Conversion of a Benzofuran Ester to an Amide through an Enamine Lactone Pathway: Synthesis of HCV Polymerase Inhibitor GSK852A

Article abstract of DOI:10.1021/acs.joc.5b01598

HCV NS5B polymerase inhibitor GSK852A (1) was synthesized in only five steps from ethyl 4-fluorobenzoylacetate (3) in 46% overall yield. Key to the efficient route was the synthesis of the highly functionalized benzofuran core 15 from the β-keto ester in

Full text of DOI:10.1021/acs.joc.5b01598

Article
pubs.acs.org/joc
Conversion of a Benzofuran Ester to an Amide through an Enamine
Lactone Pathway: Synthesis of HCV Polymerase Inhibitor GSK852A
Roy K. Bowman,† Kae M. Bullock,† Royston C. B. Copley,*,‡ Nicole M. Deschamps,†
Michael S. McClure,† Jeremiah D. Powers,† Andy M. Wolters,§ Lianming Wu,§ and Shiping Xie*,§
†Product Development, GlaxoSmithKline, 5 Moore Drive, Research Triangle Park, Durham, North Carolina 27709, United States
‡Molecular Discovery Research, GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire SG1
2NY, United Kingdom
§Product Development, GlaxoSmithKline, 709 Swedeland Road, King of Prussia, Pennsylvania 19406, United States
S
* Supporting Information
ABSTRACT: HCV NS5B polymerase inhibitor GSK852A (1) was synthesized in only five steps from ethyl 4-
fluorobenzoylacetate (3) in 46% overall yield. Key to the efficient route was the synthesis of the highly functionalized
benzofuran core 15 from the β-keto ester in one pot and the efficient conversion of ester 6 to amide 19 via enamine lactone 22.
Serendipitous events led to identification of the isolable enamine lactone intermediate 22. Single crystal X-ray diffraction and
NMR studies supported the intramolecular hydrogen bond shown in enamine lactone 22. The hydrogen bond was considered an
enabler in the proposed pathway from ester 6 to enamine lactone 22 and its rearrangement to amide 19. GSK852A (1) was
obtained after reductive amination and mesylation with conditions amenable to the presence of the boronic acid moiety which
was considered important for the desirable pharmacokinetics of 1. The overall yield of 46% in five steps was a significant
improvement to the previous synthesis from the same β-keto ester in 5% yield over 13 steps.
INTRODUCTION
■
Hepatitis C is a liver disease caused by the hepatitis C virus
(HCV). According to the World Health Organization, 130−
150 million people are chronically infected with HCV, and
more than 350 000 people die every year from HCV related
liver diseases.1 Significant discovery efforts from the pharma-
ceutical industry have led to approval of drugs such as
boceprevir, telaprevir, and sofosbuvir by the US Food and Drug
Administration. The emergence of drug resistance is still a
concern for HCV treatment: there is a need to discover and
develop new therapies which have novel modes of action and
Figure 1. Structures of HCV polymerase inhibitors 1 and 2.
resistance profiles. HCV polymerases are essential for HCV
replication. The NS5B protein, an RNA-dependent RNA
polymerase (RDRP) responsible for replicating the viral
the demonstrated potency.3 The benzofuran-3-carboxamide
genome, is a clinically validated target for therapeutic
intervention with many compounds in development.2 Com-
core has been highly optimized, and its clinical relevance has
led to many drug candidates from our program as well as from
other research groups (e.g., HCV796 or Nesbuvir, 2).4 The
focus of this Article is on the synthesis of GSK852A though
many of the findings could also be applicable to the synthesis of
pound 1 shown in Figure 1 is a HCV NS5B polymerase
inhibitor with 50% effective concentrations (EC50) in the low
nanomolar range in the genotype 1 and 2 subgenomic replicon
system as well as the infectious HCV cell culture system. Its
superior resistance profile suggests that it could be an attractive
component of an all-oral regimen for treating HCV.3 The
boronic acid moiety in 1 is an important structural feature for
Received: July 10, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Previous 13-Step Synthesis of GSK852A (1)
Figure 2. Plan for shorter synthesis of GSK852A from β-keto ester 3 or β-keto amide 17.
HCV796 and other drug candidates with the benzofuran-3-
carboxamide core.
residual Pd and the des-bromo/des-boron impurities from the
formation and hydrolysis of the pinacol ester. The use of the
benzyl bromide was also considered an issue due to potential
genotoxicity concerns and the number of steps required to
prepare it from commercially available material.
The main drawbacks of the original synthesis were the large
number of steps required and overall low yield of about 5%
from the β-keto ester.
A new synthetic strategy, shown in Figure 2, was devised for
early introduction of the benzyl fragment already containing the
boronic acid moiety. Reductive amination of commercially
available aldehyde 12 with aryl amines such as 13 and 14 would
then provide 1 or 11. This would not only be more efficient but
would also eliminate the difficult borylation step required at the
end of the synthesis. Furthermore, a one-pot synthesis of the
highly functionalized benzofuran cores such as 15 and 16 was
envisioned starting from a β-keto ester (3) or an amide (17) to
take advantage of many commercially available trihalogenben-
zenes. The synthesis of benzofurans via palladium or copper
catalyzed ring closure of 2-halo aromatic ketones is known.8,9
The 4-fluorophenyl and the ester/amide moieties were
The original synthesis of compound 1 is shown in Scheme 1.
The synthesis started with a ZnCl2 catalyzed Nenitzescu
reaction between ethyl 4-fluorobenzoylacetate (3) and 1,4-
benzoquinone to form 5-hydroxybenzofuran 4 in only 39%
yield.5 Protection of the phenol as an isopropyl ether followed
by nitration led to 5. The isopropyl protection was then
removed, and the phenol was activated as a mesylate for Pd
catalyzed coupling with cyclopropyl boronic acid to give 6.
