Discovery and Pre-Clinical Characterization of Third-Generation 4-H Heteroaryldihydropyrimidine (HAP) Analogues as Hepatitis B Virus (HBV) Capsid Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.7b00083

Described herein are the discovery and structure-activity relationship (SAR) studies of the third-generation 4-H heteroaryldihydropyrimidines (4-H HAPs) featuring the introduction of a C6 carboxyl group as novel HBV capsid inhibitors. This new series of 4-H HAPs showed improved anti-HBV activity and better drug-like properties compared to the first- and second-generation 4-H HAPs. X-ray crystallographic study of analogue 12 (HAP_R01) with Cp149 Y132A mutant hexamer clearly elucidated the role of C6 carboxyl group played for the increased binding affinity, which formed strong hydrogen bonding interactions with capsid protein and coordinated waters. The representative analogue 10 (HAP_R10) was extensively characterized in vitro (ADMET) and in vivo (mouse PK and PD) and subsequently selected for further development as oral anti-HBV infection agent.

Full text of DOI:10.1021/acs.jmedchem.7b00083

Subscriber access provided by University of Newcastle, Australia
Article
rd
Discovery and Pre-Clinical Characterization of a 3
Generation 4-H HeteroAryldihydroPyrimidine (HAP)
Analogues as Hepatitis B Virus (HBV) Capsid Inhibitors.
Zongxing Qiu, Xianfeng Lin, Weixing Zhang, Mingwei Zhou, Lei Guo, Buelent Kocer, Guolong Wu,
Zhisen Zhang, Haixia Liu, Houguang Shi, Buyu Kou, Taishan Hu, Yimin Hu, Mengwei Huang, S.
Frank Yan, Zhiheng Xu, Zheng Zhou, Ning Qin, Yue Fen Wang, Shuang Ren, Hongxia Qiu, Yuxia
Zhang, Yi Zhang, Xiaoyue Wu, Kai Sun, Sheng Zhong, Jianxun Xie, Giorgio Ottaviani, Yuan Zhou,
Lina Zhu, Xiaojun Tian, Liping Shi, Fang Shen, Yi Mao, Xue Zhou, Lu Gao, John A.T. Young,
Jim Zhen Wu, Guang Yang, Alexander V Mayweg, Hong C Shen, Guozhi Tang, and Wei Zhu
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00083 • Publication Date (Web): 24 Mar 2017
Downloaded from http://pubs.acs.org on March 25, 2017
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Xie, Jianxun; Roche Pharma Research and Early Development, Roche
Innovation Center Shanghai
Ottaviani, Giorgio; Roche Pharma Research and Early Development, Roche
Innovation Center Shanghai
Zhou, Yuan; Roche Pharma Research and Early Development, Roche
Innovation Center Shanghai
Zhu, Lina; Roche Pharma Research and Early Development, Roche
Innovation Center Shanghai
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
Tian, Xiaojun; Roche Pharma Research and Early Development, Roche
Innovation Center Shanghai
Shi, Liping; Roche Pharma Research and Early Development
Shen, Fang; Roche Pharma Research and Early Development
Mao, Yi; Roche Pharma Research and Early Development
Zhou, Xue; Roche Pharma Research and Early Development
Gao, Lu; Roche Pharma Research and Early Development, Roche
Innovation Center Shanghai
Young, John; Roche Pharma Research & Early Development, Roche
Innovation Center Basel ,
Wu, Jim; Ark Biosciences
Yang, Guang; Roche Pharma Research and Early Development
Mayweg, Alexander; Versant Ventures,
Shen, Hong; Roche R&D Center China, Medicinal chemistry
Tang, Guozhi; Ascentage Pharma,
Zhu, Wei; Senior Scientist, Medicinal Chemistry Roche Pharma Research
and Early Development, Roche Innovation Center Shanghai
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
rd
Discovery and PreꢀClinical Characterization of a 3
Generation 4ꢀH HeteroAryldihydroPyrimidine (HAP)
Analogues as Hepatitis B Virus (HBV) Capsid
Inhibitors
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ǂ,‡
ǂ,‡
ǂ,‡
ǂ,‡
ǂ,‡
ǂ,‡
Zongxing Qiu, Xianfeng Lin, Weixing Zhang, Mingwei Zhou, Lei Guo, Buelent Kocer,
ǂ,‡
ǂ,‡
ǂ,‡
ǂ,‡
ǂ,‡
ǂ,‡
ǂ,‡
Guolong Wu, Zhisen Zhang, Haixia Liu, Houguang Shi, Buyu Kou, Taishan Hu, Yimin Hu,
ǂ,‡
ǂ,†
ǂ,†
ǂ,†
ǂ,†
ǂ,ǁ
Mengwei Huang, S. Frank Yan, Zhiheng Xu, Zheng Zhou, Ning Qin, Yue Fen Wang, Shuang
ǂ,ǁ
ǂ,ǁ
ǂ,ǁ
ǂ,ǁ
ǂ,ǁ
ǂ,ǁ
ǂ,ǁ
ǂ,ǁ
Ren, Hongxia Qiu, Yuxia Zhang, Yi Zhang, Xiaoyue Wu, Kai Sun, Sheng Zhong, Jianxun Xie,
ǂ,ǁ
ǂ,§
ǂ,§
ǂ,§
ǂ,§
ǂ,§
ǂ,§
Giorgio Ottaviani, Yuan Zhou, Lina Zhu, Xiaojun Tian, Liping Shi, Fang Shen, Yi Mao, Xue
ǂ,§
ǂ,§
&,§
ǂ,§
ǂ,§
ǂ,‡
Zhou, Lu Gao, John A.T. Young, Jim Zhen Wu, Guang Yang, Alexander V. Mayweg, Hong C.
ǂ,‡,*
ǂ,‡,*
ǂ,‡,*
Shen,
Guozhi Tang,
and Wei Zhu
ǂ
&
Roche Pharma Research and Early Development, Roche Innovation Center Shanghai, Roche Innovation
‡
†
ǁ
§
Center Basel, Medicinal Chemistry, Chemical Biology, Pharmaceutical Sciences, Discovery Virology,
Bld 5, 720 Cailun Road, Shanghai, 201203 China
rd
Abstract—Described herein are the discovery and structureꢀactivity relationship (SAR) studies of the 3
generation 4ꢀH heteroaryldihydropyrimidines (4ꢀH HAPs) featuring the introduction of a C6 carboxyl
group as novel HBV capsid inhibitors. This new series of 4ꢀH HAPs showed improved antiꢀHBV activity
st
nd
and better drugꢀlike properties compared to the 1 and 2 generation 4ꢀH HAPs. Xꢀray crystallographic
study of analogue 12 (HAP_R01) with Cp149 Y132A mutant hexamer clearly elucidated the role of C6
carboxyl group played for the increased binding affinity, which formed strong hydrogen bonding
ACS Paragon Plus Environment

