A New Approach of Mitigating CYP3A4 Induction Led to the Discovery of Potent Hepatitis B Virus (HBV) Capsid Inhibitor with Optimal ADMET Profiles

Article abstract of DOI:10.1021/acs.jmedchem.9b01421

Described herein is a new approach to mitigate CYP3A4 induction. In this unconventional approach, a fine-tuning of the dihedral angle between the C4 phenyl and the dihydropyrimidine core of the heteroaryldihydropyrimidine (HAP) class of capsid inhibitors

Full text of DOI:10.1021/acs.jmedchem.9b01421

Subscriber access provided by BUPMC - Bibliothèque Universitaire Pierre et Marie Curie
Article
A New Approach of Mitigating CYP3A4 Induction Led to the Discovery of
Potent Hepatitis B Virus (HBV) Capsid Inhibitor with Optimal ADMET Profiles
Xianfeng Lin, Houguang Shi, Weixing Zhang, Zongxing Qiu, Zheng Zhou, Fabian Dey, Sheng Zhong,
Hongxia Qiu, Jianxun Xie, Xue Zhou, Guang Yang, Guozhi Tang, Hong C Shen, and Wei Zhu
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01421 • Publication Date (Web): 05 Nov 2019
Downloaded from pubs.acs.org on November 5, 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 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A New Approach of Mitigating CYP3A4 Induction
Led to the Discovery of Potent Hepatitis B Virus
(HBV) Capsid Inhibitor with Optimal ADMET
Profiles
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
Xianfeng Lin,&,†,§ Houguang Shi,&,†,§ Weixing Zhang,†,§ Zongxing Qiu,†,§ Zheng Zhou, †,∥ Fabian Dey, ‡, ∥
Sheng Zhong, †,⊥ Hongxia Qiu,†,⊥ Jianxun Xie, †,⊥ Xue Zhou,†,# Guang Yang,†,# Guozhi Tang,†,§ Hong C.
Shen,*,†,§ and Wei Zhu *,†,§
‡
§
∥
†Roche Innovation Center Shanghai, Roche Innovation Center Basel, Medicinal Chemistry, Lead
Discovery, ⊥Pharmaceutical Sciences, #Discovery Virology, Roche Pharma Research and Early
Development, Bldg 5, 720 Cailun Road, Shanghai 201203, China
&These two authors contributed equally to this work.
Abstract—Described herein is a new approach to mitigate CYP3A4 induction. In this unconventional
approach, a fine-tuning of the dihedral angle between the C2 phenyl and the dihydropyrimidine core of the
heteroaryldihydropyrimidine (HAP) class of capsid inhibitors successfully altered the structure-activity-
relationships (SARs) of the unwanted CYP3A4 induction and the desired HBV capsid inhibition to more
favorable values. This eventually led to the discovery of a new capsid inhibitor with significantly reduced
CYP3A4 induction, excellent anti-HBV activity, favorable pre-clinical PK/PD profiles and no early safety
flags.
Keywords— HBV, capsid inhibitor, HAP, CYP3A4 induction, PXR, dihedral angle, liver-to-plasma ratio,
hydrodynamic injection (HDI) mouse model.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 35
1
2
3
4
Introduction.
5
6
7
8
9
Hepatitis B virus (HBV), which chronically infects approximately 240 million people
worldwide, is an important causative agent of liver fibrosis, cirrhosis and cancer.1 Despite
widespread implementation of the vaccination program, HBV infection remains a global health
burden. However, the current standard of care (SoC), namely interferon and nucleos(t)ide
analogues, cannot lead to a substantial HBV functional cure rate (typically much less than 10%),
defined by the complete loss of HBV surface antigen (HBsAg) after a finite treatment period. It is
also known that SoCs can be associated with either viral resistance or side effects.2 Consequently,
there has been a great demand in the discovery and development of non-nucleos(t)ide, direct-acting
anti-HBV drugs with novel mechanism of action in order to improve disease cure rate, as a
standalone therapy, or more likely, in combination with other antiviral agents.3, 4
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 HBV core protein plays critical roles in the HBV lifecycle, with the most prominent
one as the essential component of the nucleocapsid.5 The core protein is conserved across all HBV
genotypes and is thus considered as an attractive target for small molecule intervention. We
previously reported a new HAP class of core protein inhibitors represented by 1 (HAP_R01) and
2 (HAP_R10), as shown in Table 1.6 These compounds prevented the HBV core proteins from
assembling into a normal nucleocapsid. The misassembled nucleocapsid was presumably
subsequently degraded through the host proteasome pathway, resulting in the interruption of HBV
replication.5 The lead compound 2 demonstrated good in vitro and in vivo anti-HBV activity,
favorable pre-clinical PK and early safety profiles. However, further development of compound 2
was hampered due to its strong CYP3A4 enzyme induction observed in human hepatocytes, which
was later confirmed in the phase I clinical trial with healthy volunteers (unpublished).
ACS Paragon Plus Environment

Page 3 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The induction of CYP3A4 enzyme represents a significant clinical issue because CYP3A4
is involved in the metabolism of more than 50% of marketed drugs.7 Specifically, CYP3A4
induction can lead to the reduction of therapeutic efficacy of a co-medication or the CYP3A4
inducer itself (a phenomenon known as auto-induction if the inducer itself is metabolized by
CYP3A4). The induction of CYP3A4 enzyme can occur through several mechanisms, with the
predominant one occurring through the activation of pregnane X receptor (PXR), a nuclear
hormone receptor.8 In the human PXR transactivation assay, both compounds 1 and 2 were found
to be strong PXR activators, indicating PXR activation as the mechanism of CYP3A4 induction
for this chemical class of capsid inhibitors. Several crystal structures of the human PXR ligand
binding domain (LBD) complexed with structurally diversified PXR activators were reported,
which revealed an unusually large and flexible binding pocket.8, 9 A plausible pharmacophore
model of PXR activators was proposed, which comprised one H-bond acceptor and at least two
flanking hydrophobic groups, in particular aromatic groups.8 The structure of the HAP class of
capsid inhibitors represented by compound 1 and 2 coincidently contains all the pharmacophore
elements favorable for hPXR binding, in which the C2 thiazoyl and the C4 phenyl play the part of
hydrophobic groups and the nitrogen of the dihydropyrimidine or the carbonyl oxygen of the C5
methyl ester accounts for the key H-bonding interaction with PXR, thus potentially explaining the
strong CYP3A4 induction effect observed for both compound 1 and 2. The well adopted strategies
to reduce hPXR activation include: (1) introducing polar substituent to the lipophilic group to
destabilize the hydrophobic interactions; (2) removing the central H-bond acceptor.8, 9 However,
from the co-crystal structure of Cp149 Y132A mutant hexamer in complex with 1 (5WRE) (Figure
1a), both the C2 thiazoyl and the C4 phenyl were located in hydrophobic pockets, where
introduction of polar substituent would essentially lead to a loss of binding affinity to core proteins.
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 35
1
2
3
4
5
6
7
8
9
Moreover, the removal of potential H-bond acceptors, e.g. replacing the dihydropyrimidine with
dihydropyridine or C5 ester with triflouromethyl, led to a complete loss of anti-HBV activity. In
this manuscript, we report a new approach to mitigate hPXR activation, which culminated with
the discovery of a new capsid inhibitor with significantly reduced CYP3A4 induction, excellent
anti-HBV activity, favorable pre-clinical PK/PD profiles, and no early safety flags.
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
Results and Discussion.
Working Hypothesis. In theory, when a ligand binds to its targeted protein, the closer the
binding conformation is to its lowest energy conformation in water, the higher binding affinity it
could exert. In the crystal structure of Cp149 Y132A in complex with compound 1 (5WRE),
compound 1 binds to core proteins in a conformation where the C4 phenyl is nearly perpendicular
to the dihydropyrimidine plane with a dihedral angle of -64.3o (Figure 1a).6 This is almost identical
to the conformation observed in the single crystal structure of compound 1a (the core structure of
compound 1), where a dihedral angle of -64.5o is observed (Figure 1b and 1c) except for an 180o
flip of the C4 phenyl ring in the co-crystal structure (the –Cl substituent will clash with the residues
forming the cap if it orients as that in the co-crystal structure of 1a).6 Interestingly, in the CSD
five out of the seven total related small molecule crystal structures (codes: 002548, 002919,
002920, FAGDAS, FAGDEW) show a similar conformation as observed in the protein/ligand
crystal structure with dihedrals in the range between -62.3 and -88.5°.10 The remaining two (codes
200608 and HAYPOP) have the phenyl flipped with the chlorine above the dihydropyrimidine
plane and the ester flipped by 180° as well.10 The concordance of the dihedral angles between C4
phenyl and dihydropyrimidine plane in water and in the binding pocket could nicely explain the
high binding affinity of compound 1 to core proteins, and thus its strong anti-HBV activity.
ACS Paragon Plus Environment

