Novel spiroketal pyrrolidine GSK2336805 potently inhibits key hepatitis C virus genotype 1b mutants: From lead to clinical compound

Article abstract of DOI:10.1021/jm4013104

Rapid clinical progress of hepatitis C virus (HCV) replication inhibitors, including these selecting for resistance in the NS5A region (NS5A inhibitors), promises to revolutionize HCV treatment. Herein, we describe our explorations of diverse spiropyrrolidine motifs in novel NS5A inhibitors and a proposed interaction model. We discovered that the 1,4-dioxa-7-azaspiro[4.4]nonane motif in inhibitor 41H (GSK2236805) supported high potency against genotypes 1a and 1b as well as in genotype 1b L31V and Y93H mutants. Consistent with this, 41H potently suppressed HCV RNA in the 20-day RNA reduction assay. Pharmacokinetic and safety data supported further progression of 41H to the clinic.

Full text of DOI:10.1021/jm4013104

Article
pubs.acs.org/jmc
Novel Spiroketal Pyrrolidine GSK2336805 Potently Inhibits Key
Hepatitis C Virus Genotype 1b Mutants: From Lead to Clinical
Compound
Wieslaw M. Kazmierski,* Andrew Maynard, Maosheng Duan,† Sam Baskaran, Janos Botyanszki,
Renae Crosby, Scott Dickerson, Matthew Tallant, Rick Grimes, Robert Hamatake, Martin Leivers,
Christopher D. Roberts, and Jill Walker
GlaxoSmithKline, 5 Moore Drive, Research Triangle Park, North Carolina 27709-3398, United States
ABSTRACT: Rapid clinical progress of hepatitis C virus (HCV) replication inhibitors, including these selecting for resistance in
the NS5A region (NS5A inhibitors), promises to revolutionize HCV treatment. Herein, we describe our explorations of diverse
spiropyrrolidine motifs in novel NS5A inhibitors and a proposed interaction model. We discovered that the 1,4-dioxa-7-
azaspiro[4.4]nonane motif in inhibitor 41H (GSK2236805) supported high potency against genotypes 1a and 1b as well as in
genotype 1b L31V and Y93H mutants. Consistent with this, 41H potently suppressed HCV RNA in the 20-day RNA reduction
assay. Pharmacokinetic and safety data supported further progression of 41H to the clinic.
a
Table 1. Potencies of 1−6b in gt1a and gt1b Replicon Assays
INTRODUCTION
■
Hepatitis C virus (HCV) affects more than 170 million
individuals worldwide.1 About 80% of all HCV infections
become chronic and can lead to fibrosis, cirrhosis, and
hepatocellular carcinoma.2,3 Until recently, a combination of
injectable peginterferon and oral ribavirin had been the standard
of care for the treatment of HCV infection. The use of interferon
can cause unpleasant flulike symptoms and fatigue, resulting in
significant treatment discontinuations and failure. The recently
introduced HCV protease inhibitors increased the sustained viral
response (SVR), considered a virologic cure, for both treatment-
naive and experienced patients. In addition, rapid clinical
development of other anti-HCV agents, including NS5A
inhibitors, promise a tantalizing opportunity for an all-oral,
non-interferon HCV therapy in the near future.4−6
replicon pEC50
compd
gt1a
gt1b
1
<4.7
<4.7
<4.7
8.8
5.6
6.9
2
3
9.6
4
11.2
11.7
11.0
6.7
5
8.8
6
10.4
6.3
6a
6b
6.1
6.9
a
pEC50 = −log EC50.
Our own medicinal chemistry efforts followed the initial hit 1
discovered through high throughput HCV replicon screen
(Figure 1).16 The combination of attractive pEC50 in gt1b HCV
replicon assay17 and low molecular weight supported further
optimization efforts. Further derivatization of 1 with the p-
cyclopropylcarboxamide moiety resulted in compound 2 and
about 10-fold potency improvement in replicon gt1b assay.
Subsequent use of the aromatic amide and spiroproline motifs in
compound 3 brought about additional and substantial potency
Daclatasvir (BMS-790052, 6) is the first-reported NS5A
inhibitor with high potency in both HCV genotype 1a (gt1a) and
HCV genotype 1b (gt1b) (Table 1).7,8 In their elegant and
insightful work, BMS scientists identified potent “dimeric” HCV
inhibitors in the mixture of biologically active species obtained by
incubating the monomeric iminothiazolidinone HCV inhibitor
BMS-824 in a cell culture medium.9 Removal of the
iminothiazolidinone moiety from dimeric products resulted in
potent trans-stilbene class of inhibitors,10 which through several
other series11,12 led to “dimeric” biphenyl scaffold in 6.7,8
Additional NS5A inhibitors from a number of laboratories have
been reported in patent and scientific literature.13−15
Special Issue: HCV Therapies
Received: August 26, 2013
Published: February 26, 2014
© 2014 American Chemical Society
2058
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
Figure 1. (a) Evolution of monomeric ligand 1 into pseudo-dimeric 4 and 5. (b) Less potent fragments: “monomer” 6a and truncated peptidic caps 6b.
increase in replicon gt1b but not gt1a. Enhancing the latter
proved challenging but could be accomplished by incorporating
the biphenyl motif in spirooxazolidines 4 and 5 (Figure 1).18,19
RC formation, resulting in inert localization of NS5A to lipid
droplet (LD) surfaces.
Consistent with the lack of observed interactions between
inhibitors and purified NS5A in isolation, no NS5A inhibitor
cocrystal structures have been published to date. Two published
structures of the N-terminal NS5A zinc finger (ZnF) domain
reveal formation of a homodimer, although the relative
arrangements of the monomer units are quite different,
suggesting the possibility of multiple protein−protein or
protein−RNA interaction roles of NS5A.24,25 Symmetry suggests
dimeric inhibitors interact with a dimeric form of NS5A,
experimentally supported by dimeric NS5A pull-down, cross-
linked with a dimeric bis-azide inhibitor (BMS-351).23 Utilizing
the Love et al. structure, which was more consistent with internal
data, manual docking of dimeric inhibitors (S. Dickerson,
unpublished) suggested a C2-symmetric orientation of inhibitors
along the dimer interface (Figure 2) that is distinct from recently
published docking studies23 utilizing the Tellinghuisen et al.
structure. To better understand the context of loop mutations
(sites 28, 30, 31), the relative orientation of the N-terminal α-
helices and differential monomeric vs dimeric inhibitor SAR, a
NS5A gt1b domain I homology model was subsequently
constructed by splicing the ZnF domain (aa 33−191)25 and N-
terminal amphiphilic α-helix (aa 1−31)26 units.14 Our model
suggests a symmetric binding mode of 5 along the dimer
PUTATIVE BINDING MODE OF DIMERIC NS5A
INHIBITORS
■
The exact mechanism(s) of HCV inhibition by NS5A inhibitors
remains a subject of ongoing investigations.14,20−23 Assigning
NS5A as the target of synthetic inhibitors has been complicated
by the lack of definitive NS5A functional assays, as well as the lack
of observed inhibitor−NS5A binding using purified NS5A, in
isolation from the HCV replication complex (RC). However,
mutation data and chimera studies map the sensitivity of HCV
inhibition to the N-terminus of NS5A, particularly residue
positions 30, 31, and 93, in the case of dimeric inhibitors.7,21,22
Moreover, application of biotin-tagged and photoaffinity labeled
inhibitors in protein pull-down experiments implicates associa-
tion between dimeric inhibitors and NS5A in the functional
context of the cellular RC.22,23 Cell-based phenotypic studies also
demonstrated that dimeric HCV inhibitors induce relocalization
of NS5A from the endoplasmic reticulum to lipid droplets in
replicon 1b cells, concomitant with the onset of replicon
inhibition.22 Inhibitor action was interpreted as the binding of a
NS5A site during de novo RC formation which conformationally
restricts NS5A, preventing sequestration to and promotion of
2059
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
interface whereby the amino acid moieties of 5 interact with
cavities formed by the connecting protein loops (aa 27−37) at
the ZnF and N-terminal α-helix junctions (Figure 2). The biaryl
spacing of the inhibitor core is also consistent with the distance
between the resistant mutations7 at the dimer interface (blue,
Figure 2). The model suggests that the peptidic “caps” of dimeric
compounds, including 5, make buried contacts at the dimer
interface, where resistant mutations are prevalent. The proposed
binding mode would restrict the movement of the N-terminal α-
helix relative to the ZnF domain, presumably impairing the cis-
acting function reported for NS5A.27 This binding mode could
lock NS5A in an inactive conformational state, as attributed to
relocalization of NS5A from RC to LD.22 Furthermore, the
model provides a rationale for dimeric inhibitors; considering the
peptidic caps as individually bound fragments, cooperative
delivery of the caps (via a biaryl core spanning the interface)
yields an exponential increase in affinity: Kd,dimer = exp-
[−ΔGdimer/(RT)] ≈ exp[−2ΔGmonomer/(RT)] = Kd,monomer
×
K
d,monomer. The (buried) peptidic caps primarily drive the
Figure 2. Putative binding mode of compound 5 with NS5A gt1b
domain I. Compound 5 (sticks, white carbons) spans the dimer
interface, positioning the peptidic caps in the cavities formed between
the ZnF domain and N-terminal α-helix units of the NS5A homodimer.
The monomer units of the homodimer are color-coded red and cyan.
Reported daclatasvir resistant mutation sites (gt1a, M28T, Q30H/R,
L31M, Y93C; gt1b, L31V, Y93H),7 color-coded blue, border the
terminal peptidic caps of dimeric inhibitors. In contrast, gt1a/gt1b
residue differences (H/Q54, R/T56, H/P58, E/Q62), color-coded
yellow, are located along the central dimer interface, flanking the
inhibitor biaryl core.
potency, while the core endows cooperative binding. Thus,
weakly bound caps (Kd,monomer ≈ 10 μM), when spatially coupled,
result in high-affinity inhibitors (Kd,dimer ≈ Kd,monomer × Kd,monomer
≈ 0.1 nM). Accordingly, a sharp drop in potency is observed with
“monomeric” (6a) or “truncated peptidic cap” (6b) compounds
(Figure 1, Table 1). This is also consistent with picomolar
potencies of dimeric HCV NS5A inhibitors containing a variety
of central cores of similar lengths, including biphenyl,
benzimidazole, napthyl, quinoline, isoquinoline, and other rigid
cores synthesized in our NS5A program.28−30 While the model
Figure 3. Synthesis of inhibitors. Reagents and conditions are as follows: (a) dioxane, Et3N, Boc2O; (b) dioxane, Pd(dppf)Cl2, bis(pinacolato)diboron,
KOAc; (c) CH2Cl2, HATU, DIEA, acid component; (d) NH4OAc, dioxane, 110 °C, 48 h or microwave 160 °C, 20 min; (e) HClaq; (f) DME, NaHCO3,
Pd(dppf)Cl2, 18h, 80 °C, 8; (g) DMF, NaH, rt, 15 min Et3N, acid component.
2060
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
Figure 4. Synthesis of 5-spiropyrrolidines. Reagents and conditions are as follows: (a) THF, triethylphosphonoacetate, NaH; (b) CH2Cl2, DIBAL-H;
(c) THF, CBr4, triphenylphosphine; (d) CH2Cl2, BnEt3NCl, CsOH, ethyl 2-(diphenylmethyleneamino)acetate; (e) CH3COOH, THF, water; (f)
CH2Cl2, Hunig’s base, 4-nitrobenzene-1-sulfonyl chloride; (g) TfOH, CHCl3; (h) K2CO3, CH3CN, 18-crown-6, benzenethiol; (i) CH2Cl2, DIEA, acetyl
chloride; (j) LiOH, THF/MeOH/water; (k) AgCN, rt, CH2Cl2, overnight, carboxymethylvaline chloride (CM-Val-Cl).
supports cooperative binding of the peptidic caps via a biaryl-like
spacer, it also suggests that the binding of both caps is not
necessarily equivalent, depending on the spacing of the caps.
Although there is strong homology (84% identity) between
gt1a and gt1b in NS5A domain I, gt1a/gt1b residue differences
M/L28, Q/R30, I/V34, V/F37, H/Q54, R/T56, H/P58, and E/
Q62 (yellow, Figure 2) are of a particular interest because of their
proximity to the putative binding pocket. H/Q54 and R/T56 are
located in the β-sheet flanking the central binding channel, while
H/P58 is strategically located in a short β-sheet loop at the
entrance of the binding cavities. Biaryl cores in dimeric inhibitors
possibly take advantage of aromatic−aromatic and/or cation−
aromatic interactions with gt1a residues H54 and R56, leading to
potency enhancement in the gt1a replicon assay. Additional
gt1a/gt1b residue variants M/L28 and Q/R30 are located in the
flexible protein loop region. We propose that gt1a/gt1b residue
differences in the binding channel (H/Q54, R/T56, and H/
P58), in concert with M/L28 and Q/R30 in the flexible loop
region, may cause the observed differential gt1a and gt1b potency
profiles of NS5A inhibitors, as observed in Table 1. As potential
experimental validation of this model, it would be interesting to
swap these residues between respective NS5A gt1a and gt1b
constructs to test if this inverts the observed gt1a and gt1b
potency profiles of NS5A inhibitors. Such experiments would
provide circumstantial validation of proposed binding models in
the absence of cocrystal data.
interface at right angles to the central core. Since both gt1b
resistant mutations L31V and Y93H result in smaller residues,
possibly causing reduced interactions with the peptidic caps in 6
(Table 6), we also speculated that bulkier pyrrolidines in dimeric
inhibitors could potentially restore interactions with these
mutants and improve potency in gt1b L31V and Y93H replicon
assays.
In this work we present the synthesis and evaluation of diverse
spiropyrrolidine motifs designed to address the potency issues
against gt1b L31V and L93H mutants, leading to the discovery of
the clinical HCV inhibitor 41H (GSK2336805).28
RESULTS AND DISCUSSION
■
5-Spiro Inhibitors. On the basis of our initial modeling
considerations and SAR, we designed additional 5-spiropyrroli-
dine analogues (Table 2). These inhibitors were synthesized by
coupling appropriate acid components to key intermediates,
amine 12 or bromide 16, followed by imidazole cyclization
(Figure 3).
The 5-spiropyrrolidines in Table 2 were synthesized in a
manner analogous to the synthesis of tetrahydropyran 22 (Figure
4). Thus, the alkylation of glycine synthon with bromide 17 was
used to secure ester 18. Next, the protecting groups were
swapped in 18 via intermediate amine 19, resulting in the
sulfonamide 20, which was then cyclized to the desired 5-
spirotetrahydropyran 21. Finally, thiol-mediated cleavage of the
sulfonamide protecting group afforded racemic 22 (Figure
4).32,33
Acylation of amine 22 with acetyl chloride proceeded as
expected, yielding 24 after ester hydrolysis. Acid 24 was then
coupled to 16 and the product cyclized to imidazole 25A.