Page 3 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
interactions with capsid protein and coordinated waters. The representative analogue 10 (HAP_R10) was
extensively characterized in vitro (ADMET) and in vivo (mouse PK and PD), and subsequently selected
for further development as oral antiꢀHBV infection agent.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Keywords— hepatitis B virus (HBV), capsid inhibitor, heteroaryldihydropyrimidine (HAP), carboxylic
acid, hydrodynamic injection (HDI) mouse model and CYP3A4 induction.
Introduction
Hepatitis B virus (HBV) infection is a major public health problem. Currently, it is estimated that
240 million people worldwide are chronically infected with HBV. Approximately oneꢀthird of these
individuals will die from the end stage of liver diseases caused by HBV infection, such as cirrhosis,
1
, 2
hepatocellular carcinoma and liver failure, if left untreated.
Current therapies include nucleos(t)ide
analogues (lamivudine, adefovir, tenofovir, telbivudine and entecavir) and interferon alpha (INFꢀα,
including nonꢀpegylated and pegylated). Although both therapies can reduce HBV DNA and normalize
liver enzymes, neither treatment offers a high rate of clinical cure as defined by the loss of HBV surface
2,
3
antigen (HBsAg) (with or without seroconversion).
In addition, IFNꢀbased therapies are poorly
tolerated; while nucleos(t)ide analogues frequently require prolonged or possibly lifeꢀlong treatment, and
some are associated with a high risk of viral resistance. As such, there remains a tremendous unmet
medical need to discover and develop efficacious and safe antiꢀHBV agents with novel mechanisms of
4, 5
action to improve disease cure rate.
HBV capsid is formed by the assembly of HBV core proteins and it protects the enclosed viral
genome and creates the environment for reverse transcription of preꢀgenomic RNA (pgRNA) to DNA.
The assembly and disassembly of HBV capsids are essential steps in the HBV life cycle and interruption
6
of the normal capsid assembly process will block the HBV replication. There are several classes of HBV
capsid assembly effectors reported to date with two distinct mechanism of actions. Most of the known
7
8
capsid assembly effectors including compound 1 (ATꢀ130) , 2 (DVRꢀ23) and NVR 3ꢀ778 (structure
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
9
unknown) act as HBV pgRNA encapsidation blockers by accelerating the assembly of normalꢀsized but
10,11
empty capsids.
By contrast, the heteroaryldihydropyrimidine (HAP) compounds represented by
6
compound 3 (Bay 41ꢀ4109) is a unique chemical class of capsid assembly inhibitor, which can prevent
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the normal assembly of core proteins and lead to an aberrant formation of capsid. The misassembled
6
, 12
capsid will be subsequently degraded by host proteasomes.
Thus, the misassembly and depletion of
HBV core proteins result in the inhibition of HBV replication. Moreover, it is hypothesized that the
13
depletion of core proteins may suppress the function of viral covalently closed circular DNA (cccDNA)
1
4, 15
and help restore the innate immune response.
HAPs thus represent a promising chemical class of antiꢀ
HBV agents with the potential to be the first new direct acting antiꢀHBV agent since the launch of
nucleosides in 1998. However, the current HAPs have not been optimized in terms of drugꢀlike properties
for oral antiꢀHBV treatment and there is a significant need existing for the discovery of new HAPs with
improved antiꢀHBV activity, pharmacokinetic and safety profiles for clinical use.
In this article, we describe the design and discovery of a new generation of 4ꢀH HAPs featuring
the introduction of a C6 carboxyl group with improved antiꢀHBV activity and better drugꢀlike properties.
Furthermore, the elaboration of this chemical series culminates in a potential clinical candidate selection.
ACS Paragon Plus Environment

Page 5 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
Br
O
F
N
O
F
O
F
O
H
N
HN
S
N
H
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
O
2
(DVR-23)
1 (AT-130)
F
F
O
Br
2
H
N
R
O
H
Cl
N
O
4
1
O
N
F
N
H
R
6
2
N
O
N
H
F
1
2
4
4
4
a (Example 31, WO 01/68640): R = 3,5-difluoro-2-pyridyl, R = Me
3
(Bay 41-4109)
1
2
b (GLS-4): R = thiazol-2-yl, R = Et
1
2
c (NVR-010-001-E2): R = thiazol-2-yl, R = Me
1st generation 4-H HAP
2nd generation 4-H HAP
Figure 1. Chemical structures of representative HBV capsid assembly effectors.
Results and Discussion
st
Design and SAR. Compound 3 is the prototype of the 1 generation HAPꢀbased capsid assembly
inhibitors. It showed moderate activity against HBV genome replication in HepG2.2.15 cells with EC50
6
values in the range of 0.05–0.08 µM (internal and literature data ). Compound 3 also demonstrated in vivo
1
6
17
antiꢀHBV efficacy in HBV transgenic mice and AlbꢀuPA/SCID mice. Further development of 3 was
not pursued with undisclosed reason. However, it was reported that 3 could induce hepatotoxicity in rats
1
8
at high doses. We identified that 3 has an unfavorable cytotoxicity profile with a CC50 of 13 µM in
HepDE19 cells. The C2 thiazolꢀ2ꢀyl HAP analogue 5 (Table 1), however, could lead to an improved CC50
and maintain a similar EC50 to that of 3. Subsequently, the 4ꢀH HAP series of capsid inhibitors were
nd
evolved with the discovery of more potent 2 generation analogues featuring the introduction of C6
1
9
20, 21
morpholinyl group, which included 4a (Example 31, WO 01/68640) , 4b (GLSꢀ4),
4c (NVRꢀ010ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
22
0
01ꢀE2) and an internal reference compound 6 (Table 1). Presumably, the C6 morpholinyl group brings
22
additional interactions with HBV capsid proteins which in turn contributes to the improved antiꢀHBV
nd
activity as observed for analogue 6 compared to 5. The 2 generation 4ꢀH HAP 6 also demonstrated a
st
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
more favorable cytotoxicity profile (CC of 6 is > 100 µM) than that of the 1 generation 4ꢀH HAPs 3
50
and 5. However, compound 6 showed high liver microsome clearance. It also demonstrated a strong
hERG inhibition with an IC of 0.48 µM, which was likely ascribed to dual effects of both the moderate
20
23
basicity of morpholine “N” (pKa = 6.3) and the high lipophilicity of 6 (logP = 3.2) as documented. In
rd
our design of the 3 generation 4ꢀH HAPs, we would like to discover more potent and drugꢀlike
nd
analogues by reducing both the basicity and lipophilicity of the 2 generation 4ꢀH HAPs through
introduction of a polar substituent such as hydroxyl, sulfonamide, carboxamide or carboxyl. However, it
remains unclear whether a polar substituent as described above on 4ꢀH HAPs can be tolerated in the
2
2, 24
largely hydrophobic binding pocket located at the interface of capsid dimers.
It turned out that polar substituents such as carboxyl are not tolerated on the C2 thiazolyl, C4
phenyl or C5 position of 4ꢀH HAPs as revealed by a dramatic potency loss for analogues 7–9 compared to
5
(additional examples in Tables 2 and 3). To our delight, the introduction of a carboxyl group at the αꢀ
position of C6 morpholine was well tolerated. Both the (S)ꢀ and (R)ꢀmorpholineꢀ3ꢀcarboxylic acid
analogues 10 and 11 demonstrated good antiꢀHBV activity. The (S)ꢀmorpholineꢀ3ꢀcarboxylic acid
analogue 10 was 5ꢀfold more potent than 6, with a mean EC50 value of 3 nM in HepG2.2.15 cells. The
tolerance of a small, lipophilic substituent on C6 morpholine of HAP can be rationalized by the analysis
2
2
of the recently disclosed coꢀcrystal structure (5E0I) of Cp149 Y132A with 4C. However, to our pleasant
surprise, this negatively charged carboxyl on C6 morpholine was tolerated and even brought increased
binding affinity to HBV capsid as reflected by the improved antiꢀHBV activity of 10 compared to 6.
The introduction of an ionizable carboxyl group significantly reduced the lipophilicity of
analogue 10 (logP = 0.3) and increased aqueous solubility (LYSA > 666 µg/mL). The CC of 10 was >
5
0
1
00 µM in HepDE19 cells and the selectivity index (SI, defined by CC /EC ) was > 33333. In the
50 50
ACS Paragon Plus Environment