Page 5 of 35
Journal of Medicinal Chemistry
1
2
3
4
Likewise, it was hypothesized that the strong CYP3A4 induction of compound 1 might also be
5
6
ascribed to its favorable binding to hPXR LBD in this conformation.
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
F
4'
2'
O
N
Cl
N
4
O
N
S
N
H
O
HO
F
F
1 (HAP_R01)
Figure 1a: Crystal structure of Cp149 Y132A in complex with compound 1 (PDB code 5WRE). Compound
1 is shown in stick with cyan carbons. The molecular surface of the protein is shown in grey in the vicinity
of the ligand. The core protein monomer forming the top of the binding pocket is omitted for clarity. The
dihedral angle constituted by the bonds labelled in red in the schematic representation was -64.3o in the co-
crystal structure.
F
Cl
O
O
N
S
N
H
N
1a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 35
1
2
3
4
Figure 1b: X-ray of 1a. Compound 1a was shown stick with the cyan carbons. The dihedral angle
5
6
constituted by the bonds labelled in red in the schematic representation was -64.5o.
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 1c: Overlay of 1 (cyan) and 1a (green).
As evident in the binding pocket (Figure 1a), there is still a bit of room around the C4
phenyl of compound 1 to allow it to move slightly around the C–C bond connecting to the
dihydropyrimidine core. As such, the SAR of capsid inhibition might tolerate small changes of the
dihedral angle between the C4 phenyl and the dihydropyrimidine plane. The medicinal chemistry
efforts were thus focused on the modulation of this dihedral angle in order to separate SARs of
CYP3A4 induction and capsid inhibition.
CYP3A4 Induction SAR. The 2’-substituent of C4 phenyl of compound 1 was assumed
to be crucial in determining the dihedral angle between the C4 phenyl and the dihydropyrimidine
plane. In the presence of an ester at the C5 position, the smaller the 2’-substituent, the smaller the
dihedral angle will likely be inferred from modelling results (vide infra). As such, we set out to
explore smaller substituents than –Cl, e.g. –H and –F.
The selected compounds were evaluated for their ability to induce CYP3A4 mRNA using
the cryopreserved primary human hepatocytes. The induction was determined in cells exposed to
ACS Paragon Plus Environment

Page 7 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the individual compounds at a concentration of 10 µM by measuring CYP3A4 mRNA levels, in
comparison with the positive control rifampicin (RIF) at the same concentration. The choice of
10 µM is for detection of induction signals as much as possible. An increase in CYP3A4 mRNA
contents that was >40% of the level stimulated by rifampicin was clearly considered as a risk for
induction in clinic based on internal experience and FDA guidance.11
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 1: Data of anti-HBV EC50 and CYP3A4 induction of compound 1–20.
O
Ar
R2
O
N
4
S
N
H
2
R1
N
EC50
CC50
CYP3A4
ID
Ar
R1
R2
(µM)a
(µM)b
induction(%)c
F
O
N
1
2
3
4
Methyl
0.003
0.003
0.010
0.002
65
109
111
123
173
F
HO
Cl
F
F
F
F
O
N
HO
Methyl
Ethyl
> 100
> 100
> 100
Cl
Cl
Cl
O
O
N
F
HO
F
O
N
O
HO
Ethyl
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 35
1
2
3
4
5
6
7
8
9
F
F
O
N
d
5
6
7
8
Methyl
Ethyl
Ethyl
Ethyl
-
> 100
> 100
62
96
135
85
F
HO
F
F
F
O
O
O
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
HO
HO
HO
0.014
0.015
0.011
O
F
N
O
F
F
N
O
> 100
41
F
O
X
N
O
d
d
HO
9a-c
Ethyl
> 1
-
-
X = Cl / Br / CN
F
O
O
O
O
F
F
N
O
HO
HO
HO
HO
10
11
12
13
Ethyl
Ethyl
Ethyl
Ethyl
0.018
0.005
0.003
0.002
> 100
49
72
105
73
F
F
F
N
O
Br
F
F
N
O
14
Cl
N
O
d
24
-
ACS Paragon Plus Environment