However, attempts to acylate 22 with more sterically hindered
acids, such as Boc-valine or carboxymethylvaline (CM-Val),
failed to yield the desired product under many standard amide
coupling conditions (not shown). We established that products,
A number of mutations in both wt gt1a (M28T, Q30H/R,
L31M, Y93C) and gt1b (L31V, Y93H) have been reported to
lead to significant viral resistance.31 In the model, these residue
positions (blue, Figure 2) surround the terminal peptidic caps.
Notably, the proline position of the inhibitor scaffold was
considered of key importance, since it is surrounded by all four
positions (28, 30, 31, and 93). The proline position also restricts
the conformational flexibility of the inhibitor scaffold, orienting
the peptidic caps to make buried contacts within the dimer
2061
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
such as 23, can be synthesized via the acyl chlorides (e.g., CM-
Val-Cl) route in the presence of silver cyanide (Figure 4), and we
used this procedure in further syntheses.
In the R2 = acetyl series, the use of spirotetrahydropyran
moiety in 25A resulted in a superior gt1b and gt1a profile
compared to the difluorocyclohexane moiety in 25B (Table 2).
Larger R2 = Cbz in 25C proved to be deleterious to gt1a potency
vs 25A. In the context of R2 = carboxymethylvaline, the sulfone
moiety in 25D clearly was a less attractive spiro partner,
compared to both spirotetrahydropyran (25E) and difluorocy-
clohexane (25F), which on the other hand were equivalent. In
addition, incremental amino acid side chain variations in the R2
(in the context of spirodifluorocyclohexanes 25F−I) had only
minor effect on potency, but similar to 25A, lower molecular
volume of alanine in 25H appeared to be somewhat advanta-
geous to the overall potency. Thus, especially in the context of
gt1a, this series favors the tetrahydropyran and difluorocyclohex-
ane spiro moieties and small R2. Compounds 25A and 25H were
more potent in gt1a replicon than lead molecules 4 and 5,
encouraging us to explore additional spirocyclic motifs.
Table 2. Gt1a and gt1b Replicon Potencies of 25A−I
4-Spiro Non-Ketal Inhibitors. We next focused on the 4-
spiropyrrolidine motif, exemplified here with inhibitors 31A−N
(Table 3). The general approach to compounds in this series is
a
Table 3. gt1a and gt1b Replicon Potencies of 31A−N
replicon pEC50
compd
X
n
gt1a
gt1b
31A
SO2
CF2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
9.4
10.4
8.0
10.4
11.6
10.4
8.8
b
31B
31C
31D
31E
31F
31G
31H
31I
N-Boc
NH
<7.5
8.0
N-Ac
9.7
N-COOMe
O
9.1
10.8
10.8
11.0
9.9
10.2
9.0
N-Boc
NH
8.5
31J
N-Ac
8.9
9.8
31K
31L
31M
31N
N-COOMe
N-CONHMe
N-SO2Me
CF2
8.7
10.3
9.6
8.2
8.9
10.3
11.4
10.0
a
All compounds in this table are diasteromeric mixtures except 31B.
31B is a single (S)-spiropyrrolidine isomer.
b
illustrated with the synthesis of 31G (Figure 5). Thus, the
reaction of enamine 27 and oxime 28 was used to generate the
oxazine 29. Subjecting the latter to the RaNi-mediated ring
contraction provided the racemic ester 30.34 Subsequent
chemistry toward 31G utilized steps described in Figure 3.
Inhibitors 31A and 31C−N were synthesized in a similar manner
as diastereomeric mixtures. We later established chromato-
graphic conditions toward separating enantiomers in 30 and via
the dipeptide 32 were able to synthesize 43A as a single isomer
(Table 5).
Toward the synthesis of inhibitor 31B, the initial conversion of
silyl ether 33 to ketone 34 and treatment with the Deoxo-Fluor
resulted in a difficult to separate mixture of 35 (and its R-
enantiomer) and the byproduct 36. To facilitate the purification,
crude mixture was treated with mCPBA, a step that converted 36
to a more readily separable epoxide (not shown), followed by
facile chiral chromatography purification. Conversion of
2062
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
Figure 5. Synthesis of 31G. Reagents and conditions are as follows: (a) CH2Cl2, DIBAL-H; (b) benzene, pyrrolidine, reflux; (c) THF, Et3N; (d) EtOH,
Raney nickel, H2; (e) CH2Cl2, CBz-Cl, Et3N; (f) chiral separation; (g) Pd(OH)2, H2, (h) CH2Cl2, HATU, DIEA, acid; (i) LiOH, THF/MeOH/water.
Figure 6. Synthesis of 31B. Reagents and conditions are as follows: (a) CH2Cl2, CBz-Cl, Et3N; (b) THF, TBAF, CH3COOH; (c) CH2Cl2, Dess−
Martin periodinane; (d) (1) Deoxo-Fluor, EtOH; (2) CH2Cl2, mCPBA, (3) chiral chromatography; (e) Pd(OH)2, EtOH, H2; (f) CH2Cl2, CM-Val,
HATU, Hunig’s base; (g) LiOH, THF−water−MeOH.
enantiomeric 35 to 37 was followed by chemistry described in
method A (Figure 3), affording 31B as single isomer (Figure 6).
Similar chemistry afforded the difluorocyclobutane analogue
31N as a diastereomeric mixture.
inhibitors 25A and 25H. We tested 31B in the transient replicon
gt1b NS5A mutant assays. While 31B was equipotent against the
wild type and L28V mutant, mutants L31V and Y93H caused
158- and 794-fold reductions in EC50 values for this compound,
respectively (Table 6). We were encouraged by the improvement
of potency as we progressed from the 5-spiro to the 4-spiro series
and felt that further explorations in the 4-spiro series could
address sensitivity of some gt1b mutants, in particular with small
and nonpolar X moieties, such as oxygen or sulfur in the
spirocyclic ring.
4-Spiroketal and Thioketal Inhibitors. We next designed
ketal and thioketal spiropyrrolidines 41A−H (Table 4).
Intermediate 38 facilitated robust syntheses of diverse (ring
size, ring substituents) 4-spirocyclic acids 40 (Figure 7), which
were then elaborated into 41A-H as described in methods A and
B (Figure 3). There was no/little ring size effect on potency
(41D and 41H) and the X = S and X = O analogues were
essentially equipotent (41A and 41H), although a somewhat
higher gt1a potency slightly favored 41H. However, methyl
Potency data in both the four- and six-membered series (Table
3) revealed that the more polar X (SO2, NH) in spirocycles 31A,
31D, and 31I were detrimental to both gt1b and, in particular, to
gt1a replicon potencies. Bulky X in 31C, 31E, 31F, 31H, 31J−M
were generally not well-tolerated. In contrast, inhibitors with
small and less polar X (O, CF2) in 31B, 31G, and 31N were very
potent in both assays. The influence of the spirocyclic ring size on
potency was somewhat context-dependent. For X = CF2 (31B
and 31N) and X = methyl carbamate (31F and 31K), the six-
membered ring was preferred over the four-membered ring.
However, for remaining X in Table 3 (X = N-Boc for 31C and
31H; X = NH for 31D and 31I; X = N-Ac for 31E and 31J) the
four-membered ring was preferred. The most potent inhibitor in
this series, 31B, featured a substantial potency improvement
especially in the gt1a assay over the most potent 5-spiro series
2063
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
substituents on the ketal moiety (41F, 41G vs 41D) and smaller
R (41A and 41C, 41D and 41E) were detrimental to gt1a (but
not gt1b) potency. This SAR contrasts with the 5-spiro SAR,
where smaller N-substituents, such as acetyl in 25A, were
associated with the potency increase. The most potent
compounds in the HCV replicon assays in this series were
41B, 41D, and 41H.
Table 4. gt1a and gt1b Replicon Potencies of 41A−H
Within the caveats and limitations of our NS5A homology
model, we considered possible interactions of 41H bound to
NS5A gt1b domain I. The model suggested that the peptidic caps
of 41H interact cooperatively and favorably with the dimer
interface (Figure 8). Both peptidic caps access the cavities
formed between the ZnF and N-terminal α-helix units of the
dimer interface in a similar, but nonequivalent, manner,
supporting a network of hydrogen bond interactions across the
dimer interface. Sterically, the 4-spiro ketal pyrrolidine motif
helps stabilize a network of intermonomer interactions (R30−
Y93, R30−I52) at the dimer interface while orienting the
carbamate terminus of 41H to make bridging hydrogen bond
interactions with the backbone NH of L31 and the side chains of
R30 and T56 (Figure 8b). Conserved (gt1a/gt1b) I52 forms a
steric buttress against the 4-spiroketal in supporting these
interactions. Although the 4-spiroketal can in principle make a
hydrogen bond contact with R30, this contact would be relatively
hydrated and therefore is considered negligible. Notably, the side
chain of conserved T95 is oriented to form a bridging hydrogen
bond between the imidazole of the 41H core and Q54. Similar to
the 4-spiroketal motif, the des-spiro cap also helps stabilize
intermonomer interactions between R30 and Y93, as well as R30
and the I52 backbone carbonyl (Figure 8c). However, here the
peptidic cap is buried more deeply in the dimer cavity and the
proline and carbamate carbonyls make bridging hydrogen bonds
to R30 and Y93, respectively. There is no obvious hydrogen bond
to the core imidazole other than the possibility of a bridging
water interaction with T56. As a class, dimeric NS5A inhibitors
have notably weaker gt1a potency than gt1b. Given their
proximity to the putative binding mode of 41H, gt1a/gt1b
residue variations M/L28, Q/R30, H/Q54, R/T56, and H/P58
would be predicted to most sensitively modulate gt1a vs gt1b
potency. For example, Q30 in gt1a would be expected to have
fewer intermonomer and/or ligand interactions than in the case
of R30 in gt1b, suggesting weaker stabilization of the dimer
interface (and less restriction of the N-terminal α-helix).
For gt1a Y93C and gt1b Y93H resistant mutations, the model
suggests that Y93 intermonomer interactions with Q/R30 would
Figure 7. Synthesis of 41A−H. Reagents and conditions are as follows: (a) HATU, Et3N, CM-Val; (b) Dess−Martin periodinane; (c) X = S or O, tosylic
acid; (d) LiOH, THF−water−MeOH.
2064
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
Figure 8. Putative binding mode of 41H (sticks, green carbons) with NS5A gt1b domain I. (a) 41H spans the homodimer interface, enabling the
peptidic caps to cooperatively bind within the cavities formed between the ZnF and N-terminal α-helix units of the dimer. The NS5A monomers are
color-coded red and cyan. Resistant mutation sites (gt1a, M28T, Q30H/R, L31M, Y93C; gt1b, L31V, Y93H),7 color-coded blue, surround the peptidic
caps, while differential gt1a/gt1b residues (H/Q54, R/T56, H/P58, E/Q62), color-coded yellow, flank the biaryl core along the central dimer interface.
Ligand contacts are depicted as CPK spheres. (b) Magnified view of the 4-spiroketal cap binding pocket depicting ligand contacts and the hydrogen-
bonding network surrounding the pocket. An intermonomer hydrogen bond between R30 and Y93 helps form the binding cavity. R30 also forms an
intermonomer hydrogen bond with backbone carbonyl of conserved I52. I52 serves as a steric buttress against the 4-spiroketal, while the terminal
carbamate of the cap makes bridging hydrogen bond contacts to the backbone of L31 and the side chains of R30 and T56. The T95 hydroxyl is oriented
to interact with the imidazole of the biaryl core. (c) Magnified view of the des-spiro cap binding pocket. R30 again makes intermonomer hydrogen bonds
with Y93 and I52. In contrast to (b), the proline and carbamate carbonyls of the cap make bridging hydrogen bonds to R30 and Y93, respectively, while
there is no apparent hydrogen bond to the core imidazole.
Figure 9. Synthesis of 43A,B. Reagents and conditions are as follows: (a) Et3N, CH3CN, 14; (b) ammonium acetate, dioxane, 110 °C, 48 h or microwave
160 °C, 20 min.
be eliminated in both cases (Figure 8b,c), effectively destabilizing
the dimer interface and relaxing cavity formation. Moreover,
L31V and Y93H mutants introduce residues with reduced
molecular volume, and we thus speculate that the added steric
bulk of the 4-spiroketal may compensate this void, establishing
new intermolecular interactions that restore the potencies of
41H and 43B against gt1b L31V and Y93H to essentially gt1b wt
level.
Symmetric 4-Spiro Inhibitors. Chiral separation of 30
enabled the synthesis of symmetric inhibitor 43A as single
isomer. Both 43A and the symmetric ketal 43B were synthesized
from dibromide 14 and acids 32 and 40 (X = O, R = H, n = 0),
respectively, followed by cyclization of diester 42 (Figure 9).
2065
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
Inhibitor 43A turned out to be somewhat more potent in gt1a
replicon than both nonsymmetric 41H and symmetric 43B. On
the other hand, both the symmetric bis-ketal 43B and
nonsymmetric (mono) ketal 41H were essentially equivalent
in gt1a and gt1b replicon assays (Table 5).
Table 7. In Vivo Pharmacokinetic Data for 41H and 43B
a
compd
animal species
CD rat
in vivo CL as % hepatic blood flow
F, %
43B
41H
58
60
8
9
13
51
13
CD rat
beagle dog
cynomolgus monkey
9
Table 5. gt1a and gt1b Replicon Potencies of 43A,B
a
Liver blood flow (mL min−1 kg−1): rat = 55; dog = 30; cyno = 45.
somewhat more bioavailable and was thus progressed to
additional PK studies in other species. Dog and cyno PK data
indicated that 41H was a low-clearance compound in both
species, with high bioavailability in dog. In the hepatocyte
stability assessment, 41H was found to be stable in dog,
cynomolgus monkey, and human hepatocytes (t1/2 ≥ 360 min, all
three species) but, consistent with its higher rat in vivo clearance,
somewhat less stable in rat hepatocytes (t1/2 = 160 min).
Inhibitor 41H was nontoxic in cellular assays up to 40 μM. It
did not inhibit cytochrome P450 enzymes 1A2, 2C9, 2D6, 2C19,
and 3A4 (IC50 > 33 μM) and was inactive in the P450 3A4 time-
dependent assay against midazolam, nifedipine, and atorvastatin,
thus displaying a low risk for drug−drug interaction, an
important factor for the anticipated combination HCV therapy.
In addition, 41H was 94% bound to human serum albumin and
86% to human α1AGP.
The ability of 41H to reduce the level of HCV RNA in the gt1b
replicon cells was evaluated over 20 days with 6 run in parallel as
a comparator (Figure 10). The replicon cells were maintained in
HCV NS5A gt1b Mutant Characterization and Rat
Pharmacokinetics of HCV Inhibitors. Daclatasvir 6 and key
inhibitors that emerged from this work, 31B, 43A, 43B, and 41H,
were evaluated in the transient replicon gt1b L28V, L31V, and
Y93H mutant assays (Table 6). The pEC50 values for all five
Table 6. Potencies (pEC50) of NS5A Compounds in Transient
gt1b Replicon Mutant Assay (% Fitness of Mutant vs Wild
Type)
transient replicon pEC50
a
b
compd WT gt1b (100%) L28V (ND ) L31V (353%) Y93H (56%)
6
11.3
11.9
11.1
11.4
10.9
11.4
11.8
11.5
11.2
10.9
9.9
9.7
9.5
9.0
31B
41H
43A
43B
10.6
10.7
10.6
10.6
9.9
Figure 10. Inhibitors 41H and 6 in HCV gt1b RNA reduction assay.
10.9
a
b
Wild type. Not determined.
concentrations of 6 or 41H that was 3 or 10 times the EC90 value
of the compound calculated from the results in the standard 2-
day assay. Initially both concentrations of 6 and 41H caused
similar declines in HCV RNA with a ∼2 log10 reduction of HCV
RNA observed at day 5. After this initial decline, compound 41H
was able to further reduce HCV RNA levels to 3 and 4 log10 by
day 20 for the 3× and 10× EC90 concentrations, respectively.
Compound 6 was able to cause further reduction in HCV RNA
levels, but the maximum reduction was between 2 and 3 log10.
Thus, at 10-fold over EC90, 41H was able to reduce HCV RNA
levels by 10-fold more than observed for 6. The combined data
supported progressing 41H to the clinic.
compounds on the L28V mutant were similar to that of wild type,
suggesting that the amino acid substitution did not impact
potency of the compounds. All compounds showed reduced
potencies against the L31V mutant, although its impact varied
significantly between compounds. Thus, the EC50 potencies of 6
and 31B were reduced ≥25-fold but ≤5-fold for 43A, 43B, and
41H, compared to the wild type virus. Evaluation of these five
compounds on the Y93H mutant showed that 6, 31B, and 43A
pEC50 values were reduced by ≥30-fold, while for 41H only 3.2-
fold reduction was observed, compared to wild type. No change
in potency was observed for 43B. These results demonstrated
that 41H and 43B had the best profile on the mutants, and both
were thus progressed to further PK studies.
In a clinical study, 41H demonstrated good pharmacokinetics
in single and multiple ascending dose studies in healthy
volunteers and was well tolerated up to 60 mg single dose and
75 mg in 14-day multiple dose study. Genotype 1 HCV infected
As seen in Table 7, both compounds exhibited a similar in vivo
clearance (about 60% hepatic blood flow), but 41H was
2066
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
than 0.25. All other stable and transient HCV replicon data were
obtained by the method described in ref 17 (http://dx.doi.org/10.1021/
jm400317w).
patients were given a single dose of up to 120 mg of 41H, and
HCV viral load was monitored. Mean HCV RNA reductions
from baseline ranged from 0.996 log10 copies/mL for patients
receiving a single dose of 1 mg up to a mean reduction of 2.95
log10 copies/mL after a single dose of 120 mg.35,36
Chemistry. Unless stated otherwise, the reagents were obtained
from commercial sources and were used directly. Reactions involving
air- or moisture-sensitive reagents were carried out under nitrogen
atmosphere. The reactions were carried out at ambient temperature
unless otherwise indicated. Silica gel (EM Science, 230−400 mesh) was
used for chromatographic purification unless otherwise indicated.
Anhydrous solvents were obtained from Aldrich (Sure Seal). 1H NMR
spectra were recorded on a Varian 300 or 400 MHz spectrometer. The
chemical shifts are reported in parts per million (ppm) relative to TMS.
The following abbreviations are used to describe peak patterns when
appropriate: b, broad, s, singlet, d, doublet, t, triplet, q, quartet, m,
multiplet. Mass spectra were obtained using electrospray (positive or
negative ion). All compounds were >95% pure as determined by LC/
MS (equipped with UV and evaporative light scattering detectors) in
both acid- and base-modified CH3CN/water gradients and 1H NMR.
2-Amino-1-(4-bromophenyl)ethanone (7). To a stirred solution
of 2-bromo-1-(4-bromophenyl)ethanone (130 g, 0.478 mol) in toluene
(2500 mL) was added hexamethylenetetramine (65.6 g 0.478 mol), and
the mixture was stirred at 40 °C for 16 h. The resulting solid was filtered
off and washed with toluene and ether to give a white solid. To a stirred
suspension of this solid in EtOH (800 mL) was added concentrated
hydrochloride acid (300 mL). The mixture was stirred at ambient
temperature for 20 h, and the solid was collected by filtration, washed
with EtOH, water, and dried in vacuo to give 7 (95 g, 92%) as a white
CONCLUSIONS
■
The clinical and preclinical results validated our medicinal
chemistry strategy. Even though the 5-spiro analogues 25A and
25H were slightly less potent in gt1a replicon than 6, their
improved potency compared to 4 and 5 encouraged us to explore
the 4-spiro series. This effort identified potent non-ketal
inhibitors 31B and 31G and established SAR of the spiro
moiety. The preference for small, relatively nonpolar X (Table 3)
discovered in the 4-spiro non-ketal series directed our
explorations into the ketal-based 4-spiropyrrolidines, leading to
potent nonsymmetric 41H and symmetric 43B NS5A inhibitors.
The PK data across species were consistent with further
progression of 41H. In addition to exhibiting good potency
profile in the NS5A gt1b L31V and Y93H mutant assay, 41H also
reduced HCV RNA in a 20-day study to a greater degree than
daclatasvir 6 (Figure 10), suggesting that it could be less
impacted by gt1b-resistant mutants. Clinical data generated to
date support further development of 41H.
1
solid, used without purification in the next step. H NMR (300 MHz,
DMSO-d6) δ ppm 8.58 (s, br, 2H), 7.96 (d, J = 8.7 Hz, 2H), 7.81 (d, J =
8.7 Hz, 2H), 4.48−4.52 (m, 2H). ES LC−MS m/z = 214, 216 (M + H)+.
1,1-Dimethylethyl {2-Oxo-2-[4-(4,4,5,5-tetramethyl-1,3,2-di-
oxaborolan-2-yl)phenyl]ethyl}carbamate (8). To a mixture of 7
(50 g, 0.2 mol) and Boc2O (48 g, 0.22 mol) in CH2Cl2 (1000 mL), TEA
(68.8 mL, 0.5 mol) was added dropwise at 0 °C. The mixture was stirred
at ambient temperature overnight and washed with 1 N HCl (300 mL ×
3) and brine, dried over Na2SO4, concentrated in vacuo to give an off-
white solid, which was further washed with petroleum ether and dried.
Pd(dppf)Cl2 (2.6 g 3.18 mmol) was added to a mixture of 1,1-
dimethylethyl [2-(4-bromophenyl)-2-oxoethyl]carbamate (20 g, 63.7
mmol), bis(pinacolato)diboron (19.4 g, 76.4 mmol), and KOAc (24.8 g,
0.254 mol) in dioxane (300 mL). The flask was purged with nitrogen
(3×) and heated to 80 °C for 16 h under nitrogen atmosphere. The
reaction mixture was diluted with hexane (300 mL), filtered,
concentrated, and the residue was purified by chromatography on silica
gel (petroleum ether/ethyl acetate = 5/1) to give 8 (13.3 g, 58%) as a
white solid. 1H NMR (300 MHz, CDCl3) δ ppm 7.90−7.93 (m, 4H),
5.55 (s, br,1H), 4.68 (s, 2H), 1.48 (s, 9H), 1.35 (s, 12H). ES LC−MS m/
z = 384 (M + Na)+.
1,1-Dimethylethyl (2S)-2-({[2-(4-Bromophenyl)-2-oxoethyl]-
amino}carbonyl)-1-pyrrolidinecarboxylate, (9). A mixture of
Boc-L-proline (50 g, 0.233 mol), HATU (106 g, 0.279 mol), and
DIEA (150 mL) in DMF (400 mL) was stirred at ambient temperature
for 10 min, followed by the addition of 7 (70 g, 0.279 mol) in DMF (500
mL). The mixture was stirred overnight, diluted with EtOAc (4 L),
washed with 1 N HCl (500 mL × 4) and brine, dried over Na2SO4, and
concentrated. The crude product was recrystallized from a mixture of
petroleum ether/ethyl acetate (2/1) to give 9 (58.4 g, 61%) as a yellow
solid. 1H NMR (300 MHz, DMSO) δ ppm 8.22 (s, 1H), 7.93 (d, J = 8.4
Hz, 2H), 7.77 (d, J = 8.4 Hz, 2H), 4.46−4.51 (m, 2H), 4.15−4.21 (m,
1H), 3.28−3.40 (m, 2H),1.78−1.90 (m, 4H), 1.29−1.41 (m, 9H). ES
LC−MS m/z = 411.1, 4113.1 (M + H)+ .
EXPERIMENTAL SECTION
■
HCV in Vitro Data. Stable HCV Replicon Data for Compounds
1−5 (Table 1). Huh-luc/neo-ET replicon cells were removed from
flasks with a 0.05% trypsin/EDTA solution and resuspended in medium
supplemented with 10% FBS, penicillin−streptomycin, nonessential
amino acids, glutamine, and geneticin. The cell density was adjusted to
7.9 × 104 cells/mL, and an amount of 95 μL of cell suspension was
placed in columns 2−12 of 96-well plates using a MultiDrop liquid
dispenser (Thermo LabSystems), for a density of 7500 cells/well. Then
95 μL of medium was dispensed to column 1 for the background
controls. Two sets of plates were prepared: white opaque plates for assay
of luciferase activity and clear plates for assay of cytotoxicity. The cell
plates were incubated overnight at 37 °C, 5% CO2. Compound plates
were prepared with 10-point, 2-fold serial dilutions of the test
compounds in duplicate, with four compounds assayed per 96-well
plate. An amount of 10 μL from the compound dilution plates was
transferred to the cell culture plates using a multichannel pipettor or a
RapidPlate 96/384 workstation (Caliper Lifesciences). The cell plates
were incubated for 48 h at 37 °C, 5% CO2. To determine luciferase
activity, an amount of 100 μL of Steady-Glo reagent (Promega) was
added to the culture medium, and the plates were shaken on a plate
shaker. The plates were incubated for 15−30 min at room temperature,
and the luminescence was read on an Analyst HT luminometer
(Molecular Devices) with a 1 s integration time. To determine
cytotoxicity, an amount of 10 μL of WST-detection reagent (Roche) was
added to the culture medium. Plates were incubated at 37 °C, 5% CO2,
for approximately 60 min. Absorbance at 440 nm was measured using
the SpectraMax Plus microplate spectrophotometer (Molecular
Devices). Data analysis for the luciferase and cytotoxicity assays was
performed using Graphpad Prism, version 4.0 (Graphpad Software,
Inc.), in conjunction with DS Accord for EXCEL 6.0 (Accelrys,
Microsoft Corp.). The percent inhibition and percent cytotoxicity values
were plotted against compound concentration, and the data were fit to a
constrained four-parameter sigmoidal fit, equivalent to the “four-
parameter logistic equation”. The curve-fit equation employed was Y =
50−X)Hillslope
1,1-Dimethylethyl (2S)-2-[4-(4-Bromophenyl)-1H-imidazol-
2-yl]-1-pyrrolidinecarboxylate (10). A mixture of compound 9
(40.0 g, 97.2 mmol) and NH4OAc (60 g, 0.778 mol) in xylene (400 mL)
was heated to 150 °C for 5 h. The reaction mixture was concentrated,
and the residue was dissolved in EtOAc (500 mL), washed with aqueous
NaHCO3 and brine. The organic phase was dried over Na2SO4,
concentrated to dryness. The crude product was purified by
chromatography on silica gel (petroleum ether/ethyl acetate = 1/1)
to give 1,1-dimethylethyl(2S)-2-[4-(4-bromophenyl)-1H-imidazol-2-
Bottom + (Top − Bottom)/[1 + (10(log EC
)], where Bottom is
the minimum Y value, Top is the maximum Y value, and Hillslope is the
slope of the linear portion of the semilog curve. Top and Bottom were
constrained to values of 100% and 0%, respectively. The minimum n
(number of independent measurements) for each compound was 2. The
average pEC50 SD in the gt1a data was less than 0.21 and in the gt1b less
2067
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
1
yl]-1-pyrrolidinecarboxylate (10) (34 g, 89%) as a brown solid. H
NMR (300 MHz, CDCl3) δ ppm 7.52 (d, J = 8.4 Hz, 2H), 7.46 (d, J = 8.4
Hz, 2H), 7.20(s, 1H), 5.58−5.71 (m, 1H), 3.38−3.42 (m, 1H), 2.80−
2.87 (m, 2H), 2.03−2.06 (m, 2H), 1.88−2.00 (m, 2H), 1.49 (s, 9H). ES
LC−MS m/z = 392, 394 (M + H)+ .
Method A. To a stirred solution of carboxylic acid component (0.25
mmol) in DMF (2 mL) were added TEA (0.77 mmol) and HATU (0.28
mmol). After the mixture was stirred for ∼3 min, the amine 12 (0.25
mmol) was introduced. After the mixture was stirred for an additional 2
h at room temperature, the mixture was directly loaded to RP HPLC
column, eluting with 5−80% CH3CN/water (0.2% NH3H2O (conc)) to
give the corresponding amide as an intermediate. To a stirred solution of
the amides (0.15 mmol) in dioxane (3 mL) was added ammonium
acetate (1.57 mmol). The mixture was heated to 110 °C in a sealed tube
overnight, cooled to room temperature, filtered to remove excess
ammonium acetate, and purified by column chromatography (silica gel,
0−15% MeOH in ethyl acetate) to give the final target compounds (as
an example 25A) as a solid.
Methyl [(1S)-1-({(2S)-2-[4-(4-Bromophenyl)-1H-imidazol-2-
yl]-1-pyrrolidinyl}carbonyl)-2-methylpropyl]carbamate (11).