Page 7 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
preliminary ADME (absorption, distribution, metabolism and excretion) and in vitro safety assessment,
compound 10 showed significantly improved human/mouse liver microsome stability and no sign of
st
nd
hERG inhibition (IC20 > 10 µM) compared to that of the representative 1 and the 2 generation 4ꢀH
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
HAP analogues likely as a result of both reduced lipophilicity and basicity (pKa is now 5.8). Moreover,
ꢀ
6
compound 10 showed a moderate to high permeability (1.4 x 10 cm/s) in PAMPA (parallel artificial
membrane permeability assay) test despite an apparently zwitterionic structure.
nd
During the optimization of the 2 generation 4ꢀH HAPs (unpublished results), we identified gemꢀ
difluoropyrrolidine as a good bioisostere of morpholine with the corresponding analogue showing
comparable antiꢀHBV activity (EC50 is ca.10 nM) to that of 6 in HepG2.2.15 cells. Recently, we have
published the 4ꢀMe HAP series where a C6 (2S)ꢀ4,4ꢀdifluoroproline analogue showed the best antiꢀHBV
25
activity within the series. In parallel with the 4ꢀMe HAP work, we also synthesized 4ꢀH HAP analogue
1
2 with a C6 (2S)ꢀ4,4ꢀdifluoroproline moiety. Compared to 10, compound 12 showed equally potent antiꢀ
HBV activity but a less favorable CC50 and microsome stability. As a result of that, in the further
rd
exploration of the 3 generation 4ꢀH HAPs, we focused on the C6 morpholine carboxylic acid analogues.
Further SAR studies around 10 revealed that both the substitution position and chain length of the
carboxyl group are crucial for antiꢀHBV activity. The corresponding morpholineꢀ2ꢀcarboxylic acid
analogue 13 was much less active than 10 in HepG2.2.15 cells. The extended carboxylic acid analogues
1
4 and 15 also demonstrated remarkably reduced antiꢀHBV activity. Additional substituents like methyl
or isopropyl can be tolerated on the C6 morpholineꢀ3ꢀcarboxylic acid group but the antiꢀHBV activity is
sensitive to both regioꢀ and stereochemistry as only analogues 17 and 18 show comparable potency to that
of 10 among the analogues explored. The αꢀmethyl analogue 16 was less active than 10 with an EC50
value of 0.02 µM. Further derivatization of carboxylic acid of 10 gave ester 19, carboxamide 20, alcohol
2
1 and sulfonamide 22, all of which maintained high antiꢀHBV activity. However, those analogues all
showed high microsomal clearance despite the lower lipophilicity than that of 6 as reflected by their logP
values. This observation highlighted the unique profile of the carboxyl group for the improved drugꢀlike
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
properties of 10. Meanwhile, bioisostere replacements of the C6 morpholineꢀ3ꢀcarboxylic acid were
explored as well. Among all the attempts, the piperidine, piperazine or Nꢀacetylpiperazine analogue 23–
25 all led to a total loss of antiꢀHBV activity. The thiomorpholine analogue 26, however, maintained a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
moderate EC value of 0.038 µM.
50
Table 1: Discovery of C6 morpholine carboxylic acid 4ꢀH HAP analogues.
and
8
9
7
d
e
EC50
CC50
HLM/MLM
LYSA
c
f
ID
X
SI
logP
a
b
−1
−1
(µM)
0.05ꢀ0.08
0.08
(µM)
13
(mL min kg ) (µg/mL)
3
5
H
H
260
6/30
38
16
4.5
4.4
83
1038
19/89
6
0.016
> 100
> 6250
21/88
195
3.2
g
7
8
9
ꢀ
ꢀ
ꢀ
> 1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
g
> 1
g
> 1
1
0
0.003
> 100
> 33333
5/10
> 666
0.3
ACS Paragon Plus Environment

Page 9 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
0.079
0.003
> 100
65
> 1266
28333
ꢀ
ꢀ
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
12
4.4/43
> 644
1.4
N
O
1
3
0.333
0.60
> 100
> 100
> 300
> 167
2/3
> 592
> 638
ꢀ
ꢀ
HO
O
14
0.1/25
1
5
0.15
> 100
> 667
0/12
> 697
ꢀ
1
1
6
7
0.02
> 100
> 100
> 5000
1.4/50
5.4/32
600
0.8
0.7
0.007
> 14286
> 667
1
8
0.005
81
16200
3/13
ꢀ
1.0
1
2
9
0
0.004
0.003
ꢀ
ꢀ
22/88
18/82
<1.0
205
2.6
1.5
> 100
> 33333
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
2
1
2
3
0.005
0.004
74
88
ꢀ
14800
22000
ꢀ
21/88
22/89
ꢀ
475
48
ꢀ
2.1
2.1
2.1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
N
g
HO
> 1
g
24
25
> 1
ꢀ
ꢀ
ꢀ
ꢀ
0.1
g
> 1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ0.2
1.2
2
6
0.038
60
1579
1.6/34
584
a
EC50 is the mean value for reduction of HBV DNA by 50% in HepG.2.2.15 cells. Experiments were run in
b
duplicate, with variation < 15%. CC50 is the mean value for cytotoxicity to HepDE19 cells. Experiments were in
c
d
duplicate runs with variation < 15%. Selectivity index (SI) is the ratio of CC /EC . Predicted hepatic clearance
50
50
−1
−1
e
obtained with human liver microsome (HLM) and mouse liver microsome (MLM) (mL min kg ). Lyophilisation
f
solubility assay (LYSA) (µg/mL). logP, predicted partition coefficient between the octanol phase and the water
g
phase (KowlogP). The highest concentration tested.
Compound 10 appeared to be a promising candidate for further ADME and in vitro safety
rd
assessment. In the meantime, we performed thorough SAR studies on C2, C4 and C5 positions of the 3
generation 4ꢀH HAPs (see structure of 3 for numbering convention) in order to identify a set of
contenders of compound 10 for comparisons in further ADME and safety assessments. The results are
shown in Tables 2 and 3.
ACS Paragon Plus Environment

Page 11 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We first explored the SAR at the C5 position of 4ꢀH HAP. The ethyl ester analogue 27 afforded a
similar or slightly better EC than that of analogue 10. However, compound 27 was prone to metabolism
50
ꢀ1
ꢀ1
in mouse liver microsome with a clearance of 55 mL min kg . It was found that the size of C5 ester
group had a significant impact on the antiꢀHBV activity as the nꢀpropyl ester analogue 28, isoꢀpropyl
ester analogue 29 and trifluoroethyl ester analogue 30 all showed reduced antiꢀHBV activity compared to
that of 10 and 27. The bulky tertꢀbutyl ester analogue 31 had a much reduced antiꢀHBV activity. The
polar C5 substituents (reflected by their logP values) including carboxylic acid (9 in Table 1),
methoxyethyl ester (32), amide (33) and nitrile (34) were not tolerated with a dramatic drop of antiꢀHBV
activities, which indicated a lipophilic environment at this binding position of capsid. Overall, the methyl
ester and ethyl ester were well tolerated in terms of antiꢀHBV activity at the C5 position of HAP.
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 C2 substituent of the 4ꢀH HAP series was indicated as an important moiety for capsid
2
2,24
binding but was less explored previously.
We decided to extensively study the substitution effect and
bioisostere replacement of the C2 thiazolyl moiety. Firstly, consistent with the previous observation for 3,
difluropyridinyl was well tolerated at the C2 position and the corresponding analogue 35 showed good
antiꢀHBV activity but also a high cytotoxicity. Most of our attempts to replace the C2 thiazolyl with other
bioisosteres did not lead to appreciable antiꢀHBV activity as exemplified by C2 oxazolyl and pyrimidinyl
analogues 36 and 37, which reflect a narrow SAR at the C2 position of 4ꢀH HAP. Nevertheless, we did
identify the Nꢀmethylimidazolyl and 5ꢀmethyloxazolyl analogues 38 and 39 with moderate antiꢀHBV
activity. Meanwhile, small substituents on the C2 thiazole were well tolerated as both the methyl and
trifluoromethyl substituted thiazole analogues 40 and 42 showed comparable antiꢀHBV activity to that of
10. However, the cyclopropyl substituted thiazole analogue 41 was less active.
rd
Table 2: Summary of SAR at the C5 and C2 positions of the 3 generation 4ꢀH HAPs.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
EC50
CC50
HLM/MLM
a
a
ID
X
R
SI
logP
a
a
−1
−1
(
µM)
(µM)
(mL min kg )
1
0
7
0.003
0.002
0.012
0.017
> 100
> 33333
> 50000
> 8333
4235
5/10
0.3
0.8
1.3
1.2
2
> 100
> 100
72
5/54
2/39
0/48
b
28
2
9
0
3
0.013
27
41
2077
ꢀ
0/21
7/78
1.2
1.7
b
c
31
> 0.1
32
0.19
> 100
> 526
6/24
0.05
c
3
3
4
> 1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ0.3
ꢀ0.2
c
3
> 1
ACS Paragon Plus Environment

Page 13 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
5
6
0.008
23
2875
6.3/23
0.5
ꢀ0.5
ꢀ0.7
ꢀ0.1
0.1
c
3
> 1
> 100
> 100
82
ꢀ
ꢀ
ꢀ
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
c
37
> 1
ꢀ
3
8
9
0.042
0.014
1952
> 7143
0.4/3.8
0.2/5.4
3
> 100
4
0
1
0.007
0.022
> 100
> 14826
0.7/46
6.9/73
0.9
1.7
4
ꢀ
ꢀ
42
0.003
> 100
> 33333
3/15
1.3
a
b
c
Definitions as the same as that described before. Diasteroisomer mixture at C4 position. The highest
concentration tested.
26
During the SAR work, the collective information gained from inꢀhouse photoaffinity study and
24
the reported 5Å coꢀcrystal structure (2G34) indicated that the C4 position of 4ꢀH HAP is located in a
conservative, largely hydrophobic pocket providing important interactions with capsid. As such, the SAR
work at the C4 position of HAP became the chemistry focus in order to discover more potent analogues.
The results are summarized in Table 3.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Firstly, bioisosteres of C4 phenyl, such as cyclohexyl, thiophenyl (substituted or unsubstituted) or
pyridyl (substituted or unsubstituted) were not tolerated with a total loss of antiꢀHBV activity for the
respective analogues. Replacement of the C4 phenyl with small substituent such as methyl or cyclopropyl
was not successful either. The SAR exploration was thus concentrated on the substituent fineꢀtunings on
the C4 phenyl as shown in Table 3.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
rd
Table 3: Summary of SAR at the C4 position of 3 generation 4ꢀH HAPs.
4
1
2
a
a
a
a
ID
R
R
EC50 (µM)
CC50 (µM)
SI
logP
1
2
4
0
7
3
Methyl
Ethyl
Ethyl
Ethyl
Ethyl
0.003
> 100
> 33333
> 50000
4133
0.3
0.002
0.015
0.014
0.001
> 100
0.8
0.2
0.4
1.1
62
44
> 100
> 7143
72000
45
72
ACS Paragon Plus Environment

Page 15 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
6
Methyl
Ethyl
0.006
0.003
0.008
> 100
> 100
> 100
> 100
> 16667
> 33333
12500
ꢀ
0.6
0.7
0.2
0.6
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
47
48
Methyl
Methyl
b
c
4
9
> 0.1
c
5
0
Methyl
Ethyl
> 0.1
> 100
> 100
> 100
ꢀ
1.0
0.6
1.2
5
1
0.014
0.003
> 7143
> 33333
52
Ethyl
5
3
4
Methyl
Ethyl
0.011
0.001
> 100
44
> 9091
44000
0.7
1.3
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
5
Ethyl
0.003
31
50
ꢀ
1.5
1.1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
c
5
6
Methyl
> 0.1
431
c
5
7
8
Ethyl
Ethyl
> 0.1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ0.5
c
5
> 0.1
0.05
59
60
Methyl
Ethyl
Ethyl
Ethyl
Ethyl
0.004
0.0008
0.0007
> 100
> 25000
0.3
0.8
1.1
1.3
0.2
> 100
125000
6
6
6
1
2
3
95
ꢀ
135714
c
> 0.1
ꢀ
ꢀ
c
> 0.1
ꢀ
64
Methyl
0.001
> 100
> 100000
0.5
ACS Paragon Plus Environment

