The Discovery of Quinoxaline-Based Metathesis Catalysts from Synthesis of Grazoprevir (MK-5172)

Article abstract of DOI:10.1021/acs.orglett.6b00070

Olefin metathesis (OM) is a reliable and practical synthetic methodology for challenging carbon-carbon bond formations. While existing catalysts can effect many of these transformations, the synthesis and development of new catalysts is essential to increase the application breadth of OM and to achieve improved catalyst activity. The unexpected initial discovery of a novel olefin metathesis catalyst derived from synthetic efforts toward the HCV therapeutic agent grazoprevir (MK-5172) is described. This initial finding has evolved into a class of tunable, shelf-stable ruthenium OM catalysts that are easily prepared and exhibit unique catalytic activity.

Full text of DOI:10.1021/acs.orglett.6b00070

Letter
pubs.acs.org/OrgLett
The Discovery of Quinoxaline-Based Metathesis Catalysts from
Synthesis of Grazoprevir (MK-5172)
Michael J. Williams,* Jongrock Kong,* Cheol K. Chung, Andrew Brunskill, Louis-Charles Campeau,
and Mark McLaughlin
Department of Process and Analytical Chemistry, Merck Research Laboratories, Rahway, New Jersey 07065, United States
*
S Supporting Information
ABSTRACT: Olefin metathesis (OM) is a reliable and practical synthetic methodology for challenging carbon−carbon bond
formations. While existing catalysts can effect many of these transformations, the synthesis and development of new catalysts is
essential to increase the application breadth of OM and to achieve improved catalyst activity. The unexpected initial discovery of
a novel olefin metathesis catalyst derived from synthetic efforts toward the HCV therapeutic agent grazoprevir (MK-5172) is
described. This initial finding has evolved into a class of tunable, shelf-stable ruthenium OM catalysts that are easily prepared and
exhibit unique catalytic activity.
lefin metathesis (OM) has emerged as a powerful carbon−
Scheme 1. RCM Approach toward Grazoprevir (MK-5172)
O
carbon bond forming methodology with applications
1
encompassing natural product total syntheses and industrial
2
3
4
processes (e.g., chemical, polymer, and pharmaceutical ). In
order to increase the application breadth of OM, catalyst
properties are continually optimized to their peak performances
(
e.g., yield, selectivity, turnover-number, functional group
compatibility), often by the introduction of new ligand
5
frameworks. We envisioned olefin metathesis as the key bond-
forming reaction to establish the macrocyclic backbone of the6
HCV NS3/4a protease inhibitor grazoprevir (MK-5172).
During these efforts, we identified the intermediacy of a novel,
isolable ruthenium complex. Herein, we present the evolution of
this finding into a new class of metathesis catalysts and
demonstrate its unique properties.
Discovery of the ligand framework emerged from a collection
of unexpected experimental findings. Initial studies on the ring
closing metathesis (RCM) of substrates 1a and 1b revealed that
the yield of macrocyclization depended greatly on the relative
length of both olefin side chains (Scheme 1). Notably, exposure
of allyl quinoxaline substrate 1a to RCM conditions failed to
afford any desired product while use of homoallyl 1b led to a high
catalyst, instead persisting as complex 5, a stable six-membered
9
chelate.
Competition experiments using 1H NMR spectroscopy
supported this hypothesis, revealing that the stoichiometric
treatment of substrate 3 with a Hoveyda−Grubbs second
generation catalyst (4) did indeed result in the appearance of a
new alkylidene triplet at 18.6 ppm vs the benzylidene singlet at
16.5 ppm in the starting material. After aging the sample
overnight, the triplet/singlet ratio was approximately 5:1
showing favorable equilibrium toward the alkylidene formation
7
yield of macrocycle 2. Similarly, attempts to perform the
intermolecular homodimerization of intermediate 3 only
generated trace product, but the system readily underwent
cross metathesis (CM) with the addition of an exogenous donor
8
olefin. These observations led us to hypothesize that homoallyl
substrate 3 undergoes initial alkylidene formation, but does not
Received: January 8, 2016
complete the catalytic cycle to form a dimer and regenerate the
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b00070
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
over the benzylidene. Complex 5 was isolated from the reaction
mixture in 80% yield after silica gel chromatography (Scheme 2).
