Design and Synthesis of Cajanine Analogues against Hepatitis C Virus through Down-Regulating Host Chondroitin Sulfate N-Acetylgalactosaminyltransferase 1

Article abstract of DOI:10.1021/acs.jmedchem.6b01301

There still remains a need to develop new anti-HCV agents with distinct mechanism of action (MOA) due to the occurrence of resistance to direct-acting antiviral agents (DAAs). Cajanine, a stilbenic component isolated from Cajanus cajan L., was identified as a potent HCV inhibitor by phenotypic screening in this work (EC₅₀ = 3.17 ± 0.75 μM). The intensive structure optimization provided significant insights into the structure-activity relationships. Furthermore, the MOA study revealed that cajanine inhibited HCV replications via down-regulating a cellular protein chondroitin sulfate N-acetylgalactosaminyltransferase 1. In consistency with this host-targeting mechanism, cajanine showed the similar magnitude of inhibitory activity against both drug-resistant and wild-type HCV and synergistically inhibited HCV replication with approved DAAs. Taken together, our study not only presented cajanine derivatives as a novel class of anti-HCV agents but also discovered a promising anti-HCV target to combat drug resistance.

Full text of DOI:10.1021/acs.jmedchem.6b01301

Subscriber access provided by La Trobe University Library
Article
Design and Synthesis of Cajanine Analogues against
Hepatitis C Virus through Down-regulating Host
Chondroitin Sulfate N-acetylgalactosaminyltransferase 1
Xingyue Ji, Jinhua Chen, Guang-Hui Zheng, Meng-Hao Huang, Lei Zhang,
Jie Jin, Hong Yi, Jian-Dong Jiang, Zong-Gen Peng, and Zhuo-Rong Li
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01301 • Publication Date (Web): 26 Oct 2016
Downloaded from http://pubs.acs.org on October 27, 2016
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 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design and Synthesis of Cajanine Analogues against
Hepatitis C Virus through Downꢀregulating Host
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
Chondroitin
Sulfate
Nꢀ
acetylgalactosaminyltransferase 1
‡
‡
XingꢀYue Ji , JinꢀHua Chen , GuangꢀHui Zheng, MengꢀHao Huang, Lei Zhang, Hong Yi, Jie Jin,
JianꢀDong Jiang, ZongꢀGen Peng*, and ZhuoꢀRong Li*
Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences and Peking Union
Medical College, Beijing 100050, China
KEYWORDS: Cajanine; AntiꢀHCV activity; SAR; CSGalNAcTꢀ1.
ABSTRACT: There still remains a need to develop new antiꢀHCV agents with distinct
mechanism of action (MOA) due to the occurrence of resistance to directꢀacting antiviral agents
(DAAs). Cajanine, a stilbenic component isolated from Cajanus cajan L., was identified as a
potent HCV inhibitor by phenotypic screening in this work (EC50 = 3.17±0.75 ꢁM). The
intensive structure optimization provided significant insights into the structureꢀactivity
relationships. Furthermore, the MOA study revealed that cajanine inhibited HCV replications via
downꢀregulating a cellular protein chondroitin sulfate Nꢀacetylgalactosaminyltransferase 1. In
consistency with this hostꢀtargeting mechanism, cajanine showed the similar magnitude of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 61
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
inhibitory activity against both drugꢀresistant and wildꢀtype HCV, and synergistically inhibited
HCV replication with approved DAAs. Taken together, our study not only presented cajanine
derivatives as a novel class of antiꢀHCV agents but also discovered a promising antiꢀHCV target
to combat drug resistance.
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
INTRODUCTION
Hepatitis C virus (HCV) infection is a major health problem, with as many as 200 million
1
individuals infected worldwide. Approximately 20% of infected individuals would progress to
cirrhosis, and 2ꢀ4% of the patient with cirrhosis are at risk of irreversible hepatic failure. In
recent years, the therapy of HCV has been revolutionized with the approval of directꢀacting
2
antiviral agents (DAAs), especially with sofosbuvir, a nucleotide analogue inhibitor of HCV
3
NS5B polymerase. Recently, sofosbuvir combined with ledipasvir (NS5A inhibitor) was
approved by FDA under the brand name of Harvoni to treat patients with genotype 1 chronic
hepatitis C. This is also the first approved regimen that is all oral and IFNꢀ or RBVꢀfree.
Undoubtedly, sofosbuvir will be the new backbone of HCV treatment, and the overall
therapeutic effect in HCVꢀinfected patients will be continuously improved. However, challenges
4
5
6
and issues still remain, including drug resistance, high cost, narrow genotype specificity, and
7
the treatment of genotype 3 infections. It is still imperative for the development of new antiꢀ
HCV agents, especially the ones with novel mechanism of action (MOA).
Compared to DAAs, hostꢀtargeting antivirals (HTAs) have an advantage of no or decreased
8
chance to induce drugꢀresistant mutations. There are a broad range of HTAs being developed in
preclinical or clinical trials, including entry inhibitors, translation inhibitors, replication
9
inhibitors and assembly inhibitors, and some promising results have been observed among these
ACS Paragon Plus Environment

Page 3 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
HTAs, especially in terms of higher barrier to resistance and synergistic effect with DAAs to
1
0
reduce viral load. For example, Alisporivir (ALV), a cyclophilin inhibitor, is effective against
all DAAsꢀresistant HCV genotype 1b variants, with the half maximal effective concentration
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
(
EC ) values in the range of 15.8ꢀ46.7 nM. In addition, the combination of ALV with
50
daclatasvir (NS5A inhibitor) constantly showed synergistic effects on GTꢀ2a and ꢀ3a HCV. In
summary, it is recognized that optimal HCV therapy regimen might be the combined treatment
1
1
of several antiꢀHCV agents targeting different steps of HCV lifecycle. Since HCV infection
requires the synergic action of viral and host factors, it is anticipated that the combination of
DAAs and HTAs would improve the overall therapeutic effect and reduce the occurrence of
DAAsꢀresistance in clinical practice.
Natural products are always a good source for the discovery of novel lead compounds. Of
the 877 smallꢀmolecule new chemical entities introduced between 1981 and 2002, nearly half
1
2
(49%) of them were derived from natural products. Cajanine (also known as cajaninstilbene
acid), a stilbene derivative isolated from the herb Cajanus cajan L., has recently drawn much
13
14
attention due to its broad range of bioactivities, including antitumor, antiꢀinflammatory, antiꢀ
1
5
16
17
18
19
osteoporotic, antioxidant, antibacterial, cytoprotective and neuroprotective effects.
1
3
Previously, we have reported the total synthesis of cajanine (1, Fig. 1A). Given that there is a
huge unmet need for effective and affordable therapies for HCV infections, 1 was subjected to a
cell based phenotype antiꢀHCV screening in our lab, and it was identified as a good inhibitor
against HCV replication with an EC value of 3.17±0.75 ꢁM, and no decrease in cell viability
5
0
was observed throughout the experiments (Fig. 1B and 1C).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 61
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
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 1. 1 inhibited HCV replication in a doseꢀdependent manner in Huh7.5 cells. (A) Structure
of 1; (BꢀC) 1 inhibited HCV replication at HCV RNA (B) and protein (C) level with no
cytotoxicity (B, MTT assay). Intracellular HCV RNA and Core protein were detected after 72 h
treatment with 1. Data were normalized to the solvent control, which was regarded as 100%.
*P<0.05 and **P<0.01 vs. solvent control.
However, as shown in Figure 1A, 1 is extremely lipophilic with a calculated AlogP value of
2
0
5
.0 (Discovery Studio 3.0), which is likely to raise poor physicochemical properties concern.
Consequently, structural simplification was made to probe the importance of the two
hydrophobic moieties (prenyl and phenyl ring B) for antiꢀHCV activity. In addition, intensive
structure modifications were also conducted, and some significant structureꢀactivity relationships
were concluded herein. Furthermore, the MOA of 1 was also studied, and the results revealed
that 1 inhibited HCV replication via a distinct mechanism, which involved a cellular protein
chondroitin sulfate Nꢀacetylgalactosaminyltransferase 1 (CSGalNAcTꢀ1). In our knowledge, the
role of CSGalNAcTꢀ1 in HCV replication has not yet been reported.
Design
ACS Paragon Plus Environment

Page 5 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Indicated structural modifications to compound 1
As shown in Figure 2, the structural modifications to
1 were primarily made to the
indicated regions I, II, III, IV and V, aiming to uncover the SARs at these positions. In
order to address the hydrophobic issue of , compounds without ring B (2i, ALogP = 4.5)
1
or prenyl (15a, ALogP = 3.4) were designed and synthesized. Meanwhile, the other two
major stilbene compounds isolated from Cajanus cajan L., namely longstyline A and C, were
also synthesized and assayed for their inhibitory activity against HCV herein.
RESULTS AND DISCUSSIONS
Chemistry
13
Class I: Compound 13 was synthesized according to our previous reported method. As shown
in scheme 1, compound 13 was condensed with various ketones or aldehydes in the presence of
NaH to yield compounds 14a-k. Selective demethylation with BCl of compounds 14a-k
3
afforded 15a-k in high yield. The sodium salt of compounds 15a-k were condensed with prenyl
bromide in anhydrous toluene to afford the desired Cꢀalkylation products 2a-2k, and the Oꢀ
alkylation products were also unavoidably formed at this step, which contributed to the moderate
yield of the Cꢀalkylation products (50%ꢀ60%). Compound 2k was further deprotected in the
presence of TBAF to give compound 2l with a free hydroxyl group. Compounds 2aꢀj and 2l
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 61
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
were then hydrolyzed to form the target compounds 1 and 2m-v. Longistyline A (3) was also
obtained when microwave irradiation was employed in the hydrolysis of 1.
a
Scheme 1. Synthesis of Class I analogues
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
Reagents and conditions: (a) NaH, ketones or aldehydes, dry DME, N , reflux, 1.5 h, 70ꢀ85%,
2
R : Phenyl, 2ꢀchlorophenyl, 2ꢀmethylphenyl, 3ꢀmethylphenyl, 4ꢀmethylphenyl, 4ꢀfluorophenyl,
1
styryl, thienyl, ethyl, 4ꢀOTBDMSꢀphenyl, or 4ꢀOHꢀphenyl; R : H (14aꢀi, and 14k), or methyl
2
(
14j); (b) BCl , ꢀ78 °C, CH Cl , 2 h, 85ꢀ95%, R : Phenyl, 2ꢀchlorophenyl, 2ꢀmethylphenyl, 3ꢀ
3 2 2 1
methylphenyl, 4ꢀmethylphenyl, 4ꢀfluorophenyl, styryl, thienyl, ethyl, 4ꢀOTBDMSꢀphenyl, or 4ꢀ
OHꢀphenyl; R : H (15aꢀi, and 15k), or methyl (15j); (c) NaH, dry toluene, prenyl bromide, 40
2
ACS Paragon Plus Environment

Page 7 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
°
C, 2 h, 50ꢀ60%; (d) TBAF, THF, rt, 2 h, 90%; (e) KOH, EtOH/H O, reflux, 2 h, 90% (or
2
KOH/EtOH, McW, 30 W, 100 °C, 40 psi, 0.5 h, 80%, for compound 3).
Class II: As shown in scheme 2, compound 15a was hydrogenated to afford compound 16 in
high yield. The sodium salt of compound 16 was reacted with prenyl bromide to yield the Cꢀ
alkylation product 4 along with its Oꢀalkylation counterpart. The hydrolysis of 4 afforded the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
1
target compound 5 (also known as amorfrutin A ) with stilbenic double bond reduced.
a
Scheme 2. Synthesis of derivative 5 with stilbenic double bond reduced
a
Reagents and conditions: (a) H , Pd/C (10%), EtOH, 50 psi, 4 h, 95%; (b) NaH, toluene, prenyl
2
bromide, 40 °C, 4 h, 63%; (c) KOH, EtOH/H O, reflux, 2 h, 92%.
2
Class III: Given that 1, 14a, 15a and 2a possessed olefinic double bond in their structures,
where the addition reaction could probably take place on the double bond along with the
2
2
13
chlorination of ring A, compound 17 was then chosen as the starting material to react with
SO Cl in diethyl ether to afford the chlorination product 18. Compound 20 was obtained after
2
2
the bromination and Arbuzov reactions. The synthesis of compound 6 was straightforward by a
similar method used to synthesize compounds of Class I (Scheme 1).
a
Scheme 3. Synthesis of class III analogue.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 61
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
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
Reagents and conditions: (a) SO Cl , ether, reflux, 4 h, 85%; (b) Br , CCl , hν, reflux, 3 h, 85%;
2
2
2
4
(
c) P(OEt) , 150 °C, 2 h, 90%; (d) PhCHO, NaH, dry DME, reflux, 2 h, 70%; (e) BCl , CH Cl , ꢀ
3
3
2
2
78 °C, 2 h, 80%; (f) NaH, toluene, prenyl bromide, 40 °C, 4 h, 50%; (g) KOH, EtOH/H O,
2
reflux, 2 h, 90%.
a
Scheme 4. Synthesis of longistyline C.
a
Reagents and conditions: (a) MeOH, Br , 0 °C, 2 h, 90%; (b) Pd(Ph P) , DMF, tributyl(3ꢀmethylbutꢀ2ꢀ
2
3
4
enyl)stannane, 110 °C, 12 h, 50%; (c) BCl , CH Cl , ꢀ78 °C, 2 h, 85%; (d) KOH/EtOH, Mcw, 30 W,
3
2
2
ACS Paragon Plus Environment