Page 17 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
6
5
Ethyl
Methyl
Ethyl
0.0006
0.001
29
80
48333
80000
1.0
0.8
1.3
0.4
0.9
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
66
67
0.0007
0.002
51
72857
6
8
9
Methyl
Ethyl
> 100
97
> 50000
161666
6
0.0006
a
b
c
Definitions as the same as that described before. Diasteroisomer mixture at C4 position The highest
concentration tested.
We firstly investigated the oꢀsubstituent of the C4 phenyl group. Replacement of the oꢀCl with oꢀ
H or oꢀF led to > 5ꢀfold potency drop (43 and 44 compared to 27). The oꢀBr and oꢀCH with a similar
3
27
Van Der Walls radius as oꢀCl were well tolerated and the corresponding analogues 45–48 gave similar
or slightly improved antiꢀHBV activity in HepG2.2.15 cells compared to that of compound 10 and 27.
However, larger substituents like trifluoromethyl or cyclopropyl were not tolerated with a significant
potency reduction (analogues 49 and 50). This SAR observation indicated that the oꢀCl of C4 phenyl of
compound 10 might tightly fit into the capsid binding pocket.
We subsequently explored the SAR of the pꢀsubstituent of the C4 phenyl. In general, the pꢀF of
C4 phenyl could be successfully replaced with pꢀCl, pꢀBr and pꢀCH with equal or even better antiꢀHBV
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
activity in some cases (52–55). However, the pꢀCl, pꢀBr analogues 54 and 55 showed moderate
cytotoxicity in HepDE19 cells. The pꢀH analogue 51 showed a 7ꢀfold potency drop with an EC value of
5
0
0
.014 µM. Analogue 56 bearing a bulky pꢀCF displayed an EC of > 0.1 µM. The analogues containing
3 50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
polar substituents such as pꢀOCH and pꢀCN gave significantly reduced antiꢀHBV activities (57 and 58 vs
3
51).
st
nd
In the 1 and the 2 generation HAP series, the mꢀposition of C4 phenyl has been barely explored.
We have identified that the introduction of a mꢀF substituent on the C4 phenyl could further improve the
antiꢀHBV activity with analogues 60 and 61 showing subꢀnM potency in HepG2.2.15 cells. However, mꢀ
Cl and mꢀCN substituents were not tolerated as analogues 62 and 63 showed EC s > 0.1 µM.
50
So far, the SAR studies suggest that the preferred substituents at each position of the C4 phenyl
are: (1) ortho position: –Cl, –Br, and methyl; (2) meta position: –H and –F; (3) para position: –F, –Cl, –
Br. With this SAR understanding in mind, several triꢀsubstituted C4 phenyl analogues were designed and
synthesized. As expected, all the analogues showed excellent antiꢀHBV activity, among which, analogues
6
5, 67 and 69 demonstrated subꢀnM EC values in HepG2.2.15 cells.
50
Xꢀray Crystal Structure. We have recently solved the cocrystal structure of Cp149 Y132A
2
6
mutant hexamer in complex with analogue 12 (named as HAP_R01, PDB code: 5WRE) , which nicely
elucidated why the introduction of a carboxyl group on the C6 substituent (e.g. gemꢀdifluoropyrrolidine
and morpholine) of 4ꢀH HAP can bring improved antiꢀHBV activity. As shown in Figure 2, the
carboxylic acid moiety forms strong bidentate hydrogen bonds with the backbone NH and the sidechain
hydroxyl of S141. The high resolution structure (1.95 Å) also reveals the water network including a wellꢀ
defined water molecule that coordinates the C6 carboxyl group and the N atom of C2 thiazolyl group (Fig.
rd
2) These hydrogen bonding interactions contribute to the increased binding affinity of the 3 generation
4ꢀH HAPs such as 10 and 12 as reflected by the improved antiꢀHBV activity compared to the
corresponding nonꢀcarboxylic acid analogues (10 vs 6 as an example). It was speculated that the reduced
ACS Paragon Plus Environment

Page 19 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
antiꢀHBV activity of the diastereoisomer 11, regioisomer 13 or the extended carboxylic acid analogues 14
and 15, relative to compound 10, could be ascribed to the complete or partial disruption of this Hꢀbonding
interaction network. On the other hand, the carboxylic acid derivatives 19–22 could at least partially
maintain the Hꢀbonding interactions and thus retain the good antiꢀHBV activity.
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 consistent with our SAR findings, the C2 thiazolyl, C4 phenyl and C5 methyl ester of 12 bind
to a largely hydrophobic pocket of capsid and make multiple van der waals (VDW) contacts contributing
to additional binding affinity. This explains why the introduction of a polar substituent at those positions
will lead to an antiꢀHBV activity drop.
S141
1
2 (HAP_R01)
Figure 2: Crystal structure of Cp149 Y132A in complex with 12. Compound 12 is highlighted in cyan stick. The
dimer forming the concave for the binding pocket is shown in light green. The dimer forming the cap for the binding
pocket is colored in dark green. Black dash line represents the key hydrogen bonds. PDB code 5WRE.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
rd
DMPK, in vivo efficacy and early safety assessment of the selected 3 generation 4ꢀH HAPs.
On the basis of antiꢀHBV activity, cytotoxicity, human/mouse metabolic stability and structural diversity,
analogues 10, 60 and 68 were selected for further physicochemical and ADME characterizations. The
data are summarized in Table 4.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 4: Physicochemical and ADME properties of 10, 60 and 68.
b
c
c
d
pKa
Lysa
HLM/MLM
PAMPA
PPB
a
ID
LogD
−1
−1
−6
e
(
acid/base)
1.8/5.8
(µg/mL) (mL min kg ) (10 cm/s) (human/mouse)
1
0
0.06
0.42
0.44
> 666
> 679
> 604
b
5/10
2/12
2/14
1.4
3.3
2.7
5.7/7.4
2.0/2.2
5.4/6.2
60
68
1.8/5.7
<2.5/5.7
a
c
d
Apparent partition coefficients at pH 7.4. Ionization constant. Definitions as the same as described before.
ꢀ6
e
Parallel Artificial Membrane Permeability Assay (10 cm/s). Plasma protein binding assay (% unbound fraction).
High solubility is a critical factor for the development of oral antiꢀHBV agents to enable
formulation development and maintain high plasma drug levels in vivo. It is also very important for HAP
analogues to have a moderate or low liver clearance, as this often leads to a good oral bioavailability and
rd
high systemic exposure. As shown in Table 4, the 3 generation 4ꢀH HAP analogues with an ionizable
carboxylic acid group have demonstrated high solubility and low human/mouse microsome clearance. In
the cryopreserved human and mouse hepatocytes incubation studies of these analogues, the unchanged
parent compounds are the major form and the detectable metabolites are mainly from the oxidation of C6ꢀ
morpholine ring and a phase II glucuronide conjugation of carboxylic acid. The hydrolysis of C5ꢀester is
almost neglectable, especially in human hepatocyte. Moreover, all three compounds show a medium to
high permeability determined in the PAMPA assay and appreciable free fraction in human/mouse plasma
as reflected by their PPB (plasma protein binding) values. Based on the favorable physicochemical (logD
and pKa) and in vitro ADME properties, these compounds were tested for in vivo PK studies in mice
ACS Paragon Plus Environment