Compound 10 (72.4 g, 185 mmol) was treated with saturated HCl in
dioxane (200 mL) and stirred at ambient temperature overnight. The
resulting solid was filtered and washed with petroleum ether to give 4-
(4-bromophenyl)-2-[(2S)-2-pyrrolidinyl]-1H-imidazole (60 g, 90%) as
a yellow solid. 1H NMR (300 MHz, DMSO-d6) δ ppm 8.10 (s, 2H), 7.88
(d, J = 6.6 Hz,2H), 7.70 (d, J = 6.6 Hz, 2H), 7.49 (s, 1H), 7.32 (s, 1H),
7.16 (s, 1H), 4.50−4.52 (m, 1H), 3.15−3.40 (m, 2H), 1.88−2.88 (m,
4H). ES LC−MS m/z = 291.1, 293.1 (M + H)+ .
Method B. 16 (95 mmol) was dissolved in DMF (800 mL) and
degassed for 15 min (N2). Acid component 15 (21.99 g, 81 mmol) was
dissolved in DMF (100 mL), followed by careful addition of NaH (2.94
g, 73.4 mmol, 60% in oil) under nitrogen over 15 min, then slowly added
over 15 min dropwise to a DMF solution of bromo compound followed
by stirring for 1 h at room temperature. Solvent volume was then
reduced in vacuo to ∼100 mL and cooled to 20 °C. An amount of 100
mL of water was slowly added, and the resulting slight gray-yellow solid
was filtered, washed with water (200 mL), hexane (200 mL), and dried
under the vacuum (12 h). The crude compound was purified on 500 g of
silica with hexane/ethyl acetate (increasing gradient from 50% to 100%
EA), yielding 30−70% of the final target compounds.
Ethyl 2-[(Diphenylmethylidene)amino]-4-(tetrahydro-4H-
pyran-4-ylidene)butanoate (18). To a solution of ethyl 2-(diphenyl-
methyleneamino)acetate (5.29 g, 19.8 mmol), benzyltriethylammonium
chloride (0.41 g, 1.8 mmol), and cesium hydroxide monohydrate (4.54
g, 27.0 mmol) in anhydrous CH2Cl2 (50 mL) was added 4-(2-
bromoethylidene)tetrahydro-2H-pyran (3.44 g, 18.0 mmol) in
anhydrous CH2Cl2 (40 mL). The mixture was stirred at room
temperature for 72 h under nitrogen. The mixture was partitioned
between CH2Cl2 and water, and the aqueous layer was extracted with
CH2Cl2. The organic layer was dried over Na2SO4 and concentrated in
vacuo. The residue was purified by silica gel chromatography, eluting
with 5−40% hexanes/EtOAc to afford the title compound as a clear oil
in quantitative yield.
A mixture of N-[(methyloxy)carbonyl]-L-valine (prepared according
to the procedure provided in patent WO2005061487) (43.1 g, 0.246
mol) and HATU (93.5 g, 0.246 mol) in CH2Cl2 (1000 mL) was stirred
for 10 min. 4-(4-Bromophenyl)-2-[(2S)-2-pyrrolidinyl]-1H-imidazole
((60 g, 0.205 mol) was introduced, followed by DIEA (82.6 mL, 0.308
mol) dropwise. The mixture was stirred at ambient temperature for 12 h,
diluted with CH2Cl2 (1000 mL), and washed with NaHCO3 and brine.
The organic phase was dried over Na2SO4, concentrated to dryness, and
purified by chromatography (silica gel, petroleum ether/ethyl acetate =
1
1/1) to give 11 (62 g, 67%) as a yellow solid. H NMR (300 MHz,
CDCl3) δ 7.43−7.51(m, 4H), 7.17 (s, 1H), 5.53−5.57 (m, 1H), 5.20−
5.22 (m, 1H), 5.29−5.33 (m, 1H), 3.64−3.71 (m, 5H), 2.99−3.03 (m,
1H), 1.88−2.31 (m, 4H), 0.88−0.92 (m, 6H). ES LC−MS m/z = 449,
451 (M + H)+.
Methyl {(1S)-1-[((2S)-2-{4-[4′-(Aminoacetyl)-4-biphenylyl]-
1H-imidazol-2-yl}-1-pyrrolidinyl)carbonyl]-2-methylpropyl}-
carbamate (12). To a mixture of 11 (62 g, 0.138 mol), 8 (47.7 g, 0.152
mol), and NaHCO3 (34.2 g, 0.414 mol) in a mixed DME (800 mL) and
water (260 mL) was added Pd(dppf)Cl2 (5.63 g, 6.9 mmol). The flask
was purged with nitrogen (3×) before being heated to 80 °C for 16 h.
The mixture was cooled to room temperature and filtered. The filtrate
was diluted with EtOAc (1000 mL), washed with NaHCO3, brine, dried
over Na2SO4, and concentrated. The residue was purified by flash
chromatography (silica gel, petroleum ether/ethyl acetate = 1/2) to give
the Boc-protected product (45 g, 54%) as a yellow solid. 1H NMR (300
MHz, CDCl3) δ ppm 8.03 (d, J = 8.4 Hz, 2H), 7.88−760 (m, 6H), 5.58
(S, br,1H), 5.42 (m, 1H), 5.28−5.30 (m,1H), 4.71 (s, 2H), 4.32−4.35
(m, 1H), 3.70−3.84 (m, 5H), 2.96 (s, br,1H), 1.96−2.11 (m, 4H), 1.49
(s, 9H), 0.88−0.92 (m, 6H). ES LC−MS m/z = 604, (M + H)+. A
solution of the Boc-protected product (3.8 g, 6.3 mmol) in CH2Cl2 (40
mL) was treated with HCl (10 mL, 4 M in dioxane) to give 12 as a light
yellow solid (3.5 g, quant).
Ethyl 2-Amino-4-(tetrahydro-4H-pyran-4-ylidene)butanoate
(19). To a solution of 18 (6.8 g, 18.0 mmol) in THF (30 mL) were
added water (30 mL) and glacial acetic acid (20 mL), and the mixture
was stirred at room temperature for 2.5 h. The mixture was concentrated
in vacuo and the residue dissolved in 0.1 N HCl. It was extracted twice
with ethyl acetate, and the organic layer was discarded. To the aqueous
layer was added solid K2CO3 until the solution gave blue pH paper. The
aqueous layer was extracted with 15% isopropanol/CH2Cl2 (3×) and
the organic layer dried over Na2SO4 and concentrated in vacuo to afford
the title compound as a clear oil (3.05 g, 79%).
(S)-2-(4′-(2-Bromoacetyl)-[1,1′-biphenyl]-4-yl)-2-oxoethyl-1-
((S)-2-((methoxycarbonyl)amino)-3-methylbutanoyl)-
pyrrolidine-2-carboxylate (16). 1,1′-(4,4′-Biphenyldiyl)bis(2-bro-
moethanone) (37.8 g, 95 mmol) was dissolved in DMF (800 mL) and
degassed for 15 min (N2). 15 (21.99 g, 81 mmol) was dissolved in DMF
(100 mL), followed by careful addition of NaH (2.94 g, 73.4 mmol, 60%
in oil) under nitrogen over 15 min, then slowly added over 15 min
dropwise to a solution of 1,1′-(4,4′-biphenyldiyl)bis(2-bromoetha-
none), followed by stirring for 1 h at room temperature. Solvent volume
was then reduced in vacuo to ∼100 mL and cooled to 20 °C. An amount
of 100 mL of water was slowly added, and resulting slight gray-yellow
solid was filtered, washed with water (200 mL), hexane (200 mL), and
dried under the vacuum (12h). The crude compound was purified on
500 g of silica with hexane/ethyl acetate (increasing gradient from 50%
to 100% EA), yielding 14.5 g (37.3%) of compound 16. 1H NMR (400
MHz, DMSO-d6) δ ppm 8.11 (dd, J = 12.0, 8.5 Hz, 4 H), 7.96 (d, J = 8.4
Hz, 4 H), 7.41 (d, J = 8.4 Hz, 1 H), 5.43−5.75 (m, 2 H), 4.99 (s, 2 H),
4.53 (dd, J = 8.6, 4.7 Hz, 1 H), 4.03 (t, J = 8.6 Hz, 1 H), 3.76−3.90 (m, 1
H), 3.60−3.73 (m, 1 H), 3.53 (s, 3 H), 2.22−2.37 (m, 1 H), 2.12−2.21
(m, 1 H), 1.85−2.06 (m, 3 H), 0.90 (dd, J = 10.7, 6.6 Hz, 6 H).
For the preparation of target compounds 25A−H/31A−N and 41A−
H, the corresponding acid components were utilzied in the following
two methods to provide the final compounds.
Ethyl 2-{[(4-Nitrophenyl)sulfonyl]amino}-4-(tetrahydro-4H-
pyran-4-ylidene)butanoate (20). To a solution of 19 (1.0 g, 4.7
mmol) in anhydrous CH2Cl2 (30 mL) was added Hunig’s base (1.6 mL,
9.4 mmol) followed by 4-nitrobenzenesulfonyl chloride (1.14 g, 5.2
mmol), and the mixture was stirred at room temperature under nitrogen
for 2 h. The mixture was diluted with CH2Cl2 and washed with 0.1 N
HCl. The organic layer was dried over MgSO4 and concentrated in
vacuo. The residue was purified by silica gel chromatography, eluting
with 10−70% hexanes/EtOAc to afford the title compound as an off-
white solid in quantitative yield.
Ethyl 1-[(4-Nitrophenyl)sulfonyl]-8-oxa-1-azaspiro[4.5]-
decane-2-carboxylate (21). To a solution of 20 (1.87 g, 4.7 mmol)
in anhydrous chloroform (47 mL) was added trifluoromethanesulfonic
acid (0.2 mL, 2.3 mmol), and the mixture was stirred at room
temperature under nitrogen for 4 h. The mixture was diluted with
CH2Cl2 and washed with saturated NaHCO3 and the organic layer dried
over MgSO4 and concentrated in vacuo. The residue was purified by
silica gel chromatography, eluting with 10−70% hexanes/EtOAc to
afford the title compound as a white solid (1.47 g, 79%). 1H NMR (400
MHz, CDCl3) δ ppm 8.19−8.36 (m, 2 H), 8.08 (d, J = 8.8 Hz, 2 H), 4.52
(dd, J = 8.5, 2.1 Hz, 1 H), 3.99−4.14 (m, 2 H), 3.87−3.99 (m, 2 H),
3.21−3.49 (m, 2 H), 2.58−2.78 (m, 2 H), 2.10−2.30 (m, 2 H), 1.89−
2068
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
2.06 (m, 2 H), 1.77 (dd, J = 13.0, 2.1 Hz, 1 H), 1.35 (dd, J = 12.9, 1.6 Hz,
1 H), 1.22 (t, J = 7.1 Hz, 3 H). LC−MS ESI (M + H)+ = 399.47.
Ethyl 8-Oxa-1-azaspiro[4.5]decane-2-carboxylate (22). To a
solution of 21 (1.47 g, 3.7 mmol) in anhydrous CH3CN (25 mL) were
added K2CO3 (0.76 g, 5.5 mmol), 18-crown-6 (0.2 g, 0.7 mmol), and
thiophenol (0.6 mL. 5.5 mmol), and the mixture was stirred at room
temperature for 18 h. The mixture was concentrated in vacuo and the
residue dissolved in 0.1 N HCl. The aqueous layer was extracted twice
with EtOAc, and the organic layer was discarded. The aqueous layer was
treated with solid K2CO3 until the solution gave blue pH paper. It was
extracted with 15% isopropanol/CH2Cl2 (3×) and the organic layer
dried over Na2SO4 and concentrated in vacuo to afford the title
compound as a clear oil (0.71 g, 90%).
1-{N-[(Methyloxy)carbonyl]-L-valyl}-8-oxa-1-azaspiro[4.5]-
decane-2-carboxylic Acid (23). To a solution of 22 (125 mg, 0.59
mmol) in anhydrous CH2Cl2 (2 mL) was added silver cyanide (98 mg,
0.73 mmol) followed by N-[(methyloxy)carbonyl]-L-valyl chloride (142
mg, 0.73 mmol) as a solution in anhydrous CH2Cl2 (3.5 mL), and the
solution was stirred at room temperature for 18 h under nitrogen. The
mixture was filtered and treated with MeOH and stirred for 5 min. The
mixture was concentrated in vacuo and the residue purified by silica gel
chromatography, eluting with 30−100% hexanes/EtOAc to afford the
ester intermediate as a white solid (43 mg, 20%).
To a solution of the above ester (80 mg, 0.2 mmol) in THF/water/
MeOH (1.0 mL/0.5 mL/0.5 mL) was added lithium hydroxide
monohydrate (13 mg, 0.3 mmol), and the mixture was stirred at room
temperature for 72 h. The mixture was treated with 1 N HCl (0.5 mL)
and the mixture partitioned between EtOAc and 0.1 N HCl. The organic
layer was separated and dried over MgSO4 and concentrated in vacuo to
afford the title compound 23 as a white solid (53 mg, 72%). 1H NMR
(400 MHz, DMSO-d6) δ ppm 12.88 (br s, 1 H), 7.45−7.61 (m, 1 H),
4.89−5.03 (m, 1 H), 3.70−3.87 (m, 2 H), 3.65 (t, J = 9.2 Hz, 1 H), 3.53
(s, 3 H), 3.02 (td, J = 12.9, 4.9 Hz, 1 H), 2.71−2.90 (m, 1 H), 2.22−2.38
(m, 1 H), 1.87−2.19 (m, 3 H), 1.53−1.73 (m, 1 H), 1.32−1.42 (m, 1 H),
1.06−1.15 (m, 1 H), 0.61−0.98 (m, 8 H). LC−MS ESI (M + H)+ =
342.95.
1-Acetyl-8-oxa-1-azaspiro[4.5]decane-2-carboxylic Acid (24).
To a solution of 22 (135 mg, 0.63 mmol) in anhydrous CH2Cl2 (2 mL)
was added Hunig’s base (0.22 mL, 1.3 mmol) followed by acetyl chloride
(0.05 mL, 0.76 mmol), and the mixture was stirred at room temperature
under nitrogen for 1.5 h. The mixture was diluted with CH2Cl2 and
washed with 0.1 N HCl and the organic layer dried over MgSO4 and
concentrated in vacuo. The residue was purified by silica gel
chromatography, eluting with 60−100% hexanes/EtOAc to afford the
acetyl compound as a yellow oil (94 mg, 58%). To a solution of this ester
(93 mg, 0.34 mmol) in THF/water/MeOH (1.1 mL/0.6 mL/0.6 mL)
was added lithium hydroxide monohydrate (29 mg, 0.69 mmol), and the
mixture was stirred at room temperature for 3 h. The mixture was treated
with 1 N HCl (0.7 mL) and partitioned between EtOAc and 0.1 N HCl.