Page 9 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
00 °C, 30 psi, 2 h, 70%.
Longistyline C (7), a natural occurring analogue of 1, was also synthesized for the first time
in this work. As shown in scheme 4, compound 14a was brominated to afford compound 24,
which was then reacted with tributylprenylstannane under the conditions of Stille coupling
reaction to give compound 25 in moderate yield. Selective demethylation of compound 25 in
the presence of BCl3 afforded compound 26 in 85% yield. The hydrolysis and
decarboxylation of compound 26 were accomplished in one step using microwave
irradiation, and compound 7 was obtained with an overall yield of 27%.
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
a
Scheme 5. Synthesis of the analogue with reduced prenylic double bond.
a
Reagents and conditions: (a) MeI, K CO , DMF, 60 °C, 2 h, 85%; (b) Br , hν, CCl , reflux, 2 h,
2
3
2
4
8
5%; (c) P(OEt) , 150 °C, reflux, 6 h, 90%; (d) Pd(Ph P) , tributyl(3ꢀmethylbutꢀ2ꢀenyl)stannane,
3 3 4
DMF, 110 °C, 12 h, 70%; (e) H , Pd/C (10%), EtOH, 50 psi, 4 h, 95%; (f) NaH, PhCHO, dry
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 61
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
DME, reflux, 2 h, 84%; (g) BCl , CH Cl , ꢀ78 °C, 2 h, 85%; (h) KOH, EtOH/H O, reflux, 2 h,
3
2
2
2
90%.
2
3
Class IV: As shown in scheme 5, dimethylation of compound 27 followed by bromination
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
and condensation with triethyl phosphite yielded compound 30. Under the conditions of Stille
coupling reaction, compound 31 was obtained in moderate yield. The target compound 9 was
then obtained through a series of reactions including hydrogenation, condensation with
benzaldehyde, selective demethylation and hydrolysis. Meanwhile, compound 10 with all
olefinic double bond reduced was also synthesized by hydrogenation of 1, as shown in scheme 6.
Additionally, as shown in scheme 7, compound 11 with (E)ꢀ3, 7ꢀdimethyloctaꢀ2,6ꢀdienyl
(geranyl) substituent at 3ꢀposition of ring A was also synthesized by a similar method used for
the synthesis of 1 (Scheme 1).
a
Scheme 6. Synthesis of the analogue with reduced olefinic double bond.
a
Reagents and conditions: (a) H , Pd/C, EtOH, 70 psi, 4 h, 95%.
2
a
Scheme 7. Synthesis of the analogue with geranyl substituent at 3ꢀposition of ring A.
ACS Paragon Plus Environment

Page 11 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Reagents and conditions: (a) NaH, toluene, geranyl bromide, 40 °C, 2 h, 50%; (b) KOH,
EtOH/H O, reflux, 2 h, 90%.
2
Class V: The carboxyl group in 1 was esterified or amidated by using N, N′ꢀ
diisopropylcarbodiimide (DIC) as the coupling reagent. In the case of esterification, the alcohols
were used both as reactant and solvent. Eventually, the target compounds 12a – d were obtained
in 65ꢀ75% yield.
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
a
Scheme 8. Synthesis of the Class V analogues.
COOH
COXR
OMe
HO
HO
a
OMe
1
12a: X = O, R = Et
1
1
1
2b: X = O, R = iso-Pro
2c: X = O, R = t-Bu
2d: X = NH, R = cyclopropanyl
a
Reagents and conditions: (a) X = O: DIC, RXH (reactant and solvent), rt, 4 h, 70ꢀ75%; Or X =
NH: DIC, CH Cl , RXH, rt, 4 h, 65%.
2
2
BIOLOGY
Anti-HCV activity in vitro
The infectious virus J6/JFH/JCꢀ1 (a recombinant HCV genotype 2a) were employed to
screen the HCV inhibition activity of the synthesized cajanine analogues in Huh7.5 cells. In the
HCVcc system, HCV can propagate with complete life cycle including an early stage of the
HCV life cycle. The Huh7.5 cells were infected with HCV viral stock (45 IU/cell), and treated
simultaneously with test compounds or the positive control telaprevir. The intracellular HCV
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 61
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
RNA was extracted and quantified with oneꢀstep qRTꢀPCR. The cytotoxicity was determined
using MTT assay. The EC and CC values were calculated with Reed & Muench methods.
50
50
Table 1. Structures and inhibitory activity of class I compounds against HCV in Huh7.5 cells.
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
COOR1
HO
R2
R3
OMe
HCV (J6/JFH gt2a full replication)
Compounds R1
R2
R3
a
b
c
CC (ꢁM)
EC (ꢁM)
SI
50
50
2
2
2
2
2
2
2
2
a
b
c
d
e
f
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
Phenyl
H
H
>200
>200
4.87 ±2.81
>41
18
7
2ꢀClꢀphenyl
10.93±3.43
25.57 ±7.49
18.31 ±8.06
32.83 ±3.71
3.18 ±0.21
>66.67
2ꢀMeꢀphenyl H
3ꢀMeꢀphenyl H
4ꢀMeꢀphenyl H
171.06 ±10.16
137.79 ±14.65
127.60±9.93
180.80±14.90
193.08 ±6.03
>200
8
4
4ꢀFꢀphenyl
styryl
H
57
<3
>16
4
g
h
H
thienyl
Et
H
12.32 ±0.15
24.76 ±1.85
2.89 ±0.95
3.17±0.75
1.03±0.04
5.92 ±0.13
1.95 ±1.03
6.54±2.01
1.49±0.08
13.40 ±3.65
2i
2j
1
H
89.27 ±10.47
102.05 ±14.71
90.99 ±8.18
39.52±0.94
55.29 ±3.67
82.45 ±9.24
82.86 ±9.12
69.22 ±3.76
96.90 ±5.99
phenyl
Phenyl
2ꢀClꢀphenyl
Me
H
35
40
39
9
2
2
2
2
2
2
m
H
H
n
o
p
q
r
H
2ꢀMeꢀphenyl H
3ꢀMeꢀphenyl H
4ꢀMeꢀphenyl H
H
42
13
46
7
H
H
4ꢀFꢀphenyl
styryl
H
H
H
ACS Paragon Plus Environment

Page 13 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
2
2
2
s
H
H
H
H
ꢀ
thienyl
Et
H
83.01 ±7.88
72.85±2.43
76.84 ±2.98
16.15±0.10
44.85 ±2.05
3.70 ±0.53
1.02 ±0.46
1.70 ±0.51
0.33 ±0.10
22
71
45
49
t
H
u
v
phenyl
Me
4ꢀOHꢀphenyl 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
Telaprevir
ꢀ
ꢀ
0.008±0.0003 5704
a: 50% cytotoxicity concentration; b: 50% effective concentration; c: selective index. The
EC and CC values were indicated as mean ± SD values, and were calculated from three
50
50
independent experiments.
Class I derivatives were designed to probe the importance of ring B for antiꢀHCV activity.
As shown in Table 1, most of the compounds exhibited definitive antiꢀHCV activity with SI
values higher than 10 in the HCVcc system, and several derivatives exhibited more potent
activity as compared to 1, especially compound 2v, which showed the most potent inhibitory
activity among this series with an EC value of 0.33 ± 0.10 ꢁM. In terms of SAR, the carboxylic
50
derivatives were generally more potent than their esterified counterparts, and hence, a hydrogen
bond donor or ionic interaction at this position may increase the inhibitory activity. On the other
hand, the esterified compounds were less cytotoxic compared to their carboxylic counterparts.
Replacing the proton at para position of ring B with a hydroxyl group (2v) significantly
increased the inhibitory activity. Substituting a chloro (2m) or fluoro (2q) group for the proton
(
1) at ortho or para position of phenyl ring B respectively increased the inhibitory activity, while
replacing the proton at ortho or para position with a methyl group (2n or 2p) decreased the
activity. The introduction of a methyl group at the meta position of ring B (2o) had little effect
on the activity, as compared to 1. Replacement of the ring B with a styryl (2r) or thienyl (2s)
decreased the inhibitory activity. Unexpectedly, substituting an ethyl group (2t) for the ring B
did not decrease the activity, but slightly increased the activity instead. This effect suggested that
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 61
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
the ring B was not indispensable for the antiꢀHCV activity. Compound 2t also showed the
highest SI value (71) among this series of analogues.
Table 2. Structures and inhibitory activity of class IIꢀ III analogues against HCV in Huh7.5 cells.
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
HCV(J6/JFH gt2a full replication)
Compounds R1
X
R2
R3
CC (ꢁM)
EC (ꢁM)
SI
50
50
4
COOMe CH
2
ꢀCH
2
2
H
prenyl
prenyl
prenyl
121.86±6.69 16.09±8.15
59.35 ±9.04 1.68 ±0.03
108.94 ±0.90 7.61 ±0.91
8
5
COOH
CH
2
ꢀCH
H
35
6
COOH
CH=CH
CH=CH
ꢀ
Cl
14
7
H
ꢀ
prenyl H
18.54 ±2.14
44.85 ±2.05
8.37 ±2.44
2.2
5704
Telaprevir
ꢀ
ꢀ
0.008±0.0003
The EC and CC values were indicated as mean± SD values, and were calculated from
5
0
50
three independent experiments.
Class II (4 and 5) and III (6 and 7) were designed to probe the importance of the stilbenic
double bond and the R group for the antiꢀHCV activity, respectively. As shown in table 2,
3
compound 5 was slightly more potent than compound 1 (Table 1), which meant the stilbenic
double bond was not necessary for the antiꢀHCV activity. The introduction of a chloro group at
4ꢀposition of ring A (6) decreased the inhibitory activity. Compound 7, another stilbene
component isolated from Cajanus cajan L., also showed a moderate level of inhibitory activity
against HCV. But its inhibitory potency was inferior to that of 1.
ACS Paragon Plus Environment

Page 15 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. Structures and inhibitory activity of class IV analogues against HCV in Huh7.5 cells.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
HCV (J6/JFH gt2a full replication)
Compounds
R1
X
R2
CC (ꢁM)
EC (ꢁM)
SI
113
>6
27
50
50
15a
8
COOMe
COOMe
COOH
COOH
COOH
ꢀ
CH=CH
CH=CH
CH=CH
H
160.26 ±27.89
>200
1.42 ±0.31
32.31±4.32
3.58±0.77
0.78 ±0.42
4.66±0.10
isoꢀpentyl
isoꢀpentyl
isoꢀpentyl
geranyl
ꢀ
9
97.76 ±3.21
72.16±6.90
46.74 ±5.17
44.85 ±2.05
1
0
1
CH
2
ꢀCH
2
92
1
CH=CH
ꢀ
10
Telaprevir
0.008±0.0003 5704
The EC and CC values were indicated as mean± SD values, and were calculated from
5
0
50
three independent experiments.
Class IV derivatives were designed and synthesized to investigate the importance of prenyl
substituent for the antiꢀHCV activity. As shown in table 3, replacement of the prenyl (1) with
isoꢀpentyl (9) or a bulkier geranyl group (11) retained the inhibitory activity, with a slight
decrease in the activity when compared to 1. Based on those results, it can be concluded that a
saturated or unsaturated alkyl substituent was well tolerated at this position. Compounds 10 with
both olefinic double bond reduced showed even more potent activity as compared to 1.
Unexpectedly, removal of the prenyl substituent at R position (15a) did not lead to the loss in
3
activity, but slightly increased the inhibitory potency instead, as compared to 1. This effect
indicated that the prenyl group was also not necessary for the antiꢀHCV activity. Actually, this
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 61
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
effect is what we would like to observe. As shown in the chemical structure of 1, its core
structure primarily consists of two hydrophobic groups (prenyl and styryl), which make this
molecule extremely lipophilic (AlogP = 5.0). It is conceivable that compound 1 may suffer from
potentially poor PK profiles including poor water solubility, poor bioavailability, and
vulnerability to metabolism, among others. Compound 15a (AlogP = 3.4), a less lipophilic
molecule compared to 1, possessed less hydrophobic burden, and hence may present better PK
profiles compared to 1. Additionally, compound 15a not only showed more potent inhibitory
activity but also displayed the highest SI value (113) among all the synthesized compounds.
Consequently, compound 15a is a more promising alternative for further optimization.
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. Structures and inhibitory activity of class V analogues against HCV in Huh 7.5 cells
.
COXR
HO
OMe
HCV (J6/JFH gt2a full replication)
Compounds
X
R
CC (ꢁM)
EC (ꢁM)
SI
47
>15
8
50
50
12a
O
O
O
NH
ꢀ
Et
195.36±7.79 4.13 ±0.27
>200 12.93 ±4.87
12b
12c
12d
3
isoꢀPro
tꢀBu
101.31±7.77 13.50 ±0.99
33.06 ±1.86 4.31 ±1.25
21.39 ±0.41 12.04 ±0.51
cyclopropyl
8
ꢀ
ꢀ
1.8
Telaprevir
ꢀ
44.85 ±2.05 0.008±0.0003 5704
The EC and CC values were indicated as mean ± SD values, and were calculated from
5
0
50
three independent experiments.
ACS Paragon Plus Environment

Page 17 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Class V derivatives were designed to investigate the importance of carboxyl group for the
antiꢀHCV activity. As shown in table 4, esterification (12a-c), amidation (12d) or removal of the
carboxyl group (3) resulted in a decrease in activity, and these results further highlighted the
importance of the carboxyl group for the antiꢀHCV activity. Additionally, with the increase in
bulk of the R group, the inhibitory activity decreased. However, the EC value of compound
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
0
12d was lower than that of compound 12b and 12c, and this effect may be attributed to the amide
proton, which could function as a hydrogen bond donor, as does the carboxyl group in 1.
Anti-HCV mechanism of 1
Having confirmed that 1 and its derivatives showed strong inhibitory activity against HCV
in Huh7.5 cells, we next studied the MOA underlying 1’s antiꢀHCV effect. 1 was initially
screened against several known viral proteins, including NS5B, NS5A and NS3/4A, and the
results showed that 1 exhibited no inhibitory effect against these proteins (data not shown).
Consequently, we speculated that 1 and its analogues might inhibit HCV replication by targeting
host proteins.
Along this clue, 1 was employed as a representative compound to study its MOA. The early
evidence from geneꢀexpression chip assay showed that treatment of Huh7.5 cells with 1
decreased mRNA level of CSGalNAcTꢀ1, which is an essential enzyme involved in the initiation
2
4
of chondroitin sulfate (CS) synthesis. This result was further validated by a doseꢀdependent
decrease in the protein level of CSGalNAcTꢀ1 after 72 h treatment of 1 either in native Huh7.5
cells (Fig. 3A, left) or in HCVꢀinfected Huh7.5 cells (Fig. 3B, left). It seems that 1 showed more
effective results in terms of decreasing CSGalNAcTꢀ1 protein levels in native Huh7.5 cells than
HCVꢀinfected ones (Figure 3A left vs Figure 3B left). Such discrepancy is presumably resulted
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 61
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
from the higher expression level of CSGalNAcTꢀ1 protein in HCV infected cells as compared to
the native ones (Fig. 3C). Unexpectedly, 1 had no effect on the mRNA level of CSGalNAcTꢀ1
after 72 h treatment of 1 either in native Huh7.5 cells (Fig. 3A, right) or HCVꢀinfected Huh7.5
cells (Fig. 3B, right), which contradicted with the results obtained from gene chip assay. The
reasons for such contradiction are possibly twoꢀfold: The short sequence of probe is not specific
enough, or the target gene obtained from gene chip assay is homologous to CSGalNAcTꢀ1, and
hence other genes are possibly also involved in the antiꢀHCV mechanism of compound 1.
Nonetheless, our results suggested that 1 may mainly downꢀregulate CSGalNAcTꢀ1 at the
protein level. Next, we tested whether 1 had effect on destabilizing CSGalNAcTꢀ1 protein in
native Huh7.5 cells. The results showed that the protease inhibitor MG132 markedly reversed 1ꢀ
induced reduction of the CSGalNAcTꢀ1 protein level (Fig. 3D), suggesting that 1 inhibited HCV
infection presumably by destabilizing CSGalNAcTꢀ1 protein. We next probed whether
compound 1 specifically act on CSGalNAcTꢀ1 protein. To this end, we tested the effect of 1 on
the expression level of other cellular proteins such as membrane protein cluster of differentiation
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
36 (CD36) and cytoplasmic protein heatꢀstress cognate 70 (Hsc70), both of which are related to
25, 26
HCV replication and are potential antiꢀHCV target.
The results showed that 1 possessed no
effect on the expression level of CD36 and Hsc70 (Figure 3E), suggesting that 1 indeed shows
some specificity to CSGalNAcTꢀ1. The inferior antiꢀHCV activity of 1 and its analogues to that
of the positive control telaprevir is presumably ascribed to this distinct hostꢀtargeting instead of
direct acting on viral protein mechanism.
Having confirmed that 1 inhibited HCV infection by targeting cellular CSGalNAcTꢀ1
protein, we next probed whether its analogues also possessed the similar MOA. Compound 15a
was chosen for such purpose, because it showed promising in vitro antiꢀHCV activity with the
ACS Paragon Plus Environment