Hydrogenolysis of the nitro group and mesylation led to bis-
mesylate 7. The ethyl ester was hydrolyzed with the concurrent
removal of one mesyl group, which was followed by coupling
with methylamine to provide the functionalized benzofuran 8.
The 10-step sequence to the benzofuran core was scaled up to
20 kg scale albeit in low overall yield of less than 12% from β-
keto ester 3. Alkylation of 8 with 1-bromo-4-(bromomethyl)-2-
fluorobenzene (9) gave aryl bromide 10. Borylation via
Pd(dba)2 catalysis, followed by harsh hydrolysis conditions
with HCl, generated GSK852A (1) in less than 50% yield due
to extensive purification required to remove large amounts of
B
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 2. Synthesis of Benzofuran Ester 6 and Attempted Conversion to Amide 19 with Methylamine
Scheme 3. Synthesis of GSK852A (1) via Intermediate 20 by Reductive Amination on an Ester Substrate
expected to contribute to a facile ring closure for a potential
one-pot synthesis. This sequence would cut the number of
steps significantly compared with the original synthesis starting
from the same β-keto ester.
in 85% of yield. The product was isolated by direct filtration
upon quenching with aqueous ammonia.
The Suzuki coupling10 with cyclopropyl boronic acid,
catalyzed by palladium acetate and 1,1′-[(oxidi-2,1-
phenylene)]bis[1,1-diphenylphosphine (DPEPhos), provided
ethyl 5-cyclopropylbenzofuran 6 in 86% yield. The DPEPhos
ligand was selected among many ligands screened because it
caused the least amount of a byproduct from debromination.11
The sequence from β-keto ester 3 to benzofuran 6 was scaled
up to 6 L.12
We herein report our findings in the execution of the new
synthetic strategy, in particular an unexpected conversion of a
benzofuran ethyl ester to an enamine lactone which rearranged
to a methyl amide through a proposed reversible pathway. The
unexpected amide formation along with the efficient one-pot
synthesis of a highly functionalized benzofuran and the clean
reductive amination and mesylation in the presence of the
boronic acid moiety were all key components in a significantly
improved synthesis of HCV polymerase inhibitor GSK852A
(1).
Ethyl ester 6 was treated with 33 wt % MeNH2 in EtOH at
room temperature. This mixing led to a clear solution which,
over just 40 min, gave rise to a substantial amount of crystalline
solids. The purported methyl amide 19, highly crystalline and
with the expected molecular weight by LCMS and the number
of protons by NMR for 19, was isolated in 76% yield by simple
filtration of the reaction mixture. The crystalline product was
then carried forward for the rest of the synthesis following the
planned sequence shown in Figure 2: hydrogenolysis of the
nitro group, reductive amination with (2-fluoro-4-
formylphenyl)boronic acid (aldehyde 12), and mesylation
with MsCl/pyridine. However, the sequence of chemistry led
to a product which had slightly shorter retention time in HPLC
compared to target product 1 despite the same molecular
weight (MW 554.4) by LCMS as 1.
RESULTS AND DISCUSSION
■
Our first attempt to synthesize a benzofuran amide is shown in
Scheme 2. Coupling of ethyl 4-fluorobenzoylacetate (3)6 and
1,4-dibromo-2-fluoro-5-nitrobenzene (18)7 in the presence of
K2CO3 led to a 2-phenyl-3-ketoester intermediate at room
temperature. Following addition of a catalytic amount of CuI,
the enol form of the coupled keto ester (see Figure 2 for the
enol structure) ring closed at 60 °C to give the highly
functionalized ethyl 5-bromobenzofuran-3-carboxylate core 15
C
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 4. Synthesis of Amide 19 from Ester 6 via Ester Hydrolysis and Coupling with Methylamine
Figure 3. Comparison of the NMR spectra of the unknown and amide 19 in DMSO-d6.
With the setback in the seemingly simple amide formation,
we took a slight detour from the route originally laid out in
Figure 2. Synthesis of GSK852A (1) was achieved in the
sequence shown in Scheme 3. The nitro group of 6 underwent
hydrogenolysis at 30 °C under slightly pressurized hydrogen
(0.02 bar) in the presence of Pd/C.13 Upon completion, the
mixture was filtered to remove the catalyst and charcoal. The
solvents were partially distilled off and replaced with MeOH for
the reductive amination of intermediate 6-aminobenzofuran in
the same reactor with boronic acid aldehyde 12. Borane/
pyridine14 complex and acetic acid were superior to many other
common reagents15 for the reductive amination. The other
reagents and conditions such as sodium triacetoxyborohydride
failed to give good yield of 20 due to low conversion or
decomposition of product with the sensitive boronic acid
moiety. Amino product 20 was isolated in 86% yield by direct
D
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 5. Synthesis of Amide 19 from Ester 6 via Enamine Lactone 22
Figure 4. Proposed pathway from ethyl ester 6 to enamine lactone 22 and rearrangement of 22 to amide 19.
filtration of the reaction mixture after quenching with aqueous
HCl.
methylenes for the cyclopropyl were more differentiated in
the unknown with one of the methylenes shifted to 0.07 ppm,
relative to 0.73 ppm in amide 19. NMR in CDCl3 showed
similar pattern of shifts. It was also obvious that no ethyl group
was found in the unknown.
Mesylation of 20 with MsCl in pyridine was slow at 35 °C
but very clean. Upon completion of the reaction after 2 days,
the reaction was quenched with water and acidified with
concentrated HCl. The mesylated product 21 was also isolated
by direct filtration in 81% yield after quenching. The rest of the
synthesis, i.e. the hydrolysis of the ethyl ester16 and the methyl
amide formation with HBTU/DIPEA,17 went smoothly to give
carboxylic acid 21 and GSK852A (1) in 76% and 79% yield,
respectively. The overall yield of 31% from β-keto ester 3 in six
steps was demonstrated in 2−6 L scale. This was a significant
improvement relative to the original synthesis of overall yield of
5% in 13 steps.
While the scaleup of the chemistry shown in Scheme 3 was
being carried out, further development of the chemistry
continued. Ester 6 was exposed to 33% MeNH2/EtOH at
room temperature as before, but the reaction was allowed to
continue for about 3 weeks whereupon crystals were generated
as expected. To our surprise, the isolated crystals from the
three-week reaction bore no resemblance to the one isolated
from the previous reaction with MeNH2/EtOH over 40 min
when analyzed by NMR. The newly isolated product was
identical to compound 19 prepared from ester 6 by a two-step
chemistry shown in Scheme 4.
Inadvertent sample switching was ruled out, and another
reaction with MeNH2/EtOH at room temperature was set up
following the exact conditions (40 min) as the first one to
provide a new sample of the unknown product for further tests.
The 1H NMR spectra of the unknown and amide 19 in DMSO-
d6 were quite different as shown in Figure 3. The unknown had
a more acidic proton at 9.69 ppm relative to 8.51 ppm for the
NH of amide 19. Overall, the aromatic protons for the
unknown were shifted upfield. In particular there was a new
peak at 5.09 ppm which appeared to be one of the aromatic
protons, severely shielded to the upper field. The two
More elaborate NMR correlation experiments on the
unknown product revealed that the para-substituted aromatic
ring was spatially close to the cyclopropyl moiety (NOESY)
and that the carbonyl carbon was not within 3 bonds distance
of any nonexchangeable protons (HMBC). On the basis of this
evidence and other NMR data discussed above, an enamine
lactone structure represented by 22 was tentatively assigned to
the thus far unknown compound (Scheme 5). The NH proton
at 9.69 ppm for 22 relative to 8.51 ppm for 19 was probably
due to a potential NH hydrogen bond with the lactone
carbonyl oxygen. The (Z)-configuration of the enamine
explained the shielding effect from the 4-fluorophenyl ring on
one of the benzofuran aryl protons and one of the methylenes
of the cyclopropyl group.
It was clear by then that the enamine lactone 22 first formed
upon treatment of ester 6 with MeNH2/EtOH at room
temperature. If the crystalline solid was not isolated, it
substantially rearranged to amide 19 over 3 weeks at room
temperature or in 1−2 days with heating. In a scale-up to 1 L,
19 was isolated as a crystalline solid in 71% yield by direct
filtration of the reaction mixture after 28 h at 57 °C. In
addition, the isolated amide 19, when submitted to fresh and
heated MeNH2/EtOH, partially reversed to enamine lactone
22. However, no further reversal to ester 6 was observed. The
equilibrium of enamine lactone 22 and amide 19 in MeNH2/
EtOH was apparently biased toward the target amide 19 with
the amide isolated in 71−81% yield. When the isolated
enamine lactone 22 was treated with MeNH2/EtOH for at 60
°C for 18 h, amide 19 was isolated in 86% yield. A reaction
pathway from ester 6 to amide 19 via enamine lactone 22 was
proposed as shown in Figure 4.
E
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
We propose the 1,4-addition as the first interaction of
MeNH2 and the benzofuran 3-carboxylate 6 to give a ketene
hemiacetal (A). A 1,2-addition, which would lead to amide 19
directly from 6, would generate a tetrahedral intermediate and
is expected to have much higher ΔH compared to the
formation of the 1,4-addition adduct (A). In all the reactions we
have monitored by HPLC upon addition of 33% methylamine/
EtOH to ester 6 at room temperature, we observed only less
than 1% of amide 19 prior to depletion of ester 6. The ratio of
enamine lactone 22 to amide 19 was typically >100:1 at the
point when all ester 6 was depleted. Ring opening from the
adduct (A) would lead to the phenolic enoate (B). The
nucleophilic addition of the phenol OH to the ester in B leads
to enamine lactone 22 after loss of EtOH. We believe that this
ring closure is aided by the hydrogen bond shown in B, and this
activates the ester carbonyl. Similarly, the hydrogen bond
shown in 22 activates the lactone carbonyl for nucleophilic
addition by methylamine to give the enamide (C). Intra-
molecular 1,4-addition from the phenol OH of C, aided by the
hydrogen bond shown, leads to the ketene hemiaminal (D).
Finally, rearomatization from D with loss of one molecule of
methylamine affords the benzofuran-3-carboxamide 19.
As discussed before, isolated amide 19 partially reversed to
enamine lactone 22 when subjected to MeNH2/EtOH in a
heated tube. The difference in the solubility of 19 and 22 as
well as the aromaticity of the benzofuran core in 19 could
explain the equilibrium biased toward 19. Experimentally, the
mother liquor, after isolation of 19 from the reaction, gave rise
to more crystals of 19 after standing at room temperature for a
few days or being heated at 60 °C and then cooled back to
room temperature. We did not observe reversibility between 22
and 6.
as a bifurcated hydrogen bond donor (Figure 5). The (Z)-
configuration of the exocyclic CC double bond allows the
intramolecular hydrogen bond between N14 and O26
[2.836(9) Å], while the crystal packing also generates a
corresponding intermolecular interaction between the same
atoms in different molecules [2.941(10) Å]. The intramolecular
hydrogen bond and the (Z)-configuration support the
observations from the NMR studies discussed previously. It
also explains the reactivity for several intermediates in the
proposed pathway shown in Figure 4.