Page 9 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
O
O
F
F
F
N
O
HO
HO
HO
14
15
16
17
18
19
20
Ethyl
Ethyl
0.005
0.015
0.046
0.022
0.028
0.01
85
69
40
94
36
52
H
F
F
N
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
N
O
Methyl
Methyl
Methyl
Methyl
Methyl
> 100
56
F
F
O
N
F
HO
F
O
O
O
N
O
d
HO
HO
HO
> 100
> 100
76
-
Cl
F
F
N
O
25
26
Br
N
O
0.01
a EC50 is the mean value for reduction of HBV DNA by 50% in HepG.2.2.15 cells. Experiments were run
in duplicate, with variation < 15%. b CC50 is the mean value for cytotoxicity to HepDE19 cells.
Experiments were in duplicate runs with variation < 15%. c mRNA level of CYP3A4 in percentage to that
of rifampicin with compound treatment at 10µM. d Not determined.
As seen in Table 1, the 2’–F analogues 5 and 6 showed slightly reduced CYP3A4 induction
compared to the 2’–Cl analogues 1 and 4. The smaller 2’–H analogue 7 exhibited more
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 35
1
2
3
4
5
6
7
8
9
pronounced effect on reducing CYP3A4 induction with a value of 85% (7 vs 4) while still
maintaining potent anti-HBV activity with an EC50 of 15 nM in HepG2.2.15 cells, which boosted
our confidence in the exploration of this novel approach to reduce CYP3A4 induction.
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
With the preliminary success, we next set out to explore the effect of 3’–substituent on
CYP3A4 induction. However, the effect on core protein binding was uncertain though there was
a bit of room in the binding pocket. As such, we prepared the –CN, –F, –Cl, –Br substituted
analogues at the 3’-position. From the SAR results, the substituents with a Van der Walla radii
larger than –F were not tolerated in the capsid binding pocket, which led to a significant loss of
anti-HBV activity (> 1 µM for all 3 analogues, Examples 9a-c). However, the 3’,4’-difluoro
substituted analogue 8 maintained potent anti-HBV activity with an EC50 of 11 nM. More
importantly it demonstrated an increased selectivity of anti-HBV activity vs CYP3A4 induction
with the latter decreased to 41% of rifampicin (an average of >3 tests). Subsequently, we
synthesized the 3’,4’,5’-trifluoro substituted analogue 10, which, however, displayed a decrease
of anti-HBV activity and an elevation of CYP3A4 induction. The introduction of one additional
–F substituent at the 2’-position could further improve anti-HBV activity with an EC50 of 5 nM
but at the cost of the increased CYP3A4 induction (Example 11).
In order to further increase the selectivity of anti-HBV activity vs CYP3A4 induction, the
substituent at the 4’-position was explored. From our previous SAR findings, we knew that the 4’–
F of C4 phenyl could be successfully replaced with –Me, –Cl and –Br with comparable or even
better anti-HBV activity.6 Indeed, all three analogues (Examples 12–14) displayed improved anti-
HBV activity in the single digit nM range. Regarding CYP3A4 induction potential, compound 12
showed a value of 73% relative to rifampicin. However, considering its cytotoxicity profile (CC50
is 14 µM in HepDE19 cells), this result may not accurately reflect its induction potential in human
ACS Paragon Plus Environment

Page 11 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hepatocytes as cytotoxicity can affect CYP3A4 induction readout also at the tested concentration
of 10 µM. Compound 13 also showed an unfavorable CC50 of 24 µM. The 4’–Me analogue 14
has potent anti-HBV EC50 of 5 nM, good CC50 of 85 µM and similar CYP3A4 induction level
(40% relative to RIF) to compound 8. The 4’–H analogue 15, however, demonstrated an increased
CYP3A4 induction potential (94% relative to RIF) albeit a similar anti-HBV EC50 to compound
8.
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
Both analogues 8 and 14 demonstrated reduced CYP3A4 induction at a level around 40%
relative to RIF, which met our preliminary goal. In order to minimize the concerns of DDI and
auto induction in clinical use, we were keen to explore whether the CYP3A4 induction could be
further reduced for this HAP chemical series. By comparing the C5 methyl ester to C5 ethyl ester
analogues (1 vs 3, and 2 vs 4), it was clear that the methyl ester analogues had lower potential of
CYP3A4 induction likely due to their reduced lipophilicity.12 As such, the corresponding C5
methyl ester analogues 16–20 were synthesized and the CYP3A4 induction data is summarized in
Table 1. Indeed, they all demonstrated reduced CYP3A4 induction in human hepatocytes as
compared to their corresponding ethyl ester analogues. As observed in previous SAR, they also
showed decreased anti-HBV activity compared to the ethyl ester analogues, of which the 4’–Cl
and 4’–Br analogues 19 and 20, however, still displayed potent anti-HBV activity. Compound 19
was the most interesting one with an anti-HBV EC50 of 10 nM, a high CC50 of >100 µM and a low
CYP3A4 induction value of 25%.
CYP3A4, CYP1A2 and CYP2B6 Induction Assessments of 19. With a low CYP3A4
induction potential at 10 µM, compound 19 was further tested in human hepatocytes for its
induction potentials in a dose dependent manner for various CYP enzymes including CYP3A4,
1A2 and 2B6. As seen in Table 2, compound 19 showed dose dependent decrease of CYP3A4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 35
1
2
3
4
5
6
7
8
9
induction from 10 to 0.1 µM (25% at 10 µM, 7.5% at 1 µM and 2% at 0.1 µM), which indicated
indeed a low risk of DDI and auto-induction in clinical use. In contrast, compound 2 showed
CYP3A4 inductions of 111%, 83% and 22% of rifampicin at 10, 1 and 0.1 µM, respectively.
Moreover, compound 19 had neither CYP1A2 nor CYP2B6 induction liabilities as determined by
the changes of individual mRNAs compared with that of corresponding positive controls (<40%
was accepted as a low risk of induction).
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: Induction data of 19 on CYP3A4, CYP1A2 and CYP2B6 mRNAs.
Enzyme subtype
CYP3A4a
Concentrations (µM)
% of positive control
0.1
1
2
7.5
25
10
0.1
1
1.0
0.9
1.7
1.3
1.5
1.4
CYP1A2b
CYP2B6c
10
0.1
1
10
a Data was reported as CYP3A4 mRNA change with the treatment of 19 in percentage to that of rifampicin
at 10µM. b Data was reported as CYP1A2 mRNA change with the treatment of 19 in percentage to that of
omeprazole at 10µM. c Data was reported as CYP2B6 mRNA change with the treatment of 19 in percentage
to that of phenobarbital at 10µM.
ACS Paragon Plus Environment

Page 13 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chemistry. The synthetic route of preparing this new series of 3’,4’-disubstituted C4
phenyl HAP analogues was shown in Scheme 1. As exemplified by the synthesis of compound 19,
a three-component condensation reaction of methyl acetoacetate, 3-fluoro-4-chlorobenzaldehyde
and thiazole-2-carboxamidine afforded the key intermediate dihydropyrimidine 19a as a racemic
mixture. An SFC (supercritical fluid chromatography) chiral separation of 19a afforded two
enantiomerically pure intermediates, of which (S)-19a was determined as the desired enantiomer
by X-ray diffraction study. (S)-19a was then reacted with NBS to give bromide 19b, which was
finally coupled with (3S)-morpholine-3-carboxylic acid to provide 19 as a single isomer after
HPLC purification.
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 1: Preparation of compound 19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 35
1
2
3
Cl
4
5
6
7
8
9
F
Cl
.HCl
S
NH
a
F
O
O
O
+
+
H2N
O
O
N
N
S
.
O
N
H
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
19a
Cl
Cl
F
F
b
O
O
+
O
N
R
O
N
S
S
S
N
H
N
H
N
N
(R)-19a
(S)-19a
O
H
Cl
Cl
Cl
.HCl
N
F
F
F
HO
O
O
O
c
O
O
N
O
N
O
N
d
S
S
S
O
N
H
N
H
N
H
N
O
N
N
Br
N
HO
(S)-19a
19
19b
o
Reagents and conditions: (a) i-PrOH, HOAc, piperidine, 48 C, 2 hours; followed by the addition of
o
thiazole-2-carboxamidine, then 90 C, 3 hours, 52%; (b) SFC chiral separation; (c) NBS, CCl4, r.t., 1 hr,
44%; (d) DIPEA, DCM, ca. 1 hr, 57%.
Putative Explanation of the Reduced CYP3A4 Induction of 19. In order to prove that
the 3’,4’-disubstituted C4 phenyl analogues such as compound 19 indeed has a smaller dihedral
angle between C4 phenyl and dihydropyrimidine plane than that of 2’,4’-disubstituted C4 phenyl
HAP analogues, we solved the single crystal structure of intermediate (S)-19a. Indeed, as shown
in Figure 3, compound (S)-19a has a much smaller dihedral angle of 1.3o than that of compound
1a (64.5o) (Figure 2a and 2b). Interestingly, with this drastic conformation change, (S)-19a shows
ACS Paragon Plus Environment

Page 15 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
an opposite optical rotation to 1a, where (S)-19a has [α]d of +50o, while 1a has [α]d of -55o.
However, in the absence of any structural information on the binding of these compounds to hPXR,
it remains speculative to ascribe the decreased CYP3A4 induction to the decreased dihedral angle
though this dihedral angle change had a consistent and pronounced effect on hPXR activation.
Two plausible explanations were proposed. On one hand, the preferable conformation of 19 in
water with a small dihedral angle between C4 phenyl and the dihydropyrimidine core may fit
poorly or lead to enthalpy penalty if the C4 phenyl has to be oriented to the conformation of a
larger dihedral angle in the binding pocket of hPXR LBD. On the other hand, the nearly planar
orientation of the C4 phenyl to the dihydropyrimidine plane might prevent either the 3-N of
dihydropyrimidine or the carbonyl of C5 ester from forming a favorable hydrogen bonding
interaction with the corresponding residues of hPXR, e.g. Ser247 or His407 as described in the
literature.8
20
20
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
Cl
F
O
O
N
S
N
H
N
(S)-19a
Figure 2a: X-ray of (S)-19a. Compound (S)-19a was shown as ball and stick with the colour of carbon in
cyan. The dihedral angle consisted by the bonds labelled in red in the schematic representation was 1.3o.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 35
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 2b: Overlay of 1a (green) and (s)-19a (purple).
ADMET and Preclinical PK/PD Studies of 19. With potent ant-HBV EC50, higher CC50
and low CYP enzymes induction potential, 19 was further profiled in ADMET and pre-clinical
PK/PD studies in order to evaluate its potential for further development. The physicochemical and
ADME data were summarized in Table 3. Compound 19 had a LogD of 0.38 and two pKa values
of 1.8 and 5.9, respectively. The ionizable carboxylic acid group of compound 19 offered the
advantage of a high solubility of >666 µg/mL. Moreover, compound 19 showed good permeability
under the conditions of the PAMPA assay at pH 6.5. This compound displayed a low human
hepatic clearance and a moderate mouse hepatic clearance. Incubation in human hepatocytes
revealed that the unchanged parent was the major form with a minor portion of a phase II
metabolite as a result of glucuronide conjugation of the carboxylic acid group (the acylglucuronide
reactivity was assessed low). Compound 19 also demonstrated a sufficient free fraction of 3.1%
and 4.0% in human and mouse plasmas, respectively.
Table 3: Physicochemical, ADME and PK properties of 19.
Physicochemical properties
LogDa:
pKab (acid/base):
0.38
1.8/5.9
ADME properties
ACS Paragon Plus Environment

Page 17 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Lysac (µg/mL):
PAMPAd (10−6 cm/s):
Human/Mouse Hep_CL (mL min−1 kg−1)e:
PPB (human/mouse)f:
>631
2.4
3.6/48.6
3.1/4.0
Mouse SDPK profilesg
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
Plasma
t1/2 (IV,H):
2.3
2.7
Vss (IV, L/kg):
CL (IV) (mL min−1 kg−1):
40.9
254
408
527
43.1
19985
38
Cmax (PO, μg/L):
AUC(0-∞) (IV, µg/L*hr):
AUC(0-∞) (PO, µg/L*hr):
F (%):
Liver
AUC(0-∞) (PO, µg/L*hr):
AUC(0-∞), liver/AUC(0-∞), plasma (PO)
a Apparent partition coefficients at pH 7.4. b Ionization constant. c Lyophilisation solubility assay (LYSA)
(µg/mL). Parallel Artificial Membrane Permeability Assay (10-6 cm/s). Predicted hepatic clearance
d
e
obtained with human and mouse hepatocytes (mL min−1 kg−1). f Plasma protein binding assay (% unbound
g
fraction, average of 3 tests). The single-dose pharmacokinetics (SDPK) study of 19 was carried out in
mice as indicated according to the standard procedures (IV: 1 mg/kg; PO.: 3 mg/kg). Major parameters,
including plasma clearance (Cl), volume of distribution at steady state (Vss), T1/2 (i.v.), Maximal
concentration (Cmax), area under the curve (AUC), and oral bioavailability (F) are reported.
In mouse SDPK studies, compound 19 showed a moderate clearance similar to the in vitro
prediction in mouse hepatocytes, a moderate half-life and oral bioavailability. Despite a relatively
low oral exposure in plasma, 19 demonstrated a high liver exposure (liver-to-plasma exposure ratio
was 38), which was a desirable property for HBV infection treatment as liver is the disease organ.
Supported by potent in vitro anti-HBV activity and favorable mouse PK profiles,
compound 19 was assessed against HBV infection in the hydrodynamic injection (HDI) mouse
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 35
1
2
3
4
5
6
7
8
9
model.13 In this study, female BALB/c mice were hydrodynamically injected with replication-
competent HBV DNA plasmid through the tail vein. Twenty hours post plasmid injection, mice
were orally dosed by blank vehicle, entecavir (ETV, 0.1 mg/kg, once daily as the positive control),
or different doses (3 and 12 mg/kg per dose, twice daily) of compound 19 for 5 days. Plasma and
liver samples were collected at the indicated time points for HBV DNA quantification by real-time
quantitative polymerase chain reaction (qPCR). As shown in Figure 3, with treatment of
compound 19, HBV DNA copy numbers were rapidly and markedly reduced in plasma and were
hardly detectable in liver tissues, demonstrating a strong anti-HBV effect of compound 19 upon
oral administration. The maximum efficacy of compound 19 was comparable to the maximum
efficacy of ETV in this model.
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: Levels of HBV DNA in plasma (A) and liver (B) of HBV HDI mice upon treatment with the
vehicle control, ETV control, and compound 19. Error bars indicate standard error of the mean.
Like its close analogue 2, compound 19 was also clean in the following in vitro safety
assessment including hERG, GSH (glutathione adduction), mutagenicity (Ames assay) and
clastogenicity (micronucleus assay). The 5-day minitox study in rats showed sufficient safety
ACS Paragon Plus Environment

Page 19 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
margin as defined by ratio of the plasma exposure (Cmax and AUC) in rat at NOAEL (no observed
adverse effect level) versus the exposure at a projected human efficacious dose, which all
supported further development of 19.
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.
In summary, we have described herein a novel approach to reduce CYP3A4 induction,
which relied on a finely designed conformational change between two close analogues in the same
chemical series. This seemingly unexpected finding might remind medicinal chemists of the
importance of looking at the ligand in a 3D (3-dimensional) way (they may look very similar in
2D, but differ dramatically in 3D!). Plausible explanations of the conformation effect on CYP3A4
induction were provided also. However, given the promiscuous nature of the ligand-binding site
of PXR which can accommodate structurally diverse small molecules in multiple ways, additional
strategies have been identified to completely abolish hPXR binding while retaining anti-HBV
activity in the HAP class of capsid inhibitors. Further research in this area will be reported in due
course.
Experimental Section
Synthetic Chemistry General Comments. All of the intermediates were purified by silica
gel chromatography using either a Biotage SP1 system or an ISCO CombiFlash chromatography
instrument. All of the final compounds were purified by preparative HPLC (prep-HPLC) on a
reversed-phase column using a Waters XBridge OBD Phenyl (30 mm × 100 mm, 5 μm) or OBD
RP18 (30 mm × 100 mm, 5 μm) column under acidic conditions (A, 0.1% formic acid in H2O; B,
0.1% formic acid in acetonitrile) or basic conditions (A, 0.1% ammonia in H2O; B, acetonitrile).
For SFC chiral separation, the intermediates were separated using a chiral column (Daicel
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 35
1
2
3
4
5
6
7
8
9
Chiralpak IC, 30 mm × 250 mm, 5 μm) on a Mettler Teledo SFC-Multigram system (solvent
system of 95% CO2 and 5% IPA (0.5% TEA in IPA), backpressure of 100 bar, UV detection at
254 nm). Optical rotation was measured using a Rudolph Autopol V automatic polarimeter at a
wavelength of 589 nm. LC–MS spectra were obtained using a MicroMass Platform LC (Waters
Alliance 2795-ZQ2000). NMR spectra were obtained using Bruker AVANCE 400 MHz
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
spectrometer, operating at 400.13 MHz (1H) and 100.62 MHz (13C). H NMR spectra were
obtained using the single pulse zg30.13C{1H} NMR spectra were obtained using composite pulse
zgpg30 for proton decoupling. The chemical shifts were referenced against internal TMS(1H,13C)
and the coupling constants were calculated as a whole of H,H and H,F. All of the starting materials
were obtained commercially. All of the final compounds had purities greater than 95% based upon
HPLC, LC–MS and 1H NMR analyses. All of the reported yields are for isolated products and are
not optimized.
General Synthetic Procedure: exemplified by (3S)-4-[[(4S)-4-(4-chloro-3-fluoro-phenyl)-5-
methoxycarbonyl-2-thiazol-2-yl-1,4-dihydropyrimidin-6-yl]methyl]morpholine-3-
carboxylic acid (19).
To a stirred solution of 4-chloro-3-fluorobenzaldehyde (0.500 g, 0.32 mmol) and methyl
acetoacetate (0.366 g, 3.2 mmol) in i-PrOH (6 mL) were added acetic acid (0.020 g, 0.034 mmol)
and piperdine (0.027 g, 0.032 mmol) at 48 oC. The resulting mixture was stirred at 48 oC for two
hours. Then thiazole-2-carboxamidine hydrochloride (0.516 g, crude, ca. 2.52 mmol) and
triethylamine (0.318 g, 3.2 mmol) were added to the reaction mixture. The mixture was heated at
o
90 C for three hours. After it was cooled to room temperature, the reaction mixture was
concentrated and the residue was dissolved in ethyl acetate and washed with brine. The organic
layer was dried over anhydrous Na2SO4. The solvent was removed and the residue was purified by
ACS Paragon Plus Environment

Page 21 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
silica gel column chromatography (ethyl acetate/petroleum ether: 1/4 to 1/2) to afford methyl 4-
(4-chloro-3-fluoro-phenyl)-6-methyl-2-thiazol-2-yl-1,4-dihydropyrimidine-5-carboxylate
(Compound 19a) as a yellow solid (600 mg, 52%). MS: calc’d (MH+) 366, measured (MH+) 366.
1H NMR (MeOD-d4, 400 MHz) δ ppm 7.96 (m, 1H), 7.77 (m, 1H), 7.42 (m, 1H), 7.18 (m, 2H),
5.69 (br s, 1H), 3.69 (s, 3H), 2.49 (s, 3H).
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 racemic compound 19a was subjected to SFC chiral separation. The desired (-)-enantiomer
methyl
(4S)-4-(4-chloro-3-fluoro-phenyl)-6-methyl-2-thiazol-2-yl-1,4-dihydropyrimidine-5-
carboxylate ((S)-19a) has a relatively short retention time. The absolute stereochemistry of (S)-
19a was determined by X-ray diffraction study (see supporting information). (S)-19a: [α]D20 +50o
(c 0.21 g/100 mL, MeOH).
To a stirred solution of compound (S)-19a (0.160 g, 0.44 mmol) in CCl4 (5 mL) was added NBS
(0.070 g, 0.39 mmol) in portions. After the reaction mixture was stirred at 70 oC for 3 hours, the
solvent was removed in vacuo and the residue was purified by column chromatography to give
methyl
(4S)-6-(bromomethyl)-4-(4-chloro-3-fluoro-phenyl)-2-thiazol-2-yl-1,4-
dihydropyrimidine-5-carboxylate (compound 19b) as a yellow solid (0.086 g, 44%). MS: calc’d
445 (MH+), measured 445 (MH+).
To a stirred solution of compound 19b (0.086 g, 0.19 mmol) and (3S)-morpholine-3-carboxylic
acid hydrochloride salt (cas: 1187929-04-9, 0.042 g, 0.25 mmol) in dichloromethane (1 mL) was
added dropwise DIPEA (0.13 mL, 0.77 mmol). The reaction mixture was stirred at room
temperature until the disappearance of 19b. The mixture was diluted with EtOAc (10 mL), washed
successively with saturated aqueous NH4Cl solution and brine. The organic layer was separated
and dried over anhydrous Na2SO4. The solvent was concentrated in vacuo and the residue was
purified by prep-HPLC to give (3S)-4-[[(4S)-4-(4-chloro-3-fluoro-phenyl)-5-methoxycarbonyl-2-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 35
1
2
3
4
thiazol-2-yl-1,4-dihydropyrimidin-6-yl]methyl]morpholine-3-carboxylic acid (compound 19) as a
5
6
o
light yellow solid (0.055 mg, 57%). Melting point: 136.5 C. MS: calc’d (MH+) 495, measured
7
8
9
(MH+) 495. HRMS: calc’d (MH+) 495.08997, measured (MH+) 495.08958. 1H NMR (DMSO-d6,
400 MHz) δ ppm 12.9–13.0 (m, 1H), 12.9–13.0 (m, 1H), 12.4–13.0 (m, 1H), 9.8–10.0 (m, 1H),
8.0–8.1 (m, 1H), 7.9–8.0 (m, 1H), 7.5–7.6 (m, 1H), 7.3–7.3 (m, 1H), 7.2–7.2 (m, 1H), 5.6–5.7 (m,
1H), 4.25 (br d, 1H, J = 17.6 Hz), 3.9–4.0 (m, 2H), 3.8–3.8 (m, 1H), 3.6–3.7 (m, 2H), 3.6–3.6 (m,
4H), 3.0–3.1 (m, 1H), 2.35 (br d, 1H, J = 12.1 Hz). 13C NMR (DMSO-d6, 400 MHz): δ ppm 172.8,
166.3, 162.4, 158.9, 156.4, 147.5, 146.7(d, 2JCF =20 Hz, FC), 145.4, 144.2, 131.2, 125.3, 124.7 (d,
3JCF =12 HZ, FC), 118.7, 118.6, 115.8, 115.6, 97.1, 69.4, 67.2, 61.9, 58.6, 54.8, 51.5, 48.5.
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 following compounds are prepared in analogy to 19 from commercially available
building blocks.
(R)-6-((S)-2-Carboxy-4,4-difluoro-pyrrolidin-1-ylmethyl)-4-(2-chloro-4-fluoro-phenyl)-2-
thiazol-2-yl-1,4-dihydro-pyrimidine-5-carboxylic acid ethyl ester (3). MS: calc’d 529 (MH+),
1
exp 529 (MH+). H NMR (MeOD-d4, 400 MHz): δ ppm 8.09 (d, J = 3.01 Hz, 1H), 8.02 (d, J =
3.01 Hz, 1H), 7.57 (dd, J = 8.78, 6.02 Hz, 1H), 7.32 (dd, J = 8.78,2.51 Hz, 1H), 7.16 (td, J = 8.41,
2.51 Hz, 1H), 6.25 (s, 1H), 4.53 (d, J = 16.06 Hz, 1H), 4.00–4.18 (m, 4H), 3.68 (q, J = 11.13 Hz,
1H), 3.20–3.30 (m,1H), 2.77–2.94 (m, 1H), 2.49–2.65 (m, 1H), 1.17 (t, J = 7.15 Hz, 3H).
(S)-4-[(R)-6-(2-Chloro-4-fluoro-phenyl)-5-ethoxycarbonyl-2-thiazol-2-yl-3,6-dihydro-
pyrimidin-4-ylmethyl]-morpholine-3-carboxylic acid (4). MS: calc’d (MH+) 509, measured
(MH+) 509. 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.08 (d, J = 3.01 Hz, 1H), 8.01 (d, J = 3.01
Hz, 1H), 7.60 (dd, J = 8.53, 6.02 Hz, 1H), 7.31 (dd, J = 8.66,2.38 Hz, 1H), 7.15 (td, J = 8.34, 2.38
Hz, 1H), 6.26 (s, 1H), 4.88 (d, J = 16.31 Hz, 1H), 4.59 (d, J = 16.56 Hz, 1H), 4.26–4.33 (m, 1H),
ACS Paragon Plus Environment

Page 23 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
4.22 (d, J = 4.02Hz, 2H), 4.11 (q, J = 6.94 Hz, 2H), 3.93–4.05 (m, 2H), 3.78 (ddd, J = 12.42, 8.28,
3.64 Hz, 1H), 3.27 (d, J = 13.05 Hz, 1H), 1.16 (t, J = 7.15 Hz, 3H).
7
8
9
(2S)-1-[[(4R)-4-(2,4-difluorophenyl)-5-methoxycarbonyl-2-thiazol-2-yl-1,4-
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
dihydropyrimidin-6-yl]methyl]-4,4-difluoro-pyrrolidine-2-carboxylic acid (5). MS: calc’d
499 (MH+), exp 499 (MH+). 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.14 (dd, J = 3.1, 12.9 Hz,
2H), 7.53 (dt, J = 6.5, 8.5 Hz, 1H), 7.0–7.1 (m, 2H), 6.04 (s, 1H), 4.54 (d, J = 16.1 Hz, 1H), 4.0–
4.2 (m, 2H), 3.70 (s, 3H), 3.5–3.7 (m, 1H), 3.2–3.3 (m, 1H), 2.8–2.9 (m, 1H), 2.5–2.6 (m, 1H).
(S)-4-[(R)-6-(2,4-Difluoro-phenyl)-5-ethoxycarbonyl-2-thiazol-2-yl-3,6-dihydro-pyrimidin-
4-ylmethyl]-morpholine-3-carboxylic acid (6). MS: calc’d 493 (MH+), exp 493 (MH+). 1H NMR
(MeOD-d4, 400 MHz): δ ppm 7.98 (d, J = 2.8 Hz, 1H), 7.79 (d, J = 2.8 Hz, 1H), 7.50–7.36 (m,
1H), 7.04–6.87 (m, 2H), 6.02–5.93 (m, 1H), 4.38–4.19 (m, 2H), 4.17–4.00 (m, 4H), 3.97–3.78 (m,
3H), 3.58 (br. s., 1H), 2.71 (br. s., 1H), 1.19 (t, J = 7.0 Hz, 3H).
(S)-4-[(S)-5-Ethoxycarbonyl-6-(4-fluoro-phenyl)-2-thiazol-2-yl-3,6-dihydro-pyrimidin-4-
ylmethyl]-morpholine-3-carboxylic acid (7). MS: calc’d 475 (MH+), exp 475 (MH+). 1H NMR
(MeOD-d4, 400 MHz): δ ppm 8.00 (d, J = 3.0 Hz, 1H), 7.84 (d, J = 3.0 Hz, 1H), 7.41 (dd, J = 5.4,
8.7 Hz, 2H), 7.07 (t, J = 8.8 Hz, 2H), 5.71 (s, 1H), 4.58 (d, J = 16.1 Hz, 1H), 4.31 (br. s., 1H),
4.21–4.03 (m, 4H), 3.98–3.81 (m, 2H), 3.71 (br. s., 1H), 3.50 (dd, J = 1.6, 3.4 Hz, 1H), 2.85 (br.
s., 1H), 1.23 (t, J = 7.2 Hz, 3H).
(S)-4-[(S)-6-(3,4-Difluoro-phenyl)-5-ethoxycarbonyl-2-thiazol-2-yl-3,6-dihydro-pyrimidin-
4-ylmethyl]-morpholine-3-carboxylic acid (8). MS: calc’d 493 (MH+), exp 493 (MH+). 1H NMR
(MeOD-d4, 400 MHz): δ ppm 8.00 (d, J = 3.26 Hz, 1H), 7.83 (d, J = 3.01 Hz, 1H), 7.14–7.31 (m,
3H), 5.71 (s, 1H), 4.54 (d, J = 17.07 Hz, 1H), 4.02–4.30 (m, 5H), 3.78–3.98 (m, 2H), 3.67 (t, J =
3.89 Hz, 1H), 3.41 (br. s., 1H), 2.70–2.86 (m, 1H), 1.24 (t, J = 7.15 Hz, 3H).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 35
1
2
3
4
5
6
7
8
9
(S)-4-[(S)-5-Ethoxycarbonyl-2-thiazol-2-yl-6-(3,4,5-trifluoro-phenyl)-3,6-dihydro-
pyrimidin-4-ylmethyl]-morpholine-3-carboxylic acid (10). MS: calc’d 511 (MH+), exp 511
(MH+). 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.01 (d, J = 3.26 Hz, 1H), 7.83 (d, J = 3.26 Hz,
1H), 7.13 (dd, J = 8.66, 6.65 Hz, 2H), 5.72 (s, 1H), 4.55 (d, J = 17.32 Hz, 1H), 4.01–4.26 (m, 5H),
3.78–3.96 (m, 2H), 3.67 (t, J = 3.76 Hz, 1H), 3.35–3.47 (m, 1H), 2.61–2.79 (m, 1H), 1.25 (t, J =
7.03 Hz, 3H).
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)-4-[(R)-5-Ethoxycarbonyl-2-thiazol-2-yl-6-(2,3,4-trifluoro-phenyl)-3,6-dihydro-
pyrimidin-4-ylmethyl]-morpholine-3-carboxylic acid (11). MS: calc’d 511 (MH+), exp 511
(MH+). 1H NMR (MeOD-d4, 400 MHz): δ ppm 7.93–8.08 (m, 1H), 7.76–7.90 (m, 1H), 7.19–7.31
(m, 1H), 7.04–7.16 (m, 1H), 6.02 (br. s., 1H), 4.35–4.54 (m, 1H), 4.10 (d, J = 4.77 Hz, 4H), 3.79–
3.97 (m, 2H), 3.56–3.70 (m, 1H), 2.61–2.81 (m, 1H), 1.09–1.30 (m, 3H).
(S)-4-[(S)-6-(4-Bromo-3-fluoro-phenyl)-5-ethoxycarbonyl-2-thiazol-2-yl-3,6-dihydro-
pyrimidin-4-ylmethyl]-morpholine-3-carboxylic acid (12). MS: calc’d 553 (MH+), exp 553
(MH+). 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.01 (d, J = 3.3 Hz, 1H), 7.83 (d, J = 3.0 Hz, 1H),
7.59 (dd, J = 7.3, 8.3 Hz, 1H), 7.22 (dd, J = 2.0, 9.8 Hz, 1H), 7.15 (dd, J = 2.0, 8.3 Hz, 1H), 5.73
(s, 1H), 4.53 (d, J = 16.8 Hz, 1H), 4.27–4.01 (m, 5H), 3.98–3.80 (m, 2H), 3.67 (br. s, 1H), 3.36
(d, J = 6.0 Hz, 1H), 2.73 (br. s., 1H), 1.24 (t, J = 7.2 Hz, 3H).
(S)-4-[(S)-6-(4-Chloro-3-fluoro-phenyl)-5-ethoxycarbonyl-2-thiazol-2-yl-3,6-dihydro-
pyrimidin-4-ylmethyl]-morpholine-3-carboxylic acid (13). MS: calc’d (MH+) 509, measured
(MH+) 509. 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.06-8.04 (m, 1H), 7.97–7.95 (m, 1H), 7.46–
7.42 (m, 1H), 7.25–7.20 (m, 2H), 5.73 (s, 1H), 4.25–4.09 (m, 5H), 3.95–3.81 (m, 3H), 3.75–3.61
(m, 3H), 1.25 (t, J = 7.2 Hz, 3H).
(3S)-4-[[(4S)-5-ethoxycarbonyl-4-(3-fluoro-4-methyl-phenyl)-2-thiazol-2-yl-1,4-
ACS Paragon Plus Environment

Page 25 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
dihydropyrimidin-6-yl]methyl]morpholine-3-carboxylic acid (14). MS: calc’d (MH+) 489,
measured (MH+) 489. 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.00 (d, J = 3.0 Hz, 1H), 7.84 (d, J
= 3.3 Hz, 1H), 7.16–7.26 (m, 1H), 6.95–7.13 (m, 2H), 5.69 (s, 1H), 4.46–4.71 (m, 1H), 4.21–4.38
(m, 1H), 4.03–4.22 (m, 4H), 3.78–3.97 (m, 2H), 3.60–3.75 (m, 1H), 3.38–3.54 (m, 1H), 2.69–2.94
(m, 1H), 2.24 (d, J = 1.5 Hz, 3H), 1.24 (t, J = 7.2 Hz, 3H).
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)-4-[(S)-5-Ethoxycarbonyl-6-(3-fluoro-phenyl)-2-thiazol-2-yl-3,6-dihydro-pyrimidin-4-
ylmethyl]-morpholine-3-carboxylic acid (15). MS: calc’d 475 (MH+), exp 475 (MH+). 1H NMR
(MeOD-d4, 400 MHz): δ ppm 8.01 (d, J = 3.0 Hz, 1H), 7.83 (d, J = 3.3 Hz, 1H), 7.35 (dt, J = 6.0,
7.9 Hz, 1H), 7.22 (d, J = 7.5 Hz, 1H), 7.10 (td, J = 2.0, 9.9 Hz, 1H), 7.01 (dt, J = 2.0, 8.4 Hz, 1H),
5.74 (s, 1H), 4.53 (d, J = 14.1 Hz, 1H), 4.34–4.02 (m, 5H), 3.98–3.80 (m, 2H), 3.66 (br. s., 1H),
3.47–3.36 (m, 1H), 2.77 (br. s., 1H), 1.24 (t, J = 7.2 Hz, 3H).
(3S)-4-[[(4S)-4-(3,4-difluorophenyl)-5-methoxycarbonyl-2-thiazol-2-yl-1,4-
dihydropyrimidin-6-yl]methyl]morpholine-3-carboxylic acid (16). MS: calc’d 499 (MH+), exp
499 (MH+). 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.57 (s, 1H), 7.99 (d, J = 3.3 Hz, 1H), 7.75
(d, J = 3.1 Hz, 1H), 7.1–7.3 (m, 3H), 5.68 (s, 1H), 3.9–4.1 (m, 3H), 3.7–3.9 (m, 3H), 3.68 (s, 3H),
3.24 (dd, J = 3.5, 8.3 Hz, 1H), 2.97 (br d, J = 11.8 Hz, 1H), 2.48 (ddd, J = 3.3, 8.9, 11.8 Hz, 1H).
(2S)-1-[[(4S)-4-(3,4-difluorophenyl)-5-methoxycarbonyl-2-thiazol-2-yl-1,4-
dihydropyrimidin-6-yl]methyl]-4,4-difluoro-pyrrolidine-2-carboxylic acid (17). MS: calc’d
479 (MH+), exp 479 (MH+). 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.07 (d, J = 3.0 Hz, 1H), 7.98
(d, J = 3.0 Hz, 1H), 7.2–7.4 (m, 3H), 5.76 (s, 1H), 4.45 (d, J = 16.3 Hz, 1H), 4.22 (d, J = 16.3 Hz,
1H), 4.04 (t, J = 8.2 Hz, 1H), 3.71 (s, 3H), 3.6–3.7 (m, 1H), 3.2–3.3 (m, 1H), 2.84 (ddt, J = 8.9,
11.5, 14.9 Hz, 1H), 2.5–2.6 (m, 1H).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 35
1
2
3
4
5
6
7
8
9
(3S)-4-[[(4S)-4-(3-fluoro-4-methyl-phenyl)-5-methoxycarbonyl-2-thiazol-2-yl-1,4-
dihydropyrimidin-6-yl]methyl]morpholine-3-carboxylic acid (18). MS: calc’d (MH+) 475,
measured (MH+) 475. 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.00 (d, J = 3.0 Hz, 1H), 7.84 (d, J
= 3.3 Hz, 1H), 7.15–7.26 (m, 1H), 6.97-7.12 (m, 2H), 5.69 (s, 1H), 4.47–4.67 (m, 1H), 4.23–4.34
(m, 1H), 4.01–4.19 (m, 2H), 3.81–3.98 (m, 2H), 3.64–3.73 (m, 1H), 3.40–3.54 (m, 1H), 2.71–2.94
(m, 1H), 2.24 (d, J = 1.5 Hz, 3H), 1.24 (t, J = 7.2 Hz, 3H).
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
(3S)-4-[[(4S)-4-(4-bromo-3-fluoro-phenyl)-5-methoxycarbonyl-2-thiazol-2-yl-1,4-
dihydropyrimidin-6-yl]methyl]morpholine-3-carboxylic acid (20). MS: calc’d (MH+) 539,
measured (MH+) 539. 1H NMR (MeOD-d4, 400 MHz): δ ppm 8.01 (d, J = 3.3 Hz, 1H), 7.84 (d, J
= 3.0 Hz, 1H), 7.59 (dd, J = 8.3, 7.3 Hz, 1H), 7.09–7.28 (m, 2H), 5.73 (s, 1H), 4.44–4.70 (m, 1H),
4.02–4.30 (m, 3H), 3.77–3.96 (m, 2H), 3.71 (s, 3H), 3.36–3.47 (m, 2H), 2.64–2.91 (m, 1H).
Cells and culture conditions: HepG2.2.15 and HepDE19 are stably-transfected cell lines
containing the HBV genome. Both cell lines are derived from the hepatoblastoma cell line Hep G2
(American Type Culture Collection, ATCC® HB-8065™) by the published procedures.14, 15 Both
cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM)-F12 medium
supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin, and
0.5 mg/mL of G418. While HepG2.2.15 cells constitutively support HBV replication and
production of virus particles, HepDE19 cells are inducible by tetracycline. Addition of 1µg/mL
tetracycline in culture medium suppresses HBV replication in HepDE19 cells, whereas switching
to tetracycline-free medium resumes this process.
Anti-HBV activity in vitro: HepG2.2.15 cells were seeded into 96-well plates (3 × 104 cells in
100 µL media per well) and incubated overnight at 37 ºC. The test compounds were serially half-
log diluted in DMSO, then diluted 100 times in culture media. 100 µL of diluted compounds were
ACS Paragon Plus Environment

Page 27 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
added into the plates to reach 0.5% final concentration of DMSO in every well. Five days after
compound treatment, culture supernatant was collected for further analysis. For quantitative PCR
detection of extracellular HBV DNA, 100 µL of culture supernatant was collected and processed
in MagNA Pure 96 Nucleic Acid Purification System (Roche Applied Science) for viral DNA
extraction. The extracted samples were subjected to HBV DNA quantification by qPCR. The
effective compound concentration at which HBV replication is inhibited by 50% (EC50) was
determined.
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
Cytotoxicity assay: HepDE19 cells were seeded into 96-well plates (5 × 103 cells per well) and
treated with compounds for CC50 determination. Five days after treatment, cell viability was
measured by addition of 20 µL of CCK-8 reagent. Two hours after incubation at 37°C, the
absorbance at wavelengths of 450 nm and 630 nm (OD450 and OD630) was recorded by a plate
reader. The concentration results in the death of 50% of the host cells (CC50) of each compound
were determined.
LYSA Solubility Assay. Samples are prepared in duplicate from 10 mM DMSO stock solution.
After evaporation of DMSO with a centrifugal vacuum evaporator, compounds are dissolved in a
0.05 M phosphate buffer (pH 6.5), stirred for one hr and shaken for two hrs. After one night, the
solutions are filtered using a microtiter filter plate. Then the filtrate and its 1/10 dilution are
analyzed by HPLC-UV. In addition, a four-point calibration curve is prepared from the 10 mM
stock solutions and used for the solubility determination of the compounds. The results are in
µg/mL. In case the percentage of sample measured in solution after evaporation divided by the
calculated maximum of sample amount is bigger than 80%, the solubility is reported as bigger than
this value.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 35
1
2
3
4
5
6
7
8
9
Hepatocytes stability assay. Cryopreserved hepatocytes suspension was diluted to 2×106 viable
cells/mL for human and 1×106 viable cells/mL for mouse in the incubation medium and transferred
50 μL into each well of 96-well plates. These plates were incubated at 37 ℃, 5% CO2 with shaking
at 900 rpm in the incubator. After 30-min pre-incubation, reactions were initiated by addition of
50μL of pre-warmed incubation medium containing 2 μM test compound to obtain a final
compound concentration of 1 μM and a final cell density of 1×106 cells/mL for human and 0.5×106
cells/ml for mouse. After 2, 10, 20, 40, 60 and 120 minutes, 200 µL of ice cold acetonitrile
containing an internal standard was added to terminate the reaction. Following precipitation and
centrifugation, the amount of compound remaining in the samples were determined by LC-MS/MS.
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
Parallel Artificial Membrane Permeability Assay. The permeation of drugs is measured using
a “sandwich” construction. Sample stock solution (DMSO) was diluted to 150 µM (DMSO%<2%)
with donor buffer, and after the filtration the solution was added into donor plate. An acceptor
plate coated with phospholipids (membrane) was placed onto the donor plate containing drug
solution. Finally, the upper plate was filled with buffer solution. The donor concentration was
determined at t-start as reference concentration. The drug concentrations in acceptor and donor
plate were detected at t-end (~ 18 hrs). Donor buffer is 0.05 M MOPSO buffer at pH 6.5 with 0.5%
(w/v) glyco cholic acid, acceptor buffer is 0.05 M MOPSO buffer at pH 6.5, and the membrane
is 10% (w/v) egg lecithin with 0.5% (w/v) cholesterol in dodecane.
Plasma Protein Binding Assay. The determination of unbound compound is performed using a
96-well format equilibrium dialysis (HTDialysis, Gales Ferry, Connecticut, USA) with a
molecular weight cutoff membrane of 12 to 14 kDa (HTDialysis, Gales Ferry, Connecticut, USA)
and Diazepam used as positive controls. Pooled human and mouse plasma were purchased from
Biopredic (Rennes, France). The equilibrium dialysis device itself is made of Teflon to minimize
ACS Paragon Plus Environment

Page 29 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
non-specific binding of the test substance. For the HTD device, the dialysis membranes are
conditioned as recommended by the supplier and the dialysis plate is assembled by placing
conditioned membranes between rows of half-wells and tightly clamping the assembled apparatus.
Compounds are tested in cassettes of 2-5 with an initial total concentration of 1000 nM, one of the
cassette compound being the positive control. By determining the unbound fraction values for the
positive control in each well, the integrity of membranes is tested. Equal volumes of matrix
samples containing substances and blank dialysis buffer (Soerensen buffer at pH 7.4) are loaded
into the donor and acceptor compartment respectively. The HTD dialysis block is sealed and kept
for 5 hrs at a temperature of 37 °C and 5% CO2 environment in an incubator. At the end of dialysis,
the plasma and buffer samples were retrieved and the drug concentrations quantified by LC-
MS/MS.
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
Pharmacokinetic (PK) Analysis. Compound was evaluated in female BALB/c mice (16–22 g. 12
mice in IV group were used for blood collection and 21 mice in PO group were used for blood and
liver collection) at IV 1 mg/kg and PO 3 mg/kg. Compound was dissolved in 5% DMSO, 40%
PEG400, 55% saline for i.v. dose and 2% Klucel, Tween 80, 0.1%Parabens in water for oral dose.
Blood samples were collected at each time point and placed into tubes containing EDTA-K2 and
centrifuged at 2000 g for 5 minutes at 4 °C to separate plasma from the samples. Following
centrifugation, the resulting plasma was transferred to clean tubes and stored frozen in dry ice and
then transferred into -80℃ fridge pending bioanalysis. Immediately after the collection of blood,
liver samples of each animal in PO group will be harvested, rinsed with cold saline, dried with
filter paper, and placed into a tube and weighed. Liver samples will be snap frozen in dry ice and
then stored at -80 ºC until bioanalysis. Plasma concentrations and liver concentration were
ACS Paragon Plus Environment