Chemical Control of Mammalian Circadian Behavior through Dual Inhibition of Casein Kinase Iα and δ

Article abstract of DOI:10.1021/acs.jmedchem.8b01541

Circadian rhythms are controlled by transcriptional feedback loops of clock genes and proteins. The stability of clock proteins is regulated by post-translational modification, such as phosphorylation by kinases. In particular, casein kinase I (CKI) phosphorylates the PER protein to regulate proteasomal degradation and nuclear localization. Therefore, CKI inhibition can modulate mammalian circadian rhythms. In the present study, we have developed novel CKIα and CKIδ dual inhibitors by extensive structural modification of N9 and C2 position of longdaysin. We identified NCC007 that showed stronger period effects (0.32 μM for 5 h period lengthening) in a cell-based circadian assay. The following in vitro kinase assay showed that NCC007 inhibited CKIα and CKIδ with an IC₅₀ of 1.8 and 3.6 μM. We further demonstrated that NCC007 lengthened the period of mouse behavioral rhythms in vivo. Thus, NCC007 is a valuable tool compound to control circadian rhythms through CKI inhibition.

Full text of DOI:10.1021/acs.jmedchem.8b01541

Subscriber access provided by WEBSTER UNIV
Article
Chemical Control of Mammalian Circadian Behavior
through Dual Inhibition of Casein Kinase I# and #
Jae Wook Lee, Tsuyoshi Hirota, Daisuke Ono, Sato Honma, Ken-ichi Honma, Keunwan Park, and Steve Kay
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01541 • Publication Date (Web): 01 Feb 2019
Downloaded from http://pubs.acs.org on February 1, 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 47
Journal of Medicinal Chemistry
1
2
3
4
Chemical Control of Mammalian Circadian Behavior through Dual Inhibition of
5
6
Casein Kinase Iα and δ
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
Jae Wook Lee1,5,#,, Tsuyoshi Hirota2,#,, Daisuke Ono3, Sato Honma3, Ken-ichi
Honma3, Keunwan Park4, Steve A. Kay2,6
1Natural Product Research Center, Korea Institute of Science and Technology,
Gangneung 25451, South Korea.
2Institute of Transformative Bio-Molecules, Nagoya University, Nagoya, 464-
8601, Japan.
3Department of Physiology, Graduate School of Medicine, Hokkaido University,
Sapporo, 060-8638, Japan.
4Natural Product Informatics Research Center, Korea Institute of Science and
Technology, Gangneung, 25451, South Korea.
5Convergence Research Center for Dementia, Korea Institute of Science and
Technology, Seoul, South Korea.
6Keck School of Medicine, University of Southern California, Los Angeles, CA
90089, USA.
#JWL and TH contributed equally.
*Correspondence: Jae Wook Lee (jwlee5@kist.re.kr) and Tsuyoshi Hirota
(thirota@itbm.nagoya-u.ac.jp)
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 47
1
2
3
4
5
6
7
8
9
Current address of D.O.: Department of Neuroscience II, Research Institute of
Environmental Medicine, Nagoya University, Furo-cho, Chikusa-ku, Nagoya
464-8601, Japan
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
2
ACS Paragon Plus Environment

Page 3 of 47
Journal of Medicinal Chemistry
1
2
3
4
Abstract
5
6
7
8
Circadian rhythms are controlled by transcriptional feedback loops of clock genes and
proteins. The stability of clock proteins is regulated by post-translational modification,
such as phosphorylation by kinases. In particular, casein kinase I (CKI) phosphorylates
PER protein to regulate proteasomal degradation and nuclear localization. Therefore, CKI
inhibition can modulate mammalian circadian rhythms. In the present study, we have
developed novel CKI and CKI dual inhibitors by extensive structural modification of
N9 and C2 position of longdaysin. We identified NCC007 that showed stronger period
effects (0.32 M for 5-hr period lengthening) in cell-based circadian assay. The
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
following in vitro kinase assay showed that NCC007 inhibited CKI and CKI with IC50
of 1.8 M and 3.6 M. We further demonstrated that NCC007 lengthened period of
mouse behavioral rhythms in vivo. Thus, NCC007 is a valuable tool compound to control
circadian rhythms through CKI inhibition.
INTRODUCTION
In mammals, circadian rhythms of behavior and physiology are controlled by
molecular circadian clock. The circadian clock is constituted by transcriptional feedback
loops of clock genes and their protein products. CLOCK (Circadian Locomotor Output
Cycles Kaput) and BMAL1 (Brain and Muscle Aryl hydrocarbon receptor nuclear
translocator-Like) transcription factors activate the transcription of other clock genes (Per
(Period) and Cry (Cryptochrome)). The PER (PER1, PER2, and PER3) and CRY (CRY1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 47
1
2
3
4
5
6
and CRY2) proteins repress their own transcription by blocking CLOCK and BMAL1
mediated transcription.1,2
7
8
9
The cellular circadian clocks are organized in a hierarchical manner. The
suprachiasmatic nucleus (SCN) constitutes the central circadian pacemaker regulating
behavioral rhythms.3 On the other hand, peripheral clocks in other tissues control
rhythmic outputs such as retinal visual processing4, hepatic glucose production5, and
heart rate6. Disturbance of clock function has been associated with various pathologies7
including circadian sleep disorder8, cardiovascular disease9, cancer10, and metabolic
disease11.
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 clock proteins are regulated by post-translational modifications such as
phosphorylation for control of stability12 and protein-protein interaction.13 Several
kinases have been reported to phosphorylate clock proteins.14 For instance, casein kinase
I (CKI)  and  phosphorylate PER protein to regulate proteasome-dependent degradation
and nuclear localization.15 Mutation of CKI or phosphorylation site of PER2 causes
familial advanced sleep phase (FASP) in humans.16,17 Considering close association of
circadian clock with kinase functions, small-molecule kinase modulators may lead to new
treatments for circadian disorder.18,19
To date, numbers of small molecules have been reported as modulators of
circadian period in cultured cells including a CDK inhibitor roscovitine20; a p38 MAPK
inhibitor SB20358021; a JNK inhibitor SP60012522; and CKI/ inhibitors PF-67046223,
IC26115, CKI-724, and D447625. We previously identified a 2-benzylaminopurine
derivative, longdaysin, as a potent modulator of circadian rhythms through high-
throughput screening.26 CKI, CKI, and ERK2 mitogen-activated protein kinase were
4
ACS Paragon Plus Environment

Page 5 of 47
Journal of Medicinal Chemistry
1
2
3
4
identified as the targets of longdaysin by affinity-based proteomic analysis. Among them,
5
6
CKI and CKI play a major role in circadian period change. Thus, the dual inhibition of
7
8
9
CKI and CKI potently modulates the mammalian circadian rhythm.26 In this study, we
designed and synthesized the focused-library of longdaysin-derived purine scaffolds.
Structure activity relationship (SAR) of the derivatives for circadian clock regulation was
investigated. A series of analogues designated as “NCC compounds” were identified with
potent circadian period-lengthening activities in cells. More importantly, a compound
NCC007 exhibited in vivo efficacy to control behavioral rhythms in mice.
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 47
1
2
3
4
RESULT AND DISCUSSION
5
6
Design, synthesis and activities of longdaysin derivatives
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
CF3
CF3
NH
NH
NH
modification of
C6
modification of
N9 and C2
6
N
N
N
N
N
N
N
N
N
N
2
9
N
N
N
N
H
Longdaysin
NCC007
Benipurine
Figure 1. Design and synthesis of longdaysin derivatives.
In the initial SAR study of benipurine, a hit compound of our cell-based circadian screen,
we identified longdaysin with better potency (Figure 1).26 The derivatives on substituent
of N9 position showed decreased activity expect isopropyl substituent. Modification of
C6 revealed that 3-trifluorobenzyl group is potent structure among various substituted
benzyl group. Especially, meta-CF3 benzyl amine produced the most potent compound26.
The compound longdaysin showed submicromolar activity for 1-hr period lengthening
(0.3 M), which was 3-fold more potent than benipurine (0.9 M for 1-hr). We aimed to
develop further potent compounds with better solubility for mouse behavioral study.
6
ACS Paragon Plus Environment

Page 7 of 47
Journal of Medicinal Chemistry
1
2
3
CF3
CF3
4
5
6
7
8
9
X
Cl
N
Cl
N
X
N
N
N
R1
N
N
c
N
N
N
N
a or b
d
N
N
N
N
N
Cl
Cl
Cl
R2
H
1
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
R1
2, R1=CH3
3, R1=cyclopropyl
R1
4, R1=CH3, X=NH
5, R1=cyclopropyl,
X=NH, NCH3, O
6, R1=CH3,X=NH,R2=amines
7, R1=cyclopropyl, X=NH, NCH3, O
R2=N1,N1-dimethylethane-1,2-diamine
Scheme 1. (a) DEAD, Ph3P, isopropyl alcohol, THF, 0oC (b) 1-cyclopropyl ethyl alcohol, DEAD, Ph3P,
THF, 0oC (c) 3-trifluoromethylbenzylamine, DIEA, n-BuOH, 60oC; or 3-trifluoromethylbenzyl
methylamine, DIEA, n-BuOH, 60oC; or 3-trifluoromethylbenzyl alcohol, NaH, DMF, 80oC (d) R2NH2,
DIEA, n-BuOH, 120oC; or R2OH, NaH, DMF, 120oC
In the present study, we conducted optimization of longdaysin by focusing on unexplored
C2 position. As shown in Scheme 1, Mitsunobu reaction of commercially available
purine 1 with isopropyl alcohol or 2-ethylcyclopropyl alcohol provided intermediate 2 or
3, which were further transformed into the desired analog 4 and 5 via a sequence of 3-
trifluorobenzylamine substitution. Analogues 6 and 7 were prepared in relatively high
yield by similar procedure using various aryl or alkyl amines.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 47
1
2
3
4
Table 1. Cellular activity of longdaysin derivatives.
5
6
7
CF3
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
NH
N
N
N
N
R2
Circadian period lengthening
Compound
R2
H
1-hr (M) a
0.39
5-hr (M) a
3.4
longdaysin
6a
N
0.09
1.03
0.9
5.7
NH
N
N
6b
O
N
N
6c
6d
6e
0.68
1.98
0.90
5.8
10.43
3.66
N
OH
N
N
O
N
6f
1.65
9.33
NH
6g
6h
6i
1.62
0.38
3.04
3.76
NH
OH
NH
Inactive
Inactive
Inactive
Inactive
NH
H3CO
6j
N
8
ACS Paragon Plus Environment

Page 9 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NH2
6k
6l
0.58
5.23
2.39
2.75
NH
0.32
0.39
O
6m
O
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
6n
6.39
23.36
O
a value is the compound concentration required to produce 1-hr and 5-hr circadian period lengthening effects. Each value is the
mean of at least two independent determinations,
We examined the effect of various functional groups with alkyl and benzyl substitution
on C2 position of purine scaffold in cell-based circadian assay (Table 1). Alkyl and
benzyl substitution derivatives was less potent than or similar to longdaysin. In contrast,
0.9 M of 6a caused 5-hr period lengthening, which was ~3-fold more potent than
longdaysin. In parallel, we designed and synthesized longdaysin derivative 7a with N9 1-
cyclopropyl ethyl group. 7a showed ~3-fold more potent effect than longdaysin (Table 2).
Based on this SAR information, we designed to combine N9 substitution of 7a and C2
substitution of 6a and synthesized “7b (NCC007)” which showed ~10-fold more potency
than longdaysin (Figure 2). To investigate the role of C6 nitrogen, we designed and
synthesized the N-methyl analogue 7c and O-analogue 7d to eliminate potential hydrogen
bond interaction (Table 2). As expected, they showed dramatic decrease in period
lengthening activity. Therefore, they can be used as negative controls. On the basis of
structure activity relationship and extended synthesis, we discovered 7b (NCC007) as the
most potent compound.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 47
1
2
3
4
Table 2. Cellular activity of NCC007 derivatives.
CF3
5
6
7
8
X
9
N
N
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
N
N
R2
Circadian period lengthening
Compound
X
R2
1-hr (M) a
5-hr (M) a
7a
7b
NH
NH
H
0.09
0.9
N
0.022
0.32
NH
(NCC007)
N
N
7c
CH3N
O
Inactive
Inactive
Inactive
Inactive
NH
NH
7d
a value is the compound concentration required to produce 1-hr and 5-hr circadian period lengthening effects. Each value is the
mean of at least two independent determinations,
Figure 2. Effects of NCC007, longdaysin and roscovitine on the period length of cellular circadian rhythms.
10
ACS Paragon Plus Environment

Page 11 of 47
Journal of Medicinal Chemistry
1
2
3
4
Inhibitory effect of NCC007 toward CKI correlates with period phenotype.
5
6
7
8
We previously revealed that longdaysin inhibits CKI and CKI.26 We analyzed the
effects of 5 µM NCC007 and longdaysin against 57 kinases from a variety of classes
(Figure 3A). Longdaysin inhibited CKIδ, CKIα, CDK2, CDK7, and MAPK1 by >30%.
The effects on CKIδ, CKIα, and CDK7 were enhanced in NCC007, while the effects on
CDK2 and MAPK1 were reduced. We then analyzed IC50 values of longdaysin, NCC007,
and roscovitine against these kinases (Figure 3B). CKIα was the most potently inhibited
kinase of NCC007 and the only kinase correlated with circadian period lengthening
activities (Figure 2) in which the activity order was NCC007>longdaysin>roscovitine.
We further conducted profiling of total 374 kinases (including 57 kinases in Figure 3A)
by using 5 µM NCC007 (Figure 3C and Table S2). Again, CKIα (CK1a1) was the most
inhibited kinase supporting an important role of CKIα in the regulation of circadian
period by NCC007. Only 9 kinases among 374 kinases were inhibited by more than 50%.
In addition to CKI family (CKIα, CKIδ, CKIα-like, and CKIε) and CDK family (CDK7
and CDK6) proteins, SRPK2, SRPK1, and CLK2 were found to be inhibited by NCC007
with much less effect than CKIα. SRPK2, SRPK1, and CLK2 phosphorylate SR proteins
to regulate mRNA splicing, and their role in circadian regulation will be investigated in
future studies. Together, NCC007 showed more selectivity against CKI, suggesting that
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
CKI plays an important role in regulation of circadian period.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 47
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 3. Effects of NCC007 and longdaysin on kinase activities. (A) Profiling of NCC007 and longdaysin
toward 57 kinases. (B) In vitro assay of CKI, CKI, CDK2, CDK7, MAPK1, and MAPK3 to Longdaysin,
12
ACS Paragon Plus Environment

Page 13 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
NCC007, and Roscovitine. (C) Profiling of NCC007 toward 374 kinases. Kinases inhibited by >50% were
highlighted in left.
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
Molecular modeling studies suggests NCC007 has stronger binding affinity toward
CKIα and CKIδ than longdaysin.
Based on the results of the kinase assay, we performed molecular docking simulations
to present plausible binding models of longdaysin and NCC007 for human CK1α.
NCC007 is a racemic mixture that has R and S enantiomers. Because of difficulties in
evaluating each enantiomer experimentally, we investigated them computationally.
RosettaLigand docking protocol was written by using RosettaScript and used to optimize
docking conformations. The lowest binding energy conformation from 5,000 docking
experiments was considered as the final docking model shown in Figure 4.
Rosetta energy scores for the final docking models were −25.28, −25.71, and −22.22
(Rosetta Energy Unit) for R-NCC007, S-NCC007, and longdaysin, respectively. Binding
energy of R- and S-forms of NCC007 was almost comparable and lower than longdaysin,
consistent with the stronger inhibition of CK1α by NCC007 than longdaysin. In the
docking model, longdaysin fit into the hydrophobic pocket formed by Ile31, Ala44,
Leu93, Pro95, Leu143, and Ile156, and especially, the benzyl amino group of londaysin
formed hydrogen bonds with the backbone of Leu93 (Figure 4A).
R- and S-NCC007 fit to the longdaysin-binding pocket, and showed more extensive
contacts to CK1α (Figures 4B and 4C). R-NCC007 fit very well into the hydrophobic
binding pocket consisted of Ile23, Ile31, Ala44, Met90, Leu93, and Leu143. Moreover,
the fluorine group in NCC007 showed additional polar interactions with Lys46, Tyr64,
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 47
1
2
3
4
5
6
7
8
9
and Asp157 in the varied binding pocket (Figure 4B) with high shape complementarity.
On the other hand, docking model of S-NCC007 showed the lowest binding energy
(−25.71) mainly by the hydrophobic interactions, which is more similar to the docking
model of longdaysin that the fluorine group is exposed to the solvent (Figure 4C). Taken
together, the docking models suggest that NCC007 likely binds to CK1α with more
favorable interactions and higher shape complementarity than longdaysin regardless of R
and S configurations.
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
In the docking model, the C2 position of longdaysin is located near the solvent-
accessible void space in the binding pocket (Figure 4A), thus 6a with adequate volume
fitting to the void space might show enhanced activity. In contrast, the derivatives of
bulky functional group such as 6d and 6f had decreased activity by clashing to the
binding pocket. Despite the chemical similarity, 6k showed much less activity compared
to 6a, probably due to more favorable contacts through the methyl groups in 6a.
Alternatively, primary amine of 6k may easily bind to cellular metabolites such as
bioaldehyde, resulting in less cellular activity compared with 6a.
Similar to CK1α, we observed lower binding energy of R-NCC007 (-26.97) and S-
NCC007 (-25.55) than longdaysin (-21.73) against CK1δ (Figure S1), consistent with
kinase assay result. However, due to high structural similarity (Cα RMSD 0.487) and
identical binding residues between CKIα and CKIδ (Figures S2 and S3), it is difficult to
explain the 2-fold activity difference of NCC007 against CKIα and CKIδ.
14
ACS Paragon Plus Environment

Page 15 of 47
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. Comparison of binding modes of (A) longdaysin (carbon atoms in magenta), (B) NCC007 R-
form (carbon atoms in yellow), and (C) NCC007 S-form (carbon atoms in orange) are presented by
RosettaLigand docking experiment. To clarify the binding modes, the protein is represented by surface and
the docked ligands are shown by sticks. All protein figures are created with PyMOL software27. Two-
dimensional interaction maps are shown together for the residues in the binding pocket by LIGPLOT
software28. Hydrogen bonds are depicted with a dotted line.
Lengthening of behavioral circadian rhythms in vivo by longdaysin and NCC007.
To extend these findings to the in vivo application, pharmacological inhibition of both
CKI and CKI by longdaysin and NCC007 were tested in mice. Compounds were
infused into the lateral ventricle using osmotic pump. This approach has advantage to
avoid possible blood brain barrier penetration problem. Mice were entrained to 12-hr
light and 12-hr dark cycles, then released to constant darkness where they show ~23.8-hr
rhythms based on endogenous circadian clock (Figure 5). After artificial cerebrospinal
fluid infusion as a control, 1 mM or 3 mM of longdaysin was continuously infused.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 47
1
2
3
4
5
6
7
8
9
Period of circadian behavior recorded by infrared sensor29 was lengthened about 0.05 hr
by 1 mM longdaysin infusion and 0.1 hr by 3 mM longdaysin infusion. We then
increased concentration of longdaysin to 5 mM or 15 mM, and 0.1 hr or 0.15 hr period
lengthening was observed, respectively. Following administration of NCC007 showed
more period lengthening effect with 0.15 hr at 5 mM and 15 mM. The period length
returned normal by buffer infusion after NCC007 treatment, demonstrating that the effect
is reversible. This result indicated that the compounds are able to control circadian
behavior in vivo.
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. Circadian behavior of longdaysin and NCC007-treated mice. Mice were treated with compounds
and their circadian behavior was recorded by infrared sensor under constant dark condition. (A) HBSS, 1
mM longdaysin, 5 mM longdaysin, and 5 mM NCC007 were sequentially infused into mouse brain. (B)
16
ACS Paragon Plus Environment

Page 17 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
HBSS, 3 mM longdaysin, 15 mM longdaysin, and 15 mM NCC007 were sequentially infused. (C) DMSO
infusion as negative control. The number in parenthesis indicates the number of mice. ***p < 0.001, **p <
0.01 against the DMSO control by two-way ANOVA, followed by a Tukey’s multiple comparisons test.
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
While longdaysin and NCC007 caused more than hours of period lengthening in cell
based assay, they showed much less period effect in vivo, although they were infused
inside of the blood-brain barrier. It is likely that enough amounts of longdaysin and
NCC007 may not be exposed into mouse brain, possibly due to low exposure and/or early
clearance of compounds. The alternate possibility could be that CKI inhibition has less
effect on circadian period in vivo. Consistently, Meng and coworkers reported that a
potent CKI inhibitor, PF-670462, also caused only small period change in vivo.30
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 47
1
2
3
4
CONCLUSION
5
6
7
8
The CKI family of serine and threonine kinases is highly conserved and ubiquitously
expressed in eukaryotes. Among this family, CKI, CKI, and CKI are known to play
important roles in modulating circadian rhythms by phosphorylation of PER protein and
stimulation of its degradation. We discovered and optimized longdaysin analogues as
novel class of CKI and CKIδ dual inhibitors that affect circadian period length. The
ethyl cyclopropyl-containing analogues showed improved period lengthening effect over
their isopropyl counterparts, of which the 3-trifluoro and benzylamine-containing
NCC007 (7b) exhibited optimal period lengthening effect. We further demonstrated that
NCC007 inhibited CKI and CKI in in vitro kinase assay and showed period
lengthening effect on mouse models. We believe that NCC007 proves a useful tool for
studies of circadian rhythms and may ultimately lead to chronotherapeutic agents.
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
EXPERIMENTAL SECTION
General Procedure. All chemicals and solvents, commercially available, were purchased
from commercial suppliers (Acros, Alfa Aesar and Aldrich) and used without further
purification. All chemical reactions were performed under nitrogen atmosphere.
Anhydrous solvents were obtained by passage through an activated alumina column.
Analytical thin-layer chromatography (TLC) was performed on pre-coated glass TLC
plates (layer 0.2 mm silica gel 60 with fluorescent indicator (UV 245: Merck). The TLC
plates was revealed by exposure of UV lamp (254/350 nm). Column chromatography was
performed using silica gel 60 (70-230 mesh; Merck). Liquid chromatography-mass
18
ACS Paragon Plus Environment

Page 19 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
spectrometry (LC-MS) were analyzed with LC-MS (Agilent Technology 1200), using a
C18 column, with 20 minutes elution using a gradient solution of CH3CN-H2O
(containing 0.05% trifluoroacetic acid), with UV detector and an electrospray ionization
source. The purity of all compounds were analyzed by the analytical LC-MS system
describe, and purity >95%. The molecular mass was evaluated using the analytical LC-
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
13
MS system. 1H and C NMR spectral data were obtained from a Bruker Avance II 400
MHz spectrometer. Chemical shifts (δ) are reported relative to internal CDCl3 (Me4Si, δ
0.0) and CD3OD (Me4Si, δ 0.0). Signal splitting patterns are described as singlet (s),
doublet (d), triplet (t), multiplet (m), or a combination thereof. Coupling constants (J) are
quoted to the nearest 0.1 Hz.
The synthesis of 2,6-dichloro-9-isopropylpurine (2). Anhydrous tetrahydrofuran (100.0
ml) is added to a flame-dried flask under N2 containing 2-chloro-6-chloropurine (2.00 g,
0.01 mol) and triphenylphosphine (5.24 g, 0.02 mol). To this 2-propanol (1.87 mL, 0.025
mol) is added, and the mixture is then cooled to -10°C by using an ethylene glycol/dry-
ice bath. After dropwise addition of diethyl azodicarboxylate (3.14 mL, 0.02 mol), the
mixture is warmed gradually to room temperature. After 12 h, the reaction is quenched
with 1 mL of water and the solvent is removed in vacuo. The resulting yellow oil is
purified by column chromatography (1.0 L of SiO2, eluted with 100% CH2Cl2). The
resulting solid is triturated with methanol to yield 1.53 g (66.5%) of a compound 2 as a
1
white solid. H NMR (400 MHz, CDCl3)  8.20 (s, 1H), 4.91 (dd, J = 13.5 Hz, 6.8 Hz,
1H), 1.66 (d, J = 6.8 Hz, 6H). 13C NMR (100 MHz, CDCl3)  152.7, 151.7, 143.6, 132.9,
131.0, 48.2, 22.5. ESI-MS: m/z calcd for C8H9Cl2N4 [M+H] = 231.0 found 231.1.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 47
1
2
3
4
5
6
7
8
The synthesis of 2,6-dichloro-9-(1-cyclopropylethyl)purine (3). The solution of
compound 1 (1.00 g, 5.32 mmol) and triphenylphosphine (2.79 g, 10.64 mmol) in
anhydrous tetrahydrofuran (50 mL) was cooling down at -10oC by using an ethylene
glycol/dry ice bath. 1-cyclopropylethyl alcohol (915.8 mg, 10.64 mmol) was slowly
added to the reaction. Diethyl azadicarboxylate (1.67 mL, 10.64 mmol) was slowly added
over a period of 20 min under nitrogen atmosphere. The reaction mixture was stirred for
48 hr at room temperature. The reaction is quenched with 0.5 mL of water and the solvent
is removed in vacuo. The residue was charged directly on silicagel column and purified
with dichloromethane to yield 742 mg (54.6%) of a compound 3 as white solid. 1H NMR
(400 MHz, CDCl3)  8.51 (s, 1H), 3.94-3.89 (m, 1H), 1.61 (d, J = 6.8 Hz, 3H), 1.36-1.30
(m, 1H), 0.77-0.71 (m, 1H), 0.56-0.50 (m, 1H), 0.49-0.41 (m, 1H), 0.37-0.29 (m, 1H). 13C
NMR (100 MHz, CDCl3)  152.6, 151.6, 148.0, 144.1, 132.9, 57.6, 22.5, 16.8, 5.0, 4.4.
ESI-MS: m/z calcd for C10H11Cl2N4 [M+H] = 257.0 found 257.1.
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
The synthesis of 2-chloro-9-isopropyl-N-(3-(trifluoromethyl)benzyl)-9H-purin-6-
amine (4). The solution of compound 2 (1.53 g, 6.65 mmol) in n-BuOH (40 mL) was
treated with m-trifluorobenzyl amine (1.28 g, 7.32 mmol) and DIEA (1.72 mL, 9.98
mmol). The reaction mixture was heated at 60 oC for 24 hr. After reaction was quenched
with 1 mL of water, solvent was evaporated in vacuo. The reaction mixture was extracted
with dichloromethane (100 mL, 3 times) and H2O (40 mL, 3 times). Organic layer was
dried with Na2SO4 and evaporated in vacuo. The residue was purified by silica gel
column chromatography (dichloromethane : methanol = 40:1) to yield 2.12 g (86.53%) of
20
ACS Paragon Plus Environment

Page 21 of 47
Journal of Medicinal Chemistry
1
2
3
4
1
a compound 4 as white solid. H NMR (400 MHz, CDCl3)  7.94 (s, 1H), 7.68 (s, 1H),
5
6
7.65-7.54 (m, 2H), 7.49 (d, J = 8.0 Hz, 1H), 5.00-4.84 (m, 3H), 1.61 (d, J = 6.8 Hz, 6H).
7
8
13
C NMR (100 MHz, CDCl3)  138.9, 137.2, 133.1, 131.4, 131.2, 130.9, 130.6, 129.2,
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
125.3, 124.8, 124.6, 124.5, 122.6, 47.6, 22.7. ESI-MS: m/z calcd for C16H16ClF3N5
[M+H] = 370.1 found 370.2.
The synthesis of 2-chloro-9-(1-cyclopropylethyl)-N-(3-(trifluoromethyl)benzyl)-9H-
purin-6-amine (5). The solution of compound 3 (500 mg, 1.95 mmol) in n-BuOH (20
mL) was treated with m-trifluorobenzyl amine (377 mg, 2.15 mmol) and DIEA (403 L,
o
2.34 mmol). The reaction mixture was heated at 60 C for 24 hr. After reaction was
quenched with 500 L of water, solvent was evaporated in vacuo. The residue was
extracted with dichloromethane (100 mL) and H2O (20mL). Organic layer was dried with
Na2SO4 and evaporated in vacuo. The residue was purified by silica gel column
chromatography (dichloromethane : methanol = 40:1) to yield 514.1 mg (66.73%) of a
1
compound 5a as white solid. H NMR (400 MHz, CDCl3)  7.79 (s, 1H), 7.58 (s, 1H),
7.56-7.44 (m, 2H), 7.39 (t, J = 7.6 Hz, 1H), 4.84 (s, 2H), 3.90-3.79 (m, 1H), 1.53 (d, J =
6.8 Hz, 3H), 1.40-1.25 (m, 1H), 0.71-0.62 (m, 1H), 0.52-0.42 (m, 1H), 0.41-0.32 (m, 1H),
13
0.31-0.22 (m, 1H). C NMR (CDCl3)  154.9, 154.4, 139.1, 138.2, 131.2, 130.8, 129.2,
125.3, 124.7, 124.4, 122.6, 118.2, 56.4, 44.2, 20.9, 16.9, 4.8, 4.2. ESI-MS: m/z calcd for
C18H18ClF3N5 [M+H] = 396.1 found 396.2.
The general synthetic procedure of compounds (6). For amine substitution reaction at
C6; The solution of a compound 4 (50 mg, 0.14 mmol) in n-BuOH (2mL) was treated
with DIEA (48.3 L, 0.28 mmol) and various amines (0.28 mmol, 2.0 eq.). The reaction
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 47
1
2
3
4
5
6
7
8
9
o
mixture was heated at 120 C for 48 hr. After quenching the reaction with 100 L of
water, solvent was evaporated in vacuo. The residue was purified by prep-HPLC with
H2O-acetonitrile to yield compounds.
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
For alcohol substitution reaction at C6; various alcohols (0.48 mmol, 2.0 eq.) in DMF (1
mL) was treated with NaH 60% (20 mg, 0.48 mmol). The reaction mixture was stirred for
o
30 min at 50 C. The solution of compound 4 (50 mg, 0.14 mmol) in DMF (1 mL) was
slowly added to the reaction mixture. The reaction mixture was heated at 120 oC for 48 hr.
After quenching the reaction with 100 L of water, solvent was evaporated in vacuo. The
residue was purified by prep-HPLC with H2O-acetonitrile to yield compounds.
N2-(2-(dimethylamino)ethyl)-9-isopropyl-N6-(3-(trifluoromethyl)benzyl)-9H-purine-
2,6-diamine (6a). The synthesis of 6a was followed by general procedure to yield 42.1
1
mg (71.4%) of a compound 6a. H NMR (400 MHz, CDCl3)  8.00-7.92 (m, 1H), 7.73-
7.66 (m, 1H), 7.65-7.54 (m, 1H), 7.53-7.47 (m, 1H), 7.44 (t, J = 8.0 Hz, 1H), 5.29 (s, 1H),
4.78 (s, 1H), 4.69 (s, 1H), 3.83-3.75 (m, 2H), 3.32-3.25 (m, 1H), 3.20-3.05 (m, 1H), 2.86
13
(s, 3H), 2.76 (s, 3H), 1.53 (s, 3H), 1.52 (s, 3H). C NMR (100 MHz, CDCl3)  163.0,
162.6, 130.9, 129.2, 125.3, 124.6, 124.1, 122.6, 56.4, 55.7, 53.8, 48.0, 47.2, 46.7, 43.1,
40.5, 36.8, 22.2. ESI-MS: m/z calcd for C20H27F3N7 [M+H] = 422.2 found 422.3.
1-(4-(9-isopropyl-6-((3-(trifluoromethyl)benzyl)amino)-9H-purin-2-yl)piperazin-1-
yl)ethan-1-one (6b). The synthesis of 6b was followed by general procedure to yield
39.92 mg (61.8%) of a compound 6b. 1H NMR (400 MHz, CDCl3)  7.66 (s, 1H), 7.60-
22
ACS Paragon Plus Environment

Page 23 of 47
Journal of Medicinal Chemistry
1
2
3
4
7.55 (m, 2H), 7.52 (d, J = 7.6 Hz, 1H), 7.44 (t, J = 7.6 Hz, 1H), 6.45 (s, 1H), 4.84 (s, 2H),
5
6
4.73-4.65 (m, 1H), 3.83-3.75 (m, 4H), 3.69-3.64 (m, 2H), 3.51-3.46 (m, 2H), 2.15 (s, 3H),
7
8
9
13
1.58 (s, 3H), 1.56 (s, 3H). C NMR (100 MHz, CDCl3)  169.2, 158.6, 154.1, 140.6,
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
135.1, 130.8, 130.7, 129.0, 124.4, 124.3, 124.0, 123.9, 115.5, 46.6, 46.1, 44.7, 44.3, 41.3,
22.5, 21.5. ESI-MS: m/z calcd for C20H27F3N7O [M+H] = 462.2 found 462.3.
9-isopropyl-2-(4-methyl-1,4-diazepan-1-yl)-N-(3-(trifluoromethyl)benzyl)-9H-purin-
6-amine (6c). The synthesis of 6c was followed by general procedure to yield 38.14 mg
(60.9%) of a compound 6c. 1H NMR (400 MHz, CDCl3)  9.04 (brs, 1H), 8.14 (brs, 1H),
7.64-7.55 (m, 2H), 7.55-7.44 (m, 2H), 4.83-4.72 (m, 3H), 4.45 (brs, 1H), 4.23 (brs, 1H),
3.68 (brs, 1H), 3.46 (brs, 2H), 2.75 (brs, 4H), 2.65 (s, 3H), 2.49 (brs, 1H), 2.17 (brs, 1H),
13
1.63 (s, 3H), 1.62 (s, 3H). C NMR (100 MHz, CDCl3)  163.4, 159.2, 152.3, 140.2,
132.4, 130.8, 129.1, 125.5, 123.8, 122.8, 57.0, 55.7, 48.4, 44.7, 44.4, 44.2, 42.1, 41.0,
23.8, 22.1. ESI-MS: m/z calcd for C22H29F3N7 [M+H] = 448.3 found 448.3.
1-(9-isopropyl-6-((3-(trifluoromethyl)benzyl)amino)-9H-purin-2-yl)piperidin-4-ol
(6d). The synthesis of 6d was followed by general procedure to yield 53.14 mg (87.4%)
of a compound 6d. 1H NMR (400 MHz, CDCl3)  8.39 (brs, 1H), 7.63 (s, 1H), 7.55 (d, J
= 7.20 Hz, 1H), 7.45 (d, J = 7.6 Hz, 1H), 7.39 (t, J = 7.6 Hz, 1H), 4.82-4.75(m, 1H), 4.73
(s, 2H), 4.33-4.23 (m, 2H), 3.90-3.80 (m, 1H), 3.65-3.40 (m, 1H), 3.28-3.12 (m, 2H),
1.89-1.78 (m, 2H), 1.57 (s, 3H), 1.56 (s, 3H), 1.40 (brs, 2H). 13C NMR (100 MHz, CDCl3)
 159.8, 152.0, 150.0, 140.5, 132.5, 131.0, 130.8, 130.5, 128.9, 125.4, 124.5, 124.4, 123.8,
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 47
1
2
3
4
5
6
122.7, 122.8, 43.9, 41.9, 40.3, 33.7, 21.9, 10.9. ESI-MS: m/z calcd for C21H26F3N6O
[M+H] = 435.2 found 435.2.
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
9-isopropyl-2-(4-methylpiperazin-1-yl)-N-(3-(trifluoromethyl)benzyl)-9H-purin-6-
amine (6e). The synthesis of 6e was followed by general procedure to yield 47.68 mg
(78.6%) of a compound 6e. 1H NMR (400 MHz, CDCl3)  8.27 (s, 1H), 7.57 (s, 1H), 7.52
(d, J = 7.6 Hz, 1H), 7.46 (d, J = 7.2 Hz, 1H), 7.41 (t, J = 7.6 Hz, 1H), 4.75-4.60 (m, 5H),
13
3.80-3.65 (m, 2H), 3.58-3.30 (m, 4H), 2.78 (s, 3H), 1.55 (d, J = 6.8 Hz, 6H). C NMR
(100 MHz, CDCl3):  158.4, 152.7, 150.0, 140.0, 134.3, 130.8, 130.7, 130.4, 129.0, 125.4,
124.1, 124.0, 123.9, 123.8, 122.7, 53.1, 43.3, 41.5, 22.0. ESI-MS: m/z calcd for
C21H27F3N7 [M+H] = 434.2 found 434.3.
9-isopropyl-N2-(2-morpholinoethyl)-N6-(3-(trifluoromethyl)benzyl)-9H-purine-2,6-
diamine (6f). The synthesis of 6f was followed by general procedure to yield 28.71 mg
(44.3%) of a compound 6f. 1H NMR (400 MHz, CDCl3)  7.85 (s, 1H), 7.74-7.66 (m, 1H),
7.65-7.59 (m, 1H), 7.56 (d, J = 7.6 Hz, 1H), 7.49 (dd, J = 8.0 Hz, 7.6 Hz, 1H), 5.34 (s,
1H), 4.85 (s, 1H), 4.76 (s, 1H), 4.00-3.92 (m, 4H), 3.90-3.79 (m, 2H), 3.20-2.96 (m, 6H),
1.57 (d, J = 6.8 Hz, 6H). ESI-MS: m/z calcd for C22H29F3N7O [M+H] = 464.2 found
464.3.
9-isopropyl-N2-(cyclopropyl)-N6-(3-(trifluoromethyl)benzyl)-9H-purine-2,6-diamine
(6g). The synthesis of 6g was followed by general procedure to yield 19.20 mg (35.2%)
1
of a compound 6g. H NMR (400 MHz, CDCl3)  7.74-7.71 (m, 1H), 7.66-7.61 (m, 1H),
24
ACS Paragon Plus Environment

Page 25 of 47
Journal of Medicinal Chemistry
1
2
3
4
7.58-7.54 (m, 1H), 7.51-7.46 (m, 1H), 5.91 (brs, 1H, -NH), 5.35-5.33 (m. 1H), 4.90-4.80
5
6
(m, 1H), 4.73-4.65 (m, 1H), 1.59 (d, J = 6.8 Hz, 3H, -CH3), 1.56 (d, J = 6.8 Hz, 3H, -
7
8
9
13
CH3), 1.01-0.99 (m, 1H), 0.90-0.80 (m, 2H), 0.70-0.60 ppm (m, 2H). C NMR (100
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
MHz, CDCl3)  163.1, 162.7, 149.4, 138.2, 137.8, 132.1, 131.0, 129.3, 125.3, 117.8,
114.8, 58.1, 42.7, 29.7, 22.4, 22.2, 7.1, 7.0. ESI-MS: m/z calcd for C19H22F3N6 [M+H] =
391.2 found 391.2.
(R)-2-((9-isopropyl-6-((3-(trifluoromethyl)benzyl)amino)-9H-purin-2-
yl)amino)butan-1-ol (6h). The synthesis of 6h was followed by general procedure to
1
yield 23.9 mg (40.4%) of a compound 6h. H NMR (400 MHz, CDCl3)  9.11 (s, 1H),
8.26 (s, 1H), 7.75 (s, 1H), 7.68 (d, J = 7.2 Hz, 1H), 7.62 (d, J = 7.6 Hz, 1H), 7.57 (t, J =
7.6 Hz, 1H), 4.78 (brs, 1H), 4.73 (brs, 1H), 4.61 (dd, J = 13.2 Hz, 6.8Hz, 1H), 3.84 (s,
1H), 3.52-3.45 (m, 1H), 1.70-1.60 (m. 1H), 1.51 (d, J = 3.6 Hz, 3H), 1.49 (d, J = 3.2 Hz,
3H), 1.48 (s, 3H), 0.88-0.82 (m, 2H). 13C NMR (100 MHz, CDCl3)  159.1, 131.5, 129.4,
129.2, 128.9, 125.5, 124.0, 123.9, 123.7, 122.8, 62.2, 54.5, 47.2, 43.0, 23.6, 21.8, 21.7,
10.3. ESI-MS: m/z calcd for C20H26F3N6O [M+H] = 423.2 found 423.3.
N2-benzyl-9-isopropyl-N6-(3-(trifluoromethyl)benzyl)-9H-purine-2,6-diamine (6i).
The synthesis of 6i was followed by general procedure to yield 34.5 mg (56.9%) of a
compound 6i. 1H NMR (400 MHz, CDCl3)  8.47 (brs, 1H), 8.17 (s, 1H), 7.71 (s, 1H),
7.60 (d, J = 7.6 Hz, 2H), 7.55-7.47 (m, 1H), 7.33-7.16 (m, 5H), 4.70 (brs, 2H), 4.59 (dd, J
= 13.6 Hz, 6.8 Hz, 1H), 4.47 (s, 2H), 1.47 (s, 3H), 1.46 (s, 3H). 13C NMR (100 MHz,
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 47
1
2
3
4
CDCl3)  131.5, 129.3, 128.8, 128.0, 127.3, 126.5, 123.9, 47.0, 44.5, 21.8. ESI-MS: m/z
5
6
calcd for C23H24F3N6 [M+H] = 441.2 found 441.3.
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
9-isopropyl-2-(2-(methoxymethyl)pyrrolidin-1-yl)-N-(3-(trifluoromethyl)benzyl)-9H-
purin-6-amine (6j). The synthesis of 6j was followed by general procedure to yield
1
31.49 mg (50.2%) of a compound 6j. H NMR (400 MHz, CDCl3)  7.73 (s, 1H), 7.72-
7.48 (m, 5H), 4.76 (s, 1H), 4.65-4.56 (m, 2H), 3.50-3.44 (m, 1H), 3.25-3.14 (m, 2H),
1.91-1.85 (m, 4H), 1.85-1.76 (m, 2H), 1.50 (d, J = 6.8 Hz, 3H, -CH3), 1.49 (d, J = 6.8 Hz,
13
3H, -CH3). C NMR (100 MHz, CDCl3)  158.1, 157.8, 142.0, 133.2, 132.2, 132.0,
131.5, 131.4, 129.2, 128.8, 128.7, 123.3, 72.4, 58.2, 47.3, 46.9, 40.4, 28.1, 22.9, 22.5,
21.7, 21.6, 10.9. ESI-MS: m/z calcd for C23H29F3N5O [M+H] = 449.2 found 449.3.
N2-(2-aminoethyl)-9-isopropyl-N6-(3-(trifluoromethyl)benzyl)-9H-purine-2,6-
diamine (6k). The synthesis of 6k was followed by general procedure to yield 21.9 mg
1
(39.8%) of a compound 6k. H NMR (400 MHz, CDCl3)  7.86 (s, 1H), 7.72-7.68 (m,
1H), 7.65-7.58 (m, 1H), 7.54 (d, J = 8.0 Hz, 1H), 7.47 (dd, J = 8.0 Hz, 7.6 Hz, 1H), 5.31
(s, 1H), 4.90-4.80 (m, 1H) 4.70-4.62 (m, 1H), 1.54 (s, 3H), 1.53 (s, 3H), 1.43-1.39 (m,
2H), 1.38-1.36 (m, 2H). 13C NMR (100 MHz, CDCl3)  153.0, 149.0, 138.0, 133.0, 131.2,
129.3, 125.1, 124.7, 53.7, 40.6, 38.5, 22.2, 17.3. ESI-MS: m/z calcd for C18H23F3N7
[M+H] = 394.2 found 394.3.
9-isopropyl-2-methoxy-N-(3-(trifluoromethyl)benzyl)-9H-purin-6-amine (6l). The
synthesis of 6l was followed by general procedure to yield 18.7 mg (36.6%) of a
26
ACS Paragon Plus Environment