N-heterocyclic carbene (NHC) if the ligand were to adopt the
cis-configuration (Figure 1) and (2) electronic factors.
12e
Scheme 2. Initial Discovery and Isolation of Quinoxaline
Based Complexes
With complex 5 as the basis of our catalyst design, we sought to
pursue an accessible, simplified version with greater modularity.
We prepared a set of complexes that probed the structure−
activity relationship around the chelating quinoxaline ligand by
replacing the proline moiety with alternative substituents (Table
1). Compounds were easily synthesized featuring electron-
Figure 1. Crystal structure of complex 9a with thermal ellipsoids drawn
at the 50% probability level. Selected bond distances (Å) and angles
Table 1. Synthesis of Quinoxaline Complexes (9a−9g)
(deg): Ru1−Cl1 = 2.381, Ru1−Cl2 = 2.378, Ru1−N1 = 2.158, Ru1−
C1C = 2.039, Ru1−C1L = 1.814; Cl1−Ru1−Cl2 = 166.53°, C1L−
Ru1−N1 = 89.42°, C1C−Ru1−N1 = 172.77°.
We examined the metathesis activity of our new catalysts using
the heterocyclic precursor, tert-butyl diallylcarbamate (10, eq 1),
as an RCM substrate (Figure 2). Initial observations revealed
a
Isolated yields.
Figure 2. RCM of tert-butyl diallylcarbamate 10 with complexes 9a, 9c,
9
f, and 9g. Conditions: 10 (0.1 M), catalyst (1 mol %; 0.001 M), in
1
CD Cl at 30 °C. Conversion was measured by H NMR spectroscopy.
2
2
donating and -withdrawing substituents (9a−9f), and two
alkylidene chain lengths (9a and 9g), highlighting the quinoxa-
line ligand’s modularity. Under optimized conditions, all the
complexes described could be isolated via crystallization directly
competent metathesis activity with faster kinetic profiles using
10
from the relevant reaction mixtures in good to excellent yields.
catalysts featuring an electron-withdrawing substituent (cf.
13
These complexes are strikingly stable, as demonstrated by a series
complex 9f vs complexes 9a and 9c). Much like the analogous
observation with macrocycle 1a, little to no metathesis activity
was observed when the alkylidene chain length was shortened
from homoallyl to allyl: complex 9g promoted <1% conversion
of compound 10 to the desired RCM product 11 when treated
under the same conditions.
11
of solution phase and solid-state experiments. In particular,
complex 9a was observed to not undergo trans → cis
12
rearrangement during handling. We suggest this stability is
attributable to (1) the formation of a strong steric repulsion
between the quinoxaline moiety and the mesitylene rings of the
B
DOI: 10.1021/acs.orglett.6b00070
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Given that electron-deficient catalysts were more active in the
RCM reaction, we envisioned that the reaction kinetics could be
accelerated by protonation of the quinoxaline ligand with an acid
co-catalyst. Protonation should create a more electron-deficient
Table 2. Thermal Plot from Lewis and Brønsted Acid Co-
catalysts HTE Screening
14
ligand and concomitantly decrease the Ru−N bond strength.
We were pleased to find that the addition of catalytic Brønsted
acid (5 mol % benzenesulfonic acid, BSA) to a mixture of
complex 9a (1 mol %) and diethyl diallylmalonate (12) resulted
in a significant enhancement of the reaction rate (Figure 3).
a
b
c
HCl in dioxane. ZnCl in THF (0.5 M). AlCl in THF (0.5 M).
2
3
Conditions: 14 (0.1 M), 100 μL solvent, and 0.5 mol % of complex
with 10 mol % of additive. Chart numbers represent solution yield (%)
as measured by UPLC-MS analysis, and colors indicate yield (highest
yield = green; lowest yield = red). See the Supporting Information for
complete details. HOAc = acetic acid, BSA = benzenesulfonic acid,
TFA = trifluoroacetic acid, TFMSA = trifluoromethanesulfonic acid,
CSA = camphorsulfonic acid, TCA = trichloroacetic acid; DCE = 1,2-
dichloroethane.
Table 3. Metathesis Reactions Using Complex 9a
Figure 3. RCM of diene 12 with catalysts 9a and 9f in the presence and
absence benzenesulfonic acid (5 mol %); Conditions: CD Cl (0.1 M),
2
2
1
catalyst loading (0.005 M), at 30 °C. Conversion was measured by H
NMR spectroscopy. BSA = benzenesulfonic acid.
When BSA was added to a mixture of previously inactive complex
9
g and malonate 12, metathesis activity was then observed
(
∼10% conversion over 0.5 h). Despite achieving only a modest
level of conversion under these conditions, the ability to initiate
metathesis in the presence of BSA demonstrated the potential
latent utility of these complexes.
Encouraged by the positive result of using BSA as an activator,
we pursued a high-throughput experimentation (HTE) approach
to define an optimal co-catalyst. An array of Brønsted and Lewis
acids, along with three Ru complexes, were screened in the RCM
of diene 14 to form dihydropyrrole 15. This screen revealed that
AlCl , BSA, and HBF were found to have the greatest benefit to
a
3
4
Conditions: 14 or 16 (0.1 M) in DCM or tol, 1 mol % 9a, AlCl ·THF
3
b
the reaction rate relative to control reactions (Table 2).
The reactivity of catalyst 9a was fully evaluated in the presence
complex (0.5 M in THF, 20 mol %), 30 °C. Conditions: 14 or 16
c
(0.1 M) in DCM or tol, 1 mol % 9a, 30 °C. Conditions: 18 (0.1 M)
and absence of catalytic AlCl for metathesis reactions (Table
in CD Cl , 0.1 mol % 9a, 0 or 2 mol % AlCl ·THF complex (0.5 M in
3
2
2
3
1
5
1
d
3). The RCM reactions with precursors 14 and 16 proceeded
THF), 30 °C; conversion based upon H NMR analysis. Conditions:
0 (0.1 M) in tol-d , 1 mol % 9a, 0 or 20 mol % AlCl ·THF complex
2
in excellent yields in the absence of co-catalyst (91.9−98.7%,
8
3
1
(
0.5 M in THF), 23 °C; conversion based upon H NMR analysis.
entries 3−4 and 7−8) in both DCM and toluene. However, both
e
Isolated yield. See the Supporting Information for details. Ts = para-
reactions, when treated with catalytic AlCl , proceeded to
3
toluenesulfonyl, DCM = dichloromethane, tol = toluene.
completion in 0.5 h versus 6 h without (94.5−99.8% yield, entries
1
−2 and 5−6). The ring opening metathesis polymerization
16
(ROMP) of cyclooctadiene (18) was shown to proceed in
capable of inducing efficient latent catalysis. Consistent with
the results from Table 2, we observed only marginal effects using
acidic activation for the CM of 1-hexene (20) to 5-decene (21)
but were delighted to find that thermal activation indeed led to
the desired latency effect (Table 4). Less than 5% conversion was
observed after 24 h at ambient temperature, and only 8%
conversion at 50 °C, but 90% conversion to the desired product
was observed upon increasing the temperature to 95 °C. This
excellent conversion in the presence and absence of catalytic
AlCl in CD Cl (95.6−99.1% conversion, entries 5 and 6). The
3
2
2
CM of 1-hexene (20) also proceeded to high conversion (97−
9
9%, entries 11 and 12) in the presence and absence of catalytic
AlCl3.
Having demonstrated the broad applicability of complex 9a,
we then determined whether the less reactive complex 9g was
C
DOI: 10.1021/acs.orglett.6b00070
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Applications of Olefin Metathesis Polymerizations. In Olefin Metathesis
Theory and Practice; Grela, K., Ed.; John Wiley & Sons: Hoboken, NJ,
United States, 2014.
Table 4. CM of 1-Hexene (20) in the Presence of Complex 9g
(4) (a) Kong, J.; Chen, C.-y.; Balsells-Padros, J.; Cao, Y.; Dunn, R. F.;
Dolman, S. J.; Janey, J.; Li, H.; Zacuto, M. J. J. Org. Chem. 2012, 77, 3820.
(b) Shu, C.; Zeng, X.; Hao, M.-H.; Wei, X.; Yee, N. H.; Busacca, C. A.;
Han, Z.; Farina, V.; Senanayake, C. H. Org. Lett. 2008, 10, 1303.
a
b
entry
temp (°C)
conversion (%)
1
2
3
4
23
50
75
95
<5
8
(
5) For representative reviews on ligand classes, see corresponding
chapters in ref 2a.
6) See: Kuethe, J.; Zhong, Y.-L.; Yasuda, N.; Beutner, G.; Linn, K.;
27
90
(
Kim, M.; Marcune, B.; Dreher, S. D.; Humphrey, G.; Pei, T. Org. Lett.
2013, 15, 4174 and references therein.
a
Conditions: The concentration of 1-hexene (20) in toluene-d was
8
0
.5 M, and the reactions were carried out in a sealed tube under a
(
7) The manuscript related to this subject matter is currently in
preparation and will be reported in due time.
8) The reaction proceeded to 100% conversion over 1 h with 5 equiv
of 1,4-diacetoxy-cis-2-butene at 80 °C.
9) The serendipitous discovery of novel OM complexes has occurred
nitrogen atmosphere. Conversion was determined via 1H NMR
spectroscopy where product was cleanly observed.
b
(
(
experiment highlights one of the unique properties of this
catalyst in its ability to modulate reactivity based on reaction
temperature.
previously. See: (a) Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P. J.,
Jr.; Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 791. (b) Tallarico, J. A.;
Bonitatebus, P. J., Jr.; Snapper, M. L. J. Am. Chem. Soc. 1997, 119, 7157.
(10) A variety of commercial N-heterocyclic carbene (NHC)-
containing OM catalysts were utilized as the starting material.
In summary, we have transformed a serendipitous discovery of
an intermediate Ru-complex derived from our work on the
synthesis of grazoprevir (MK-5172) into a highly tunable
metathesis catalyst class. The ruthenium complexes are isolated
in good to excellent yields, are shelf- and solution-stable for
extended periods, and exhibit wide-ranging metathesis activities
including active to latent catalysis. Furthermore, we have
successfully designed and explored a strategy for modulating
metathesis reactivity using acids as competent co-catalysts or by
modulating the reaction temperature.
(11) Complex 9a demonstrated solution stability up to 30 days in
CD Cl at room temperature under nitrongen, and the solid material
2
2
showed shelf stability up to 6 months.
12) (a) Szadkowska, A.; Gstrein, X.; Burtscher, D.; Jarzembska, K.;
Wozniak, K.; Slugovc, C.; Grela, K. Organometallics 2010, 29, 117.
b) Barbasiewicz, M.; Szadkowska, A.; Bujok, R.; Grela, K. Organo-
(
(
metallics 2006, 25, 3599. (c) Ung, T.; Hejl, A.; Grubbs, R. H.; Schrodi, Y.
Organometallics 2004, 23, 5399. (d) Puentener, K.; Scalone, M. New
ruthenium complexes as catalysts for metathesis reactions. Eur. Pat.
Appl. EP2008154367A, 2008. (e) Diesendruck, C. E.; Tzur, E.; Ben-
Asuly, A.; Goldberg, I.; Straub, B. F.; Lemcoff, N. G. Inorg. Chem. 2009,
ASSOCIATED CONTENT
Supporting Information
■
4
8, 10819.
*
S
(13) Alternative reaction kinetics can be obtained by modulating the
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b00070.
electronics and sterics of the alkoxybenzylidene ligand. For
representative examples, see: (a) Engle, K. M.; Lu, G.; Luo, S.-X.;
Henling, L. M.; Takase, M. K.; Liu, P.; Houk, K. N.; Grubbs, R. H. J. Am.
Chem. Soc. 2015, 137, 5782. (b) Michrowska, A.; Grela, K. Pure Appl.
Chem. 2008, 80, 31.
Crystallographic data (CIF)
Experimental details, characterization data, and NMR
spectra (PDF)
(14) For promotion of metathesis with acid, see: (a) Keitz, B. K.;
Bouffard, J.; Bertrand, G.; Grubbs, R. H. J. Am. Chem. Soc. 2011, 133,
8498 and references therein. For references regarding the effect of
Brønsted acids on polytopal rearrangements of Ru complexes, see:
AUTHOR INFORMATION
Corresponding Authors
■
*
*
(
b) Gulajski, L.; Michrowska, A.; Bujok, R.; Grela, K. J. Mol. Catal. A:
Chem. 2006, 254, 118.
15) The AlCl ·THF complex was chosen because of the potential
E-mail: michael.williams222@icloud.com.
E-mail: jongrock.kong@merck.com.
(
3
Notes
application toward Brønsted acid sensitive functionality.
16) For reviews on latent OM catalysis, see: (a) Reference 2a.
b) Monsaert, S.; Vila, A. L.; Drozdzak, R.; Van Der Voort, P.; Verpoort,
(
The authors declare no competing financial interest.
(
F. Chem. Soc. Rev. 2009, 38, 3360. For select examples, see: (c) Rouen,
M.; Queval, P.; Falivene, L.; Allard, J.; Toupet, L.; Crevisy, C.; Caijo, F.;
Basle, O.; Cavallo, L.; Mauduit, M. Chem. - Eur. J. 2014, 20, 13716.
(d) Thomas, R. M.; Fedorov, A.; Keitz, B. K.; Grubbs, R. H.
Organometallics 2011, 30, 6713. (e) Hejl, A.; Day, M. W.; Grubbs, R.
H. Organometallics 2006, 25, 6149.
ACKNOWLEDGMENTS
■
We thank Becky Ruck (Merck) for helpful discussions,
suggestions, and editing. We thank Erin Guidry (Merck) for
helpful discussions. Finally we thank Renee K. Dermenjian
(
Merck) and Claudia Monteiro (Merck) for help obtaining
characterization data.
REFERENCES
■
(
2
1) (a) Werrel, S.; Walker, J. C. L.; Donohoe, T. J. Tetrahedron Lett.
015, 56, 5261. (b) Olszewski, T. K.; Bieniek, M.; Skowerski, K.; Grela,
K. Synlett 2013, 24, 903. (c) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D.
Angew. Chem., Int. Ed. 2005, 44, 4490.
(
&
2) (a) Olefin Metathesis Theory and Practice; Grela, K., Ed.; John Wiley
Sons: Hoboken, NJ, United States, 2014. (b) Handbook of Metathesis;
Grubbs, R. H., Ed.; Wiley-VCH: Weinheim, 2003.
3) (a) Knall, A.-C.; Slugovc, C. Olefin Metathesis Polymerization. In
(
Olefin Metathesis Theory and Practice; Grela, K., Ed.; John Wiley & Sons:
Hoboken, NJ, United States, 2014. (b) Slugovc, C. Industrial
D
DOI: 10.1021/acs.orglett.6b00070
Org. Lett. XXXX, XXX, XXX−XXX