Page 21 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
following single dose intravenous (IV) and oral (PO) administrations. The results are summarized in
Table 5.
a
Table 5. Singleꢀdose pharmacokinetics of 10, 60 and 68 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
Example
t1/2 (IV, h)
V (IV, L/kg)
10
1.5
60
3.0
68
1.4
1.9
44
2.5
4.1
ss
−
1
−1
CL (IV) (mL min kg )
19
66
Plasma
C
max (PO, ꢁg/L)
413
878
962
37
783
253
624
82
283
379
344
30
AUC(0ꢀ∞) (IV, µg/L*hr)
AUC(0ꢀ∞) (PO, µg/L*hr)
F (%)
Liver
AUC(0ꢀ∞) (PO, µg/L*hr)
11800
12.3
2005
3.2
9210
27
AUC(0ꢀ∞), liver/AUC(0ꢀ∞), plasma (PO)
a
The singleꢀdose pharmacokinetics (SDPK) study of selected compounds were 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 (V ), T1/2 (i.v.), Maximal concentration (Cmax), area under the curve
ss
(AUC), and oral bioavailability (F) are reported.
4
ꢀH HAP analogues such as 10 and 60 showed good oral absorption as reflected by their
respective values of oral Cmax, AUC and F%. Analogue 68 showed relatively lower exposure as expressed
by Cmax and AUC than that of compounds 10 and 68. Interestingly, all three compounds, especially
compounds 10 and 68, demonstrated a higher liver exposure than their corresponding plasma levels which
is considered as a desirable attribute for liver targeting drugs such as HBV capsid inhibitors. The high
liverꢀtoꢀplasma exposure ratio (from 3.2 to 27) might point to a potential involvement of uptake hepatic
transporters in governing the preferable liver disposition of these compounds and it could also explain the
underestimation of the in vivo clearance from in vitro mouse microsomes.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compound 10 was the most advanced analogue and was further assessed against HBV infection
28
in the hydrodynamic injection (HDI) mouse model. In this study, female BALB/c mice were
hydrodynamically injected with replicationꢀcompetent 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 12.5 mg/kg per dose, twice daily) of compound 10
for 5 days. Plasma and liver samples were collected at the indicated time points for HBV DNA
quantification by realꢀtime quantitative polymerase chain reaction (qPCR). As shown in Figure 3, with
treatment of compound 10, HBV DNA copy numbers were rapidly and markedly reduced by about 100ꢀ
fold in plasma and were no longer detectable in liver tissues, demonstrating a strong antiꢀHBV effect of
10 upon oral administration. The maximum efficacy of 10 was comparable to the maximum efficacy of
ETV in this model.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. The levels of HBV DNA in plasma (A) and liver (B) of HBV HDI mice upon treatment with the vehicle
control, ETV control, and compound 10. The HBV DNA copy numbers were quantified by realꢀtime qPCR. Error
bars indicate standard error of the mean (6 mice per group).
Compound 10 was further profiled in an effort to identify any early safety flags and to define DDI
(
drugꢀdrug interaction) liabilities. The in vitro toxicology profiles of compound 10 were favourable with
no signs of hERG, GSH (glutathione adduction), mutagenicity (Ames assay) or clastogenicity
ACS Paragon Plus Environment

Page 23 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
micronucleus assay) liability. A 5ꢀday in vivo toxicology assessment in rats supported further
development of compound 10 that showed supportive safety 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. In the in vitro DDI assessment, compound 10 displayed IC values of >
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
50
50 µM for CYP inhibition in a panel of five enzymes (CYP3A4, 2D6, 2C9, 2C19 and 1A2). Compound
0 had no Time Dependent Inhibition (TDI) flags to CYP3A4 inhibition. However, this compound was
identified as a strong CYP3A4 inducer in primary human hepatocytes as determined by the CYP3A4
1
2
9
mRNA levels (111% increase compared to the positive control rifampicin at 10 µM). This observation
rd
of CYP3A4 induction is likely related to 4ꢀH HAP core as over 250 compounds of the 3 generation 4ꢀH
HAPs were screened and most of them demonstrated more or less CYP3A4 induction liabilities. The data
of a few selected analogues is shown in Table 6.
Table 6: Relative induction values of CYP3A4 mRNA by selected analogues
ID
10
% of Positive control (10 µM rifampicin)
111
109
81
1
1
2
2
7
7
183
72
42
6
6
0
8
144
56
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
rd
Chemistry. As exemplified by the synthesis of compound 10, the 3 generation 4ꢀH HAP
analogues described herein were mostly prepared according to the procedure shown in Scheme 1. Firstly,
amidine 10Aꢀ1 was prepared by reacting thiazoleꢀ2ꢀcarbonitrile with sodium methoxide followed by
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
treatment with NH Cl. Next, a threeꢀcomponent reaction of methyl acetoacetate, 2ꢀchloroꢀ4ꢀ
4
fluorobenzaldehyde and amidine 10Aꢀ1 was conducted to give dihydropyrimidine 10Aꢀ2. The racemic
1
0Aꢀ2 was then subjected to SFC (supercritical fluid chromatography) chiral separation to afford the
enantiomerically pure (R)ꢀ10Aꢀ2 as the desired intermediate (compound 5 in Table 1) and its antipode
S)ꢀ10Aꢀ2. The absolute stereochemistry of 5 was unambiguously determined by Xꢀray diffraction
(
3
0
study. Bromination of 5 gave 10A, which was finally reacted with (3S)ꢀmorpholineꢀ3ꢀcarboxylic acid
hydrochloride salt to provide 10 after HPLC purification. This short and concise synthetic route was
rd
successfully applied to the synthesis of the 3 generation 4ꢀH HAPs described in this paper, which greatly
facilitated and expedited our SAR studies.
Scheme 1:
ACS Paragon Plus Environment

Page 25 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
HCl
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
b
.
1
0A-1
10A-2
c
(
R)-10A-2 (5)
(S)-10A-2
.HCl
d
e
(
R)-10A-2
10A
10
Reagents and conditions: (a) NaOMe, MeOH, r.t., 3 hrs; then NH Cl, overnight, 100% (crude); (b) KOAc,
4
CF CH OH, reflux, 16 hrs, 60% for two steps from thiazoleꢀ2ꢀcarbonitrile; (c) SFC chiral separation; (d) NBS,
3
2
CCl , r.t., 1 hr, 77%; (e) DIPEA, DCM, ca. 1 hr, 55%.
4
rd
Conclusions. In summary, we have discovered the 3 generation 4ꢀH HAPs which feature the
introduction of a carboxyl group at the C6 position of 4ꢀH HAP. This carboxyl group brings an increased
st
binding affinity with capsid which leads to an improved antiꢀHBV activity comparing to that of the 1
nd
and 2 generation 4ꢀH HAPs. The high resolution xꢀray crystal structure of analogue 12 bound into
Cp149 Y132A mutant hexamer reveals that this carboxyl group facilitates the formation of a strong
hydrogen bonding interaction network between 12, coordinated waters and capsid. Besides the improved
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
st
nd
rd
antiꢀHBV activity, compared to the 1 and 2 generation 4ꢀH HAPs, the 3 generation 4ꢀH HAP
compounds represented by 10 are more drugꢀlike with an improved solubility and metabolic stability. The
favourable in vitro biological activity and ADME properties of compound 10 are well translated in vivo as
demonstrated in mice PK and PD studies, where a low clearance, good oral bioavailability, preferential
liver distribution and strong antiꢀHBV effects were demonstrated. The favourable in vitro and short term
rat tox profiles also support the nomination of 10 (named as HAP_R10) for further development.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
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 (prepꢀHPLC) on a reversedꢀphase column
using a Waters XBridge OBD Phenyl (30 mm × 100 mm, 5 ꢁm) or OBD RP18 (30 mm × 100 mm, 5 ꢁm)
column under acidic conditions (A, 0.1% formic acid in H O; B, 0.1% formic acid in acetonitrile) or basic
2
conditions (A, 0.1% ammonia in H O; B, acetonitrile). For SFC chiral separation, the intermediates were
2
separated using a chiral column (Daicel Chiralpak IC, 30 mm × 250 mm, 5 ꢁm) on a Mettler Teledo SFCꢀ
Multigram system (solvent system of 95% CO and 5% IPA (0.5% TEA in IPA), backpressure of 100 bar,
2
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 2795ꢀZQ2000). NMR spectra were obtained using Bruker AVANCE 400 MHz
1
13
1
spectrometer, operating at 400.13 MHz ( H) and 100.62 MHz ( C). H NMR spectra were
1
3
1
obtained using the single pulse zg30. C{ H} NMR spectra were obtained using composite pulse
1
13
zgpg30 for proton decoupling. The chemical shifts were referenced against internal TMS( H, C)
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,
1
LC–MS and H NMR analyses. All of the reported yields are for isolated products and are not optimized.
ACS Paragon Plus Environment

Page 27 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
rd
General Synthetic Procedures for the synthesis of 3 generation 4ꢀH HAPs with (3S)ꢀ4ꢀ[[(4R)ꢀ4ꢀ(2ꢀ
Chloroꢀ4ꢀfluoroꢀphenyl)ꢀ5ꢀmethoxycarbonylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀdihydropyrimidinꢀ6ꢀ
yl]methyl]morpholineꢀ3ꢀcarboxylic acid (10) exemplified.
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
To a stirred solution of thiazoleꢀ2ꢀcarbonitrile (1.50 g, 14 mmol) in 5 mL of dry MeOH was added
dropwise a solution of sodium methoxide (0.74 g, 14 mmol) in 10 mL of dry methanol. The reaction
mixture was stirred at room temperature until the disappearance of starting material monitored by LC/MS.
After that, ammonium chloride (1.50 g, 28 mmol) was added in one portion and the reaction mixture was
stirred overnight. The undissolved material was removed by filtration and the filtrate was concentrated in
vacuo to afford the thiazoleꢀ2ꢀcarboxamidine hydrochloride (Compound 10Aꢀ1) as a grey solid which
+
was used directly in the next step without further purification. MS: calc’d 128 (MH ), measured 128
+
(MH ).
To a stirred solution of thiazoleꢀ2ꢀcarboxamidine hydrochloride (10Aꢀ1, 0.15 g, crude, ca. 1.0 mmol),
methyl acetoacetate (0.12 g, 1.0 mmol) and 2ꢀchloroꢀ4ꢀfluorobenzaldehyde (0.16 g, 1.0 mmol) in
CF CH OH (8 mL) was added potassium acetate (0.20 g, 2.0 mmol). The reaction mixture was refluxed
3
2
for 16 hrs in a nitrogen atmosphere. 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 Na SO . The solvent was removed and the residue was purified by silica gel column
2
4
chromatography (ethyl acetate/petroleum ether: 1/4 to 1/2) to afford methyl 4ꢀ(2ꢀchloroꢀ4ꢀfluoroꢀphenyl)ꢀ
ꢀmethylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀdihydropyrimidineꢀ5ꢀcarboxylate (Compound 10Aꢀ2) as a yellow solid (219
6
+
+
1
mg, 60%). MS: calc’d (MH ) 366, measured (MH ) 366. H NMR (DMSOꢀd , 400 MHz): δ ppm 9.98 (s,
6
1H), 7.97 (d, J = 4.0 Hz, 1H), 7.90 (d, J = 4.0 Hz, 1H), 7.41 (dd, J = 8.0, 4.0 Hz, 1H), 7.35 (dd, J = 8.0,
8.0 Hz, 1H), 7.18 (td, J = 8.0, 4.0 Hz, 1H), 5.98 (s, 1H), 3.53 (s, 3H), 2.47 (s, 3H).
The racemic compound 10Aꢀ2 was subjected to SFC chiral separation. The desired (ꢀ)ꢀenantiomer methyl
(
4R)ꢀ4ꢀ(2ꢀchloroꢀ4ꢀfluoroꢀphenyl)ꢀ6ꢀmethylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀdihydropyrimidineꢀ5ꢀcarboxylate
((R)ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
0Aꢀ2) has a relatively short retention time. The absolute stereochemistry of (R)ꢀ10Aꢀ2 was determined
by Xꢀray diffraction study (see supporting information).
20
(
R)ꢀ10Aꢀ2: [α]D ꢀ55.0 (c 0.845, MeOH).
20
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(S)ꢀ10Aꢀ2: [α]D +44.6 (c 0.175, MeOH).
To a stirred solution of compound (R)ꢀ10Aꢀ2 (0.37 g, 1.0 mmol) in CCl (5 mL) was added NBS (0.20 g,
4
1
.1 mmol) in portions. After the reaction mixture was stirred at room temperature for 1 hour, the solvent
was removed in vacuo and the residue was purified by column chromatography to give methyl (4R)ꢀ6ꢀ
(
(
bromomethyl)ꢀ4ꢀ(2ꢀchloroꢀ4ꢀfluoroꢀphenyl)ꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀdihydropyrimidineꢀ5ꢀcarboxylate
+
+
compound 10A) as a yellow solid (342 mg, 77%). MS: calc’d 445 (MH ), measured 445 (MH ).
To a stirred solution of compound 10A (0.049 g, 0.11 mmol) and (3S)ꢀmorpholineꢀ3ꢀcarboxylic
acid hydrochloride salt (cas: 1187929ꢀ04ꢀ9, 0.029 g, 0.17 mmol) in dichloromethane (5 mL) was added
dropwise DIPEA (0.078 mL, 0.45 mmol). The reaction mixture was stirred at room temperature until the
disappearance of 10A. The mixture was diluted with EtOAc (10 mL), washed successively with saturated
aqueous NH Cl solution and brine. The organic layer was separated and dried over anhydrous Na SO .
4
2
4
The solvent was concentrated in vacuo and the residue was purified by prepꢀHPLC to give (3S)ꢀ4ꢀ[[(4R)ꢀ
4ꢀ(2ꢀchloroꢀ4ꢀfluoroꢀphenyl)ꢀ5ꢀmethoxycarbonylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀdihydropyrimidinꢀ6ꢀ
yl]methyl]morpholineꢀ3ꢀcarboxylic acid (compound 10) as a light yellow solid (30 mg, 55%). Melting
o
+
+
+
point: 136ꢀ137 C. MS: calc’d (MH ) 495, measured (MH ) 495. HRMS: calc’d (MH ) 495.0899,
+
1
measured (MH ) 495.0893. H NMR (DMSOꢀd , 400 MHz): δ ppm 12.88 (br. s, 1H), 9.87 (s, 1H), 8.04
6
(
d, J = 4.0 Hz, 1H), 7.95 (d, J = 4.0 Hz, 1H), 7.44–7.39 (m, 2H), 7.15 (td, J = 8.76, 2.62 Hz, 1H), 6.04 (s,
H), 4.25 (d, J = 20 Hz, 1H), 4.07–3.97 (m, 2H), 3.86–3.61 (m, 4H), 3.51 (s, 3H), 3.12–3.07 (m, 1H),
1
13
1
2.45–2.39 (m, 1H). C NMR (DMSOꢀd , 100 MHz): δ ppm 172.8, 166.2, 162.5, 161.1 (d, J
=
CF
6
4
3
2
46.33Hz, FC), 148.0, 144.8, 144.1, 138.7 (d, J = 2.83Hz, FC), 133.1 (d, J = 10.52Hz, FC), 131.4 (d,
CF CF
3
2
2
J = 8.56Hz, FC), 125.1, 117.1 (d, J = 25.74Hz, FC), 115.3 (d, J = 20.42Hz, FC), 96.3, 69.4, 67.2,
CF
CF
CF
62.0, 56.3, 54.8, 51.4, 48.7.
ACS Paragon Plus Environment

