Discovery of ABT-267, a pan-genotypic inhibitor of HCV NS5A

Article abstract of DOI:10.1021/jm401398x

We describe here N-phenylpyrrolidine-based inhibitors of HCV NS5A with excellent potency, metabolic stability, and pharmacokinetics. Compounds with 2S,5S stereochemistry at the pyrrolidine ring provided improved genotype 1 (GT1) potency compared to the 2R,5R analogues. Furthermore, the attachment of substituents at the 4-position of the central N-phenyl group resulted in compounds with improved potency. Substitution with tert-butyl, as in compound 38 (ABT-267), provided compounds with low-picomolar EC₅₀ values and superior pharmacokinetics. It was discovered that compound 38 was a pan-genotypic HCV inhibitor, with an EC₅₀ range of 1.7-19.3 pM against GT1a, -1b, -2a, -2b, -3a, -4a, and -5a and 366 pM against GT6a. Compound 38 decreased HCV RNA up to 3.10 log₁₀ IU/mL during 3-day monotherapy in treatment-naive HCV GT1-infected subjects and is currently in phase 3 clinical trials in combination with an NS3 protease inhibitor with ritonavir (r) (ABT-450/r) and an NS5B non-nucleoside polymerase inhibitor (ABT-333), with and without ribavirin.

Full text of DOI:10.1021/jm401398x

Subscriber access provided by DUESSELDORF LIBRARIES
Article
Discovery of ABT-267, a pan-genotypic inhibitor of HCV NS5A
David Allen DeGoey, John Randolph, Dachun Liu, John Pratt, Charles Hutchins, Pamela Donner,
Chris Krueger, Mark Matulenko, Sachin Patel, Christopher Motter, Lissa Nelson, Ryan Keddy,
Michael Tufano, Daniel Caspi, Preethi Krishnan, Neeta Mistry, Gennadiy Koev, Thomas Reisch,
Rubina Mondal, Tami Pilot-Matias, Gao Yi, David Beno, Clarence Maring, Akhter Molla, Emily
Dumas, Andrew Campbell, Laura Williams, Christine Collins, Rolf Wagner, and Warren Kati
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 08 Jan 2014
Downloaded from http://pubs.acs.org on January 8, 2014
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 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of ABTꢀ267, a panꢀgenotypic inhibitor of
HCV NS5A
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
David A. DeGoey,* John T. Randolph, Dachun Liu, John Pratt, Charles Hutchins, Pamela
Donner, A. Chris Krueger, Mark Matulenko, Sachin Patel, Christopher E. Motter,†Lissa Nelson,
Ryan Keddy, †† Michael Tufano,‡ Daniel D. Caspi, Preethi Krishnan, Neeta Mistry, Gennadiy
Koev, Thomas J. Reisch, Rubina Mondal,# Tami Pilot-Matias, Yi Gao, David W. A. Beno,¶
Clarence J. Maring, Akhter Molla, Emily Dumas, Andrew Campbell, Laura Williams, Christine
Collins, Rolf Wagner, and Warren M. Kati.
Research and Development, AbbVie Inc., North Chicago, IL 60064, USA
KEYWORDS. Hepatitis C, antiviral, NS5A, pyrrolidine, prolinamide.
ABSTRACT. We describe here Nꢀphenylpyrrolidineꢀbased inhibitors of HCV NS5A with
excellent potency, metabolic stability and pharmacokinetics. Compounds with 2S,5S
stereochemistry at the pyrrolidine ring provided improved genotype 1 (GT1) potency compared
to the 2R,5R analogs. Furthermore, the attachment of substituents at the 4ꢀposition of the central
Nꢀphenyl group resulted in compounds with improved potency. Substitution with tertꢀbutyl, as in
compound 38 (ABTꢀ267), provided compounds with lowꢀpicomolar EC50 values and superior
pharmacokinetics. It was discovered that compound 38 was a panꢀgenotypic HCV inhibitor, with
an EC50 range of 1.7 – 19.3 pM against GT1a, 1b, 2a, 2b, 3a, 4a, and 5a, and 366 pM against GT
6a. Compound 38 decreased HCV RNA up to 3.10 log10 IU/mL during 3ꢀday monotherapy in
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 40
1
2
3
4
5
6
7
8
9
treatmentꢀnaïve HCV GT1ꢀinfected subjects and is currently in Phase 3 clinical trials in
combination with an NS3 protease inhibitor with ritonavir (r) (ABTꢀ450/r) and an NS5B nonꢀ
nucleoside polymerase inhibitor (ABTꢀ333), with and without ribavirin.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Introduction
Once treatable only through lengthy, poorly tolerated and marginally efficacious
regimens using pegylated interferon alpha (pegIFNꢀα) and ribavirin (RBV), chronic hepatitis C
virus (HCV) infection has now been cured with high rates in clinical trials using IFNꢀfree
combinations of directꢀacting antivirals (DAA).1 HCV is a bloodꢀborne virus, transmitted largely
through needle sharing and blood transfusions that occurred prior to 1992 before routine
screening was implemented.2 Left untreated, chronic HCV infection can result in cirrhosis, an
increased incidence of hepatocellular carcinoma and ultimately liver failure. Complications due
to chronic HCV infection have now become the predominant driver for liver transplantation.3
The widespread incidence of HCV infections worldwide (estimated at 170ꢀ200 million), together
with a large number of baby boomers in the US with more difficult to treat HCV genotype 1
(GT1) infections (approximately 2.1 million), have contributed to the urgent need to identify
more tolerable and efficacious treatments with oral administration. For more than a decade,
HCV treatment consisted of a lengthy (48 weeks) regimen of pegIFNꢀα and ribavirin (RBV), a
therapy associated with significant tolerability issues and poor efficacy (40ꢀ50% cure rate in GT1
patients).4 Recently, the approval of the first DAAs, the NS3/4A protease inhibitors telaprevir or
boceprevir, led to a significant improvement in the cure rate (up to 75% in GT1 treatment naïve
patients) and their use in combination with pegIFNꢀα and RBV is the current standard of care.5,6
While these new regimens provide improved efficacy, tolerability remains suboptimal due to
side effects associated with pegIFNꢀα and RBV as well as gastrointestinal symptoms and
ACS Paragon Plus Environment
2

Page 3 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
significant anemia associated with telaprevir and boceprevir. Clearly, there is a compelling need
for new agents that will bolster the cure rate for patients infected with HCV, while eliminating
IFN as a requirement for successful treatment.
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 genome of HCV encodes several nonstructural (NS) proteins. While inhibition of the
enzymatic activities of NS3/4A protease and NS5B polymerase has been the subject of intense
drug discovery efforts, the NS5A protein has no known intrinsic enzymatic activity, although it
is essential for viral RNA replication.7 The discovery of compounds that select mutations in
NS5A using the replicon assay established it as a potential drug target, with mutations at Tyr93
conferring resistance to multiple classes of compounds.8,9 Because adaptive mutations in NS5A
that increase replication facilitated the development of the replicon cell culture assay,7,10 it
remained to be demonstrated that compounds active in the replicon assay would exert an
antiviral effect in vivo against the native virus. The observation of robust viral load declines in
human clinical trials with NS5A inhibitors has now firmly established NS5A as a viable drug
target.11 Recent efforts at AbbVie to study HCV inhibitors yielded potent naphthyridine
inhibitors such as 1 that lacked activity against HCV protease and polymerase (Figure 1).12,13
Efforts to improve potency and metabolic stability culminated in the discovery of compound 2,
which demonstrated a robust viral load decline in an GT1b HCVꢀinfected chimpanzee. 12 As was
the case with many of the early NS5A chemical series, however, the compounds demonstrated
weaker activity against GT1a compared to GT1b.
Here we describe the structureꢀactivity relationships (SAR) of a novel symmetric series
of Nꢀphenylpyrrolidineꢀbased inhibitors (Figure 1) leading to the discovery of compound 38
(ABTꢀ267), which is currently undergoing phase 3 clinical trials in combination with an NS3
protease inhibitor with ritonavir (r) (ABTꢀ450/r)14 and an NS5B nonꢀnucleoside polymerase
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 40
1
2
inhibitor (ABTꢀ333) ,14 with and without RBV. Xꢀray crystallographic studies of domain I
fragments of NS5A have yielded dimeric structures, although the biological relevance of the
dimeric forms has not been firmly established.7,15 NS5A inhibitors with symmetric dimer or
dimerꢀlike structures have recently become the subject of intense research due to their potency,
presumably from improved interactions with the dimeric form of the NS5A protein.16ꢀ21 For
example, BMSꢀ790052 (Figure 1) was inspired from the isolation of dimeric impurities formed
as byproducts of the initial lead compound BMSꢀ824 and which were determined to be the active
species.17,18 Other approaches to symmetryꢀbased NS5A inhibitors have also been actively
studied.9,16ꢀ21 We explored inhibitors with chiral pyrrolidineꢀbased cores (Figure 1), which
projected Nꢀaryl functional groups orthogonal to the vector linking the proline amide groups that
greatly impacted the potency and pharmacokinetics. We also describe the effect of the
pyrrolidine ring stereochemistry on the compound potency and pharmacokinetics.
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
ACS Paragon Plus Environment
4

