Development of a practical, asymmetric synthesis of the hepatitis C virus protease inhibitor MK-5172

Article abstract of DOI:10.1021/ol401864t

The development of a practical, asymmetric synthesis of the hepatitis C virus (HCV) protease inhibitor MK-5172 (1), an 18-membered macrocycle, is described.

Full text of DOI:10.1021/ol401864t

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Development of a Practical, Asymmetric
Synthesis of the Hepatitis C Virus
Protease Inhibitor MK-5172
Jeffrey Kuethe,* Yong-Li Zhong,* Nobuyoshi Yasuda,* Gregory Beutner,
Katherine Linn, Mary Kim, Benjamin Marcune, Spencer Douglas Dreher,
Guy Humphrey, and Tao Pei
Process Chemistry, Merck Research Laboratories, P.O. Box 2000, Rahway,
New Jersey 07065-0200, United States
Jeffrey_kuethe@merck.com; yongli_zhong@merck.com; nobuyoshi_yasuda@merck.com
Received July 3, 2013
ABSTRACT
The development of a practical, asymmetric synthesis of the hepatitis C virus (HCV) protease inhibitor MK-5172 (1), an 18-membered macrocycle,
is described.
Affecting an estimated 130ꢀ170 million people world-
wide, the hepatitis C virus (HCV) is the leading cause of
chronic liver infection and transplants.1 Therapy for
chronic HCV infection consists of a regimen of pegylated
interferon-R and ribavirin (P/R).2 These agents provide
nonspecific immunostimulatory and antiviral effects. A
virologic cure is achieved in only 40 to 75% of patients,
depending on the virus genotype, patient characteristics,
and the severity of liver damage at the time of treatment.
Therapy may be accompanied by significant side effects.
Recently, compounds that directly target key HCV
proteins such as HCV NS3/4A protease have been
discovered. The first generation of such agents, including
Merck’s boceprevir,3,4 have substantially improved the
probability of a cure among patients with HCV G1 infec-
tion when added to P/R. Despite this advance, patients
continue to experience treatment related failure, often due
to emergence of a resistant virus. Efficacy in genotypes
other than G1 has not been defined. Therapy requires
three-time daily administration, with food, and the total
duration of therapy typically ranges from 24 to 48 weeks.
Seeking to further optimize the pharmacokinetic profile
and offer broader activity across genotypes and resistant
HCV variants, medicinal chemists at Merck have discov-
ered MK-5172 (1) which has the potential to be the
cornerstone of an all-oral treatment for HCV (Figure 1).5
In this Letter we describe the development of a practical
and highly efficient asymmetric synthesis of MK-5172 (1).
(1) (a) WHO. Wkly. Epidemiol. Rec. 1999, 74, 425ꢀ427. (b) Liang,
T. J.; Heller, T. Gastroenterology 2004, 127, S62–S71. (c) Lavanchy, D.
Liver Int. 2009, 29 (suppl 1), 74–81.
(2) Hadziyannis, S. J.; Sette, H.; Morgan, T. R. Ann. Intern. Med.
2004, 140, 346–355.
(3) Poordad, F.; McCone, J., Jr.; Bacon, B. R.; Bruno, S.; Manns,
M. P.; Sulkowski, M. S.; Jacobson, I. M.; Reddy, K. R.; Goodman,
Z. D.; Boparai, N.; DiNubile, M. J.; Sniukiene, V.; Brass, C. A.;
Albrecht, J. K.; Bronowicki, J.-P. New Engl. J. Med. 2011, 364, 1195.
(4) Bacon, B. R.; Gordon, S. C.; Lawitz, E.; Marcellin, P.; Vierling,
J. M.; Zeuzem, S.; Poordad, F.; Goodman, Z. D.; Sings, H. L.; Boparai,
N.; Burroughs, M.; Brass, C. A.; Albrecht, J. K.; Esteban, R. New Engl.
J. Med. 2011, 364, 1207.
(5) Harper, S.; McCauley, J. A.; Rudd, M. T.; Ferrara, M.; DiFilippo,
M.; Crescenzi, B.; Koch, U.; Petrocchi, A.; Holloway, M. K.; Butcher,
J. W.; Romano, J. J.; Bush, K. J.; Gilbert, K. F.; McIntyre, C. J.;
Nguyen, K. T.; Nizi, E.; Carrol, S. S.; Ludmerer, S. W.; Burlein, C.;
DiMuzio, J. M.; Graham, D. J.; Mchale, C. M.; Stahlhut, M. W.; Olsen,
D. B.; Monteagudo, E.; Cianetti, S.; Giuliano, C.; Pucci, V.; Trainor, N.;
Fandozzi, C. M.; Rowley, M.; Coleman, P. J.; Vacca, J. P.; Summa, V.;
Liverton, N. J. Med. Chem. Lett. 2012, 3, 332–336.
r
10.1021/ol401864t
XXXX American Chemical Society