The 3-(benzylidene)benzofuran-2(3H)-one core structure of
22 was previously known and present in marginalin isolated
from the pygidial glands of the water beetle Dytiscus marginalis
(Coleoptera).18 The structure of marginalin is (E)-5-hydroxy-3-
(4-hydroxybenzylidene)benzofuran-2(3H)-one. Barbier re-
ported the synthesis of this natural product and determined it
to be the (E)-configuration.19 Subsequently, it was showed that
marginalin rearranged to strongly fluorescent 5-hydroxy-2-(4′-
hydroxyphenyl)-3-benzo[b]furan carboxylic acid or the methyl
ester under basic conditions such as potassium carbonate or
sodium methoxide in very low yields.20 It was reported that the
two phenolic OHs in 2-phenylbenzo[b]furans were important
for the rearrangement.20b Although not mentioned in the
paper, a careful examination of the structures of marginalin and
the rearranged products under basic conditions revealed the
involvement of oxidation in the reported rearrangement,
probably through the 1.4-benzoquinone from air oxidation.
This would explain the importance of the two phenolic OHs in
the rearrangement. However, this was not the case for our more
elaborate and nitrogen containing structure (22), which did not
need the OH in the 5-position of the benzofuran core for the
rearrangement to occur. The amino moiety of the enamine
lactone (22) allowed nonoxidative rearrangement to the
benzofuran-3-carboxamide. Aniline and bulkier alkylamines
such as i-butylamine, however, were less efficient for the
rearrangement, leading to incomplete and less clean reactions.
The final two steps of the five-step synthesis of 1 are shown
in Scheme 6. Benzofuran-3-carboxamide 19, prepared from
ester 6 in one step with methylamine, was hydrogenolyzed. The
amino intermediate after removal of the catalyst by filtration
was subjected to the reductive amination. The boronic acid
product (23) was isolated in 87% yield by direct filtration of the
HCl quenched reaction. Mesylation of amide 23 with MsCl in
pyridine was more difficult than with ester 20, and MsCl had to
be used in large excess as a cosolvent with pyridine to achieve a
clean and complete reaction at 0 °C. GSK852A (1) was isolated
in 89% yield.
More recently, we obtained the structure of 22 from an X-ray
diffraction study on a single crystal (Figure 5). Although the
diffraction intensities were weaker than normal owing to the
small crystal size, the structure was highly informative. With the
exception of a subset of the carbons in the cyclopropyl and
fluorophenyl rings, the remaining non-hydrogen atoms lay
essentially in a single plane. The NH group of the amine acts
The overall yield from β-keto ester 3 in five steps was 46%
for this amide route. This was better than the ester route shown
in Scheme 3 with 31% overall yield in six steps, and
substantially better than the original synthesis shown in
Scheme 1 with 5% overall yield in 13 steps. The amide route
was selected for further scaleup to deliver GSK852A in the pilot
plant.
CONCLUSIONS
■
Figure 5. View of two molecules of 22 from the crystal structure,
highlighting the hydrogen bonding (dashed lines) and the orientation
of the 4-fluorophenyl ring, shielding a methylene of the cyclopropyl
group and a methine of the fused ring system. The non-hydrogen
atoms in the crystallographic asymmetric unit are numbered.
Anisotropic atomic displacement ellipsoids for the non-hydrogen
atoms are shown at the 50% probability level. Hydrogen atoms are
displayed with an arbitrarily small radius.
GSK852A (1) was synthesized in only five steps from ethyl 4-
fluorobenzoylacetate (3) in 46% overall yield, compared to the
original synthesis from the same β-keto ester in 5% yield over
13 steps. The efficiency was achieved with a one-pot synthesis
of the highly functionalized benzofuran core 15 from β-keto
ester 3 and conversion of benzofuran-3-carboxylate 6 to
benzofuran-3-carboxamide 19 via the unique rearrangement
F
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 6. Synthesis of GSK852A via Reductive Amination with a 6-Aminobenzofuran-3-carboxamide
(s, 1 H), 8.13 (s, 1 H), 8.13 (t, J = 8.0 Hz, 2 H), 7.25 (t, J = 8.0 Hz, 2
H), 4.48 (q, J = 8.0 Hz, 2 H), 1.47 (t, J = 8.0 Hz, 3 H). 13C NMR (100
MHz, CDCl3): δ 165.9, 164.5, 163.4, 162.5, 150.9, 146.4, 132.2 (d, J =
10.0 Hz), 128.5, 124.3, 115.7 (d, J = 21.0 Hz), 109.7, 109.5, 108.2,
81.5, 14.3. HRMS (APCI) m/z calcd for C17H12BrFNO5 (MH+)
407.9877, found 407.9855.
Ethyl-5-cyclopropyl-2-(4-fluorophenyl)-6-nitrobenzofuran-3-car-
boxylate (6). To a 6 L reactor were successively added 360 g (882
mmol) of 15, 91.0 g (1.06 mmol) of cyclopropyl boronic acid, 3.96 g
(17.6 mmol) of Pd(OAc)2, 9.50 g (17.6 mmol) of DPEPhos, and 305
g (2.21 mol) of K2CO3 under nitrogen, followed by 2.9 L of toluene
and 720 mL of water. The reaction was stirred at 70 °C for 2 h. After
being cooled to 20 °C, the mixture was filtered through 360 g of Celite
545 to remove a small amount of emulsions. The Celite pad was
washed with 720 mL of toluene and 1.8 L of water. The organic layer
was concentrated in vacuo to give a slurry. After addition of 1.8 L of
MeOH, the mixture was cooled to 0 °C and stirred overnight. The
slurry was filtered, washed with 720 mL of MeOH, and dried at 65 °C
in vacuo to give 279 g (86%) of 6 as a crystalline solid: mp 133.1 °C.
1H NMR (400 MHz, CDCl3): δ 8.12 (s, 1 H), 8.11 (t, J = 8.0 Hz, 2
H), 7.96 (s, 1 H), 7.23 (t, J = 8.0 Hz, 2 H), 4.46 (q, J = 7.8 Hz, 2 H),
2.54 (m, 1 H), 1.46 (t, J = 7.8 Hz, 3 H), 1.11 (dt, J = 6.0, 7.8 Hz, 2 H),
0.77 (dt, J = 6.0, 7.8 Hz, 2 H). 13C NMR (100 MHz, CDCl3): δ 165.6,
163.7, 163.1, 150.4, 148.4, 134.5, 132.0 (d, J = 9.0 Hz), 131.2, 124.8,
122.6, 115.7 (d, J = 22.0 Hz), 108.6, 107.9, 81.2, 14.2, 13.6, 7.5. HRMS
(ESI) m/z calcd for C20H17FNO5 (MH+) 370.1085, found 370.1084.
Anal. Calcd for C20H16FNO5: C, 65.04; H, 4.37; N, 3.79. Found: C,
65.08; H, 4.47; N, 3.73.
(4-(((5-Cyclopropyl-3-(ethoxycarbonyl)-2-(4-fluorophenyl)-
benzofuran-6-yl)amino)methyl)-2-fluorophenyl)boronic Acid (20).
To a 2 L reactor were added 50.0 g (135 mmol) of 6, 5.00 g of
palladium on charcoal (10 wt %, dry basis, ∼ 50% water), 900 mL of
MeOH, and 450 mL of CH2Cl2. After three cycles of vacuum (−0.49
bar)/nitrogen (0.01 bar) purges, the mixture was heated to 30 °C and
evacuated to −0.50 bar. Hydrogen was introduced via a gas regulator
to 0.02 bar. Upon completion of the hydrogenolysis over 15 h, the
heterogeneous reaction mixture was purged with nitrogen and heated
to 40 °C, and the intermediate hydrogenolysis product fully dissolved.
The mixture was filtered through 50 g of Celite 545 and washed with
200 mL of CH2Cl2. The filtrate was transferred back to the reactor and
concentrated in vacuo to 750 mL. After being diluted with 150 mL of
MeOH, the solution was used directly for the reductive amination as
follows.
of enamine lactone 22. A single crystal X-ray diffraction study
and NMR analysis revealed the configuration of the enamine
lactone with the intramolecular hydrogen bond, which was
considered an enabler for the proposed pathway from ester 6 to
amide 19. The structural cores of 6-amino-5-alkyl-2-aryl-N-
methylbenzofuran-3-carboxamide, represented by 19, have
been highly optimized in structure−activity relationship
(SAR) studies and are frequently seen in antiviral drug
candidates. We have also developed the simple and clean
reductive amination and mesylation in the presence of the
sensitive boronic acid, which has been considered important for
the biological activities of GSK852A. The methods should also
be applicable to other drug candidates containing a boronic acid
moiety.
EXPERIMENTAL SECTION
■
General Procedures. All reactions were run under nitrogen.
Unless otherwise specified, concentration by rotary evaporation or
distillation was carried out under house vacuum of 12−50 Torr.
Melting points were not corrected. NMR spectra were recorded on a
400 MHz spectrometer. The accurate mass measurements were
performed on an Orbitrap mass spectrometer. The heated electrospray
ionization (HESI) source was operated in the positive ion mode under
the following conditions: spray voltage, −4.0 kV; source heater
temperature, 200 °C; heated capillary temperature, 250 °C; S lens RF
level, 60%; sheath gas flow, 50 arbitrary units. The atmospheric
pressure chemical ionization (APCI) source was operated in the
positive ion mode under the following conditions: corona discharge
current, 5 μA; source heater temperature, 300 °C; heated capillary
temperature, 250 °C; S lens RF level, 60%; sheath gas flow, 50
arbitrary units. The single crystal X-ray diffraction data for 22 were
collected with a diffractometer at 150(2) K using Cu Kα X-radiation (λ
= 1.541 78 Å). Crystal data for 22 are given in the relevant section
below. A description of the refinement and the full tables associated
with the crystal structure are given in the Supporting Information. A
crystallographic information file has been deposited with the
Cambridge Crystallographic Data Centre. CCDC 1404059 contains
the supplementary crystallographic data for this paper.
Ethyl-5-bromo-2-(4-fluorophenyl)-6-nitrobenzofuran-3-carboxy-
late (15). To a 6 L reactor were successively added 300 g (1.03 mmol)
of 1,4-dibromo-2-fluoro-5-nitrobenzene, 253 g (1.20 mol) of 4-
fluorobenzoylacetate (crude from evaporation of a 51% solution in
EtOH), 416 g (3.01 mol) of K2CO3, and 2.4 L of DMF. The nearly
homogeneous mixture was stirred at room temperature for 20 h. After
being heated to 45 °C, the mixture was treated with 38.2 g (201
mmol) of CuI. An additional 300 mL of DMF was added as the
reaction mixture became thick. The reaction was heated to 60 °C and
stirred for 20 h. After being cooled to 0 °C, the reaction was quenched
with 1.2 L of 28% ammonium hydroxide via an additional funnel at
such a rate that the mixture was maintained below 10 °C. The mixture
was further treated with 1.2 L of water and stirred for about 30 min at
0 °C. The slurry was filtered, washed with 3.0 L of water and 600 mL
of MeOH, and dried in vacuo at 65 °C to give 350 g (85%) of 15 as a
To the above solution were successively added 25.0 g (149 mmol)
of boronic acid aldehyde 12 and 15.5 mL (271 mmol) of acetic acid at
ambient temperature. After being stirred for 15 min, the mixture was
treated with 12.6 g (135 mmol) of borane pyridine complex over 5
min. Upon completion of the reaction in 1 h, the reaction was
quenched with 6.80 mL (81.2 mmol) of concentrated HCl over 3 min.