Page 5 of 40
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 1. Structures of HCV NS5A inhibitors.
Chemistry
As shown in Scheme 1, compounds 12ꢀ46 (Table 1) were synthesized starting from the
diketone 3, which was prepared by reaction of 2ꢀbromoꢀ1ꢀ(4ꢀnitrophenyl)ethanone with 1ꢀ(4ꢀ
nitrophenyl)ethanone in the presence of zinc(II) chloride.22 Chiral diols 4a and 4b were prepared
by asymmetric reduction of 3 using N,Nꢀdiethylaniline borane in combination with (S)ꢀ(ꢀ)ꢀα,αꢀ
diphenylꢀ2ꢀpyrrolidinemethanol or (R)ꢀ(ꢀ)ꢀα,αꢀdiphenylꢀ2ꢀpyrrolidinemethanol, respectively, in
the presence trimethyl borate in > 99% ee.23 The mixture of diols 4c was generated using sodium
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 40
1
2
3
4
5
6
7
8
9
borohydride. Generation of the dimesylate followed by treatment with the aniline of choice
provided pyrrolidines 6a-6c. Reduction of the nitro groups to give 7a-c could be affected under a
variety of conditions, including heating with iron and ammonium chloride or catalytic
hydrogenation. Hydrogenation with Raney nickel or platinum (IV) oxide was best suited for
enantiomerically pure compounds 6a and 6b in order to minimize stereochemical isomerization.
Compound 7c could be readily separated into cisꢀpyrrolidine and racemic transꢀpyrrolidine
isomers by column chromatography on silica gel. The proline amide side chains were attached
by peptide coupling with BocꢀProline to give 8, followed by deprotection and peptide coupling
with methyl carbamateꢀcapped amino acids to give the final products 9. Final compounds
prepared using 6c were separated into the trans-pyrrolidine and cis-pyrrolidine analogs by
reversed phase HPLC. Further separation of the diastereomeric mixture of transꢀpyrrolidines was
accomplished by chiral chromatography on a Chiralpak ADꢀH column. Alternatively,
diastereomerically pure trans-pyrrolidine compounds were synthesized starting from the
enantiomerically pure intermediates 6a and 6b.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Synthesis of pyrrolidineꢀbased HCV NS5A inhibitors
ACS Paragon Plus Environment
6

Page 7 of 40
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
a
Reagents and conditions: (a) ZnCl2, tꢀBuOH, Et2NH; (b) (S)ꢀ(ꢀ)ꢀα,αꢀdiphenylꢀ2ꢀ
pyrrolidinemethanol or (R)ꢀ(ꢀ)ꢀα,αꢀdiphenylꢀ2ꢀpyrrolidinemethanol, PhNEt2•BH3, B(OMe)3; (c)
NaBH4; (d) MsCl, EtN(i-Pr)2; (e) RC6H4NH2; (f) Fe, NH4Cl; (g) PtO2 or Raney nickel, H2; (h)
BocꢀPro, HATU, EtN(i-Pr)2; (i) TFA or HCl; (j) MeOC(O)NHCH(R2)CO2H, HATU or EDCI
and HOBT, EtN(i-Pr)2.
Results and Discussion
In vitro biological data for these compounds are summarized in Table 1. The compounds
were tested in GT1a and GT1b subgenomic replicon assays with and without the addition of 40%
human plasma (HP), as binding to human plasma proteins can reduce the pharmacological
activity of antiviral compounds via reduction of the free drug levels.24,25 Thus, the replicon EC50
values obtained in the presence of 0% HP more closely represent the intrinsic potencies of the
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 40
1
2
3
4
5
6
7
8
9
compounds, whereas values obtained in the presence of 40% HP may be more closely correlated
with clinical efficacy.26,27 As indicated in Table 1, the compounds were tested either as mixtures
of the trans-pyrrolidine isomers or as the pure 2S,5S or 2R,5R diastereomers. Testing the
separated trans-pyrrolidine isomers 12 and 13 revealed a stereochemical dependence for their
replicon potencies. The 2S,5S isomer 12 was approximately 10ꢀfold more potent than the 2R,5R
isomer 13 against GT1a, while 12 was only 3.5ꢀfold more potent than 13 against GT1b.
Comparison of the other pairs of trans-pyrrolidine diastereomers in Table 1 revealed a similar
trend of increased potency for 2S,5S isomers against GT1a and GT1b compared to the 2R,5R
isomers. The cisꢀpyrrolidine isomer 14 was significantly less active against GT1a (55ꢀfold) and
GT1b (7.5ꢀfold) than the 2S,5S trans-pyrrolidine compound 12.
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 comparison of the 2S,5S trans-pyrrolidine compound 12 with acyclic analog 10 (Figure
2) revealed the effect of the pyrrolidine ring formation on potency, with 12 demonstrating 96ꢀ
and 33ꢀfold improved potency against GT1a and GT1b, respectively. Likewise, comparison of
12 with the unsubstituted pyrrolidine analog 11 (Figure 2) revealed that 12 was 639ꢀ and 23,500ꢀ
fold more potent against GT1a and GT1b, respectively.
The addition of 40% HP to the replicon assay resulted in a 17ꢀfold loss of potency on
average for the compounds in Table 1 against GT1a and GT1b. Compounds bearing more
lipophilic groups generally demonstrated larger shifts than less lipophilic compounds. For
example, the tertꢀbutyl glycine capped compound 15 (LogD = 5.2) demonstrated 59ꢀ and 88ꢀfold
shifts in potency with the addition of 40% HP against GT1a and GT1b, respectively, while the
valine capped compound 12 (LogD = 4.6) demonstrated 9ꢀ and 15ꢀfold shifts against GT1a and
GT1b, respectively. This observation is consistent with the findings of others who have noted
that increasing lipophilicity (log D) within a series increases plasma protein binding.28,29
A
ACS Paragon Plus Environment
8

Page 9 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
survey of amino acid capping groups was conducted on the 4ꢀfluorophenyl substituted
pyrrolidine core (compare compounds 12ꢀ26). In addition to valine (12, 13), other lipophilic
amino acids were well tolerated, including tertꢀbutylglycine (15, 16), isoleucine (19),
alloisoleucine (20) and phenylglycine (21), while the 2ꢀaminobutyric acid analog 18 was less
potent. The introduction of more polar amino acid capping groups, such as 3ꢀhydroxyꢀvaline
compound 22, was also tolerated. The incorporation of amino acids bearing ether side chains,
such as Oꢀmethylꢀthreonine (23 and 24) and THFꢀglycine (25 and 26), resulted in compounds
with more uniform GT1a and GT1b potencies, with replicon potencies in the 1 to 2 nanomolar
range in the presence of 40% HP.
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
9

Journal of Medicinal Chemistry
Page 10 of 40
1
2
3
4
F
5
6
7
8
9
H
N
H
N
N
H
N
N
H
N
O
N
O
O
O
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
O
O
O
10
GT1a EC50 = 38.0 nM
GT1b EC50 = 0.40 nM
H
N
H
N
H
N
N
H
N
N
H
N
O
O
O
O
O
O
O
O
11
GT1a EC50 = 253 nM
GT1b EC50 = 282 nM
F
H
N
H
N
N
H
N
N
H
N
O
N
O
O
O
O
O
O
O
12
GT1a EC50 = 0.396 nM
GT1b EC50 = 0.012 nM
Figure 2. Structures of acyclic analog 10, unsubstituted analog 11 and Nꢀphenyl pyrrolidine 12.
Modification of the 4ꢀposition of the central Nꢀphenyl ring also proved to have a
significant effect on the GT1a and GT1b potencies. While hydrophobic substitutions of the
fluoro substituted analog 12 with methyl (27), chloro (28), bromo (29) and methoxy (30)
provided compounds with similar potency, the more polar phenol 31 was significantly less
active. Notably, the trifluoromethyl analog 32 demonstrated picomolar activity against GT1a and
GT1b, while retaining subnanomolar potency in the presence of 40% HP. Substitution with
ACS Paragon Plus Environment
10

Page 11 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
isopropyl (36), sometimes considered to be isosteric with the trifluoromethyl group, produced
similar potency, while the cyclopropyl analog 37 was slightly less active. With the larger tꢀbutyl
analog 38, additional potency improvement in GT1a compared to 32 was observed. Compound
38 demonstrated uniform picomolar GT1a and GT1b activity, including picomolar activity in the
presence of 40% HP. Once again, the 2S,5S trans-pyrrolidine isomer 38 proved to be 10ꢀfold
more potent against GT1a and 1.8ꢀfold more potent against GT1b than the 2R,5R isomer 39. The
tertꢀbutyl glycine capped analog 40, although similarly potent against GT1a and 1b, exhibited a
greater potency loss with the addition of 40% HP. Exploration of other large groups with more
polarity, such as hydroxylated tertꢀbutyl (42) and the methyl sulfone (43), resulted in a loss of
potency, compared to 38. While the phenyl substituted compound 44 retained high potency
against GT1a and GT1b, the introduction of polar heterocycles such as pyridine (45) and
morpholine (46) resulted in a loss of potency compared to 38.
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
Table 1. In vitro activity against HCV GT1a and GT1b in the replicon assay.
Compound
R1
Replicon EC50 (nM)
GT1a 40%
GT1b 40%
HP (fold a)
0.184 (15)
0.208 (5)
0.300 (3)
0.528 (88)
0.162 (20)
0.278 (9)
1.00 (7)
no.
Configuration
R2
GT1a
GT1b
HP (fold a)
3.68 (9)
48
49
50
51
52
53
54
55
56
57
58
59
60
12
13
14
15
16
17
18
2S,5S
2R,5R
cis
F
F
F
F
F
F
F
i-Pr
i-Pr
i-Pr
tꢀBu
tꢀBu
tꢀBu
Et
0.396
2.53
22.0
0.263
1.18
6.22
4.60
0.012
0.042
0.090
0.006
0.008
0.032
0.140
11.1 (4)
81.0 (4)
15.5 (59)
20.8 (18)
103.0 (17)
34.0 (7)
2S,5S
2R,5R
cis
trans b
19
trans b
F
1.31
0.030
9.79 (7)
0.171 (6)
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 40
1
2
3
4
5
6
20
21
trans b
trans b
F
F
0.202
1.06
0.010
6.56 (32)
22.8 (22)
0.090 (9)
Ph
0.002
0.050
0.073 (37)
2.60 (52)
7
8
9
22
23
24
trans b
2S,5S
2R,5R
F
F
F
0.600
0.168
0.794
18.0 (30)
1.40 (8)
1.95 (2)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.086
0.134
0.539 (6)
1.33 (10)
25
26
2S,5S
F
F
0.102
0.281
0.054
0.106
1.39 (14)
1.92 (7)
1.27 (24)
0.917 (9)
2R,5R
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
trans b
trans b
trans b
trans b
trans b
2S,5S
2R,5R
2S,5S
2R,5R
trans b
2R,5R
2S,5S
2R,5R
2S,5S
2R,5R
Me
Cl
Br
OMe
OH
CF3
CF3
CF3
CF3
i-Pr
cꢀPr
tꢀBu
tꢀBu
tꢀBu
tꢀBu
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
tꢀBu
tꢀBu
i-Pr
i-Pr
i-Pr
i-Pr
tꢀBu
tꢀBu
0.720
0.370
0.286
0.992
23.0
0.057
0.299
0.099
0.829
0.033
0.164
0.014
0.138
0.021
0.123
0.040
0.030
0.0068
0.076
4.36 (6)
4.07 (11)
4.46 (16)
5.92 (6)
242.0 (11)
0.650 (11)
2.55 (9)
4.14 (42)
14.3 (17)
0.706 (21)
1.68 (10)
0.186 (13)
0.986 (7)
1.06 (50)
4.31 (35)
0.204 (5)
0.300 (10)
0.089 (13)
0.447 (6)
6.54
59.5 (9)
0.0071
0.0062
0.0088
0.011
0.0079
0.0087
0.005
0.0088
0.0022
0.0084
0.072 (10)
0.077 (12)
0.211 (24)
0.184 (17)
0.086 (11)
0.114 (13)
0.056 (11)
0.090 (10)
0.117 (53)
0.344 (41)
42
2S,5S
i-Pr
0.116
0.095
1.47 (13)
0.948 (10)
OH
43
2R,5R
tꢀBu
0.659
0.226
27.6 (42)
3.42 (15)
44
45
trans b
trans b
Ph
i-Pr
i-Pr
0.089
0.143
0.011
0.069
1.69 (19)
2.57 (18)
0.256 (23)
0.597 (9)
46
2R,5R
tꢀBu
1.24
0.211
11.2 (9)
1.60 (8)
a Fold shift in potency upon addition of 40% HP. b Tested as a mixture of 2S,5S+ 2R,5R isomers.
Based on their in vitro potencies, compounds were selected for pharmacokinetic studies
(Table 2). In vitro physicochemical data (including solubility and LogD) and metabolic stability
ACS Paragon Plus Environment
12

Page 13 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
in human and rat liver microsomes (HLM and RLM, respectively) for the set of compounds
appear in Table 3. The compounds generally demonstrated low aqueous solubility (< 10 ꢁM) and
compounds with higher observed LogD values demonstrated the lowest solubility (i.e., 40 and
41). Due to the low aqueous solubility of the compounds, permeability data generated using in
vitro assays such as PAMPA proved to be of low utility for predicting oral absorption in vivo.
Human and rat microsomal stability were generally similar for the compounds. Oral
pharmacokinetic data were obtained using solutions in organic solvents in order to aid in
dissolution and absorption of the sparingly soluble compounds. In rat, the tertꢀbutyl glycineꢀ
capped compound 16 showed a longer halfꢀlife and higher oral bioavailability than the
corresponding valineꢀcapped compound 13, consistent with the higher metabolic stability for 16
observed in RLM. Both 13 and 16 showed higher oral bioavailability in dog (F = 62 and 94,
respectively) than in rat, while both compounds demonstrated high clearance and low oral
bioavailability in monkey. The Oꢀmethyl threonine compounds 23 and 24 along with the THFꢀ
glycine analogs 25 and 26 were of high interest due to their potency in the presence of human
plasma. However, the compounds demonstrated short halfꢀlives, high clearance and low oral
bioavailability, despite their moderate stability in RLM. The ether oxygen containing compounds
23ꢀ26 also had higher total polar surface areas (TPSA) of 197.2 compared to the other
compounds in Table 2 which had TPSA values of 178.7, and this factor may have also
contributed to their lower oral bioavailability. The potent trifluoromethylꢀsubstituted compounds
32 and 33 and the tertꢀbutylꢀsubstituted compound 38 demonstrated high stability in RLM and
long halfꢀlives in rat (6, 4, and 9 hours, respectively), although they suffered from low oral
bioavailability. However, compound 38 demonstrated significantly higher oral bioavailability
and exposures in the dog compared to compound 32 and 33. Unlike compounds 13 and 16,
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
13