Scheme 1. Retrosynthesis of 1
Figure 1. Major liabilities on the Med Chem Route.5
Our initial efforts began by examining the medicinal
chemistry approach which utilized an ring-closing metath-
esis (RCM) strategy to construct the challenging 18-mem-
bered ring of 1.5 It was recognized that this approach had a
number of liabilities. Major liabilities included instability
of TMS enol ether 7, poor regioselectivity during the
cyclopropanation of 7 (separation of double cyclopro-
panated byproducts from the desired intermediate 8 was
only accomplished after the RCM step), and the instability
of vinyl quinoxaline 9 that resulted in poor yields for the
RCM reaction. Therefore, we proposed an alternative
approach centered on construction of the macrocycle via
macrolactamization (Scheme 1). Retrosynthetically, we
envisioned appending the western cyclopropyl side chain
through a Sonogashira coupling. Deconstruction of the
rest of the molecule leads to a regioselective SNAr reaction
to install the hydroxyproline moiety and two amide bond-
forming reactions. The requisite building blocks would
include cyclopropanol 2,6 dichloroquinoxaline 3,7 com-
mercially available hydroxyproline 4, and amine 5.8
levels of 12 proved problematic during the isolation of 10
without recourse to chromatography. We then set out to
fully optimize the reaction in terms of solvent, tempera-
ture, and base and discovered that a combination of DBU
in either DMSO, NMP, or dimethylacetamide (DMAc)
routinely gave the highest level of regioselectivity between
10:11 (∼20:1) with minimal formation of 12 when con-
ducted at 50 °C (Table 1, entries 7ꢀ9). The final conditions
selected involved reaction of 3 with 1.05 equiv of 4 in the
presence of 1.5 equiv of DBU at 50 °C in DMAc giving 10
in 88% HPLC assay yield.11 It should also be pointed out
that no detectable epimerization of the proline ester was
detected in the HPLC analysis of any of the crude reaction
mixtures as confirmed by comparison to an authentic
sample. Compound 10 was isolated by crystallization from
Our investigations began by examining the SNAr reac-
tion between 3 and 4 (Scheme 2).9 For example, reaction of
3 and 4 in the presence of Cs2CO3 in DMF at room
temperature afforded primarily the desired product 10
(Table 1, entry 1). Also detected in the crude reaction
mixture was the corresponding regioisomer 11 and double
addition product 12.10 The high regioselectivity (17:1) favor-
ing the desired product was encouraging, but the high
Scheme 2. Preparation of Heterocyclic Core
(6) Bassan, E. M.; Baxter, C. A.; Beutner, G. L.; Emerson, K. M.;
Fleitz, F. J.; Johnson, S.; Keen, S.; Kim, M. M.; Kuethe, J. T.; Leonard,
W. R.; Mullens, P. R.; Muzzio, D. J.; Roberge, C.; Yasuda, N. Org.
Process Res. Dev. 2012, 16, 87–95.
(7) Sarges, R.; Howard, H. R.; Browne, R. G.; Lebel, L. A.; Seymour,
P. A.; Koe, B. K. J. Med. Chem. 1990, 33, 2240–2254.
(8) Li, J.; Smith, D.; Wong, H. S.; Cambell, J. A.; Meanwell, N. A.;
Scola, P. M. Synlett 2006, 725–728.
(9) There is precedence for regioselective displacement of 3 with
methoxide; see ref 8.
(10) Compounds 11 and 12 were observed spectroscopically (HPLC,
LCMS, and NMR) but were not isolated.
B
Org. Lett., Vol. XX, No. XX, XXXX