Page 19 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
highest SI value (116), and also possessed the most different chemical structure among the other
analogues as compared to 1. As shown in Figure 3F, compound 15a also decreased the level of
CSGalNAcTꢀ1 in a concentration dependent manner, and showed no effect on CD36 and Hsc70
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
(
3G), suggesting that removal of the prenyl group from 1 did not make a difference in terms of
antiꢀHCV MOA.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 61
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
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. 1 inhibited HCV replication via destabilizing CSGalNAcTꢀ1 protein. (AꢀB) 1
downꢀregulated cellular CSGalNAcTꢀ1 protein (left) with no role at the mRNA level of
CSGalNAcTꢀ1 (right) in native Huh7.5 cells (A) or HCVꢀinfected Huh7.5 cells (B); (C) HCV
infection upꢀregulated cellular CSGalNAcTꢀ1 protein. Huh7.5 cells were infected with HCV (45
IU/cell) for 48 hours; (D) 1 destabilized the CSGalNAcTꢀ1 protein. MG132 (1ꢁM) reversed the
1 (25ꢁM) ꢀinduced degradation of CSGalNAcTꢀ1, **P < 0.01 vs. 1 alone; (E) Compounds 1 had
no effect on CD36 or Hsc70; (F) Compound 15a decreased the protein level of cellular
CSGalNAcTꢀ1 in naive Huh7.5 cells. (G) Compounds 15a had no effect on CD36 or Hsc70. *P
<
0.05 and **P < 0.01 vs. control. Error bars represented the standard deviation of three
independent experiments
ACS Paragon Plus Environment

Page 21 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In order to better understand the MOA of 1, we investigated the role of CSGalNAcTꢀ1 in
HCV lifecycle. Huh7.5 cells were transfected with expression plasmid pcDNA3.1(+)ꢀ
CSGalNAcTꢀ1 followed by HCV infection, and the incubation was continued for another 72 h.
The results showed that intracellular HCV RNA and Core protein were significantly higher in the
Huh7.5 cells transfected with pcDNA3.1(+)ꢀCSGalNAcTꢀ1, as compared to the control plasmid
pcDNA3.1(+) (Fig. 4A, left for RNA level, right for protein level). Meanwhile, after the silence
of endogenous CSGalNAcTꢀ1 with specific siRNA, the intracellular HCV RNA load was largely
reduced, and the HCV Core protein was declined as well (Fig. 4B, left for RNA level, right for
protein level). Taken together, it can be concluded that CSGalNAcTꢀ1 is essential for HCV
replication.
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 22 of 61
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
Figure 4. CSGalNAcTꢀ1 was a coꢀfactor for HCV replication. (A) Highꢀexpression of
CSGalNAcTꢀ1 with expression vector increased HCV RNA level (left) and HCV core protein
(right); (B) Silence of CSGalNAcTꢀ1 with siRNA decreased HCV RNA level (left) and HCV
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
core protein (right). *P < 0.05 and **P < 0.01 vs. control. Error bars represented the standard
deviation of three independent experiments.
ACS Paragon Plus Environment

Page 23 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 5. CSGalNAcTꢀ1 facilitated HCV replication at early steps in the viral life cycle. (A)
1
didn’t inhibit HCV replication in GS4.3 cells (HCV subꢀgenome replicon cell); (B)
CSGalNAcTꢀ1 facilitated HCV replication at early steps in the viral life cycle at RNA level (left)
and protein level (right); (C) 1 didn’t directly inhibited HCV replication at early steps in the viral
life cycle at RNA level (left) and protein level (right). Data were normalized to the control,
which was regarded as 100%. *P<0.05 and **P<0.01, vs control. Error bars represented the
standard deviation of three independent experiments.
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 explore the possible action step(s) of 1 in the viral lifecycle, we first evaluated its antiꢀ
27
HCV activity in GS4.3 replicon cell line carrying an HCV subgenomic replicon I377ꢀ3’del.S ,
where HCV can replicate without early steps such as attachment, absorption and fusion. As
depicted in Figure 5A, no inhibitory effect was observed in 1ꢀtreated GS4.3 cells, while the
positive control telaprevir exhibited strong antiꢀHCV activity, suggesting that 1 did not inhibit
HCV replication at the replicative steps of HCV lifecycle. Then, we conducted further
experiment to probe whether 1 inhibited HCV replication at early steps. Huh7.5 cells were
transfected with plasmid pcDNA3.1(+)ꢀCSGalNAcTꢀ1. After 48 h, the cells were infected with
HCV (150 IU/cell) for 2 h. After washing with PBS, the cells were cultured with fresh medium.
At 24 h postꢀinfection, intracellular proteins and RNAs were extracted to detect CSGalNAcTꢀ1
expression, or the cells were passaged and cultured with fresh medium for 72 h. The results
showed that the intracellular HCV RNA and HCV Core protein was remarkably increased in
Huh7.5 cells transfected with pcDNA3.1(+)ꢀCSGalNAcTꢀ1 when compared with that in cells
transfected with control plasmid pcDNA3.1(+) (Fig. 5B, left for RNA level, right for protein
level), suggesting that CSGalNAcTꢀ1 is important to support HCV replication at the early stage.
We then studied whether 1 directly inhibited HCV replication at early steps in the viral life
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 61
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
cycle. Huh7.5 cells were infected with HCV and simultaneously treated with 1 for 2 h, and then
the intracellular proteins were detected. Results showed that 1 did not directly inhibit HCV
replication at the early steps of HCV life cycle (Fig. 5C, left for RNA level, right for protein
level). Collectively, 1 might inhibit HCV replication at early steps of the viral life cycle via
downꢀregulating of cellular CSGalNAcTꢀ1.
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
Anti-HCV activity of 1 against variants resistant to DAAs
Since HTAs show advantages over DAAs in terms of combating DAAsꢀdrug resistance, we
next probe the antiviral activities of
1 against HCV variants resistant to DAAs. As reported,
D168V and A156T mutants showed the most frequent resistance to NS3/4A protease inhibitors,
28
29
such as simeprevir and telaprevir, and S282T mutant was resistant to sofosbuvir. Therefore,
1
was assayed for its inhibitory activity against these mutant viruses. As expected, telaprevir,
simeprevir, and sofosbuvir showed decreased inhibitory activity against these mutant variants,
however,
1 exhibited equivalent inhibitory activities against both wildꢀtype and mutant HCV
virus (Table 5). Collectively,
HCV treatment.
1 showed promising potential for combating drug resistance in
Table 5. The inhibitory activity of
1 against wild type or drugꢀresistant HCV
EC (ꢁM)
5
0
a
Compounds
FOR
Wildꢀtype
3.03
A156T
D168V
S282T
1
6.44
3.80
8.59
1.3~2.8
31.0
Telaprevir
Simeprevir
0.02
0.62
ꢀ
0.93
ꢀ
ꢀ
ꢀ
0.04
ꢀ
ꢀ
23.3
Sofosbuvir
0.064
0.29
4.5
a
FOR: Fold of resistance. The antiꢀHCV activity was determined with qRTꢀPCR in Huh7.5
cells infected with HCV (J6/JFH/JCꢀ1) or mutant variant. A156T, D168V and S282T were
mutant variants from wildꢀtype J6/JFH/JCꢀ1.
ACS Paragon Plus Environment

Page 25 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synergism of 1 plus DAAs against HCV
Since 1 inhibited HCV replication via targeting host factor CSGalNAcTꢀ1, which is distinct
from the MOA of DAAs, we then tested whether 1 possessed any synergistic effects against
HCV infection with approved DAAs, including simeprevir (NS3/4A inhibitor), sofosbuvir
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
(
NS5B inhibitor) and daclatasvir (NS5A inhibitor). As shown in table 6, 1 synergistically
inhibited HCV replication with simeprevir, sofosbuvir or daclatasvir at the RNA level with
combination index (CI) < 1.0 and dose reduction index (DRI) > 1. The similar trend was also
observed at the protein level (Fig. 6). Altogether, 1 showed synergistic effects with DAAs to
reduce HCV viral load.
Table 6. Synergism of 1 plus DAAs against HCV
b
DRI
a
Compounds
EC (µM) CI value
50
1
DAAs
1
12.02
0.04
0.11
Simeprevir
Sofosbuvir
Daclatasvir
c
17.67
1
+ Simeprevir
Combination 1 + Sofosbuvir
+ Daclatasvir
5.89+0.02
2.18+0.06
0.953
0.693
0.760
2.0
2.2
1.9
2.3
5.5
3.1
c
1
3.87+7.74
The antiꢀHCV activity was determined with qRTꢀPCR in Huh7.5 cells infected with HCV
a
(J6/JFH/JCꢀ1). Combination index (CI) value at EC50 was calculated by the ChouꢀTalalay
b
method using CompuSyn version 1.0; Folds of dose reduction allowed for each drug caused by
synergism at a given effect level (e.g., the dose reduction index (DRI) values at EC effect level);
50
c
the concentration unit of daclatasvir was pM.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 61
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
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 6. 1 enhanced the inhibitory activities against HCV of DAAs. Huh7.5 cells were
infected with HCV (45 IU/cell) and simultaneously treated with simeprevir (A, 0.025 ꢁM),
sofosbuvir (B, 0.1 ꢁM) or daclatasvir (C, 16 pM) alone or with 1 (6.25 ꢁM). *P < 0.05 vs. 1
#
##
alone; P < 0.05 and P < 0.01 vs. simeprevir, sofosbuvir or daclatasvir alone. All of the
experiments were at least triplicated. The protein bands showed the results of a representative
experiment. The Data presented are mean ± standard deviation and Student’s tꢀtest was used.
CONCLUSIONS
1 was identified as a potent HCV inhibitor for the first time in this work. A series of novel
cajanine derivatives along with its natural occurring counterpart 7 were synthesized and assayed
for antiꢀHCV activity in vitro. The intensive structure modifications led to several novel
compounds with more potent inhibitory activity compared to 1, especially compounds 2v and 10
with EC50 values in the range of submicromolar. Additionally, some significant SARs were
uncovered, especially that both the hydrophobic moieties (ring B and prenyl) were not required
for the inhibitory activity. Such SARs provided valuable implications for further lead
optimizations. Compound 15a, which is analogous to 1 without prenyl substitution, showed
ACS Paragon Plus Environment

Page 27 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
comparable inhibitory activity to 1, and it possessed less hydrophobic burden with molecular
weight and ALogP value being 284.3 and 3.4, respectively. Moreover, it shared the same MOA
to 1 by targeting CSGalNAcTꢀ1, despite that it possessed simplified structure as compared to 1.
Therefore, compound 15a represented a more promising scaffold for further optimization.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The MOA study showed 1 inhibited HCV replication by downꢀregulating cellular
CSGalNAcTꢀ1, which is demonstrated to be essential for viral replication in the early life cycles.
In line with this distinct MOA, 1 showed the similar magnitude of inhibitory activity against both
wildꢀtype and DAAsꢀresistant variants, and synergistically inhibited HCV replication with
approved DAAs, including simeprevir, sofosbuvir or daclatasvir. All these results highlighted
cellular CSGalNAcTꢀ1 as a potentially promising target to combat drug resistance. However, the
detailed mechanism of how 1 destabilized CSGalNAcTꢀ1 remains unclear at this stage, and other
mechanism might also be involved. The detailed studies is ongoing in our lab.
In summary, our study not only presented 1 and its derivatives as a novel class of antiꢀHCV
agents, but also discovered a novel antiꢀHCV target to combat drug resistance.
EXPERIMENTAL SECTION
General. All reagents and solvent were purchased from commercial sources. THF, toluene
and DME were distilled from sodium and benzophenone before use. CCl , CH Cl , CH CN were
4
2
2
3
distilled from P O . All reactions were carried out inflameꢀdried glassware and monitored by thin
2
5
1
13
layer chromatography using aluminum TLC plate 60F254D (Merck Millipore). H NMR and C
NMR spectra were recorded in CDCl or DMSOꢀd using a Varian Inova 400MHz, 500MHz or
3
6
600MHz spectrometer (Varian, SanFrancisco, CA, USA). Chemical shifts are reported in parts
per million relative to tetramethylsilane as internal standard. The mass spectra (MS) were
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 61
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
recorded on a Thermo Scientific LTQ ORBITRAP instrument with an ESI mass selective
detector. Melting points were determined using an X6 microscope melting point apparatus and
are uncorrected. The microwave experiment was conducted using a Discover microwave oven
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
(CEM, Matthews, NC, USA). The purities of tested compounds were assessed as being at least
95% with analytical HPLC, which was performed using a C18 (5 ꢁm, 250 mm ×4.6 mm)
column, UV detected at 254 nm and with elution 15% (0.01 M KH PO solution) / 85% (MeOH)
2
4
for the carboxylic compounds, or 2% (0.01M KH PO solution) / 98% (MeOH) for the nonꢀ
2
4
carboxylic compounds at 1 mL/min. For the synthesis and spectra data of compounds 13, 14a,
21
15a, 2a, 1 and 3, please refer to our previous report.
General procedure A for the prenylation of demethylation compounds.
Compound 15aꢀk (1 equiv.) was dissolved in dry toluene (30 mL), and NaH (1.3 equiv.) was
added at room temperature. The reaction mixture was stirred at room temperature for 0.5h, after
which a portion of the toluene was evaporated away and prenyl bromide(1.5 equiv.) was added at
room temperature. The mixture was stirred for 2 h at 40 °C, and quenched by the addition of ice
water. The organic layer was separated and dried over anhydrous MgSO . It was then filtered
4
and concentrated to give a residue, which was purified over silica gel to give a crude product,
which was recrystallized from petroleum ether to afford the target compounds.
(E)-Methyl 6-(2-chlorostyryl)-2-hydroxy-4-methoxy-3-prenylbenzoate (2b). Yield: 50%.
1
White solid, mp 94ꢀ96 °C. H NMR (400 MHz, CDCl ) δ (ppm): 1.68 (s, 3H), 1.79 (s, 3H), 3.38
3
(d, J = 6.8 Hz, 2H), 3.93 (s, 3H), 5.22 (t, J = 6.8 Hz, 1H), 6.64 (s, 1H), 7.15 (d, J = 16.0 Hz,
1H), 7.19 (t, J = 7.6 Hz, 1H), 7.28 (t, J = 7.6 Hz, 1H),7.40 (d, J = 7.6 Hz, 1H), 7.64 (d. J = 7.6
1
3
Hz, 1H), 7.69 (d, J = 16.0 Hz, 1H), 11.66 (s, 1H). C NMR (126 MHz, DMSOꢀd ) δ (ppm):
6
ACS Paragon Plus Environment

Page 29 of 61
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
7.8, 22.1, 25.9, 52.1, 55.4, 103.3, 104.6, 117.2, 121.8, 126.1, 126.4, 126.7, 128.8, 129.8, 131.7,
ꢀ
133.0, 133.3, 135.4, 140.0, 161.6, 171.7. ESI-HRMS cacld for C H O Cl [MꢀH] 385.12011,
22 22
4
found 385.12070. HPLC:t = 9.99 min, normalization method purity 98.36%.
R
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
(
E)-Methyl 6-(2-methylstyryl)-2-hydroxy-4-methoxy-3-prenylbenzoate (2c). Yield: 50%.
1
White solid, mp 89ꢀ91 °C. H NMR (400 MHz, CDCl ) δ (ppm): 1.68 (s, 3H), 1.79 (s, 3H), 2.42
3
(s, 3H), 3.38 (d, J = 7.2 Hz, 2H), 3.93 (s, 6H), 5.22 (t, J = 7.2 Hz, 1H), 6.60 (s, 1H), 6.96 (d, J =
1
1
1
1
6.0 Hz, 1H), 7.19ꢀ7.24 (m, 3H), 7.56 (d, J = 6.8 Hz, 1H), 7.60 (d, J = 16.0 Hz, 1H), 11.69 (s,
1
3
H). C NMR (126 MHz, CDCl ) δ (ppm): 17.8, 19.9, 22.1, 25.8, 52.2, 55.6, 103.1, 104.6,
3
16.7, 122.1, 125.6, 126.2, 127.6, 128.0, 130.4, 131.8, 131.96 135.80 136.53 140.79 161.35
ꢀ
61.49 172.0. ESI-HRMS cacld for C H O [MꢀH] 365.17474, found 365.17454. HPLC:t =
2
3
25
4
R
8.43 min, normalization method purity 97.62%.
(E)-Methyl 6-(3-methylstyryl)-2-hydroxy-4-methoxy-3-prenylbenzoate (2d). Yield: 50%.
1
White solid, mp 64ꢀ66°C. H NMR (400 MHz, CDCl ) δ (ppm): 1.68 (s, 3H), 1.79 (s, 3H), 2.38
3
(s, 3H), 3.37 (d, J = 7.2 Hz, 2H), 3.92 (s, 3H), 3.94 (s, 3H), 5.21 (t, J = 7.2 Hz, 1H), 6.60 (s,
1
H), 6.75 (d, J = 16.0 Hz, 1H), 7.10 (d, J = 6.8 Hz, 1H), 7.30 (m, 3H), 7.70 (d, J = 16.0 Hz, 1H),
13
1
1.67 (s, 1H). C NMR (126 MHz, CDCl ) δ (ppm): 17.8, 21.2, 22.2, 25.8, 52.2, 55.6, 102.8,
3
104.5, 116.5, 122.1, 123.6, 127.5, 128.5, 128.6, 130.2, 130.5, 131.8, 137.5, 138.3, 140.4, 161.4,
ꢀ
1
61.5, 171.9. ESI-HRMS cacld for C H O [MꢀH] 365.17474, found 365.17496. HPLC:t =
2
3
25
4
R
8
.21 min, normalization method purity 95.04%.
(E)-Methyl 6-(4-methylstyryl)-2-hydroxy-4-methoxy-3-prenylbenzoate (2e). Yield: 50%.
1
White solid, mp 95ꢀ97°C. H NMR (400 MHz, CDCl ) δ (ppm): 1.68 (s, 3H), 1.79 (s, 3H), 2.37
3
(s, 3H), 3.37 (d, J = 7.2 Hz, 2H), 3.92 (s, 6H), 5.22 (t, J = 7.2 Hz, 1H), 6.60 (s, 1H), 6.76 (d, J =
ACS Paragon Plus Environment