The mixture was stirred at 15 °C for 2 h. The yellow slurry was
filtered, washed with 2 × 85 mL of MeOH, and dried at 65 °C in
vacuo to give 57.5 g (86%) of 20 as a light yellow solid: mp 176.0 °C.
1H NMR (400 MHz, DMSO-d6): δ 8.04 (s, 1 H), 7.94 (m, 2 H), 7.70
(m, 1 H), 7.52 (m, 2 H), 7.30 (m, 2 H), 7.26 (m, 1 H), 7.09 (m, 1 H),
6.53 (s, 1 H), 6.31 (m, 1 H), 4.52 (m, 2 H), 4.29 (m, 2 H), 1.82 (m, 1
H), 1.31 (t, J = 6.0 Hz, 3 H), 1.02 (m, 2 H), 0.58 (m, 2 H). 13C NMR
1
crystalline solid: mp 139.2 °C. H NMR (400 MHz, CDCl3): δ 8.45
G
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(100 MHz, CDCl3): δ 169.6, 167.1, 164.5 (d, J = 25.0 Hz), 162.1,
157.1, 154.3, 146.1, 145.4 137.1 (m), 131.2, 126.3, 124.9, 122.9, 122.2,
116.9 (d, J = 4.0 Hz), 115.1 (d, J = 21.0 Hz), 113.2 (m), 108.8, 91.9,
80.5, 50.8, 47.7, 14.3, 11.8, 5.4. HRMS (ESI) m/z calcd for
C27H25BF2NO5 (MH+) 492.1788, found 492.1792.
solid: mp 135.6 °C. 1H NMR (400 MHz, DMSO-d6): δ 13.1 (s, 1 H),
8.12 (s, 1 H), 8.05 (t, J = 8.0 Hz, 2 H), 7.84 (s, 1 H), 7.52 (m, 1 H),
7.45 (m, 1 H), 7.36 (m, 3 H), 7.02 (dd, J = 10.0, 8.0 Hz, 2 H), 4.90
(ABq, ΔδAB = 0.14, JAB = 16.0 Hz, 2 H), 3.24 (s, 3 H), 2.27 (m, 1 H),
0.98 (m, 1 H), 0.75 (m, 2 H), 0.23 (m, 1 H). 13C NMR (100 MHz,
DMSO-d6): δ 167.0, 164.7, 162.4, 160.4, 151.4, 140.8, 136.6, 135.9,
132.4 (d, J = 10.0 Hz), 127.6, 126.0, 124.8, 116.8, 115.9 (d, J = 22.0
Hz), 112.2, 109.4, 56.5, 54.9, 38.5, 19.0, 12.0, 11.3, 9.1. HRMS (APCI)
calcd for C26H23BF2NO7S (MH+) 542.1251, found 542.1253.
(4-((N-(5-Cyclopropyl-3-(ethoxycarbonyl)-2-(4-fluorophenyl)-
benzofuran-6-yl)methylsulfonamido)methyl)-2-fluorophenyl)-
boronic Acid (11). To a 2 L reactor were successively added 57.0 g
(116 mmol) of 20 and 375 mL of pyridine at ambient temperature.
The resultant solution was treated with 27.8 mL (360 mmol) of MsCl.
The reaction was heated to 35 °C and stirred for 20 h. An additional
8.90 mL (116 mmol) of MsCl was added due to incomplete reaction
overnight. After being stirred for 20 h, the reaction was cooled to 25
°C and quenched with 350 mL of water, followed by 350 mL of
EtOAc. The mixture was acidified with 358 mL (4.29 mol) of
concentrated HCl over 10 min. After the mixture was stirred for 15
min, an additional 150 mL of EtOAc and 150 mL of water were added,
but very little further dissolution was observed. The heterogeneous
mixture was then filtered, washed with 2 × 200 mL of water, and dried
in vacuo at 60 °C to give 47.1 g (71%) of 11 as an off-white solid.
Layers in the filtrate were separated, and the aqueous layer was back
extracted with 150 mL of EtOAc. The combined organic layers were
washed with 200 mL of brine and evaporated to 25.0 g. A slurry was
made by addition of 20 mL of acetone, 40 mL of ethanol, and 15 mL
of water to the 25.0 g of crude product. Filtration and drying at 60 °C
gave 11.2 g (17%) of 11 as an off-white solid.