The aqueous layer was extracted twice with EtOAc, and the combined
organic layers were dried over MgSO4 and concentrated in vacuo to
afford the title compound 24 as a white solid (79 mg, 95%).
Benzyl 2-(5-(4′-(2-((S)-1-((S)-2-((Methoxycarbonyl)amino)-3-
methylbutanoyl)pyrrolidin-2-yl)-1H-imidazol-5-yl)-[1,1′-bi-
phenyl]-4-yl)-1H-imidazol-2-yl)-8-oxa-1-azaspiro[4.5]decane-
1-carboxylate (25C). 1H NMR (400 MHz, MeOH-d4) δ ppm 8.17 (br
s, 1 H), 7.80 (d, J = 8.21 Hz, 1 H), 7.64−7.77 (m, 9 H), 7.28−7.40 (m, 3
H), 7.13 (br s, 2 H), 6.98 (d, J = 8.40 Hz, 2 H), 5.13−5.22 (m, 2 H),
4.15−4.26 (m, 1 H), 3.86−4.01 (m, 4 H), 3.64 (s, 3 H), 3.43−3.60 (m, 3
H), 3.01−3.19 (m, 1 H), 2.77 (br s, 1 H), 2.27−2.39 (m, 4 H), 2.12−
2.26 (m, 2 H), 2.02−2.08 (m, 3 H), 1.71 (br s, 1 H), 1.44 (br s, 1 H),
0.95−1.02 (m, 1 H), 0.91 (dd, J = 17.10, 6.74 Hz, 6 H).
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[1-((2S)-3-methyl-
2-{[(methyloxy)carbonyl]amino}butanoyl)-8,8-dioxido-8-thia-
1-azaspiro[4.5]dec-2-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-
imidazol-2-yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate
(25D). 1H NMR (400 mHz, CDCl3) δ 7.92−7.52 (m, 9 H), 5.39 (br s, 4
H), 4.36 (br s, 1 H), 3.98−3.82 (m, 1 H), 3.82−3.60 (m, 10 H), 3.19 (br
s, 4 H), 2.88−2.70 (m, 1 H), 2.33 (br s, 4 H), 2.02 (s, 4 H), 1.81 (br s, 5
H), 1.06 (s, 1 H), 0.89 (d, J = 7.0 Hz, 6 H), 0.74 (d, J = 6.5 Hz, 3 H),
0.44−0.34 (m, 3 H). HRMS for C44H57N8O8S (M + H)+ calcd,
857.4020; found, 857.4018.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[1-((2S)-3-methyl-
2-{[(methyloxy)carbonyl]amino}butanoyl)-8-oxa-1-azaspiro-
[4.5]dec-2-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-
yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate (25E). 1H NMR
(400 MHz, MeOH-d4) δ ppm 7.56−7.86 (m, 9 H), 7.16−7.44 (m, 2
H), 6.88−7.09 (m, 1 H), 5.72 (br s, 1 H), 5.09−5.21 (m, 1 H), 4.03−
4.28 (m, 1 H), 3.80−4.04 (m, 6 H), 3.42−3.71 (m, 9 H), 2.11−2.56 (m,
8 H), 1.88−2.11 (m, 6 H), 0.80−1.06 (m, 12 H). HRMS for
C44H57N8O7 (M + H)+ calcd, 809.4350; found, 809.4347.
Methyl [(1S)-1-({(2S)-2-[4-(4′-{2-[8,8-Difluoro-1-((2S)-3-meth-
yl-2-{[(methyloxy)carbonyl]amino}butanoyl)-1-azaspiro[4.5]-
dec-2-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-
pyrrolidinyl}carbonyl)-2-methylpropyl]carbamate (25F). 1H
NMR (400 mHz, MeOH-d4) δ 7.92−7.51 (m, 8 H), 7.29 (br s, 2 H),
5.71 (br s, 1 H), 5.32 (br s, 1 H), 4.21 (d, J = 7.4 Hz, 1 H), 4.15−3.80 (m,
4 H), 3.77−3.41 (m, 8 H), 3.18 (br s, 1 H), 3.12−2.82 (m, 1 H), 2.63−
1.40 (m, 16 H), 1.05−0.84 (m, 10 H), 0.78−0.64 (m, 1 H), 0.44−0.28
(m, 1 H). HRMS for C45H57F2N8O6 (M + H)+ calcd, 843.4369; found,
843.4368.
Methyl [(1S)-1-({8,8-Difluoro-2-[4-(4′-{2-[(2S)-1-((2S)-3-meth-
yl-2-{[(methyloxy)carbonyl]amino}butanoyl)-2-pyrrolidinyl]-
1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-azaspiro-
1
[4.5]dec-1-yl}carbonyl)propyl]carbamate (25G). H NMR (400
mHz, CDCl3) δ 7.84 (d, J = 7.8 Hz, 3 H), 7.72−7.41 (m, 6 H), 7.31−
7.09 (m, 3 H), 5.73−5.17 (m, 4 H), 5.12 (br s, 1 H), 4.41−4.06 (m, 3 H),
3.97−3.56 (m, 11 H), 3.37−2.76 (m, 4 H), 2.60−1.55 (m, 9 H), 1.46−
1.20 (m, 2 H), 1.06 (d, J = 6.0 Hz, 2 H), 1.00−0.79 (m, 5 H), 0.56 (d, J =
7.3 Hz, 1 H). HRMS for C44H55F2N8O6 (M + H)+ calcd, 829.4213;
found, 829.4211.
Methyl ((1S)-2-{8,8-Difluoro-2-[4-(4′-{2-[(2S)-1-((2S)-3-meth-
yl-2-{[(methyloxy)carbonyl]amino}butanoyl)-2-pyrrolidinyl]-
1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-azaspiro-
1
[4.5]dec-1-yl}-1-methyl-2-oxoethyl)carbamate (25H). H NMR
(400 mHz, CDCl3) δ 10.97−10.14 (m, 2 H), 7.91−7.40 (m, 8 H), 7.34−
7.16 (m, 2 H), 5.68−5.15 (m, 3 H), 5.08 (d, J = 6.0 Hz, 1 H), 4.34 (d, J =
8.0 Hz, 2 H), 3.95−3.54 (m, 8 H), 3.37−2.69 (m, 4 H), 2.60−1.57 (m,
13 H), 1.56−1.18 (m, 3 H), 1.06 (d, J = 6.8 Hz, 1 H), 0.89 (d, J = 6.5 Hz,
6 H). HRMS for C43H53F2N8O6 (M + H)+ calcd, 815.4056; found,
815.4061.
Methyl [(1S)-1-({8,8-Difluoro-2-[4-(4′-{2-[(2S)-1-((2S)-3-meth-
yl-2-{[(methyloxy)carbonyl]amino}butanoyl)-2-pyrrolidinyl]-
1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-azaspiro-
[4.5]dec-1-yl}carbonyl)-3-methylbutyl]carbamate (25I). 1H
NMR (400 mHz, CDCl3) δ 11.07−9.94 (m, 2 H), 8.17−7.42 (m, 10
H), 7.33−7.08 (m, 2 H), 5.72−5.19 (m, 3 H), 5.14−4.99 (m, 1 H), 4.86
(br s, 1 H), 4.50−4.09 (m, 2 H), 4.02−3.50 (m, 8 H), 3.45 (s, 1 H),
3.33−2.71 (m, 4 H), 2.59−1.47 (m, 11 H), 1.40 (br s, 3 H), 1.19−0.67
(m, 11 H), 0.63−0.40 (m, 1 H).). HRMS for C46H59F2N8O6 (M + H)+
calcd, 857.4526; found, 857.4531.
Methyl ((1S)-1-{[(2S)-2-(4-{4′-[2-(1-Acetyl-8-oxa-1-azaspiro-
[4.5]dec-2-yl)-1H-imidazol-4-yl]-4-biphenylyl}-1H-imidazol-2-
yl)-1-pyrrolidinyl]carbonyl}-2-methylpropyl)carbamate (25A).
1
Yield, 19 mg, 17%. H NMR (400 MHz, MeOH-d4) δ ppm 7.51−
7.90 (m, 8 H), 7.17−7.52 (m, 2 H), 5.08−5.26 (m, 2 H), 3.81−4.03 (m,
4 H), 3.64 (s, 3 H), 3.41−3.61 (m, 3 H), 2.21−2.51 (m, 5 H), 2.01−2.21
(m, 4 H), 1.97 (d, J = 19.9 Hz, 5 H), 1.77−1.91 (m, 1 H), 1.64−1.75 (m,
1 H), 1.31−1.42 (m, 1 H), 0.79−1.04 (m, 8 H). HRMS for C39H48N7O5
(M + H)+ calcd, 694.3717; found, 694.3718.
Methyl ((1S)-1-{[(2S)-2-(4-{4′-[2-(1-Acetyl-8,8-difluoro-1-
azaspiro[4.5]dec-2-yl)-1H-imidazol-4-yl]-4-biphenylyl}-1H-imi-
dazol-2-yl)-1-pyrrolidinyl]carbonyl}-2-methylpropyl)-
carbamate (25B). 1H NMR (400 mHz, CDCl3) δ 11.17−10.07 (m, 2
H), 8.16−7.40 (m, 10 H), 7.35−7.03 (m, 2 H), 5.85−4.98 (m, 2 H),
4.59−4.12 (m, 1 H), 3.72 (s, 5 H), 3.46 (br s, 1 H), 3.33−2.72 (m, 2 H),
2.76−1.39 (m, 17 H), 1.27−0.74 (m, 5 H). HRMS for C40H48F2N7O4
(M + H)+ calcd, 728.3736; found, 728.3736.
2-{N-[(Methyloxy)carbonyl]-L-valyl}-8-oxa-2-azaspiro[4.5]-
decane-3-carboxylic Acid (32). To a solution of 30 (200 mg, 0.94
mmol), HATU (392 mg, 1.03 mmol), and N-[(methyloxy)carbonyl]-L-
2069
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
valine (181 mg, 1.03 mmol) in anhydrous CH2Cl2 (6 mL) was added
Hunig’s base (0.33 mL, 1.88 mmol), and the solution was stirred at room
temperature under nitrogen. After 2 h, the mixture was concentrated in
vacuo and purified by C18 RP chromatography, eluting with 10−90%
ACN/water/0.2% NH4OH to afford the ester as a yellow oil (313 mg,
90%). To a solution of this ester (310 mg, 0.84 mmol) in a 2:1:1 mixture
of THF/water/MeOH (6 mL) was added lithium hydroxide
monohydrate (70 mg, 1.67 mmol), and the mixture was stirred at
room temperature for 2 h. The mixture was treated with 1 N HCl (1.6
mL), partitioned between EtOAc and water (30 mL each), and the
organic layer was extracted with EtOAc (30 mL), dried (MgSO4), and
concentrated to give 32 as a white foam (211 mg, 74%). This material
was used in subsequent steps without additional purification. 1H NMR
(400 MHz, DMSO-d6) δ ppm 12.43 (br s, 1 H), 7.19−7.41 (m, 1 H),
4.05−4.28 (m, 1 H), 3.91−4.05 (m, 2 H), 3.40−3.71 (m, 6 H), 3.20−
3.28 (m, 1 H), 1.81−1.98 (m, 2 H), 1.70−1.80 (m, 1 H), 1.46−1.69 (m,
2 H), 1.32−1.46 (m, 1 H), 0.65−1.03 (m, 10 H).
3-Ethyl 2-(Phenylmethyl)-8-oxo-2-azaspiro[4.5]decane-2,3-
dicarboxylate (34). To a solution of ethyl 33 (10.61 g, 31.3 mmol)
dissolved in dry CH2Cl2 was added Et3N (10.8 mL, 78 mmol). The
mixture was cooled to 0 °C followed by the addition of Cbz-Cl (6.2 mL,
43.5 mmol), and the mixture was stirred at 0 °C for 5 min at room
temperature for 1 h. The mixture was diluted with CH2Cl2 (700 mL),
washed with 0.1 N HCl (700 mL), dried (MgSO4), and concentrated in
vacuo. The residue was purified by silica gel chromatography, eluting
with 10−60% hexanes/EtOAc to afford the cbz protected compound as
a yellow oil (5.73 g, 39%). To a solution of this intermediate (5.73 g,
12.05 mmol) in THF (60 mL) was added glacial acetic acid (1.38 mL,
24.0 mmol) followed by TBAF (24 mL) as a 1 M solution in THF. The
mixture was stirred at room temperature for 72 h. The mixture was
partitioned between EtOAc and water (250 mL) each. The organic layer
was separated and washed with saturated NaHCO3 (100 mL) and dried
(MgSO4) and concentrated in vacuo. The reaction was found to be
incomplete by TLC. The residue was dissolved in dry THF (60 mL),
cooled to 0 °C, and treated with HF−pyridine (1.6 mL, 18.0 mmol),
warmed to room temperature, and stirred for 2 h under nitrogen. The
mixture was poured into saturated NaHCO3 (100 mL), and solid K2CO3
was added until gas evolution ceased. Extraction was with EtOAc (2 ×
150 mL). The organic layers were combined and washed with 0.1 N HCl
(100 mL), dried (MgSO4), and concentrated in vacuo to afford the title
compound in quantitative yield as a yellow oil which was used in
subsequent reactions without additional purification. To a solution of
the intermediate from above (4.36 g, 120.05 mmol) in dry CH2Cl2 (60
mL) was added Dess−Martin periodinane (10.22 g, 24.1 mmol), and the
mixture was stirred at room temperature under nitrogen for 18 h. The
mixture was poured into 10% aqueous sodium thiosulfate (150 mL) and
saturated NaHCO3 (150 mL) and stirred for 10 min. Extraction was
with CH2Cl2 (2 × 150 mL). The mixture was dried (MgSO4) and
concentrated in vacuo. The residue was purified by silica gel
chromatography, eluting with 25−80% hexanes/EtOAc to afford 34
as a pale yellow oil (2.89 g, 67%).
(632 mg, 23%). The absolute configurations were determined by
vibrational circular dichroism (VCD). 1H NMR (400 MHz, CDCl3) δ
ppm 7.28−7.41 (m, 5 H), 4.98−5.23 (m, 2 H), 4.37 (ddd, J = 19.3, 8.1,
8.0 Hz, 1 H), 4.22 (q, J = 7.2 Hz, 1 H), 3.93−4.12 (m, 1 H), 3.45−3.70
(m, 1 H), 3.36 (dd, J = 10.8, 2.0 Hz, 1 H), 2.22 (dd, J = 12.8, 8.5 Hz, 1 H),
1.79−2.03 (m, 4 H), 1.59−1.78 (m, 4 H), 1.56 (br s, 1 H), 1.28 (t, J = 7.1
Hz, 1 H), 1.22 (d, J = 6.1 Hz, 1 H), 1.13 (t, J = 7.1 Hz, 1 H). LC−MS ESI
(M + H)+ = 381.68.
Ethyl (3S)-8,8-Difluoro-2-azaspiro[4.5]decane-3-carboxylate
(37). To 35 (630 mg, 1.65 mmol) in absolute EtOH (12 mL) was added
20% Pd(OH)2 on carbon (65 mg), and the mixture was hydrogenated
on a Fisher-Porter apparatus for 18 h at 60 psi. The mixture was filtered
through Celite and concentrated in vacuo to afford the title compound
as a clear oil (380 mg, 93%).
(3S)-8,8-Difluoro-2-{N-[(methyloxy)carbonyl]-L-valyl}-2-
azaspiro[4.5]decane-3-carboxylic Acid. To a solution of 37 (380
mg, 1.54 mmol) in anhydrous CH2Cl2 (10 mL) were added HATU (614
mg, 1.6 mmol), N-[(methyloxy)carbonyl]-L-valine (283 mg, 1.6 mmol)
followed by Et3N (0.43 mL, 3.1 mmol), and the mixture was stirred at
room temperature under nitrogen for 1 h. The mixture was concentrated
in vacuo and the residue purified by silica gel chromatography, eluting
with 15−80% hexanes/EtOAc. Appropriate fractions were combined
and concentrated in vacuo. The residue was dissolved in THF/water/
MeOH (5 mL/2.5 mL/2.5 mL), and lithium hydroxide monohydrate
was added (119 mg, 2.8 mmol). The solution was stirred at room
temperature for 30 min. The mixture was treated with 1 N HCl (3.5 mL)
and partitioned between EtOAc and water (50 mL each). The aqueous
layer was extracted with ethyl acetate (50 mL). The organic layers were
combined and dried (MgSO4) and concentrated in vacuo. The residue
was triturated in diethyl ether and concentrated in vacuo to afford the
1
title compound as a white solid (542 mg, 93%). H NMR (400 MHz,
DMSO-d6) δ ppm 12.47 (br s, 1 H), 7.41 (d, J = 8.0 Hz, 1 H), 4.24 (t, J =
8.6 Hz, 1 H), 3.95−4.14 (m, 2 H), 3.52 (s, 3 H), 3.25−3.33 (m, 1 H),
2.22 (dd, J = 12.3, 8.4 Hz, 1 H), 1.79−2.11 (m, 5 H), 1.58−1.77 (m, 3
H), 1.41−1.59 (m, 2 H), 0.93 (dd, J = 12.7, 6.6 Hz, 6 H). LC−MS ESI
(M + H)+ = 377.23. Method B was utilized for the synthesis of 31B.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[2-((2S)-3-methyl-
2-{[(methyloxy)carbonyl]amino}butanoyl)-8,8-dioxido-8-thia-
2-azaspiro[4.5]dec-3-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-
imidazol-2-yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate
(31A). 1H NMR (400 MHz, MeOH-d4) δ ppm 7.64 (br s, 10 H), 7.32
(br s, 2 H), 5.05−5.24 (m, 1 H), 4.04−4.29 (m, 1 H), 3.99 (br s, 1 H),
3.81 (br s, 1 H), 3.55−3.73 (m, 6 H), 3.03−3.25 (m, 4 H), 1.80−2.47
(m, 14 H), 0.68−1.06 (m, 16 H). HRMS for C44H57N8O8S (M + H)+
calcd, 857.4020; found, 857.4020.
Methyl [(1S)-1-({(2S)-2-[4-(4′-{2-[8,8-Difluoro-2-((2S)-3-meth-
yl-2-{[(methyloxy)carbonyl]amino}butanoyl)-2-azaspiro[4.5]-
dec-3-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-
pyrrolidinyl}carbonyl)-2-methylpropyl]carbamate (31B). 1H
NMR (400 MHz, MeOH-d4) δ ppm 7.69−7.82 (m, 3 H), 7.63 (br s,
5 H), 7.17−7.37 (m, 2 H), 5.03−5.19 (m, 1 H), 4.20 (br s, 1 H), 4.06−
4.16 (m, 1 H), 3.96 (br s, 1 H), 3.85 (br s, 1 H), 3.62 (d, J = 3.1 Hz, 6 H),
2.24−2.38 (m, 3 H), 2.15 (br s, 3 H), 1.85−2.09 (m, 8 H), 1.77−1.85
(m, 2 H), 1.53−1.76 (m, 3 H), 0.96 (br s, 4 H), 0.87 (d, J = 6.3 Hz, 12
H). HRMS for C45H57N8O6F2 (M + H)+ calcd, 843.4369; found,
843.4371.
1,1-Dimethylethyl 2-{N-[(Methyloxy)carbonyl]-L-valyl}-3-(4-
{4′-[2-((2S)-1-{N-[(methyloxy)carbonyl]-L-valyl}-2-pyrrolidinyl)-
1H-imidazol-4-yl]-4-biphenylyl}-1H-imidazol-2-yl)-2,8-
diazaspiro[4.5]decane-8-carboxylate (31C). 1H NMR (400 MHz,
MeOH-d4) δ ppm 7.56−7.87 (m, 10 H), 7.31 (br s, 2 H), 5.04−5.23 (m,
1 H), 4.11−4.32 (m, 2 H), 3.93−4.04 (m, 1 H), 3.80−3.92 (m, 1 H),
3.63 (s, 6 H), 3.33−3.61 (m, 6 H), 1.86−2.49 (m, 9 H), 1.68 (br s, 1 H),
1.56 (br s, 2 H), 1.37−1.49 (m, 9 H), 0.81−1.08 (m, 15 H). HRMS for
C49H65N9O8 (M + H)+ calcd, 908.5034; found, 908.5031.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[2-((2S)-3-methyl-
2-{[(methyloxy)carbonyl]amino}butanoyl)-2,8-diazaspiro[4.5]-
dec-3-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-
pyrrolidinyl}carbonyl)propyl]carbamate (31D). 1H NMR (400
MHz, MeOH-d4) δ ppm 7.56−7.83 (m, 10 H), 7.31 (br s, 2 H), 5.00−
5.22 (m, 2 H), 4.12−4.30 (m, 3 H), 4.01 (br s, 1 H), 3.80−3.91 (m, 1 H),
3-Ethyl 2-(Phenylmethyl)-(3S)-8,8-difluoro-2-azaspiro[4.5]-
decane-2,3-dicarboxylate (35). To a solution of 34 (2.89 g, 7.13
mmol) in anhydrous CH2Cl2 (50 mL) was added Deoxo-Fluor (2.2 mL,
12.1 mmol) followed by a catalytic amount of EtOH, and the mixture
was stirred at room temperature under nitrogen. After 2.5 h the mixture
was poured into saturated NaHCO3 (150 mL) and stirred for 10 min.
Extraction was with CH2Cl2 (2 × 150 mL), and the organic layer was
washed with 0.1 N HCl (100 mL), dried (MgSO4), and concentrated in
vacuo to afford the desired compound (as a racemic mixture) as a yellow
oil (3.01 g) which was found to be contaminated with 36 in a 1:1 ratio.
The residue was dissolved in dry CH2Cl2 (35 mL) and treated with
mCPBA (77%, 1.66 g, 7.45 mmol) and stirred under nitrogen for 18 h.
The mixture was poured into saturated NaHCO3 (40 mL) and 10%
aqueous sodium thiosulfate (40 mL) and stirred for 10 min. Extraction
was with CH2Cl2 (100 mL). The mixture was dried (MgSO4) and
concentrated in vacuo. The residue was purified by silica gel
chromatography, eluting with 5−50% hexanes/EtOAc. The racemate
was then separated by chiral HPLC on a 10 μm o.d. column, eluting with
25% isopropanol in hexanes to afford the title compound as a clear oil
2070
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
3.57−3.70 (m, 6 H), 2.76−3.07 (m, 4 H), 1.89−2.48 (m, 10 H), 1.69−
1.81 (m, 1 H), 1.61 (br s, 2 H), 0.82−1.09 (m, 15 H). HRMS for
C44H58N9O6 (M + H)+ calcd, 808.4510; found, 808.4509.
phenylyl)-1H-imidazol-2-yl]-1-pyrrolidinyl}carbonyl)propyl]-
carbamate (31L). 1H NMR (400 MHz, MeOH-d4) δ ppm 7.64 (br s,
10 H), 7.33 (br s, 2 H), 5.06−5.22 (m, 1 H), 4.03−4.46 (m, 3 H), 3.73−
4.04 (m, 8 H), 3.54−3.73 (m, 6 H), 3.46 (q, J = 7.0 Hz, 2 H), 2.52−2.76
(m, 5 H), 2.10−2.53 (m, 4 H), 2.02 (br s, 3 H), 0.70−1.09 (m, 12 H).
HRMS for C44H57N10O7 (M + H)+ calcd, 837.4412; found, 838.4416.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[6-((2S)-3-methyl-
2-{[(methyloxy)carbonyl]amino}butanoyl)-2-(methylsulfonyl)-
2,6-diazaspiro[3.4]oct-7-yl]-1H-imidazol-4-yl}-4-biphenylyl)-
1H-imidazol-2-yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate
(31M). 1H NMR (400 MHz, MeOH-d4) δ ppm 7.50−7.86 (m, 10 H),
7.33 (br s, 2 H), 5.06−5.21 (m, 1 H), 4.04−4.26 (m, 2 H), 3.75−4.05
(m, 6 H), 3.56−3.72 (m, 6 H), 3.38−3.55 (m, 2 H), 2.84−3.04 (m, 3 H),
2.70−2.86 (m, 1 H), 2.40−2.64 (m, 1 H), 2.10−2.40 (m, 3 H), 1.83−
2.09 (m, 2 H), 1.20−1.37 (m, 2 H), 1.07−1.20 (m, 2 H), 0.81−1.07 (m,
12 H). HRMS for C43H56N9O8S (M + H)+ calcd, 858.3973; found,
858.3975.
Methyl [(1S)-1-({(2S)-2-[4-(4′-{2-[8-Acetyl-2-((2S)-3-methyl-2-
{[(methyloxy)carbonyl]amino}butanoyl)-2,8-diazaspiro[4.5]-
dec-3-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-
pyrrolidinyl}carbonyl)-2-methylpropyl]carbamate (31E). 1H
NMR (400 MHz, MeOH-d4) δ ppm 7.54−7.88 (m, 10 H), 7.32 (br s,
2 H), 5.05−5.23 (m, 2 H), 4.03−4.34 (m, 2 H), 3.71−4.03 (m, 3 H),
3.56−3.71 (m, 8 H), 3.34−3.56 (m, 2 H), 2.14−2.46 (m, 4 H), 1.89−
2.14 (m, 6 H), 1.46−1.83 (m, 4 H), 1.19−1.32 (m, 2 H), 0.78−1.07 (m,
14 H). HRMS for C46H60N9O7 (M + H)+ calcd, 850.4616; found,
850.4617.
Methyl 2-{N-[(Methyloxy)carbonyl]-L-valyl}-3-(4-{4′-[2-((2S)-
1-{N-[(methyloxy)carbonyl]-L-valyl}-2-pyrrolidinyl)-1H-imida-
zol-4-yl]-4-biphenylyl}-1H-imidazol-2-yl)-2,8-diazaspiro[4.5]-
decane-8-carboxylate (31F). 1H NMR (400 MHz, MeOH-d4) δ ppm
7.51−7.92 (m, 10 H), 7.32 (br s, 2 H), 5.02−5.25 (m, 1 H), 4.03−4.31
(m, 2 H), 3.75−4.05 (m, 2 H), 3.61−3.72 (m, 7 H), 3.34−3.60 (m, 4 H),
1.82−2.50 (m, 7 H), 1.70 (br s, 2 H), 1.44−1.63 (m, 4 H), 1.27 (br s, 4
H), 0.77−1.10 (m, 14 H). HRMS for C46H60N9O8 (M + H)+ calcd,
866.4565; found, 850.4564.
Methyl [(1S)-1-({(2S)-2-[4-(4′-{2-[(7S)-2,2-Difluoro-6-((2S)-3-
methyl-2-{[(methyloxy)carbonyl]amino}butanoyl)-6-azaspiro-
[3.4]oct-7-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-
yl]-1-pyrrolidinyl}carbonyl)-2-methylpropyl]carbamate (31N).
1H NMR (400 MHz, DMSO-d6) δ ppm 11.42−12.37 (m, 2 H),
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[2-((2S)-3-methyl-
2-{[(methyloxy)carbonyl]amino}butanoyl)-8-oxa-2-azaspiro-
[4.5]dec-3-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-
yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate (31G). 1H NMR
(400 MHz, MeOH-d4) δ ppm 7.55−7.87 (m, 8 H), 7.32 (br s, 4 H),
5.02−5.23 (m, 3 H), 4.13−4.34 (m, 4 H), 3.93−4.04 (m, 1 H), 3.88 (br
s, 2 H), 3.49−3.84 (m, 8 H), 1.86−2.54 (m, 10 H), 1.45−1.84 (m, 4 H),
0.69−1.12 (m, 12 H). HRMS for C44H57N8O7 (M + H)+ calcd,
809.4350; found, 809.4346.