MTBE/heptanes in 71% yield and in analytically pure
form.
much cleaner reaction profile. After an extractive workup,
the desired carbamate 13 was isolated as a solution in
cyclopentyl methyl ether (CPME) in 70% HPLC assay
yield. There was no detectable amounts of dimer15formed
under these reaction conditions, and the product was
sufficiently pure for use in the next transformation without
further purification.
Table 1. Optimization of the Regioselective SNAr Displacement
4
t
conv
(%)
ratio12
entry
base
(equiv) solvent (°C)
(10:11:12)
The Sonogashira cross-coupling between chloride 10
and carbamate 13 was extensively investigated with the
aid of high throughput experimentation where catalyst/
ligand combinations, solvent, temperature, and base were
explored (Scheme 4). The optimal conditions that were
selected involved addition of 10 and crude 13 to a slurry of
Pd(OAc)2 (3 mol %), P(t-Bu)3BF4 (6 mol %), and K2CO3
(2.5 equiv) in a solvent mixture of CPME/MeCN (2.4:1)
and heating at 85 °C for 2.5 h. After an aqueous workup,
product 16 was obtained in 98% HPLC assay yield.
Compound 16 was sufficiently pure for use in the next step
without the need for purification. Catalytic hydrogenation
of 16 was conducted with Pd/C in IPAc/MeOH and gave
the reduced product 17, which was not isolated, in 89%
HPLC assay yield.
1
2
3
4
5
6
7
8
9
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
DABCO
DBU
1.25
1.25
1.25
1.05
1.05
1.25
1.05
1.05
1.05
DMSO
DMSO
DMAc
NMP
rt
100
100
<5
85:5:10
85:4:11
NA
35
rt
50
50
rt
89
91:7:2
85:8:7
NA
DMSO
DMSO
DMSO
NMP
90
0
50
50
50
91
92:6:2
93:5:2
94:4:2
DBU
100
100
DBU
DMAc
With a suitable method in hand for the preparation of
10, the construction of the macrocyclic ring of 1 was next
investigated. Our strategy centered on a Sonogashira/
macrolactamization approach and required the prepara-
tion of the acetylene precursor 13 (Scheme 3). The synthe-
sis of 13 was extensively examined. Initial experiments
involved reaction of 2 with CDI in DMF followed by
addition of L-tert-leucine 6 and heating to 90 °C for 12 h.
Under these conditions, 13 could be isolated routinely in
60ꢀ65% yield after an extractive workup. However, also
detected in the crudeproduct were carbamate14and dimer
15 in varying levels which were extremely difficult to
remove.13 Carbamate 14 arises from hydrolysis of DMF
and subsequent reaction with the intermediate acylimida-
zole with dimethylamine.
Scheme 4. Sonogashira/Macrolactamization
Scheme 3. L-tert-Butyl Leucine Carbamate
Perhaps the single most challenging aspect of the synthe-
sis involved the formation of the 18-membered ring sys-
tem of 1 via macrolactamization. In addition, control of
the impurity profile was critical considering the fact that
macrolactam 18 was the first isolated crystalline intermedi-
ate after 10 steps from commercially available 5-chloro-
propyne, the starting material for cyclopropanol 2. We
devised a very effective and highly productive procedure
for the deprotection, macrolactamization, and isolation of
18 (Scheme 4). Removal of the Boc-protecting group was
accomplished by treating crude 17 with PhSO3H in 8 mL/g
of MeCN at 50 °C for 3 h to give the corresponding
benzenesulfonate salt. All attempts to crystallize this
Attempted activation of 2 with phosgene or triphosgene
followed by reaction with 6 in a number of solvent systems
led to unsatisfactory results, presumably due to the insol-
ubility of 6 in all solvents other than water. It was
discovered that activation of 2 with CDI employing
€
Hunig’s base as the reaction solvent followed by the slow
addition of 6 and heating to 95 °C for 2.5 h resulted in a
(11) HPLC assay yield refers to quantitative HPLC analysis employ-
ing an analytical standard.
(12) Ratio was determined by HPLC at a wavelength detection of
210 nm.
(13) Compounds 14 and 15 were identified in the NMR of the crude
isolated product and were not isolated.
€
salt were unsuccessful. Direct addition of Hunig’s base
Org. Lett., Vol. XX, No. XX, XXXX
C