The two solids (47.1 + 11.2 g) were combined, treated with 60 mL
of acetone and 180 mL of ethanol, and heated to 40 °C for 15 min,
followed by addition of 60 mL of water. The mixture was cooled to 20
°C, stirred for 30 min, filtered, washed with 2 × 60 mL of 1:1 EtOH/
water and 60 mL of EtOH, and dried at 65 °C in vacuo to give 53.2 g
(81% yield for reaction, 91% recovery for recrystallization) of 11 as an
off-white crystalline solid: mp 165.7 °C. 1H NMR (400 MHz, CDCl3):
δ 8.04 (m, 2 H), 7.76 (t, J = 8.0 Hz, 1 H), 7.55 (s, 1 H), 7.31 (s, 1 H),
7.19 (t, J = 8.0 Hz, 2 H), 7.10 (dd, J = 10.0, 8.0 Hz, 2 H), 5.04 (d, J =
4.0 Hz, 2 H), 4.92 (ABq, ΔδAB = 0.25, JAB = 8.0 Hz, 2 H), 4.42 (q, J =
7.0 Hz, 2 H), 3.06 (s, 3 H), 2.24 (m, 1 H), 1.43 (t, J = 6.0 Hz, 3 H),
1.10 (m, 1 H), 0.96 (m, 2 H), 0.59 (m, 1 H). 13C NMR (100 MHz,
CDCl3): δ 165.3, 163.6, 162.8, 161.4, 151.2, 142.1 (d, J = 9.0 Hz),
139.8, 137.2 (d, J = 7.0 Hz), 135.0, 131.8 (d, J = 6.0 Hz), 128.1, 125.5,
117.8, 115.7 (d, J = 20.0 Hz), 115.4 (d, J = 22.0 Hz), 112.8, 108.5,
80.8, 50.1, 39.7, 14.2, 11.4, 11.0, 9.2. HRMS (ESI) m/z calcd for
C28H27BF2NO7S (MH+) 570.1564, found 570.1564. Anal. Calcd for
C28H26BF2NO7S: C, 59.06; H, 4.60; N, 2.46, S, 5.63. Found: C, 58.94;
H, 4.67; N, 2.64, S, 5.73.
6-(N-(4-Borono-3-fluorobenzyl)methylsulfonamido)-5-cycloprop-
yl-2-(4-fluorophenyl)benzofuran-3-carboxylic Acid (21). To a 2 L
reactor were added 53.0 g (93.1 mmol) of 11, 530 mL of MeOH, and
265 mL of water at ambient temperature. The white slurry was treated
with 11.4 g (205 mmol) of KOH. The resultant clear solution was
heated to 60 °C and stirred for 4 h. The reaction was cooled to
ambient temperature and stirred overnight. An additional 3.12 g (5.56
mmol) of KOH was added due to the incompletion, and the mixture
was stirred at 60 °C for 1.5 h.
The mixture was concentrated to about 400 mL by distillation in
vacuo. After being cooled to ambient temperature, the mixture was
treated with 500 mL of water and 500 mL of EtOAc. Layers were
separated, and the aqueous layer was washed with 2 × 200 mL of
EtOAc. The organic layers were discarded, and the aqueous layer was
cooled to 10 °C and acidified to pH 1.0 with ∼200 mL of 1 N HCl
and 5.5 mL of concentrated HCl. The resultant solids seemed not
suitable for filtration, so the mixture was extracted with 350 and 200
mL of EtOAc successively. The combined EtOAc extracts were washed
with 200 mL of brine and concentrated in vacuo to dryness. After
addition of 150 mL of EtOH, 50 mL of water, 15 mL of acetone, 2.0
mL of acetic acid, and 5 mg of seeds of 21 previous prepared in smaller
scale, the solution was stirred overnight at ambient temperature for
crystallization. The mixture was cooled to 0 °C, stirred for 1 h, filtered,
washed with 80 and 50 mL of 1:1 EtOH/water successively, and dried
at 60 °C in vacuo to give 38.3 g (76%) of 21 as an off-white crystalline
5-Cyclopropyl-2-(4-fluorophenyl)-N-methyl-6-nitrobenzofuran-3-
carboxamide (19). A. Amidation from 6 via Enamine Lactone 22 in
Methylamine/EtOH. To a 1 L reactor calorimetry vessel were
successively added 35.0 g (94.7 mmol) of ester 6, 245 mL of 33 wt
% methylamine in ethanol, and 105 mL of absolute ethanol at ambient
temperature. The well-stirred heterogeneous mixture was heated at 57
°C for 28 h. The pressure was about 0.7 bar due to the heating of the
methylamine. After being cooled to 0 °C over 30 min, the mixture was
stirred for 1 h, filtered, and washed with 2 × 70 mL of MeOH. The
filtering cake was dried at 60 °C in vacuo to give 23.7 g (71%) of
1
amide product 19 as a light yellow crystalline solid: mp 249.8 °C. H
NMR (400 MHz, CDCl3): δ 8.07 (s, 1 H), 7.92 (t, J = 8.0 Hz, 2 H),
7.74 (s, 1 H), 7.25 (t, J = 8.0 Hz, 2 H), 5.82 (s, 1 H), 3.04 (d, J = 4.0
Hz, 3 H), 2.49 (m, 1 H), 1.09 (dt, J = 6.0, 7.8 Hz, 2 H), 0.76 (dt, J =
6.0, 7.8 Hz, 2 H). 13C NMR (100 MHz, CDCl3): δ 165.4, 163.5, 162.9,
157.7, 150.6, 148.5, 134.4, 131.5, 130.6 (d, J = 8.0 Hz), 124.7, 121.1,
116.4 (d, J = 21.0 Hz), 112.5, 108.0, 26.7, 13.6, 7.5. HRMS (ESI) calcd
for C19H16FN2O4 (MH+) 355.1089, found 355.1084. Anal. Calcd for
C19H15FN2O4: C, 64.40; H, 4.27; N, 7.91. Found: C, 64.17; H, 4.27;
N, 7.80.
B. Amidation from 6 through Hydrolysis and Coupling with
Methyamine Hydrochloride. A mixture of 500 mg (1.35 mmol) of 6
and 152 mg (2.71 mmol) of KOH in 15 mL of 2:1 EtOH/water was
heated to 80 °C. The solution was stirred for 20 min, cooled to room
temperature, and evaporated in vacuo to remove most of the ethanol.