1,1-Dimethylethyl 6-{N-[(Methyloxy)carbonyl]-L-valyl}-7-(4-
{4′-[2-((2S)-1-{N-[(methyloxy)carbonyl]-L-valyl}-2-pyrrolidinyl)-
1H-imidazol-4-yl]-4-biphenylyl}-1H-imidazol-2-yl)-2,6-
diazaspiro[3.4]octane-2-carboxylate (31H). Pale yellow solid (315
mg, 80%). 1H NMR (400 MHz, DMSO-d6) δ ppm 11.63−12.35 (m, 1
H), 7.47−7.92 (m, 9 H), 7.19−7.42 (m, 1 H), 5.08 (br s, 2 H), 3.94−
4.23 (m, 4 H), 3.62−3.93 (m, 7 H), 3.56 (s, 6 H), 2.24−2.45 (m, 2 H),
2.14 (br s, 2 H), 1.77−2.07 (m, 6 H), 1.37 (d, J = 5.5 Hz, 9 H), 0.76−
1.02 (m, 12 H). HRMS for C47H62N9O8 (M + H)+ calcd, 880.4721;
found, 880.4725.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[6-((2S)-3-methyl-
2-{[(methyloxy)carbonyl]amino}butanoyl)-2,6-diazaspiro[3.4]-
oct-7-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-
pyrrolidinyl}carbonyl)propyl]carbamate (31I). 1H NMR (400
MHz, MeOH-d4) δ ppm 7.43−7.86 (m, 8 H), 7.04−7.42 (m, 2 H),
5.20−5.41 (m, 1 H), 5.01−5.19 (m, 2 H), 4.30−4.43 (m, 1 H), 4.12−
4.27 (m, 3 H), 3.78−4.03 (m, 5 H), 3.65−3.78 (m, 4 H), 3.61−3.64 (m,
6 H), 3.39−3.52 (m, 1 H), 2.54−2.77 (m, 1 H), 2.38−2.54 (m, 1 H),
2.09−2.39 (m, 2 H), 1.89−2.13 (m, 4 H), 0.81−1.13 (m, 12 H). HRMS
for C42H54N9O6 (M + H)+ calcd, 780.4197; found, 780.4200.
Methyl [(1S)-1-({(2S)-2-[4-(4′-{2-[2-Acetyl-6-((2S)-3-methyl-2-
{[(methyloxy)carbonyl]amino}butanoyl)-2,6-diazaspiro[3.4]-
oct-7-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-1-
pyrrolidinyl}carbonyl)-2-methylpropyl]carbamate (31J). 1H
NMR (400 MHz, DMSO-d6) δ ppm 11.67−12.33 (m, 1 H), 7.19−
7.99 (m, 12 H), 5.07 (br s, 1 H), 3.92−4.25 (m, 4 H), 3.67−3.94 (m, 4
H), 3.45−3.66 (m, 8 H), 2.20−2.46 (m, 3 H), 2.17 (br s, 2 H), 1.80−
2.06 (m, 5 H), 1.60−1.82 (m, 3 H), 0.78−1.02 (m, 12 H). HRMS for
C44H56N9O7 (M + H)+ calcd, 822.4303; found, 822.4300.
Methyl 6-{N-[(Methyloxy)carbonyl]-L-valyl}-7-(4-{4′-[2-((2S)-
1-{N-[(methyloxy)carbonyl]-L-valyl}-2-pyrrolidinyl)-1H-imida-
zol-4-yl]-4-biphenylyl}-1H-imidazol-2-yl)-2,6-diazaspiro[3.4]-
octane-2-carboxylate (31K). 1H NMR (400 MHz, DMSO-d6) δ ppm
11.38−12.37 (m, 1 H), 7.58−7.93 (m, 8 H), 7.19−7.59 (m, 3 H), 4.93−
5.17 (m, 1 H), 3.63−4.22 (m, 7 H), 3.45−3.64 (m, 6 H), 2.24−2.48 (m,
4 H), 2.08 (br s, 4 H), 1.79−2.07 (m, 6 H), 0.99−1.18 (m, 1 H), 0.75−
0.97 (m, 14 H). HRMS for C44H56N9O8 (M + H)+ calcd, 838.4252;
found, 838.4252.
7.56−7.90 (m, 8 H), 7.51 (s, 2 H), 7.19−7.46 (m, 2 H), 4.73−5.39 (m, 2
H), 3.91−4.23 (m, 3 H), 3.82 (br s, 2 H), 3.54 (s, 6 H), 2.55−2.86 (m, 4
H), 2.20−2.45 (m, 2 H), 2.14 (br s, 2 H), 1.73−2.09 (m, 5 H), 0.57−
0.99 (m, 12 H). HRMS for C43H53N8O6F2 (M + H)+ calcd, 815.4056;
found, 815.4059.
Intermediates 38, 39A−H, 40A−H and products 41A−H were
synthesized as previously described.28
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[(8S)-7-((2S)-3-
methyl-2-{[(methyloxy)carbonyl]amino}butanoyl)-1,4-dithia-7-
azaspiro[4.4]non-8-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imi-
dazol-2-yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate (41A). 1H
NMR (400 MHz, CDCl3) δ ppm 10.07−11.08 (m, 2 H), 7.82 (br s, 3
H), 7.58 (d, J = 6.02 Hz, 5 H), 7.10−7.30 (m, 2 H), 5.54 (d, J = 9.29 Hz,
2 H), 5.17−5.42 (m, 3 H), 4.01−4.48 (m, 2 H), 3.23−4.03 (m, 10 H),
2.70−3.19 (m, 2 H), 2.30−2.49 (m, 1 H), 1.91−2.29 (m, 4 H), 1.19−
1.56 (m, 4 H), 1.07 (dd, J = 10.54, 7.03 Hz, 2 H), 0.58−0.97 (m, 10 H).
HRMS (M + H)+ calcd, 829.3530; found, 829.3534.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[(8S)-7-((2S)-2-
{[(methyloxy)carbonyl]amino}butanoyl)-1,4-dithia-7-azaspiro-
[4.4]non-8-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-
yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate (41B). 1H NMR
(400 MHz, CDCl3) δ ppm 9.99−11.64 (m, 2 H), 7.53 (br s, 8 H), 7.12−
7.34 (m, 2 H), 5.26 (br s, 2 H), 3.18−4.64 (m, 17 H), 2.79 (br s, 2 H),
1.43−2.62 (m, 8 H), 0.70−1.19 (m, 9 H). HRMS (M + H)+ calcd,
815.3373; found, 815.3373.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[(8S)-7-
({[(methyloxy)carbonyl]amino}acetyl)-1,4-dithia-7-azaspiro-
[4.4]non-8-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-
yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate (41C). 1H NMR
(400 MHz, CDCl3) δ ppm 9.86−11.31 (m, 2 H), 7.50 (br s, 8 H), 6.91−
7.27 (m, 2 H), 5.41−6.31 (m, 2 H), 5.24 (br s, 2 H), 3.12−4.45 (m, 18
H), 2.77 (br s, 2 H), 1.77−2.50 (m, 4 H), 0.58−1.15 (m, 6 H). HRMS
(M + H)+ calcd, 787.3055; found, 787.3056.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[(3S)-2-((2S)-3-
methyl-2-{[(methyloxy)carbonyl]amino}butanoyl)-6,10-dioxa-
2-azaspiro[4.5]dec-3-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-
imidazol-2-yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate
1
(41D). H NMR (400 MHz, CDCl3) δ ppm 10.20−11.06 (m, 2 H),
7.35−8.02 (m, 8 H), 7.22 (d, J = 7.78 Hz, 2 H), 4.97−5.89 (m, 4 H),
3.22−4.58 (m, 15 H), 2.83−3.20 (m, 1 H), 2.58−2.83 (m, 1 H), 2.29−
2.58 (m, 1 H), 1.57−2.31 (m, 8 H), 0.64−1.22 (m, 12 H). HRMS (M +
H)+ calcd, 811.4143; found, 811.4142.
Methyl ((1S)-1-Methyl-2-{(3S)-3-[4-(4′-{2-[(2S)-1-((2S)-3-
methyl-2-{[(methyloxy)carbonyl]amino}butanoyl)-2-pyrroli-
dinyl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imidazol-2-yl]-6,10-
1
dioxa-2-azaspiro[4.5]dec-2-yl}-2-oxoethyl)carbamate (41E). H
NMR (400 MHz, CDCl3) δ ppm 10.04−11.11 (m, 2 H), 7.37−8.02 (m,
8 H), 7.18−7.36 (m, 2 H), 5.60 (br s, 2 H), 5.17−5.40 (m, 2 H), 3.17−
4.73 (m, 14 H), 2.79−3.19 (m, 1 H), 2.45−2.81 (m, 1 H), 2.29−2.45
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[2-[(methylamino)-
carbonyl]-6-((2S)-3-methyl-2-{[(methyloxy)carbonyl]amino}-
butanoyl)-2,6-diazaspiro[3.4]oct-7-yl]-1H-imidazol-4-yl}-4-bi-
2071
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
(m, 1 H), 1.49−2.29 (m, 8 H), 1.22−1.47 (m, 3 H), 0.73−1.16 (m, 6 H).
HRMS (M + H)+ calcd, 783.3830; found, 783.3832.
Present Address
†M.D.: HD Biosciences (China) Co., Ltd., Shanghai, China.
Methyl [(1S)-1-({(2S)-2-[4-(4′-{2-[(3S)-8,8-Dimethyl-2-((2S)-3-
methyl-2-{[(methyloxy)carbonyl]amino}butanoyl)-6,10-dioxa-
2-azaspiro[4.5]dec-3-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-
imidazol-2-yl]-1-pyrrolidinyl}carbonyl)-2-methylpropyl]-
carbamate (41F). 1H NMR (400 MHz, DMSO-d6) δ (ppm) 11.84 (br
s, 1 H), 11.80 (br s, 1H), 7.90−7.60 (br m, 9H), 7.52 (s, 1H), 7.36−7.25
(br m, 2H), 5.09 (br m, 1H), 4.98 (m, 1H), 4.49 (br m, 1H), 4.14−4.02
(br m, 2H), 3.81 (br, 2H), 3.73−3.40 (br m, 11H), 2.65 (m, 1H), 2.43−
2.09 (br m, 3H), 2.07−1.81 (br m, 4H), 1.05−0.75 (br m, 18H). ES
LC−MS m/z = 839 (M + H)+.
Notes
All authors are current or former employees of GlaxoSmithKline.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank all members of the program team, especially Luz H.
Kryn for replicon screening, Amy Wang for virological
characterization of compounds, Joel Cooper, Robert McFadyen,
and Aniko Redman for expert chemistry support, and Andrew
Spaltenstein and Zhi Hong for encouragement and supporting
this work.
Methyl [(1S)-1-({(2S)-2-[4-(4′-{2-[(3S,7S,9S)-7,9-Dimethyl-2-
((2S)-3-methyl-2-{[(methyloxy)carbonyl]amino}butanoyl)-6,10-
dioxa-2-azaspiro[4.5]dec-3-yl]-1H-imidazol-4-yl}-4-biphenyl-
yl)-1H-imidazol-2-yl]-1-pyrrolidinyl}carbonyl)-2-
1
methylpropyl]carbamate (41G). HNMR (400 MHz, CDCl3) δ
ppm 10.23−11.01 (m, 1 H), 7.31−8.08 (m, 8 H), 7.23 (d, J = 8.03 Hz, 2
H), 5.13−5.89 (m, 4 H), 3.34−4.69 (m, 13 H), 2.84−3.31 (m, 2 H),
2.63−2.84 (m, 1 H), 2.29−2.53 (m, 1 H), 1.85−2.29 (m, 4 H), 1.56−
1.85 (m, 4 H), 1.16−1.47 (m, 6 H), 0.63−1.16 (m, 12 H). HRMS (M +
H)+ calcd, 835.4456; found, 835.4458.
Methyl [(1S)-2-Methyl-1-({(2S)-2-[4-(4′-{2-[(8S)-7-((2S)-3-
methyl-2-{[(methyloxy)carbonyl]amino}butanoyl)-1,4-dioxa-7-
azaspiro[4.4]non-8-yl]-1H-imidazol-4-yl}-4-biphenylyl)-1H-imi-
dazol-2-yl]-1-pyrrolidinyl}carbonyl)propyl]carbamate (41H).
1H NMR (400 MHz, CDCl3) δ ppm 10.00−11.36 (m, 2 H), 7.56 (br
s, 10 H), 7.02−7.34 (m, 2 H), 5.05−5.89 (m, 4 H), 3.76−4.65 (m, 6 H),
3.53−3.83 (m, 6 H), 2.77−3.54 (m, 2 H), 2.26−2.70 (m, 2 H), 1.45−
2.26 (m, 6 H), 0.61−1.25 (m, 12 H). HRMS (M + H)+ calcd, 797.3986;
found, 797.3981.
ABBREVIATIONS USED
■
HCV, hepatitis C virus; gt1a, genotype 1a; gt1b, genotype 1b;
NS5A, nonstructural protein 5A; CM, carboxymethyl; wt, wild
type
REFERENCES
■
(1) Alter, M. J. Epidemiology of hepatitis C virus infection. World J.
Gastroenterol. 2007, 13, 2436−2441.
(2) Kamal, S. M.; Fouly, A. E.; Kamel, R. R.; Hockenjos, B.; Al Tawil,
A.; Khalifa, K. E.; He, Q.; Koziel, M. J.; El Naggar, K. M.; Rasenack, J.;
Afdhal, N. H. Peginterferon alfa-2b therapy in acute hepatitis C: impact
of onset of therapy on sustained virologic response. Gastroenterology
2006, 130, 632−638.
(3) Tong, M. J.; El-Farra, N. S.; Reikes, A. R.; Co, R. L. Clinical
outcomes after transfusion-associated hepatitis C. N. Engl. J. Med. 1995,
332, 1463−1466.
(4) Adeel, A. B.; Fasiha, K. Boceprevir and telaprevir in the
management of hepatitis C virus-infected patients. Clin. Infect. Dis.
2012, 54, 96−104.
(5) Lemon, S. M.; McKeating, J. A.; Pietschmann, T.; Frick, D. N.;
Glenn, J. S.; Tellinghuisen, T. L.; Symons, J.; Furman, P. A.
Development of novel therapies for hepatitis C. Antiviral Res. 2010,
86, 79−92.
(6) Poenisch, M.; Bartenschlager, R. New insights into structure and
replication of the hepatitis C virus and clinical implications. Semin. Liver
Dis. 2010, 30, 333−347.
(7) Gao, M.; Nettles, R. E.; Belema, M.; Snyder, L. B.; Nguyen, V. N.;
Fridell, R. A.; Serrano-Wu, M. H.; Langley, D. R.; Sun, J.-H.; O’Boyle, D.
R., II; Lemm, J. A.; Wang, C.; Knipe, J. O.; Chien, C.; Colonno, R. J.;
Grasela, D. M.; Meanwell, N. A.; Hamann, L. G. Chemical genetics
strategy identifies an HCV NS5A inhibitor with a potent clinical effect.
Nature 2010, 465, 96−100.
(8) Fridell, R. A.; Qiu, D.; Wang, C.; Valera, L.; Gao, M. Resistance
analysis of the hepatitis C virus NS5A inhibitor BMS-790052 in an in
vitro replicon system. Antimicrob. Agents Chemother. 2010, 54, 3641−
3650.
(9) Lemm, J. A.; Leet, J. E.; O’Boyle, D. R.; Romine, J. L.; Huang, X. S.;
Schroeder, D. R.; Alberts, J.; Cantone, J. L.; Sun, J. H; Nower, P. T.;
Martin, S. W.; Serrano-Wu, M. H.; Meanwell, N. A.; Snyder, L. B.; Gao,
M. Discovery of potent hepatitis C virus NS5A inhibitors with dimeric
structures. Antimicrob. Agents Chemother. 2011, 55, 3795−3802.
(10) Romine, J. L.; St. Laurent, D. R.; Leet, J. E.; Martin, S. W.; Serrano-
Wu, M. H.; Yang, F.; Gao, M.; O’Boyle, D. R.; Lemm, J. A.; Sun, J.-H.;
Nower, P. T.; Huang, X.; Deshpande, M. S.; Meanwell, N. A.; Snyder, L.
B. Inhibitors of HCV NS5A: from iminothiazolidinones to symmetrical
stilbenes. ACS Med. Chem. Lett. 2011, 2, 224−229.
Acids 32 and 40 were synthesized as previously described.28
Step A. To a solution of 14 (0.39 mmol) and the carboxylic acid
component (0.86 mmol) in anhydrous CH3CN (4 mL) was added Et3N
(0.19 mL, 1.4 mmol), and the solution was stirred at room temperature
under nitrogen for 3.5 h. The mixture was partitioned between EtOAc
and 0.1 N HCl (40 mL each). The organic layer washed with brine and
dried (MgSO4) and concentrated in vacuo. The residue was purified by
silica gel chromatography, eluting with 20−100% hexanes/EtOAc to
afford the diester product (yield 60−70%).
Step B. A mixture of diesters (97.2 mmol) and NH4OAc (0.778 mol)
in dioxane (50 mL) was heated to 160 °C for 20 min in the microwave
reactor. The reaction mixture was concentrated and residue dissolved in
EtOAc (500 mL) and washed with aqueous NaHCO3 and brine. The
organic phase was dried over Na2SO4 and concentrated to dryness. The
crude product was purified by chromatography on silica gel (petroleum
ether/ethyl acetate = 1/1) to give the product as a brown solid.
Dimethyl (4,4′-Biphenyldiylbis{1H-imidazole-4,2-diyl(3S)-8-
oxa-2-azaspiro[4.5]decane-3,2-diyl[(2S)-3-methyl-1-oxo-1,2-
butanediyl]})biscarbamate (43A). 1H NMR (400 MHz, MeOH-d4)
δ ppm 7.60−7.79 (m, 10 H), 7.33 (s, 2 H), 5.11 (dd, J = 9.9, 7.9 Hz, 2H),
4.30 (d, J = 10.2 Hz, 2 H), 4.17 (d, J = 8.2 Hz, 2 H), 3.70−3.87 (m, 6 H),
3.56−3.71 (m, 10 H), 2.40 (dd, J = 12.9, 7.8 Hz, 2 H), 2.13 (dd, J = 12.6,
10.3 Hz, 2H), 1.88−2.00 (m, 2H), 1.70−1.84 (m, 4H), 1.46−1.69 (m, 6
H), 0.83−0.97 (m, 12 H). HRMS for C48H63N8O8 (M + H)+ calcd,
879.4769; found, 879.4769.
Dimethyl (4,4′-Biphenyldiylbis{1H-imidazole-4,2-diyl(8S)-
1,4-dioxa-7-azaspiro[4.4]nonane-8,7-diyl[(2S)-3-methyl-1-oxo-
1,2-butanediyl]})biscarbamate (43B). 1H NMR (400 MHz, CDCl3)
δ ppm 11.10−11.23 (m, 1 H), 10.10−11.63 (m, 2 H), 7.40−8.15 (m, 8
H), 7.27−7.40 (m, 2 H), 5.23−5.88 (m, 2 H), 4.31 (dd, J = 8.7, 6.5 Hz, 2
H), 3.83−4.21 (m, 10 H), 3.57−3.83 (m, 6 H), 3.05−3.46 (m, 3 H), 2.47
(dd, J = 13.5, 8.6 Hz, 2 H), 1.51−2.20 (m, 2 H), 1.08 (d, J = 6.8 Hz, 2H),
0.60−0.97 (m, 12H). HRMS (M + H)+ calcd, 855.4041; found,
855.4039.
(11) Lopez, O. D.; Nguyen, V. N.; St. Laurent, D. R.; Belema, M.;
Serrano-Wu, M. H.; Goodrich, J. T.; Yang, F.; Qiu, Y.; Ripka, A. S.;
Nower, P. T.; Valera, L.; Liu, M.; O’Boyle, D. R.; Sun, J. H.; Fridell, R. A.;
Lemm, J. A.; Gao, M.; Good, A. C.; Meanwell, N. A.; Snyder, L. B. HCV
NS5A replication complex inhibitors. Part 3: Discovery of potent
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 919-483-9462. E-mail: wieslaw.m.kazmierski@gsk.com.
2072
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073

Journal of Medicinal Chemistry
Article
analogs with distinct core topologies. Bioorg. Med. Chem. Lett. 2013, 23,
779−784.
(27) Fridell, R. A.; Qiu, D.; Valera, L.; Wang, C.; Rose, R. E.; Gao, M.
Distinct functions of NS5A in hepatitis C virus RNA replication
uncovered by studies with the NS5A Inhibitor BMS-790052. J. Virol.
2011, 85, 7312−7320.
(28) Chen, P.; Couch, R.; Duan, M.; Grimes, R. M.; Kazmierski, W. M.;
Norton, B. A.; Tallant, M. Preparation of biphenyls end capped with
amino acids or peptide derivatives for treating HCV infection. WO
2011/028596, 2010.
(29) Baskaran, S.; Botyanszki, J.; Cooper, J. P.; Duan, M.; Kazmierski,
W. M.; Mcfadyen, R. B.; Redman, A. Preparation of phenylene,
pyridinylene, furanylene and thiofuranylene linked benzimidazoles and
derivatives end capped with peptide derivatives for treating HCV
infection. WO 2011/050146, 2010.
(30) Baskaran, S.; Dickerson, S. H.; Duan, M.; Kazmierski, W. M.;
McFadyen, R. B. Preparation of naphthalenylene, quinolinylene and
isoquinolinylene linked benzimidazole-imidazoles and analogs end
capped with peptide derivatives for treating HCV infection. WO 2011/
091446, 2011.
(31) Wang, C.; Huang, H.; Valera, L.; Sun, J.-H.; O’Boyle, D. R., II;
Nower, P. T; Jia, L.; Qiu, D.; Huang, X.; Altaf, A.; Gao, M.; Fridella, R. A.
Hepatitis C virus RNA elimination and development of resistance in
replicon cells treated with BMS-790052. Antimicrob. Agents Chemother.
2012, 56, 1350−1358.
(32) Siebum, A. H. G.; Woo, W. S.; Raap, J.; Lugtenburg, J. Access to
any site-directed isotopomer of methionine, selenomethionine, cysteine,
and selenocysteineuse of simple, efficient modular synthetic reaction
schemes for isotope incorporation. Eur. J. Org. Chem. 2004, 13, 2905−
2913.
(12) St Laurent, D. R; Belema, M.; Gao, M.; Goodrich, J.; Kakarla, R.;
Knipe, J. O; Lemm, J. A; Liu, M.; Lopez, O. D; Nguyen, V. N; Nower, P.
T.; O’Boyle, D.; Qiu, Y.; Romine, J. L.; Serrano-Wu, M. H.; Sun, J. H.;
Valera, L.; Yang, F.; Yang, X.; Meanwell, N. A.; Snyder, L. B. HCV NS5A
replication complex inhibitors. Part 2: Investigation of stilbene
prolinamides. Bioorg. Med. Chem. Lett. 2012, 22, 6063−6.
(13) Schmitz, U.; Tan, S.-L. NS5Afrom obscurity to new target for
HCV therapy. Recent patents on anti-infective. Drug Discovery 2008, 3,
77−92.
(14) Cordek, D. G.; Bechtel, J. T.; Maynard, A. T.; Kazmierski, W. M.;
Cameron, C. E. Targeting the protein NS5A protein of HCV: an
emerging option. Drugs Future 2011, 36, 691−711.
(15) Hamatake, R.; Maynard, A.; Kazmierski, W. M. HCV Inhibition
mediated through the nonstructural protein 5A (NS5A) replication
complex. Annu. Rep. Med. Chem. 2012, 47, 331−345.
(16) Schmitz, F. U.; Roberts, C. D.; Abadi, A. D. M.; Griffith, R. C.;
Leivers, M. R.; Slobodov, I.; Rai, R. N-Heterocyclylpyrrolidine
carboxamide derivatives, processes for preparing them, pharmaceutical
compositions containing them, and their use as antiviral agents. WO
2007/070600, 2006.
(17) Maynard, A.; Crosby, R. M.; Ellis, B.; Hamatake, R.; Hong, Z.;
Johns, B. A.; Kahler, K. M.; Koble, C.; Leivers, A.; Leivers, M. R.; Mathis,
A.; Peat, A. J.; Pouliot, J. J.; Roberts, C. D.; Samano, V.; Schmidt, R. M.;
Smith, G. K.; Spaltenstein, A.; Stewart; E. L.; Thommes, P.; Turner, E.
M.; Voitenleitner, C.; Walker, J. T.; Waitt, G.; Weatherhead, J.; Weaver,
K.; Williams, S.; Wright, L.; Xiong, Z. Z.; Haigh, D.; Shotwell, J. B.
Discovery of a potent boronic acid derived inhibitor of the HCV RNA-
dependent RNA polymerase. J. Med. Chem. 2013, DOI: 10.1021/
jm400317w.
(18) Leivers, M. R.; Schmitz, F. U.; Griffith, R. C.; Roberts, C. D.;
Dehghani Mohammad Abadi, A.; Chan, S. A.; Rai, R.; Slobodov, I.; Ton,
T. L. Preparation of thiazolidine derivatives as antiviral agents. WO
2008/064218, 2007.
(19) Schmitz, F. U.; Rai, R.; Roberts, C. D.; Kazmierski, W.; Grimes, R.
Preparation of biphenyls and biheteroaryls end-capped with amino acid
or peptide derivatives as antivirals for treating Flaviviridae family virus
infection. WO 2010/062821, 2009.
(20) Guedj, J.; Dahari, H.; Rong, L.; Sansone, N. D.; Nettles, R. E.;
Cotler, S. J.; Layden, T J.; Uprichard, S. L.; Perelson, A. S. Modeling
shows that the NS5A inhibitor daclatasvir has two modes of action and
yields a shorter estimate of the hepatitis C virus half-life. Proc. Natl. Acad.
Sci. U.S.A. 2013, 110, 3991−3996.
(21) Lemm, J. A.; O’Boyle, D.; Liu, M.; Nower, P.; Colono, R.;
Deshpande, M.; Snyder, L.; Martin, S.; St. Laurent, D.; Serrano-Wu, M.;
Romine, J.; Meanwell, N.; Gao, M. Identification of hepatitis C virus
NS5A inhibitors. J. Virol. 2010, 84, 482−491.
(33) Henriksen, B. S.; Anderson, J. L.; Hrycyna, C. A.; Gibbs, R. A.
Synthesis of desthio prenylcysteine analogs: sulfur is important for
biological activity. Bioorg. Med. Chem. Lett. 2005, 15, 5080−5083.
(34) Sot
Balogh, Z.; Szan
compounds. LV. Synthesis of ( )-desmethoxy cuanzine. Tetrahedron
1991, 47, 271−296.
́
i, F.; Kajtar
́
-Peredy, M.; Keresztury, G.; Incze, M.; Kardos-
́
tay, C. Synthesis of vinca alkaloids and related
(35) Walker, J.; Crosby, R.; Wang, A.; Woldu, E.; Vamathevan, J.; You,
S.; Voitenleitner, C.; Remlinger, K.; Duan, M.; Kazmierski, W.;
Hamatake, R. Preclinical characterization of GSK2336805, a novel
inhibitor of hepatitis C virus replication that selects for resistance in
NS5A. Antimicrob. Agents Chemother. 2014, 58, 38−47.
(36) Spreen, W.; Wilfret, D.; Bechtel, J.; Adkison, K. K.; Lou, Y.; Jones,
L. Willsie, S. K.; Glass, S. J.; Roberts, C. D. GSK2336805 HCV NS5A
inhibitor demonstrates potent antiviral activity in chronic genotype 1
infections: results from a first time in human (FTIH) single and repeat
dose study. Presented at the 62nd Annual Meeting of the American
Association for the Study of Liver Diseases, San Francisco, CA, U.S.,
Nov 6−9, 2011; Hepatology, 2011, 54, 400A.
(22) Targett-Adams, P.; Graham, E. J. S.; Middleton, J.; Palmer, A.;
Shaw, S.; Lavender, H.; Brain, P.; Tran, T. D.; Jones, L.; Wakenhut, F.;
Stammen, B.; Pryde, D.; Pickford, C.; Westby, M. Small molecules
targeting hepatitis C virus-encoded NS5A cause subcellular redistrib-
ution of their target: insights into compound modes of action. J. Virol.
2011, 85, 6353−6358.
(23) O’Boyle, D. R.; Sun, J. H.; Nower, P. T.; Lemm, J. A.; Fridell, R. A.;
Wang, C.; Romine, J. L.; Belema, M.; Nguyen, V. N.; Laurent, D. R.;
Serrano-Wu, M.; Synder, L. B.; Meanwell, N.; Langley, D.; Gao, M.
Characterizations of HCV NS5A replication complex inhibitors.
Virology 2013, 444, 343−354.
(24) Tellinghuisen, T. L.; Marcotrigiano, J.; Rice, C. M. Structure of the
zinc-finding comain of an essential replicase component of hepatisis C
virus reveals a novel fold. Nature 2005, 435, 374−379.
(25) Love, R. A.; Brodsky, O.; Hickey, M. J.; Wells, P. A.; Cronin, C. N.
Crystal structure of a novel dimeric form of NS5A domain I protein from
hepatitis C virus. J. Virol. 2009, 83, 4395−4403.
(26) Penin, F.; Brass, V.; Appel, N.; Ramboarina, S.; Montserret, R.;
Ficheux, D.; Blum, H. E.; Bartenschlager, R.; Moradpour, D. Structure
and function of the membrane anchor domain of hepatitis C virus
nonstructural protein 5A. J. Biol. Chem. 2004, 279, 40835−40843.
2073
dx.doi.org/10.1021/jm4013104 | J. Med. Chem. 2014, 57, 2058−2073