Journal of Medicinal Chemistry
Page 14 of 40
1
2
3
4
5
6
7
8
9
compound 38 also demonstrated a reasonable half‐life with low clearance and moderate oral
bioavailability in monkey. The tertꢀbutyl glycine analogs 40 and 41 showed comparable rat PK
to the valine analogs 38 and 39, including low oral bioavailability, likely a result of their high
lipophilicity (LogD = 5.5) and low aqueous solubility.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Pharmacokinetic parameters for selected compounds. a
IV
Oral
Cmax
0.19
1.38
0.42
0.05
0.11
0.96
1.07
0.12
0.07
0.12
0.02
0.03
0.03
0.07
0.20
0.05
0.01
0.64
0.29
0.02
0.09
0.09
no.
Species
t1/2
2.6
3.3
2.0
0.8
ND b
7.5
9.1
4.6
0.6
0.4
0.9
0.8
6.0
6.0
4.2
4.4
9.9
7.9
4.4
5.5
8.2
10.2
Vss
2.4
0.5
1.9
1.5
Cl
t1/2
3.0
2.8
Tmax
1.0
1.3
1.2
1.5
3.0
2.3
2.7
1.7
0.7
0.5
0.4
0.3
2.2
4.7
1.5
1.7
3.7
3.3
3.3
4.7
5.3
2.5
AUC
0.75
6.36
1.28
0.08
1.85
10.21
18.26
0.49
0.12
0.17
0.02
0.02
0.50
0.85
1.36
0.37
0.29
8.00
2.40
0.22
1.25
1.39
F
22
22
62
10
23
39
94
18
14
15
1.5
3.2
16
5.5
13
4.2
6.2
57
12
rat
rat
dog
monkey
rat
0.90
0.13
1.3
13
15
16
2.2
3.1
0.6
ND b < 0.39
ND b
7.0
rat
1.0
1.6
4.1
1.3
0.9
0.5
1.5
8.9
1.3
1.3
1.5
5.9
1.8
1.5
2.2
2.3
1.7
0.10
0.13
0.93
3.4
2.7
1.9
dog
monkey
rat
rat
rat
rat
rat
dog
rat
dog
rat
dog
monkey
rat
10.3
5.4
1.1
0.9
0.6
0.5
12.1
5.8
4.9
5.5
15.9
7.3
5.0
8.6
11.0
12.4
23
24
25
26
4.0
1.2
32
33
0.16
0.29
0.29
0.65
0.18
0.38
0.43
0.17
0.17
38
35
39
40
41
3.1
7.2 c
7.2 c
rat
rat
a Units: t1/2 (h); Cl (L/h/kg); Cmax (ꢁg/mL); AUC0ꢀ24h (ꢁg*h/mL); F (%); Doses: 3 mg/kg IV and
oral for rat, 1 mg/kg IV and 2.5 mg/kg oral for dog and monkey; Vehicle: Capmul
b
PG8/Cremophor RH40 (90/10) for oral; DMSO/PEGꢀ400 (10/90) for IV. Unable to calculate
from 24 hour study. c Vehicle: PEGꢀ400/Tween/Poloxamer (70/20/10).
Table 3. In vitro physicochemical and metabolic stability data.
Solubility a
(ꢁM)
RLM
HLM
no.
LogD b
(% Remaining c) (% Remaining c)
ACS Paragon Plus Environment
14

Page 15 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
12
13
15
16
23
24
25
26
32
33
38
39
40
41
4.1
6.0
4.4
3.9
9.6
7.7
9.7
8.5
3.4
3.2
2.7
1.8
1.2
2.2
4.6
4.6
5.2
5.2
3.0
3.0
3.2
3.2
4.8
4.9
5.4
5.2
5.5
5.6
49
38
56
69
47
46
42
46
56
58
73
65
87
83
44
38
52
54
54
48
41
47
52
58
60
NAd
78
80
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
Kinetic solubility determined by chemiluminescent nitrogen detection (CLND) in 10 mM
phosphate buffer (pH 7.2). Determined by HPLC method (pH 7.4, 25 °C). Percentage of
parent compound remaining after 30 minute incubation. d Not available.
b
c
As a result of its picomolar potency and good pharmacokinetics, compound 38 was
selected for further in vitro virologic evaluation.30 As shown in Table 4, 38 demonstrated
picomolar potency not only against GT1a and GT1b, but also against GT2 through GT6.
Replicon resistance selection experiments in GT1 with compound 38 identified variants at
positions 28, 30 and 93 as the predominant resistance associated variants, although additional
minor variants were also observed (Table 5). In GT1a, variants M28V, L31V, and H58D
conferred 58ꢀ to 243ꢀfold resistance to 38. Single variants M28T, Q30R, and Y93C/S conferred
800ꢀ to 8965ꢀfold resistance, while Y93H/N conferred >40,000 fold resistance to 38. In GT1b,
the predominant variant selected in vitro with 38 was Y93H, which conferred 77ꢀfold resistance.
Double variants R30Q+Y93H and L31M+Y93H conferred 142ꢀ to 284ꢀfold resistance, whereas
all other double substitutions including Y93H in combination with substitutions at positions 28,
31, or 58 conferred more than 400ꢀfold resistance to 38. This resistance profile exhibits
considerable overlap with BMSꢀ790052, although there are some differences in foldꢀresistance
and variants selected.31 These observations of highly resistant mutants selected in vitro
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 40
1
2
3
4
5
6
underscores the need for combination therapy with DAAs that inhibit viral replication through
different mechanisms of action in order to increase the barrier to resistance.
7
8
9
Table 4. Antiviral activity of 38 in HCV replicon cell lines containing NS5A from GT1ꢀ6
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 Replicon
Genotype
Mean EC50
Nb
± Std. Dev. (pM)
1aꢀH77
9
14.1 ± 6.8
1bꢀCon1
2ac
9
6
6
6
6
5.0 ± 1.9
12.4 ± 2.7
4.3 ± 1.2
2bc
3ac
19.3 ± 5.8
1.71 ± 0.88
4ac
5ac
6ac
5
9
4.3 ± 0.9
415 ± 97
a
b
c
Assay contains 5% fetal bovine serum. Number of independent replicates. Chimeric 1bꢀ
Con1 replicon containing NS5A gene from a treatmentꢀnaïve patient isolate of the indicated GT.
Table 5. Selection of Resistant Variants in NS5A by 38 in GT1 Replicon Cell Lines
Genotype
Variant
M28T
Fold Resistancea Replication Efficiency
8965
58
100%
87%
60%
66%
24%
18%
25%
17%
127%
86%
73%
104%
M28V
Q30R
H58D
Y93C
Y93H
Y93N
800
243
1675
41,382
66,739
661
10
8
77
415
1a
L28T
L31F
L31V
Y93H
1b
L28M + Y93H
ACS Paragon Plus Environment
16

Page 17 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
R30Q + Y93H
L31F + Y93H
L31M + Y93H
L31V + Y93H
284
10,270
142
60%
35%
11%
24%
12,323
a Fold resistance is the ratio of variant EC50/wildꢀtype EC50.
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 assess the safety, tolerability, pharmacokinetics and antiviral activity of compound 38,
it was dosed for 3 days in HCV GT1ꢀinfected treatment naïve patients, with patients receiving
onceꢀdaily doses of 38 for three days at doses ranging from 5 mg to 200 mg.32 On day 3, doseꢀ
normalized Cmax and AUC values were similar across doses. Compound 38 demonstrated Cmax
values ranging from 5.7 to 442 ng/mL and a halfꢀlife ranging from 25 to 32 hours across the dose
groups. As shown in Figure 3, 38 decreased HCV RNA up to 3.10 log10 IU/mL during 3ꢀday
monotherapy with a nearly 3 log reduction observed in all dose groups. Compound 38 was
generally well tolerated at all doses and there were no serious or severe adverse events, no
clinically significant laboratory abnormalities, and no subjects discontinued. Most adverse events
were mild and were not dose related.32
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 40
1
2
3
4
5
6
7
8
9
1
0.5
0
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
ꢀ0.5
ꢀ1
Placebo
5 mg Compound 38
50 mg Compound 38
200 mg Compound 38
ꢀ1.5
ꢀ2
ꢀ2.5
ꢀ3
ꢀ3.5
0
1
2
3
Time (Days)
Figure 3. Mean decreases in HCV RNA from baseline during 3ꢀday monotherapy with
compound 38 in HCV GT1ꢀinfected treatmentꢀnaïve subjects.
Conclusions
There is a tremendous unmet medical need to discover new treatments for HCV infection
with improved efficacy, tolerability and convenience relative to the current standard of care. The
addition of inhibitors of HCV NS5A to the arsenal of DAAs that are available for use in
combination therapy holds significant promise for delivering such therapies. We discovered
ACS Paragon Plus Environment
18

Page 19 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compound 38, based on a chiral pyrrolidine core, which demonstrates picomolar potency against
GT1a and GT1b. Compound 38 is a panꢀgenotypic HCV inhibitor, with an EC50 range of 1.7 –
19.3 pM against genotypes 1a, 1b, 2a, 2b, 3a, 4a, and 5a, and 366 pM against genotype 6a.
Subsequent to its discovery, compound 38 has demonstrated excellent human pharmacokinetics
with a long 25 to 32 hour halfꢀlife, consistent with onceꢀdaily dosing. Compound 38 decreased
HCV RNA up to 3.10 log10 IU/mL during 3ꢀday monotherapy in treatmentꢀnaïve HCV GT1ꢀ
infected subjects, and it is currently undergoing phase 3 clinical trials in combination with an
NS3 protease inhibitor and an NS5B nonꢀnucleoside polymerase inhibitor, with and without
ribavirin.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
Biological Evaluation. The GT1a replicon construct (GenBank accession number NC004102)
contained the 5’ nontranslated region (NTR) from 1aꢀH77 followed by a firefly luciferase
reporter gene and the neomycin phosphotransferase (Neo) gene, which together comprised the
first cistron of the bicistronic replicon construct. This was followed by the EMCV IRES and
then the second cistron containing the 1aꢀH77 NS3ꢀNS5B coding region with adaptive mutations
E1202G, K1691R, K2040R, and S2204I, and finally the 1aꢀH77 3’ NTR. The GT1bꢀCon1
replicon construct (GenBank accession number AJ238799) contained the 5’ nontranslated region
(NTR) from 1bꢀCon1 followed by a firefly luciferase reporter gene and the neomycin
phosphotransferase (Neo) gene, which together comprised the first cistron of the bicistronic
replicon construct. This was followed by the EMCV IRES and the second cistron containing the
1bꢀCon1 NS3ꢀNS5B coding region with adaptive mutations K1609E, K1846T, and Y3005C, and
finally the 1bꢀCon1 3’ NTR. In addition, the 1bꢀCon1 replicon construct contained a poliovirus
IRES between the HCV 5’ NTR and the firefly luciferase gene. In order to assess the ability of
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 40
1
2
3
4
5
6
7
8
9
compounds to inhibit NS5A from nonꢀGT1 HCV, a 1bꢀCon1 replicon shuttle vector was
constructed that containing restriction sites just upstream of NS5A in the Cꢀterminus of NS4B,
and just after NS5A amino acid 214. Six chimeric subgenomic replicon cell lines were generated
for evaluation of the activity of compounds. The NS5A regions for generation of chimeric
replicons were derived from GT 2a, 2b, 3a, 4a, 5a, and 6a HCVꢀinfected patient sera. RTꢀPCR
was conducted on the viral RNA from each of these patient serum samples to generate a DNA
fragment encoding the first 214 amino acids of NS5A. The fragment was ligated into the replicon
shuttle vector, and in vitro transcribed into replicon RNA. This RNA was introducing into
human hepatoma cells (Huhꢀ7) to create stable replicon cell lines. The inhibitory effect of
compounds on HCV replication was determined (in the presence or absence of 40% HP), by
measuring decrease in luciferase signal. The percent inhibition of HCV replicon replication was
calculated for each compound concentration and the EC50 value was calculated using nonlinear
regression curve fitting to the 4ꢀparameter logistic equation in the GraphPad Prism 4/5 software.
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 methods describing the measurement of the effects of individual amino acid variants
on the activity of an inhibitor in HCV replicon cell culture assays were described previously.33
Briefly, the resistanceꢀassociated variants in NS5A were each introduced into GT 1aꢀH77, GT
1bꢀCon1, or chimeric replicons. In transient assays, the replicon containing the variant was
transfected via electroporation into a Huhꢀ7 derived cell line. The percent inhibition of HCV
replicon replication as determined by decreased luciferase signal was calculated for each
compound concentration, and the EC50 values were calculated as described above. Replication
capacity was calculated as a percentage of wildꢀtype replication using the following equation,
100x{(variant 4 day luciferase counts/wildꢀtype 4 day luciferase counts)/(variant 4 hr luciferase
counts/wildꢀtype 4 hr luciferase counts)}.
ACS Paragon Plus Environment
20

Page 21 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chemistry. General Procedures. Unless otherwise noted, all materials were obtained from
commercial suppliers and used without further purification. Anhydrous solvents were obtained
from Aldrich (Milwaukee, WI) and used directly. All reactions involving airꢀ or moistureꢀ
sensitive reagents were performed under a nitrogen or argon atmosphere. NMR spectra were
collected on either a Varian Inova 400 MHz or 500 MHz spectrometer as specified with
chemical shifts given in ppm (δ) and are referenced to an internal standard of tetramethylsilane
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(δ 0.00). H – 1H couplings are assumed to be firstꢀorder and peak multiplicities are reported in
the usual manner. All final compounds were purified to >95% purity as determined by two
independent HPLC analyses. Reverse phase HPLC purity determinations were performed on a
Waters e2695 Separation Module equipped with a Waters 2489 UVꢀVisible Detector and a
YMC ODSꢀA, 5.5 ꢂm, 120Å, 4.6 X 150 mm column and an XTerra RP18, 5 ꢂm, 4.6 X 150 mm
column. Silica gel chromatography purifications were performed using either glass columns
packed with silica gel 60 (230ꢀ400 mesh) or prepacked silica gel cartridges (various vendors /
various automated chromatography systems). Mass spectral (MS) ESI data were determined on a
ThermoꢀFinnigan SSQ7000 spectrometer. Mass spectral (MS) APCI data were determined on
either a ThermoꢀFinnigan Navigator or ThermoꢀFinnigan MSQꢀPlus mass spectrometer.
Analytical LCMS was performed on an Agilent Series 1100 HPLC system using a Phenomenex
Luna CombiꢀHTS C8(2) 5 ꢂm 100Å (2.1 mm × 50 mm) column. The system was equipped with
an autosampler and coupled to a Finnigan Thermoquest atmospheric pressure chemical
ionization (APCI) mass spectrometer.
1,4-Bis(4-nitrophenyl)butane-1,4-dione (3). Anhydrous zinc (II) chloride (2.73 g, 20.0 mmol)
was stirred in dry benzene (15 mL) while diethylamine (1.6 mL, 15.0 mmol) and tꢀbutanol (1.4
mL, 15.0 mmol) were added, and the resulting mixture was stirred at room temperature for 90
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 40
1
2
3
4
5
6
7
8
9
minutes to give a cloudy solution. To this mixture was added 2ꢀbromoꢀ1ꢀ(4ꢀnitrophenyl)ethanone
(2.44 g, 10.0 mmol) and 1ꢀ(4ꢀnitrophenyl)ethanone (2.48 g, 15.0 mmol), and the resulting
mixture was stirred at room temperature for 16 hours. The mixture was poured into water (50
mL) and extracted with ethyl acetate (3 x 50 mL). The combined organic layers were dried over
Na2SO4, filtered and concentrated. The resulting residue was triturated with dichloromethane to
give an orange solid that was collected by filtration and dried to give the product (2.0 g, 61%
yield). 1H NMR (500 MHz, CDCl3) δ 3.46 (s, 4H), 8.09 – 8.16 (m, 4H), 8.27 – 8.32 (m, 4H).
1,4-Bis(4-nitrophenyl)butane-1,4-diol (4c). To a solution of 3 (1.0 g, 3.05 mmol) in anhydrous
THF (30 mL) at 0 °C was added sodium borohydride (0.357 g, 9.44 mmol). The resulting
mixture was stirred at 50 °C for 16 hours. The cooled mixture was poured into water, extracted
with ethyl acetate, dried over Na2SO4, filtered and concentrated in vacuo. The resulting solid was
triturated with dichloromethane to give a tan solid that was collected by filtration and dried to
give 4c (0.82 g, 81% yield). 1H NMR (500 MHz, DMSOꢀd6) δ 1.52 – 1.75 (m, 4H), 4.62 – 4.77
(m, 2H), 5.47 – 5.55 (m, 2H), 7.51 – 7.60 (m, 4H), 8.15 – 8.22 (m, 4H).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1,4-Bis(4-nitrophenyl)butane-1,4-diyl dimethanesulfonate (5c). To a stirred mixture of 4c
(0.950 g, 2.86 mmol) in anhydrous dichloromethane (30 mL) at 0 °C was added triethylamine
(1.2 mL, 8.58 mmol), followed by dropwise addition of methanesulfonyl chloride (0.56 mL, 7.15
mmol). The resulting mixture was stirred at 0 °C for 30 minutes, during which time 4c slowly
went into solution. After stirring an additional 1 hour at 0 °C, saturated aqueous NH4Cl (4 mL)
was added, and the resulting mixture was stirred at room temperature for 20 minutes and then
transferred to a separatory funnel and washed with water (2 x 10 mL). The organic layer was
separated, dried over Na2SO4, and filtered. Hexanes (10 mL) was added to give an orange
ACS Paragon Plus Environment
22

Page 23 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
precipitate that was collected by filtration and dried to give the product (1.4 g, quantitative)
which was used without further purification.
7
8
9
1-(4-Fluorophenyl)-2,5-bis(4-nitrophenyl)pyrrolidine (47). To a solution of 5c (1.4 g, 2.86
mmol) in anhydrous N,Nꢀdimethylformamide (4 mL) was added 4ꢀfluoroaniline (2.0 mL, 20.8
mmol), and the resulting mixture was stirred at 50 °C for 6 hours. The resulting mixture was
cooled and partitioned between 1 N aqueous HCl (40 mL) and ethyl acetate (3 x 40 mL), and the
combined organic layers were dried over Na2SO4. The drying agent was filtered off, and the
solvent was evaporated to give a crude product that was purified by column chromatography on
silica gel using a solvent gradient of 0ꢀ50% ethyl acetate in hexanes to give 47 as a 2.4 to 1 ratio
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
of transꢀ to cisꢀsubstituted pyrrolidine isomers (0.385 g, 33%). H NMR (400 MHz, CDCl3) δ
1.84 – 2.04 (m, 2H), 2.50 – 2.66 (m, 2H), 4.82 – 4.90 (m, 0.6H), 5.27 – 5.34 (m, 1.4H), 6.14 –
6.21 (m, 1.4H), 6.27 – 6.34 (m, 0.6H), 6.71 – 6.89 (m, 2H), 7.33 – 7.40 (m, 2.8H), 7.63 – 7.69
(m, 1.2H), 8.17 – 8.23 (m, 2.8H), 8.25 – 8.31 (m, 1.2H).
4,4’-(trans-1-(4-Fluorophenyl)pyrrolidine-2,5-diyl)dianiline (48a) and 4,4’-(cis-1-(4-
Fluorophenyl)pyrrolidine-2,5-diyl)dianiline (48b). To a solution of 47 (0.38 g, 0.93 mmol) in
THF (4 mL) and ethanol (4 mL) were added a solution of NH4Cl (0.15 g, 2.80 mmol) in water (1
mL) and iron powder (0.26 g, 4.66 mmol). The resulting mixture was stirred at 80 °C for 90
minutes. The cooled mixture was filtered through Celite® and concentrated in vacuo.
Purification of the crude product by column chromatography on silica gel using a solvent
gradient of 0ꢀ100% ethyl acetate in hexanes allowed for the separation of the transꢀpyrrolidine
and cisꢀpyrrolidine isomers. The first eluting component was the transꢀpyrrolidine isomer 48a
(55 mg, 17%). 1H NMR (400 MHz, DMSOꢀd6) δ 1.48 – 1.65 (m, 2H), 2.31 – 2.44 (m, 2H), 4.88
(s, 4H), 4.96 (d, J = 6.7 Hz, 2H), 6.16 – 6.27 (m, 2H), 6.47 (d, J = 8.3 Hz, 4H), 6.71 – 6.78 (m,
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 40
1
2
3
4
2H), 6.82 (d, J = 8.3 Hz, 4H). The second eluting component was the cisꢀpyrrolidine isomer 48b
5
6
1
(52 mg, 16%). H NMR (400 MHz, DMSO) δ 1.70 – 1.83 (m, 2H), 2.20 – 2.31 (m, 2H), 4.48 (t,
7
8
9
J = 5.3 Hz, 2H), 4.93 (s, 4H), 6.33 – 6.43 (m, 2H), 6.56 (d, J = 8.4 Hz, 4H), 6.83 (t, J = 8.9 Hz,
2H), 7.12 (d, J = 8.3 Hz, 4H).
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
(2S,2'S)-tert-Butyl 2,2'-(4,4'-(trans-1-(4-fluorophenyl)pyrrolidine-2,5-diyl)bis(4,1-
phenylene))bis(azanediyl)bis(oxomethylene)dipyrrolidine-1-carboxylate (49). A mixture of
the transꢀpyrrolidine isomer of 48a (0.050 g, 0.144 mmol), (S)ꢀ1ꢀ(tert-
butoxycarbonyl)pyrrolidineꢀ2ꢀcarboxylic acid (0.077 g, 0.360 mmol), Hünig’s base (0.075 mL,
0.432 mmol), and HATU (0.137 g, 0.360 mmol) in anhydrous dimethylsulfoxide (1 mL) was
stirred at room temperature for 90 minutes. The mixture was partitioned between ethyl acetate
and water, and the organic layer was dried over Na2SO4. The drying agent was filtered off, and
the solvent was removed in vacuo. The crude product was purified by column chromatography
on silica gel using a solvent gradient of 0ꢀ100% ethyl acetate in hexanes to give 49 as a light
1
orange solid (0.107 g, quant.). H NMR (400 MHz, DMSOꢀd6) δ 1.24 – 1.44 (m, 18H), 1.59 –
1.71 (m, 2H), 1.72 – 1.96 (m, 8H), 2.07 – 2.27 (m, 2H), 3.37 – 3.50 (m, 2H), 4.13 – 4.28 (m,
2H), 4.34 (t, J = 5.1 Hz, 2H), 5.17 (d, J = 4.2 Hz, 2H), 6.15 – 6.28 (m, 2H), 6.78 (t, J = 8.9 Hz,
2H), 7.14 (d, J = 7.9 Hz, 4H), 7.46 – 7.56 (m, 4H), 9.87 – 9.97 (m, 2H); MS (ESI) m/z 742
(M+H)+.
Dimethyl (2S,2'S)-1,1'-((2S,2'S)-2,2'-(4,4'-((2S,5S)-1-(4-fluorophenyl)pyrrolidine-2,5-
diyl)bis(4,1-phenylene))bis(azanediyl)bis(oxomethylene)bis(pyrrolidine-2,1-diyl))bis(3-
methyl-1-oxobutane-2,1-diyl)dicarbamate (12) and Dimethyl (2S,2'S)-1,1'-((2S,2'S)-2,2'-
(4,4'-((2R,5R)-1-(4-fluorophenyl)pyrrolidine-2,5-diyl)bis(4,1-
phenylene))bis(azanediyl)bis(oxomethylene)bis(pyrrolidine-2,1-diyl))bis(3-methyl-1-
ACS Paragon Plus Environment
24

Page 25 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
oxobutane-2,1-diyl)dicarbamate (13). A solution of 49 (107 mg, 0.148 mmol) in
dichloromethane (1 mL) and trifluoroacetic acid (2 mL) was stirred at room temperature for 45
minutes and then concentrated in vacuo. The residue was partitioned between dichloromethane
and saturated aqueous NaHCO3, and the organic layer was dried over Na2SO4, filtered and
concentrated in vacuo. The residue was dissolved in anhydrous dimethylsulfoxide (1.5 mL), and
(S)ꢀ2ꢀ(methoxycarbonylamino)ꢀ3ꢀmethylbutanoic acid (65 mg, 0.37 mmol), HATU (141 mg,
0.37 mmol) and Hünig’s base (0.077 mL, 0.44 mmol) were added. The resulting mixture was
stirred at room temperature for 90 minutes, and was partitioned between water and ethyl acetate
(3x). The organic layer was dried over Na2SO4, filtered and concentrated in vacuo, and the crude
product was purified by column chromatography on silica gel using a solvent gradient of 0ꢀ5%
methanol in dichloromethane to give a mixture of transꢀpyrrolidine isomers (110 mg, 87%). The
mixture was separated on a Chiralpak ADꢀH column eluting with a mixture of 1 part (2:1
isopropanol:ethanol) and 1 part hexanes (0.1% TFA). Compound 12 was the first of 2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
stereoisomers to elute. H NMR (400 MHz, DMSOꢀd6) δ 0.84 – 0.97 (m, 12H), 1.64 (d, J = 5.6
Hz, 2H), 1.88 (m, 6H), 1.95 – 2.05 (m, 2H), 2.08 – 2.19 (m, 2H), 3.52 (s, 6H), 3.58 – 3.66 (m,
2H), 3.76 – 3.85 (m, 2H), 4.02 (t, J = 8.5 Hz, 2H), 4.42 (dd, J = 8.0, 4.9 Hz, 2H), 5.15 (d, J = 6.5
Hz, 2H), 6.20 (dd, J = 9.2, 4.4 Hz, 2H), 6.78 (t, J = 8.9 Hz, 2H), 7.13 (d, J = 8.5 Hz, 4H), 7.31
(d, J = 8.4 Hz, 2H), 7.50 (d, J = 8.5 Hz, 4H), 9.99 (s, 2H); MS (ESI) m/z 856.9 (M+H)+.
Compound 13 was the second of 2 stereoisomers to elute. 1H NMR (400 MHz, DMSOꢀd6) δ
0.82 – 0.97 (m, 12H), 1.65 (d, J = 5.7 Hz, 2H), 1.80 – 2.05 (m, 8H), 2.08 – 2.20 (m, 2H), 3.52 (s,
6H), 3.57 – 3.68 (m, 2H), 3.76 – 3.87 (m, 2H), 4.01 (t, J = 8.5 Hz, 2H), 4.42 (dd, J = 8.0, 4.9 Hz,
2H) 5.16 (d, J = 6.4 Hz, 2H), 6.20 (dd, J = 9.2, 4.5 Hz, 2H), 6.77 (t, J = 9.0 Hz, 2H), 7.12 (d, J =
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 40
1
2
3
4
5
6
8.6 Hz, 4H), 7.30 (d, J = 8.4 Hz, 2H), 7.50 (d, J = 8.5 Hz, 4H), 9.98 (s, 2H). MS (ESI) m/z 856.4
(M+H)+.
7
8
9
Dimethyl (2S,2'S)-1,1'-((2S,2'S)-2,2'-(4,4'-((2S,5R)-1-(4-fluorophenyl)pyrrolidine-2,5-
diyl)bis(4,1-phenylene))bis(azanediyl)bis(oxomethylene)bis(pyrrolidine-2,1-diyl))bis(3-
methyl-1-oxobutane-2,1-diyl)dicarbamate 2,2,2-trifluoroacetate (14). Compound 14 was
prepared using the methods described for the synthesis of 12 and 13, substituting a mixture of
both cisꢀpyrrolidine 48a and transꢀpyrrolidine of 48b. Compound 14 was separated from the
transꢀpyrrolidine component by Cꢀ18 HPLC using a gradient of 10ꢀ100% acetonitrile in water
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(0.1% TFA), and was the second of 2 isomers to elute. H NMR (400 MHz, DMSOꢀd6) δ 1.15
(dd, J = 6.7, 2.1 Hz, 6H) 1.22 (dd, J = 6.8, 2.1 Hz, 6H) 3.87 – 3.97 (m, 2H) 3.84 (s, 6H) 3.99 –
4.09 (m, 2H) 4.42 – 4.52 (m, 2H) 4.67 – 4.78 (m, 2H) 4.87 – 4.97 (m, 2H) 6.23 (d, J = 8.4 Hz,
2H) 6.61 – 6.71 (m, 2H) 7.01 (t, J = 9.0 Hz, 2H) 7.68 – 7.73 (m, 4H) 7.77 – 7.83 (m, 4H) 8.26 (s,
2H) 9.35 (s, 2H). MS (ESI) m/z 856.1 (M+H)+.
1-(4-tert-Butylphenyl)-2,5-bis(4-nitrophenyl)pyrrolidine (50). A mixture of 5c (3.67 g, 7.51
mmol) and 4ꢀtert-butylaniline (11.86 mL, 75 mmol) in anhydrous N,Nꢀdimethylformamide (40
mL) was stirred at 50 °C for 4 hours. The mixture was partitioned between 1M aqueous HCl and
ethyl acetate, and the organic layer was washed with brine and dried over Na2SO4. The drying
agent was filtered off, and the solution was concentrated in vacuo to give a crude product that
was purified by column chromatography on silica gel using a solvent gradient of 5ꢀ30% ethyl
acetate in hexanes to give 50 as a 3:1 mixture of transꢀpyrrolidine and cisꢀpyrrolidine isomers
1
(1.21 g, 36%). H NMR (400 MHz, DMSOꢀd6) δ 1.11 (s, 3H), 1.15 (s, 6H), 1.69 – 1.79 (m,
0.5H), 1.79 – 1.91 (m, 1.5H), 2.49 – 2.61 (m, 2H), 4.94 (t, J = 5.4 Hz, 1.5H), 5.47 (d, J = 7.1 Hz,
0.5H), 6.19 (d, J = 8.8 Hz, 0.5H), 6.28 (d, J = 8.8 Hz, 1.5H), 6.99 (d, J = 8.8 Hz, 0.5H), 7.08 (d,
ACS Paragon Plus Environment
26

Page 27 of 40
Journal of Medicinal Chemistry
1
2
3
4
5
6
J = 8.8 Hz, 1.5H), 7.53 (d, J = 8.7 Hz, 1H), 7.81 (d, J = 8.7 Hz, 3H), 8.21 (d, J = 8.7 Hz, 1H),
8.27 (d, J = 8.7 Hz, 3H).
7
8
9
4,4'-(trans-1-(4-tert-butylphenyl)pyrrolidine-2,5-diyl)dianiline (51). A mixture of 50 (1.10 g,
2.47 mmol) and platinum (IV) oxide (0.22 g, 0.97 mmol) in 1:1 ethanol:THF (40 mL) was stirred
under hydrogen gas at 30 psi for 30 minutes. The resulting mixture was filtered and
concentrated in vacuo to give a crude product which was purified by column chromatography on
silica gel using a solvent gradient of 20ꢀ60% ethyl acetate in hexanes. The desired transꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
pyrrolidine isomer was the first of 2 stereoisomers to elute, providing 51 (510 mg, 54%). H
NMR (400 MHz, DMSOꢀd6) δ 1.12 (s, 9H), 1.47 – 1.62 (m, 2H), 2.32 – 2.44 (m, 2H), 4.86 (s,
4H), 4.96 (d, J = 6.7 Hz, 2H), 6.19 (d, J = 8.8 Hz, 2H), 6.47 (d, J = 8.3 Hz, 4H), 6.82 (d, J = 8.3
Hz, 4H), 6.91 (d, J = 8.8 Hz, 2H).
(2S,2'S)-tert-Butyl 2,2'-(4,4'-(trans-1-(4-tert-butylphenyl)pyrrolidine-2,5-diyl)bis(4,1-
phenylene)bis(azanediyl)bis(oxomethylene))dipyrrolidine-1-carboxylate (52). A mixture of
51 (250 mg, 0.648 mmol), (S)ꢀ1ꢀ(tert-butoxycarbonyl)pyrrolidineꢀ2ꢀcarboxylic acid (307 mg,
1.427 mmol), HATU (542 mg, 1.427 mmol) and Hünig's base (0.453 mL, 2.59 mmol) in
dimethylsulfoxide (10 mL) was stirred at room temperature for 1 hour. The mixture was
partitioned between ethyl acetate and water, and the organic layer was washed with brine, dried
with Na2SO4, filtered and concentrated in vacuo. The residue was purified by column
chromatography on silica gel eluting with a solvent gradient of 10ꢀ50% ethyl acetate in hexanes
1
to give compound 52 as a colorless solid (0.500 g, 99%). H NMR (400 MHz, DMSOꢀd6) δ 1.11
(s, 9H), 1.28 (m, 11H), 1.39 (m, 7H), 1.58 – 1.69 (m, 2H), 1.71 – 1.95 (m, 8H), 2.06 – 2.25 (m,
2H), 2.42 – 2.48 (m, 2H), 3.35 – 3.47 (m, 2H), 4.12 – 4.28 (m, 2H), 5.16 (d, J = 5.4 Hz, 2H),
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 40
1
2
3
4
5
6
6.19 (d, J = 8.3 Hz, 2H), 6.94 (d, J = 8.6 Hz, 2H), 7.08 – 7.22 (m, 4H), 7.50 (t, J = 7.8 Hz, 4H),
9.91 (m, 2H); MS (ESI) m/z 780.4 (M+H)+.
7
8
9
Dimethyl (2S,2'S)-1,1'-((2S,2'S)-2,2'-(4,4'-((2S,5S)-1-(4-tert-butylphenyl)pyrrolidine-
2,5-diyl)bis(4,1-phenylene))bis(azanediyl)bis(oxomethylene)bis(pyrrolidine-2,1-diyl))bis(3-
methyl-1-oxobutane-2,1-diyl)dicarbamate (38) and Dimethyl (2S,2'S)-1,1'-((2S,2'S)-2,2'-
(4,4'-((2R,5R)-1-(4-tert-butylphenyl)pyrrolidine-2,5-diyl)bis(4,1-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
phenylene))bis(azanediyl)bis(oxomethylene)bis(pyrrolidine-2,1-diyl))bis(3-methyl-1-
oxobutane-2,1-diyl)dicarbamate (39). A solution of 52 (498 mg, 0.638 mmol) in
trifluoroacetic acid (6 mL) and dichloromethane (4 mL) was stirred at room temperature for 1
hour. The solution was concentrated to dryness, and the residue was partitioned between
saturated aqueous NaHCO3 and a 3:1 mixture of chloroform and isopropanol (2x). The
combined organic layers were dried over Na2SO4, filtered and concentrated in vacuo. A portion
of this material (232 mg, 0.400 mmol) was dissolved in DMF (10 mL) and cooled to 0 °C,
followed by addition of (S)ꢀ2ꢀ(methoxycarbonylamino)ꢀ3ꢀmethylbutanoic acid (210 mg, 1.200
mmol), 1ꢀhydroxybenzotriazole hydrate (184 mg, 1.200 mmol), Nꢀmethylmorpholine (0.220 mL,
2.000 mmol), and Nꢀ(3ꢀdimethylaminopropyl)ꢀN’ꢀethylcarbodiimide hydrochloride (230 mg,
1.200 mmol). The resulting mixture was stirred at room temperature for 16 hours. The mixture
was partitioned between ethyl acetate and water, and the organic layer was washed with
saturated aqueous NaHCO3, brine (2x), and dried with Na2SO4. The drying agent was filtered
off and the solution was concentrated in vacuo to give a crude product that was purified by
column chromatography on silica gel eluting with a solvent gradient of 2ꢀ8% methanol in
dichloromethane to give a 1:1 mixture of transꢀpyrrolidine isomers (290 mg, 96%). The mixture
was separated on a Chiralpak ADꢀH column eluting with a mixture of 1 part (2:1
ACS Paragon Plus Environment
28

Page 29 of 40
Journal of Medicinal Chemistry
1
2
3
4
isopropanol:ethanol) and 2 parts hexanes (0.1% TFA). Compound 38 was the first of 2
5
6
7
1
stereoisomers to elute (101 mg, 99% ee by chiral HPLC). H NMR (400 MHz, DMSOꢀd6) δ
8
9
0.88 (d, J = 6.61 Hz, 6H), 0.93 (d, J = 6.72 Hz, 6H), 1.11 (s, 9H), 1.63 (d, J = 5.42 Hz, 2H), 1.80
– 2.04 (m, 8H), 2.09 – 2.19 (m, 2H), 2.44 – 2.47 (m, 2H), 3.52 (s, 6H), 3.59 – 3.66 (m, 2H), 3.77
– 3.84 (m, 2H), 4.02 (t, J = 8.40 Hz, 2H), 4.42 (dd, J = 7.86, 4.83 Hz, 2H), 5.14 (d, J = 6.18 Hz,
2H), 6.17 (d, J = 8.67 Hz, 2H), 6.94 (d, J = 8.78 Hz, 2H), 7.13 (d, J = 8.46 Hz, 4H), 7.31 (d, J =
8.35 Hz, 2H), 7.50 (d, J = 8.35 Hz, 4H), 9.98 (s, 2H); MS (ESI) m/z 894.9 (M+H)+. 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
1
39 was the second of 2 stereoisomers to elute. H NMR (400 MHz, DMSOꢀd6) δ 0.87 (d, J =
6.51 Hz, 6H), 0.92 (d, J = 6.72 Hz, 6H), 1.11 (s, 9H), 1.63 (d, J = 5.53 Hz, 2H), 1.82 – 2.04 (m,
8H), 2.09 – 2.18 (m, 2H), 2.41 – 2.47 (m, 2H), 3.52 (s, 6H), 3.58 – 3.67 (m, 2H), 3.75 – 3.84 (m,
2H), 4.02 (t, J = 7.26 Hz, 2H), 4.43 (dd, J = 7.92, 4.88 Hz, 2H), 5.14 (d, J = 6.18 Hz, 2H), 6.17
(d, J = 8.78 Hz, 2H), 6.94 (d, J = 8.67 Hz, 2H), 7.12 (d, J = 8.46 Hz, 4H), 7.31 (d, J = 8.35 Hz,
2H), 7.49 (d, J = 8.46 Hz, 4H), 9.98 (s, 2H). MS (ESI) m/z 895.0 (M+H)+.
(1R,4R)-1,4-Bis(4-nitrophenyl)butane-1,4-diol (4a). To a stirred mixture of (S)ꢀ(ꢀ)ꢀα,αꢀ
diphenylꢀ2ꢀpyrrolidinemethanol (2.71 g, 10.70 mmol) in anhydrous THF (15 mL) at 23 °C was
added trimethyl borate (1.44 g, 13.86 mmol), and the resulting solution was stirred at room
temperature for 1 hour. The solution was cooled to 16ꢀ19 °C, and N,Nꢀdiethylaniline borane
(21.45 g, 132 mmol) was added dropwise via syringe over 3ꢀ5 minutes (caution: vigorous H2
evolution), while the internal temperature was maintained at 16ꢀ19 °C. After 15 minutes, the H2
evolution had ceased. The resulting borane mixture was cooled to 11°C before being transferred
via cannula over 3ꢀ5 minutes to a separate flask that contained a stirred slurry of 3 (22.04 g, 63.8
mmol) in anhydrous THF (80 mL). During the transfer period, the internal temperature of the
slurry rose to 16 °C. After the addition was complete, the reaction was maintained at room
ACS Paragon Plus Environment
29