Page 29 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
rd
For the following 3 generation 4ꢀH HAPs, except described specifically, all the analogues are
prepared in analogy to 10 from the commercially available building blocks.
(
(3R)ꢀ4ꢀ[[(4R)ꢀ4ꢀ(2ꢀChloroꢀ4ꢀfluoroꢀphenyl)ꢀ5ꢀmethoxycarbonylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀ
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
+
dihydropyrimidinꢀ6ꢀyl]methyl]morpholineꢀ3ꢀcarboxylic acid (11). MS: calc’d (MH ) 495, measured
+
1
(MH ) 495. H NMR (DMSOꢀd , 400 MHz): δ ppm 12.86 (br s, 1H), 9.88 (s, 1H), 8.07–7.85 (m, 2H),
6
7
.47–7.29 (m, 2H), 7.16 (td, J = 8.5, 2.6 Hz, 1H), 6.03 (s, 1H), 4.18–4.02 (m, 2H), 3.93–3.80 (m, 2H),
3.77–3.58 (m, 2H), 3.47–3.24 (m, 4H), 3.14–2.99 (m, 1H), 1.38–1.12 ppm (m, 1H).
(
2S)ꢀ1ꢀ[[(4R)ꢀ4ꢀ(2ꢀChloroꢀ4ꢀfluoroꢀphenyl)ꢀ5ꢀmethoxycarbonylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀ
dihydropyrimidinꢀ6ꢀyl]methyl]ꢀ4,4ꢀdifluoroꢀpyrrolidineꢀ2ꢀcarboxylic acid (12). Melting point: 106ꢀ
o
+
+
+
+
107 C. MS: calc’d (MH ) 515, measured (MH ) 515. HRMS: calc’d (MH ) 515.0762, measured (MH )
1
515.0770. H NMR (Methanolꢀd , 400 MHz): δ ppm 8.21 (s, 2H), 7.63–7.60 (m, 1H), 7.38–7.35 (m, 1H),
4
7
.22–7.18 (m, 1H), 6.29 (s, 1H), 4.57 (d, J = 16 Hz, 1H), 4.13–4.08 (m, 2H), 3.70–3.67 (m, 4H), 3.33–
3.28 (m, 1H), 2.92–2.82 (m, 1H), 2.63–2.56 (m, 1H).
4
ꢀ[[(4R)ꢀ4ꢀ(2ꢀChloroꢀ4ꢀfluoroꢀphenyl)ꢀ5ꢀmethoxycarbonylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀdihydropyrimidinꢀ6ꢀ
+
+
1
yl]methyl]morpholineꢀ2ꢀcarboxylic acid (13). MS: calc’d (MH ) 495, measured (MH ) 495. H NMR
(Methanolꢀd , 400 MHz): δ ppm 8.02 (d, J = 3 Hz, 0.5H), 8.01 (d, J = 3 Hz, 0.5H), 7.91 (dd, J = 3 Hz,
4
0
.5H), 7.90 (d, J = 3 Hz, 0.5H), 7.54 (dd, J = 6.0, 9.0 Hz, 0.5H), 7.53 (dd, J = 6.0, 9.0 Hz, 0.5H), 7.30 (dd,
J = 3, 9 Hz, 1H), 7.15–7.09 (m, 1H), 6.21 (s, 1H), 4.82–4.55 (m, 3H), 4.31–4.24 (m, 1H), 4.12–3.91 (m,
2H), 3.74–3.65 (m, 1H), 3.66 (s, 3H), 3.55–3.37 (m, 2H).
2
ꢀ[(3R)ꢀ4ꢀ[[(4R)ꢀ4ꢀ(2ꢀChloroꢀ4ꢀfluoroꢀphenyl)ꢀ5ꢀmethoxycarbonylꢀ2ꢀthiazolꢀ2ꢀylꢀ1,4ꢀ
dihydropyrimidinꢀ6ꢀyl]methyl]morpholinꢀ3ꢀyl]acetic acid (14). The synthetic procedure was described
+
+
1
in the supporting information. MS: calc’d (MH ) 509, measured (MH ) 509. H NMR (Methanolꢀd , 400
4
MHz): δ ppm 7.97 (d, J = 3.01 Hz, 1H), 7.77 (d, J = 3.14 Hz, 1H), 7.44 (dd, J = 8.60, 6.21 Hz, 1H), 7.24
(
dd, J = 8.78, 2.64 Hz, 1H), 7.06 (td, J = 8.38, 2.57 Hz, 1H), 6.17 (s, 1H), 4.32 (br d, J = 17.44 Hz, 1H),
ACS Paragon Plus Environment