimethoxy-3,4-dihydroisoquinoline-2(1H)-
methanal (3). This compound (a white solid) was prepared by replacement of 5a with
1
7a using a similar synthetic procedure of product 1. H NMR (500 MHz, CDCl3): δ
8.32 (brs, 1H), 8.23 (brs, 1H), 7.68-7.56 (m, 1H), 7.43-7.34 (m, 1H), 7.25 – 7.00 (m,
3H), 6.59, 6.57 (2 × s, 1H), 6.43, 6.42 (2 × s, 1H), 5.53-5.46, 4.62 – 4.53 (2 × m, 1H),
4.50 – 4.40, 3.80-3.70 (2 × m, 1H), 3.86, 3.85(2 × s, 3H), 3.76, 3.70 (2 × s, 3H), 370-
13
3.60, 3.22-3.09 (2 × m, 1H), 3.02 – 2.66 (m, 4H), 2.38-2.16(m, 2H). C NMR (125
MHz, CDCl3): δ 162.04, 161.88, 148.07, 147.86, 147.84, 147.67, 136.73, 136.50,
26
ACS Paragon Plus Environment

Page 27 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
128.81, 128.76, 127.28, 127.08, 125.65, 124.59, 122.19, 122.07, 121.96, 121.86,
119.38, 119.13, 118.80, 118.72, 115.37, 114.29, 111.61, 111.49, 111.37, 111.24,
110.04, 109.63, 56.18, 56.02, 55.92, 55.88, 50.51, 40.38, 36.98, 36.25, 34.33, 29.42,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
27.56, 22.30, 22.15. HRMS (ESI): exact mass calculated for C22H25N2O3 [M+H]+:
365.1860, found: 365.1862.
6,7-dimethoxy-3,4-dihydroisoquinoline-2(1H)-carbaldehyde (4). To a solution of
formic acid (20 mL), was slowly added compound 1a (1.81 g, 10 mmol) at 0 °C for 10
min, then polyformaldehyde (1.8 g, 12 mmol) was added. The mixture was stirred at
o
50 C for 12 h. The reaction mixture was cooled to room temperature, and alkalified
pH to 9 by 1 M NaOH. The mixture was quenched with water and extracted with
dichloromethane, washed with brine, dried over Na2SO4, filtered and concentrated. The
residue was dissolved in ethyl formate (20 mL), added trimethylamine (catalytic
amount, 0.101 g). The mixture was stirred at 50 °C for 12 h. The reaction mixture was
extracted with ethyl acetate. The organic layer was washed with brine, dried over
Na2SO4, filtered and concentrated under vacuum. The residue was purified by the
column chromatography (petroleum ether/ethyl acetate=1:1) to afford the titled
compound 4 (1.76 g, 80%) as a white solid. 1H NMR (400 MHz, CDCl3): δ 8.25, δ 8.19
(2 × s, 1H), 6.62, 6.60 (2 × s, 1H), 6.60, 6.57 (2 × s, 1H), 4.61, 4.48 (2 × s, 2H), 3.86
(s, 6H), 3.78, 3.63 (2 × t, J = 6.0 Hz, 2H), 2.88 – 2.74 (m, 2H). 13C NMR (125 MHz,
CDCl3): δ 161.60, 161.14, 148.12, 148.10, 147.83, 147.82, 126.33, 125.35, 123.91,
123.64, 111.82, 111.54, 109.20, 108.67, 56.03, 55.99, 55.96, 47.09, 43.40, 41.98, 38.04,
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 66
1
2
3
4
5
+
29.27, 27.54. HRMS (ESI): exact mass calculated for C12H16NO3 [M+H]+:222.1052 ,
6
7
found:222.1123 .
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7-ethoxy-6-methoxy-3,4-dihydroisoquinoline-2(1H)-carbaldehyde (5). This
compound (a white solid) was prepared by replacement of 1a with 1b using a similar
synthetic procedure of product 4. 1H NMR (600 MHz, CDCl3): δ 8.27, 8.21 (2 × s, 1H),
6.64, 6.62 (2 × s, 1H), 6.61, 6.60 (2 × s, 1H), 4.62, 4.48 (2 × s, 2H), 4.08 (q, J = 7.0 Hz,
2H), 3.87, 3.86 (22 × s, 3H), 3.80, 3.65 (2 × t, J = 6.0 Hz, 2H), 2.84, 2.81 (2 × t, J = 6.0
Hz, 2H), 1.52 – 1.44 (m, 3H). 13C NMR (150 MHz, CDCl3): δ 161.59, 161.13, 148.45,
148.12, 147.38, 147.12, 126.35, 125.33, 123.86, 123.59, 112.05, 111.77, 110.64,
110.19, 64.56, 64.46, 56.01, 47.07, 43.41, 41.96, 38.04, 29.27, 27.55, 14.81, 14.76.
+
HRMS (ESI): exact mass calculated for C13H18NO3 [M+H]+: 236.1281, found:
236.1277.
6-ethoxy-7-methoxy-3,4-dihydroisoquinoline-2(1H)-carbaldehyde (6). This
compound (a white solid) was prepared by replacement of 1a with 1c using a similar
synthetic procedure of product 4. 1H NMR (600 MHz, CDCl3): δ 8.24, 8.18 (2 × s, 1H),
6.62, 6.59 (2 × s, 1H), 6.59, 6.57(2 × s, 1H), 4.60, 4.46 (2 × s, 2H), 4.09-4.02 (m, 2H),
3.84, 3.83 (2 × s, 3H), 3.77, 3.62 (2 × t, J = 6.0 Hz, 2H), 2.81, 2.77 (2 × t, J = 6.06.0
Hz, 2H), 1.48-1.42 (m, 3H). 13C NMR (150 MHz, CDCl3): δ 161.59, 161.14, 148.40,
148.12, 147.43, 147.12, 126.29, 125.30, 123.88, 123.64, 113.24, 113.03, 109.44,
108.91, 64.44, 64.42, 56.09, 56.05, 47.10, 43.42, 41.99, 38.05, 29.25, 27.52, 14.80,
28
ACS Paragon Plus Environment

Page 29 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
+
14.78. HRMS (ESI): exact mass calculated for C13H17NO3 [M+H]+: 236.1281, found:
6
7
236.1282.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-1-((1H-indol-3-yl)methyl)-6,7-dimethoxy-3,4-dihydroisoquinoline-2(1H)-
methanal (7). This compound (a white solid) was prepared by replacement of 5a with
1
6b using a similar synthetic procedure of product 1. H NMR (400 MHz, CDCl3): δ
8.37, 8.21 (2 × s, 1H), 8.15, 7.57 (2 ×s, 1H), 7.67-7.53 (m, 1H), 7.41-7.30 (m, 1H),
7.25-7.04 (m, 2H), 6.96-6.87 (m, 1H), 6.68, 6.55 (2 × s, 1H), 6.64, 6.31 (2 × s, 1H),
5.71-5.61, 4.76-4.64 (2 × m, 1H), 4.54-4.44, 3.61-3.54 (2 × m, 1H), 3.88, 3.84 (2 × s,
3H), 3.86, 3.55 (2 × s, 3H), 3.45 – 3.10 (m, 3H), 2.96 – 2.62 (m, 2H). 13C NMR (125
MHz, CDCl3): δ 161.56, 161.51, 148.30, 147.84, 147.67, 147.31, 136.44, 136.06,
127.91, 127.87, 127.80, 126.85, 126.12, 125.19, 123.59, 123.00, 122.39, 122.07,
119.80, 119.64, 119.09, 118.13, 112.05, 111.65, 111.42, 111.22, 110.98, 110.58,
110.03, 57.44, 56.12, 55.95, 55.88, 55.63, 51.20, 40.82, 33.94, 33.34, 31.80, 29.22,
+
27.79. HRMS (ESI): exact mass calculated for C21H23N2O3 [M+H]+: 351.1703, found:
351.1704.
(S)-1-(3-(1H-indol-3-yl)propyl)-6,7-dimethoxy-3,4-dihydroisoquinoline-
2(1H)-carbaldehyde (8). This compound (a white solid) was prepared by replacement
1
of 5a with 6c using a similar synthetic procedure of product 1. H NMR (400 MHz,
DMSO): δ 10.87, 10.74 (2 × s, 1H), 8.18, 8.15 (2 × s, 1H), 7.43-7.23 (m, 2H), 7.08-
6.90 (m, 2H), 6.82-6.72 (m, 1H), 6.64-6.52 (m , 2H), 5.55-5.46, 5.06-4.95 (2 × m, 1H),
29
ACS Paragon Plus Environment