(4.6 equiv) was followed by slow addition of the reaction
mixture to a solution of HATU14 (1.25 equiv)15 in 7 mL/g
of MeCN (based on the amount of 17). Slow addition to
HATU was required in order to minimize dimerization
and maintain high productivity. The HPLC assay yield of
18 was 76% for the two-step protocol with dimer forma-
tion being controlled to <2%.16 Concentration of the
reaction mixture to 4 mL/g, based on the amount of 18,
resulted in crystallization of 18 which was isolated in
analytically pure form in 65% overall yield.17
Scheme 5. End Game
The preparation of MK-5172 is illustrated in Scheme 5.
Saponification of 18withLiOH in aqueous THF furnished
the corresponding acid 19 as a crystalline solid in 98%
isolated yield. The final coupling with side chain 20
required a degree of reaction optimization. For example,
the coupling was effective and gave high conversions when
EDC was employed by itself; however, significant amounts
of epimerization adjacent to the proline carboxylic acid
was observed leading to the formation of 21 (up to 10%).
The removal of 21 to acceptable levels during the crystal-
lization of 1 proved to be challenging, and the use of
additional additives was explored in order to suppress
the formation of 21. After examining a number of solvents
(DCM, EtOAc, MeCN, DMF, DMAc, DMSO, and NMP)
and additives (HOBT, 2-hydroxypyridine N-oxide, N-hydr-
oxysuccinamide, and CuCl2) and the effect of base load-
ing,18 we found better conditions to minimize the forma-
tion of 21. Thus, EDC coupling in MeCN with an excess of
pyridine gave full conversion of acid 19 to the desired
product 1 with <0.2% of isomer 21 observed. After an
acidic quench, compound 1 was obtained in 93% yield and
with >99.7% purity.
In conclusion, we have outlined a highly efficient asymmetric
synthesis of 1 that centered on a Sonogashira/macrolactamiza-
tion approach for construction of a challenging 18-mem-
bered macrocycle. Key transformations included a regio-
selective SNAr reaction for the formation of core heterocycle
10 and a volume efficient macrolactamization protocol
leading to 16. The synthesis proceeded in 42% overall
yield from 3, required no chromatographic purifications,
and was amenable to the kilogram scale production of 1.
(14) O-(7-Azabenzotriazol-1-yl)-N,N,N0,N0-tetramethyluronium
hexafluorophosphase; CAS [148893-10-1].
Acknowledgment. We appreciated support from our
NMR, HRMS specialty groups. We also thank Drs. Jeremy
Scott and Carl Baxter of Merck & Co., Inc. for helpful
discussions during the preparation of this manuscript.
(15) Other common amide coupling reagents such as EDC, EDC/
HOBt, EDC/HOAt, EDC/HOPO, EDC/pyridine, pivolate mixed anhy-
dride, etc. resulted in lower conversion and generated a large amount of
the dimerization byproduct.
(16) Dimerization of des-Boc of 17 was observed under various other
conditions, but was effectively controlled under the described reaction
conditions.
(17) There was no detectable amounts of epimerization at any of the
stereogenic centers during macrolactamization as confirmed by HPLC
analysis versus previously prepared reference standards.
(18) Cheng, C.-Y.; Frey, L. F.; Shultz, S.; Wallace, D. J.; Marcantonio,
K.; Payack, J. F.; Vazquez, E.; Springfield, S. A.; Zhou, G.; Liu, P.;
Kieczykowski, G. R.; Chen, A. M.; Phenix, B. D.; Singh, U.; Strine, J.;
Izzo, B.; Krska, S. W. Org. Process Res. Dev. 2007, 616–623.
Supporting Information Available. Experimental de-
tails and characterization data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX