Enantioselective synthesis of an HCV NS5a antagonist

Article abstract of DOI:10.1021/ol500971c

A concise, enantioselective synthesis of the HCV NS5a inhibitor MK-8742 (1) is reported. The features of the synthesis include a highly enantioselective transfer hydrogenation of an NH imine and a dynamic diastereoselective transformation. The synthesis of this complex target requires simple starting materials and nine linear steps for completion.

Full text of DOI:10.1021/ol500971c

Letter
pubs.acs.org/OrgLett
Enantioselective Synthesis of an HCV NS5a Antagonist
Ian K. Mangion,* Cheng-yi Chen,* Hongmei Li, Peter Maligres, Yonggang Chen, Melodie Christensen,
Ryan Cohen, Ingyu Jeon, Artis Klapars, Shane Krska, Hoa Nguyen, Robert A. Reamer, Benjamin D. Sherry,
and Ilia Zavialov
Department of Process Chemistry, Merck and Co., Inc., P.O. Box 2000, Rahway, New Jersey 07065, United States
*
S Supporting Information
ABSTRACT: A concise, enantioselective synthesis of the
HCV NS5a inhibitor MK-8742 (1) is reported. The features of
the synthesis include a highly enantioselective transfer
hydrogenation of an NH imine and a dynamic diastereose-
lective transformation. The synthesis of this complex target
requires simple starting materials and nine linear steps for
completion.
hronic hepatitis C virus (HCV) infection is estimated to
afflict over 170 million people worldwide. Untreated
1
C
chronic HCV patients are at high risk of succumbing to
2
cirrhosis, liver cancer, or liver failure. Standard of care for HCV
patients has generally been a combination of pegylated
interferon-α and ribavirin; unfortunately, this combination has
3
been shown to have limited efficacy and tolerability. Moreover,
achieving sustained virologic response is dependent upon the
genotype of the virus and the duration of treatment. There has
been considerable interest in the development of direct-acting
4
antiviral agents to treat HCV in order to provide an interferon-
free regimen with efficacy for a broader population of patients
while improving tolerability.
One of the most promising new targets for development of
direct-acting HCV therapies is the nonstructural viral protein 5a
5
(
NS5a). A recent clinical study provided validation of NS5a as
a therapeutic target, during which a single dose of a NS5a
5
a
inhibitor resulted in significant viral load reduction. Although
the NS5a protein has no known enzymatic function, it has been
shown to serve a number of functions in the life cycle of the
5c
Figure 1. Retrosynthesis and stereochemical relay strategy.
virus, including viral replication. MK-8742 (1) is a potent and
6
selective NS5a inhibitor recently disclosed by Merck & Co.
8
aryl hydrogen of the indoline. As a result, the resident
Herein is described a concise, enantioselective synthesis of MK-
stereochemistry of the indoline would then direct the attack of
the phenol selectively to the Si face of the iminium, ensuring
stereocontrol. We further envisioned that indoline 4 could be
derived from an acyclic amine via an intramolecular C−N aryl
bond coupling.
To begin investigations into the stereochemical relay
strategy, we first sought an efficient synthesis of indoline 4.
In the first step, 2,5-dibromophenylacetic acid was converted to
its corresponding acid chloride and coupled with 3-
bromophenol to provide crystalline ester 7 in 91% yield
8
742.
The critical and most challenging structural feature of MK-
742 from a synthetic perspective is the remote hemiaminal
8
stereocenter. We were attracted by the intriguing possibility of
using stereochemical relay to introduce the hemiaminal
7
stereoselectively (Figure 1). In this approach, introduction of
the symmetrical heterocyclic fragments (2) at a late stage via
cross-coupling reactions places pentacyclic indole 3 as a
retrosynthetic target. Indole 3 could then be derived from the
corresponding indoline (4) via a net dehydrogenation.
We anticipated that chiral indoline 5 could condense
diastereoselectively with benzaldehyde, installing the hemi-
aminal stereocenter. We reasoned that iminium 6 would
preferentially adopt an (E)-geometry to minimize van der
Waals interactions between the phenyl ring and the proximal
(
Scheme 1). Ester 7 then underwent a Fries rearrangement
mediated by neat methanesulfonic acid desiccated by Ms O to
2
Received: January 22, 2014
Published: April 11, 2014
©
2014 American Chemical Society
2310
dx.doi.org/10.1021/ol500971c | Org. Lett. 2014, 16, 2310−2313

Organic Letters
Letter
Scheme 1. Synthesis of o-Hydroxyaryl Imine 9
Table 1. Asymmetric Reduction Optimization
a
b
entry
metal
ligand
solvent
yield (%)
ee (%)
c
1
Ir
11a
11b
11c
11d
11e
11f
11f
11f
11f
11f
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
i-PrOH
THF
13
22
30
19
15
81
83
95
96
87
80
80
56
78
78
85
94
98
95
c
2
Ir
d
3
Rh
Ru
Ru
Ru
Ru
Ru
Ru
Ru
e
4
5
6
7
8
9
furnish ketone 8 as a crystalline intermediate in 82% yield.
Condensing ketone 8 with dry ammonia in methanol provided
crystalline o-hydroxyaryl imine 9 (94% yield), a potential
substrate for asymmetric reduction.
9
CH Cl2
2
f
1
0
CH Cl2
82
2
a
b
Assay yield of 12 as determined by HPLC. Enantiomeric excess
A number of reports have described asymmetric hydro-
c
determined by HPLC. 4 mol % of [Ir(COD)Cl] , 500 psi H , 25 °C.
10
2
2
genations of imines, and a recent study from Merck detailed
the success of iridium catalysis for the enantioselective
d
e
4
mol % of [Rh(COD)Cl] , 500 psi H , 50 °C. Commercially
2 2
available DPEN-RuCl catalysts used at 0.3 mol % loading, 2.0 equiv of
NH CO H, 70 °C. 1:1 Et N:HCO H used as the hydrogen source.
10c
f
hydrogenation of unprotected N−H imines. However, we
anticipated that the o-phenol functional group of imine 9 might
stabilize the imine by hydrogen bonding, reducing its affinity
for transition metals and therefore attenuating its reactivity.
Nevertheless, we began surveying the reduction of imine 9 with
catalytic [Ir(COD)Cl]2 in conjunction with a range of
4
2
3
2
11
bisphosphine ligands (Table 1). To our delight, we found
that amine 10 could be accessed in enantiomeric excesses as
great as 87% with select ligands (Table 1, entries 1−2);
however, yields were modest due to competing hydrolysis of
the imine back to the ketone, and subsequent reduction of the
1
2
13
ketone to the corresponding alcohol. Rhodium catalysts and
14
traditional Noyori-type Ru transfer hydrogenation catalysts
were also evaluated, with promising enantioselectivities, but,
unfortunately, the same limitations in yield (entries 3−5).
However, it was found that structurally related “tethered”
mediated the C−N bond coupling selectively to provide
15
17
TsDPEN catalysts pioneered by Wills provided not only
superior enantioselectivity, but also significantly improved the
yield of amine 10 (entries 6−10). Although the magnitude of
improvement provided by the Wills catalyst was surprising, it is
consistent with studies suggesting a role for the tethered ligand
in stabilizing the catalyst and providing faster Ru hydride
regeneration, and perhaps providing a stabilizing interaction in
indoline 5 in 91% yield. There was no evidence of competing
intramolecular C−N bond coupling products, though indoline
5 can eventually proceed to react intermolecularly to form
dimeric species if the reaction is not quenched. A number of
ligands for Cu were evaluated but found to have no impact on
reactivity or yield, perhaps indicating that amine 10 is itself
precoordinating the Cu catalyst.
17
c
the transition state of the reduction that aids selectivity.
We then surveyed reaction conditions for the proposed
stereochemical relay by condensing benzaldehyde with indoline
5. Gratifyingly, it was found that the pentacyclic hemiaminal 4
could be formed in the presence of a mild acid catalyst, with
diastereoselectivity on the order of 7:1. Unfortunately, the
diastereomeric ratio and yield degraded on prolonged reaction
times due to reversibility of the hemiaminal formation.
However, by carefully optimizing solvent and temperature to
favor crystallization during the course of the reaction, we were
Several solvents, including MeOH and iPrOH, provided
good yields but only moderate enantioselecitvity (entries 6 and
7
9
, 78−85% ee). Subsequently, it was found that THF (entry 8,
4% ee) provided improved reactivity and selectivity. Our best
results were achieved with 0.3 mol % (R,R) Teth-TsDPEN-
RuCl in CH Cl using ammonium formate as the hydrogen
2
2
source (entry 9, 98% ee). This catalyst system ultimately
afforded chiral amine 10 in a crude 96% yield and 98% ee,
which upon crystallization increased to 99% ee (88% isolated
18
able to promote a dynamic diastereoselective transformation.
yield). A mixture of Et N and formic acid can be used as an
That is, by taking advantage of the reversibility of the
hemiaminal formation in solution and the greater stability of
the crystalline phase of 4 relative to its diastereomer, we could
drive the isolated dr of 4 from 7:1 to >99:1 in 93% yield using
acetonitrile as the solvent and TFA as the acid catalyst.
We began to look extensively at conditions for effecting a net
dehydrogenation of hemiaminal 4 to access indole 3. Various
low-valent transition metals known to mediate dehydrogen-
ation of unsaturated systems were evaluated, but provided only
3
alternative hydrogen source, but we found that the yield was
diminished with hydrolysis of the imine becoming more
16
prevalent (entry 10).
With a reliable synthesis of amine 10 in hand, we then sought
to complete the indoline ring system by way of an
intramolecular aryl C−N bond coupling (Scheme 2). Although
19
1
0 features three aryl bromides that could participate in a
transition metal insertion, we found that catalytic CuI smoothly
2
311
dx.doi.org/10.1021/ol500971c | Org. Lett. 2014, 16, 2310−2313

Organic Letters
Letter
Scheme 2. Completion of the Synthesis of MK-8742 (1)
2
0
modest reactivity. A number of conventional oxidants did
prove capable of promoting this transformation; however, we
observed significant levels of racemization of the hemiaminal
stereocenter depending on the nature of the oxidant and the
solvent system. Following exhaustive screening efforts, we
ASSOCIATED CONTENT
Supporting Information
■
*
S
Experimental procedures, compound characterization, and
1
13
copies of H and C NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
found KMnO effects the desired dehydrogenation without
racemization, providing indole 3 in 83% yield and >99% ee.
Indole 3 was then subjected to a palladium mediated bis-
borylation and a subsequent bis-Suzuki coupling with imidazole
4
21
AUTHOR INFORMATION
■
*
*
Corresponding Authors
E-mail: ian_mangion@merck.com.
E-mail: cheng88chen@gmail.com.
2
2
12
in a through process, after which the product was
crystallized as its 4-nitrobenzoic acid salt. Although there are
four chemical transformations in this borylation and Suzuki
through process, the overall isolated yield of the Suzuki product
Notes
The authors declare no competing financial interest.
1
3 is 82%, representing an average yield of 95% per
ACKNOWLEDGMENTS
■
transformation. To our surprise, the rate of the Suzuki reaction
in the through process is greater than that seen when the steps
are decoupled with an intermediate purification, suggesting that
one or more components of the borylation reaction stream
We thank Dr. Rebecca Ruck (Merck) for helpful discussions.
REFERENCES
■
23
(1) (a) Kim, W. R. Microbes Infect. 2002, 4, 1219. (b) Shepard, C. W.;
Finelli, L.; Alter, M. J. Lancet Infect. Dis. 2005, 5, 558.
accelerates the subsequent Suzuki coupling.
Suzuki product 13 was then deprotected with aqueous HCl
in MeOH, providing diamine 14 as a crystalline tetra-HCl salt
in 92% yield. EDC-mediated coupling of diamine 13 with 2
equiv of L-Moc-valine (15) then provided crude MK-8742 (1),
which was then crystallized from ethanol in 93% overall yield.
In conclusion, a concise, enantioselective synthesis of MK-
(
2) Alter, M. J.; Kruszon-Moran, D.; Nainan, O. V.; McQuillan, G.
M.; Gao, F.; Moyer, L. A.; Kaslow, R. A.; Margolis, H. S. N. Engl. J.
Med. 1999, 341, 556.
(
3) (a) Hoofnagle, J. H.; Seeff, L. B. N. Engl. J. Med. 2006, 355, 2444.
b) Pearlman, B. L. World J. Gastroenterol. 2008, 14, 3621.
4) (a) Asselah, T.; Benhamou, Y.; Marcellin, P. Liver Int. 2009, 29,
7. (b) Pawlotsky, J. M. Gastroenterology 2011, 140, 746. (c) Sheridan,
(
(
5
8
742, a potent and selective NS5a inhibitor for the treatment of
C. Nat. Biotechnol. 2011, 29, 553. (d) Vermehren, J.; Sarrazin, C. Eur. J.
Med. Res. 2011, 16, 303.
chronic HCV infection, has been developed. This approach
features a highly enantioselective asymmetric hydrogenation
and a directed stereochemical relay strategy that leverages a
dynamic diastereoselective condensation to produce the
challenging hemiaminal stereocenter. The entire synthesis
requires only 10 linear steps for completion in the longest
linear sequence and proceeds in 30% overall yield without the
need for chromatography.
(5) (a) 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.; Lemm, J. A.; Wang, C.; Knipe, J. O.; Chien, C.;
Colonno, R. J.; Grasela, D. M.; Meanwell, N. A.; Hamann, L. G. Nature
2
010, 465, 96. (b) Lee, C. Arch. Pharm. Res. 2011, 34, 1403. (c) Gao,
M. Curr. Opin. Virol. 2013, 3, 514. (d) Pol, S. Clin. Invest. 2013, 3, 191.
e) Guedj, J.; Dahari, H.; Rong, L.; Sansone, N. D.; Nettles, R. E.;
(
Cotler, S. J.; Layden, T. J.; Uprichard, S. L.; Perelson, A. S. Proc. Nat.
Acad. Sci. U.S.A. 2013, 110, 3991.
2
312
dx.doi.org/10.1021/ol500971c | Org. Lett. 2014, 16, 2310−2313

Organic Letters
Letter
(
6) Coburn, C. A.; Meinke, P. T.; Chang, W.; Fandozzi, C. M.;
Graham, D. J.; Hu, B.; Huang, Q.; Kargman, S.; Kozlowski, J.; Liu, R.;
McCauley, J. A.; Nomeir, A. A.; Soll, R. M.; Vacca, J. P.; Wang, D.; Wu,
H.; Zhong, B.; Olsen, D. B.; Ludmerer, S. W. ChemMedChem. 2013, 8,
1
930.
7) (a) Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93,
307. (b) Mengel, A.; Reiser, O. Chem. Rev. 1999, 99, 1191−1223.
c) Kozmin, S. A. Org. Lett. 2001, 3, 755. (d) Clayden, J.; Vassiliou, N.
Org. Biomol. Chem. 2006, 4, 2667.
8) Kunz, R. K.; MacMillan, D. W. C. J. Am. Chem. Soc. 2005, 127,
240.
(
1
(
(
3
(
9) Jeon, I.; Mangion, I. K. Synlett 2012, 23, 1927.
(
10) (a) Carpentier, J. F.; Bette, V. Curr. Org. Chem. 2002, 6, 913.
(
b) Tang, W.; Zhang, X. Chem. Rev. 2003, 42, 5472. (c) Hou, G.;
Gosselin, F.; Li, W.; McWilliams, J. C.; Sun, Y.; Weisel, M.; O’Shea, P.
D.; Chen, C.; Davies, I. W.; Zhang, X. J. Am. Chem. Soc. 2009, 131,
9
882.
11) For reports of enantioselective organocatalytic reductions of o-
hydroxyaryl imines, see: (a) Nguyen, T. B.; Bousserouel, H.; Wang,
(
Q.; Guer
́
itte, F. Org. Lett. 2010, 12, 4705. (b) Nguyen, T. B.; Wang,
itte, F. Chem.Eur. J. 2011, 17, 9576.
12) Various dessicants were evaluated to suppress ketone formation
Q.; Guer
(
́
without success. For a successful application of dessicants, see: Chi, Y.;
Zhou, Y.-G.; Zhang, X. J. Org. Chem. 2003, 68, 4120.
(
13) (a) Hsiao, Y.; Rivera, N. R.; Rosner, T.; Krska, S. W.; Njolito, E.;
Wang, F.; Sun, Y.; Armstrong, J. D.; Grabowski, E. J. J.; Tillyer, R. D.;
Spindler, F.; Malan, C. J. Am. Chem. Soc. 2004, 126, 9918. (b) Hansen,
K. B.; Hsiao, Y.; Xu, F.; Rivera, N.; Clausen, A.; Kubryk, M.; Krska, S.;
Rosner, T.; Simmons, B.; Balsells, J.; Ikemoto, N.; Sun, Y.; Spindler, F.;
Malan, C.; Grabowski, E. J. J.; Armstrong, J. D. J. Am. Chem. Soc. 2009,
131, 8798.
(
14) (a) Uematsu, N.; Fujii, A.; Hashiguchi, S.; Ikariya, T.; Noyori, R.
J. Am. Chem. Soc. 1996, 118, 4916. (b) Kadyrov, R.; Riermeier, T. H.
Angew.Chem., Int. Ed. 2003, 42, 5472. (c) Boggs, S. D.; Cobb, J. D.;
Gudmundsson, K. S.; Jones, L. A.; Matsuoka, R. T.; Millar, A.;
Patterson, D. E.; Samano, V.; Trone, M. D.; Xie, S.; Zhou, X. Org.
Process Res. Dev. 2007, 11, 539.
(
15) (a) Hayes, A. M.; Morris, D. J.; Clarkson, G. J.; Wills, M. J. Am.
Chem. Soc. 2005, 127, 7318. (b) Matharu, D. S.; Morris, D. J.;
Kawamoto, A. M.; Clarkson, G. J.; Wills, M. Org. Lett. 2005, 7, 5489.
(
c) Cheung, F. K.; Lin, C.; Minissi, F.; Lorente Criville,
M. A.; Fox, D. J.; Wills, M. Org. Lett. 2007, 9, 4659.
16) Other reductants such as PMHS or Et SiH were evaluated but
́
A.; Graham,
(
3
found to provide limited reactivity.
(
17) For a related example with palladium catalysis, see: (a) Katkevics,
M.; Kukosha, T.; Trufilkina, N. Synlett 2011, 17, 2525. (b) Anderson,
J. C.; Noble, A.; Tocher, D. A. J. Org. Chem. 2012, 77, 6703.
(18) Caddick, S.; Jenkins, K. Chem. Soc. Rev. 1996, 25, 447.
(19) Dobereiner, G. E.; Crabtree, R. H. Chem. Rev. 2010, 110, 681.
(20) Crabtree, R. H. J. Chem. Soc., Dalton Trans. 2001, 2437.
(21) DDQ was the best alternative oxidant we found, providing
indole 3 in an optimized 91% ee. Catalytic Mn oxidation systems such
as Mn(OAc) were investigated in conjunction with various oxidants
2
but were found to suffer from significant racemization.
(
22) See the Supporting Information for details on the synthesis of
imidazole 12.
23) A number of additives were evaluated in the Suzuki coupling to
(
mimic the effect of the crude borylation stream, including methanol,
pinacol, and inorganic salts; however, no single additive evaluated to
date replicates this phenomenon.
2
313
dx.doi.org/10.1021/ol500971c | Org. Lett. 2014, 16, 2310−2313