Page 27 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
compound 6l. H NMR (400 MHz, DMSO-d6)  8.47 (s, 1H), 8.08 (s, 1H), 7.73 (s, 1H),
7.71-7.65 (m, 1H), 7.63-7.53 (m, 2H), 4.71 (s, 2H), 4.65-4.60 (m, 1H), 3.79 (s, 3H), 1.50
(s, 3H), 1.49 (s, 3H). 13C NMR (100 MHz, DMSO-d6)  142.0, 139.5, 137.6, 132.0, 131.5,
131.4, 129.3, 128.8, 128.7, 123.8, 123.4, 53.8, 46.3, 42.7, 22.0, 10.9. ESI-MS: m/z calcd
for C17H19F3N5O [M+H] = 366.2 found 366.2.
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
2-isopropoxy-9-isopropyl-N-(3-(trifluoromethyl)benzyl)-9H-purin-6-amine
(6m).
The synthesis of 6m was followed by general procedure to yield 17.9 mg (32.5%) of a
compound 6m. 1H NMR (400 MHz, CDCl3)  7.74 (s, 1H), 7.66 (s, 1H), 7.59 (d, J = 7.6
Hz, 1H), 7.54 (d, J = 7.6 Hz, 1H), 7.46 (t, J = 7.6 Hz, 1H), 5.25 ppm (dd, J = 12.4 Hz, 6.4
Hz, 1H), 4.90 (brs, 2H), 4.80 (dd, J = 13.2 Hz, 6.8 Hz, 1H), 1.60 (s, 3H), 1.58 (d, J = 6.8
Hz, 6H), 1.37 (d, J = 2.8 Hz, 3H), 1.36 (d, J = 2.8 Hz, 3H). 13C NMR (100 MHz, CDCl3)
 139.9, 130.9, 129.1, 129.0, 125.0, 124.4, 124.2, 59.9, 51.0, 46.8, 41.0, 23.5, 22.6, 21.9,
11.2. ESI-MS: m/z calcd for C19H23F3N5O [M+H] = 394.2 found 394.2.
2-butoxy-9-isopropyl-N-(3-(trifluoromethyl)benzyl)-9H-purin-6-amine (6n). The
synthesis of 6n was followed by general procedure to yield 22.4 mg (39.3%) of a
compound 6n. 1H NMR (400 MHz, CDCl3)  8.21 (brs, 1H), 7.66 (brs, 1H), 7.61-7.58 (m,
1H), 7.56-7.52 (m, 1H), 7.46-7.40 (m, 1H), 4.83 (s, 2H), 4.29 (brs, 2H), 4.15-4.05 (m,
1H), 1.78-1.69 (m, 2H), 1.59 (s, 3H), 1.57 (s, 3H), 1.52-1.40 (m, 2H), 0.99-0.94 (m, 2H),
0.92-0.90 (m, 3H). ESI-MS: m/z calcd for C20H25F3N5O [M+H] = 408.2 found 408.3.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 47
1
2
3
4
5
6
The general synthetic procedure of compounds (7). The solution of compound 5 (50
mg, 0.126 mmol) in n-BuOH (6 mL) was treated with DIEA (30.2 L, 0.19 mmol) and
7
8
9
o
various amines (33.5 mg, 0.38 mmol). The reaction mixture was heated at 120 C for 48
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
hr. Solvent was evaporated in vacuo. The residue was purified by prep-HPLC with H2O-
acetonitrile to afford compounds as oil.
9-(1-cyclopropylethyl)-N2-(2-(dimethylamino)ethyl)-N6-(3-(trifluoromethyl)benzyl)-
9H-purine-2,6-diamine (7b, NCC007). The synthesis of 7a was followed by general
1
procedure to yield 23.1 mg (41.0%) of a compound 7a. H NMR (400 MHz, CD3OD) 
8.53 (s, 1H), 8.05 (s, 1H), 7.80 (s, 1H), 7.73 (d, J = 7.6 Hz, 1H), 7.57 (d, J = 7.6 Hz, 1H),
7.49 (dd, J = 8.0 Hz, 7.6 Hz, 1H), 5.70 (s, 2H), 4.89 (dd, J = 13.6 Hz, 6.8 Hz, 1H), 1.63 (s,
13
3H), 1.62 (m, 5H). C NMR (100 MHz, CD3OD)  163.0, 162.8, 162.4, 158.0, 157.5,
134.4, 130.8, 129.3, 125.0, 117.8, 115.7, 114.5, 72.2, 56.6, 43.4, 40.6, 37.0, 20.4, 16.5,
5.0, 4.2. ESI-MS: m/z calcd for C22H29F3N7 [M+H] = 448.2 found 448.3.
9-(1-cyclopropylethyl)-N2-(2-(dimethylamino)ethyl)-N6-methyl-N6-(3-
(trifluoromethyl)benzyl)-9H-purine-2,6-diamine (7c). The synthesis of 7b was
1
followed by general procedure to yield 25.4 mg (45.9%) of a compound 7b. H NMR
(400 MHz, CDCl3)  9.89 (brs, 1H, NH), 7.77 (s, 1H), 7.62-7.47 (m, 4H), 5.69 (s, 1H),
5.15 (s, 1H), 4.12-3.90 (m, 2H), 3.89-3.75 (m, 2H), 3.40-3.25 (m, 3H), 3.18 (s, 1H), 2.91
(s, 3H, -N(CH3)3), 2.73 (s, 3H, -N(CH3)3) 1.63 (d, J = 6.8 Hz, 3H, -CH3), 1.26 (dd, J =
17.6 Hz, 4.4 Hz, 1H, cyclopropy-H), 0.85-0.75 (m, 1H), 0.62-0.56 (m, 2H), 0.52-0.44 (m,
13
1H). C NMR (125 MHz, CDCl3)  163.1, 162.7, 151.8, 131.5, 131.1, 130.9, 129.5,
28
ACS Paragon Plus Environment

Page 29 of 47
Journal of Medicinal Chemistry
1
2
3
4
5
6
125.3, 122.6, 117.8, 114.9, 57.3, 55.5, 43.2, 40.9, 37.0, 20.9, 17.1, 4.7, 4.0. ESI-MS: m/z
calcd for C23H31F3N7 [M+H] = 462.3 found 462.3.
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
N1-(9-(1-cyclopropylethyl)-6-((3-(trifluoromethyl)benzyl)oxy)-9H-purin-2-yl)-
N2,N2-dimethylethane-1,2-diamine (7d). The synthesis of 7c was followed by general
1
procedure to yield 31.2 mg (55.3%) of a compound 7c. H NMR (400 MHz, CDCl3) 
7.72 (s, 1H), 7.68 (s, 1H), 7.63 (d, J = 7.60 Hz, 1H), 7.49 (d, J = 8.0 Hz, 1H), 7.42 (dd, J
= 7.6 Hz, 1H), 5.51 (s, 2H, Ph-CH2-), 3.73-3.71 (m, 1H), 3.64 (t, J = 6.0 Hz, 2H, -HN-
CH2-), 3.04-2.97 (m, 2H, -HN-CH2-CH2-N(CH3)3), 2.65 (s, 6H), 1.52 (d, J = 6.8 Hz, 3H),
13
1.30-1.20 (m, 1H), 0.70-0.62 (m, 1H), 0.50-0.42 (m, 1H), 0.37-0.23 (m, 2H). C NMR
(100 MHz, CDCl3)  160.3, 158.1, 153.8, 137.8, 137.7, 131.3, 129.0, 125.4, 124.8, 124.7,
124.6, 122.7, 115.0, 66.9, 57.5, 56.0, 43.9, 37.7, 20.3, 16.8, 4.5, 4.0. ESI-MS: m/z calcd
for C22H28F3N6O [M+H] = 449.2 found 449.3.
Cell-based circadian assay. The cell-based circadian assay was carried out as described
previously26. Bmal1-dLuc U2OS cells were cultured in culture medium (DMEM
supplemented with 10% fetal bovine serum, 0.29 mg/ml L-glutamine, 100 units/ml
penicillin, and 100 mg/ml streptomycin). The cells were suspended and seeded onto 384-
well white color solid bottom plates with 20 µL (2,000 cells) per well. After incubation
for two days, 50 µL of the explant medium (DMEM supplemented with 2% B27, 10 mM
HEPES, 0.38 mg/ml sodium bicarbonate, 0.29 mg/ml L-glutamine, 100 units/ml
penicillin, 100 mg/ml streptomycin, 0.1 mg/ml gentamicin, and 1 mM luciferin, pH 7.2)
was added to each well, followed by the treatment of 500 nL of compounds (8 µM to 1
29
ACS Paragon Plus Environment