After being treated with 5 mL of 1 N HCl, the resultant solids were
filtered, washed with 2 × 5 mL water, and dried at 50 °C to give 480
mg of the carboxlic acid intermediate as a white solid. To this solid
were successviely added 700 mg (1.85 mmol) of O-benzotriazole-
N,N,N′,N′-tetramethyluronium hexafluorophosphate (HBTU), 142
mg (2.11 mmol) of methylamine hydrochoride, and 3.5 mL of
DMF. The mixture was stirred for 3 min at room temperature to form
a solution, followed by addition of 0.60 mL (3.43 mmol) of N,N-
diisopropylethylamine. After being stirred for 30 min, the reaction was
quenched with 8 mL of water, filtered, washed with 2 × 5 mL of water,
and dried at 60 °C to give 470 mg (97%) of 19 as a solid which was
1
identical to the same product prepared from method A by H NMR
and HPLC. The compound prepared with method B was also
converted to target compound 1.
(Z)-5-Cyclopropyl-3-((4-fluorophenyl)(methylamino)methylene)-
6-nitrobenzofuran-2(3H)-one (22). To a mixture of 3.00 g (8.12
mmol) of 6 and 9 mL of EtOH was added 21 mL of 33 wt %
methylamine in EtOH at room temperature for 5 h. The reaction
remained heterogeneous throughout the reaction. After being cooled
to 0 °C, the mixture was filtered, washed with 2 × 5 mL of EtOH, and
dried at 60 °C to give 2.36 g (82%) of 22 as a yellow solid.
Alternatively, the reaction was run with 1.00 g (2.70 mmol) of 6 in 10
mL of 33 wt % methylamine in EtOH without additional EtOH added.
The reaction was faster without the added EtOH and reached
completion after 40 min. Filtering, washing with 5 mL of MeOH, and
drying at 60 °C provided 734 mg (76%) of 22 isolated as a crystalline
1
solid: mp 220.7 °C. H NMR (400 MHz, CDCl3): δ 9.38 (s, 1 H),
7.47 (s, 1 H), 7.23−7.24 (m, 4 H), 5.25 (s, 1 H), 2.82 (d, J = 4.0 Hz, 3
H), 2.12 (m, 1 H), 0.65 (dt, J = 6.0, 7.8 Hz, 2 H), 0.00 (dt, J = 6.0, 7.8
Hz, 2 H). 13C NMR (100 MHz, CDCl3): δ 170.6, 165.3, 164.7, 162.8,
146.2, 145.5, 134.2, 131.4, 129.6 (d, J = 8.0 Hz), 127.1, 117.4 (d, J =
22.0 Hz), 115.7, 106.6, 89.4, 31.5, 13.4, 7.4. HRMS (ESI) calcd for
C19H16FN2O4 (MH+) 355.1089, found 355.1083. Anal. Calcd for
C19H15FN2O4: C, 64.40; H, 4.27; N, 7.91. Found: C, 64.34; H, 4.41;
N, 7.77. Crystal data: C19H15FN2O4; M = 354.33; yellow block;
crystallization from synthetic route in ethanol; 0.08 × 0.05 × 0.03
H
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
mm3; monoclinic; space group, P21/c (No. 14); unit cell dimensions, a
= 7.9217(15) Å, b = 20.462(6) Å, c = 9.881(2) Å, β = 100.377(18)°, V
= 1575.5(6) Å3; Z = 4; dcalc = 1.494 Mg m−3; and μ(Cu Kα, λ =
1.541 78 Å) = 0.959 mm−1.
(4-(((5-Cyclopropyl-2-(4-fluorophenyl)-3-(methylcarbamoyl)-
benzofuran-6-yl)amino)methyl)-2-fluorophenyl)boronic Acid (23).
To a solution of 500 mg (1.4 mmol) of 19 in 7.5 mL of THF was
added 5% Pt/alumina (0.05 g). A hydrogen filled balloon was inserted,
and the mixture was stirred at room temperature for 24 h. The
reaction was heated to 60 °C to ensure full dissolution of the product,
and the catalyst was removed via filtration through a short pad of
Celite. The Celite pad was washed with 2.5 mL of hot THF. HPLC
analysis of the filtrate showed 99.5% product. That was carried into the
next reaction without isolation.
A solution of crude amino intermediate from a hydrogenolysis
reaction (4.58g, 14.7 mmol, assuming 100% yield from the
hydrogenolysis) in THF was concentrated in vacuo to 25 mL. To
the solution were successively added 2.72 g (16.2 mmol) of aldehyde
12, 2.52 mL (44.2 mmol) of acetic acid, and 35 mL of ethanol. After
reduction of the volume to 15 mL via atmospheric distillation, 35 mL
of ethanol was added. The volume was again reduced to 15 mL via
atmospheric distillation, followed by addition of 35 mL of ethanol. GC
analysis showed 95% ethanol in the mixture. The mixture was treated
with 1.57 g (16.9 mmol) of borane-pyridine complex at 70 °C.
Analysis by HPLC showed the reaction mostly complete at the end of
the addition. An additional 0.2 g (2.2 mmol) was added, and the
reaction was stirred at 70 °C for 30 min. The reaction was quenched
with 9.2 mL of water and stirred at 70 °C for 1 h, followed by addition
of 20 mL of ethanol to improve the stirring. After being cooled to
room temperature, the mixture was filtered, washed with 25 mL of
ethanol, and dried in vacuo at 50 °C to provide 5.94 g (87%) of 23 as a
white solid: mp 196.5 °C. 1H NMR (400 MHz, DMSO-d6): δ 8.90 (s,
1 H), 8.27 (m, 1 H), 7.83 (m, 2 H), 7.51−7.71 (m, 1 H), 7.20−7.29
(m, 3 H), 7.15 (s, 1 H), 7.09 (dd, J = 18.0, 8.0 Hz, 2 H), 6.52 (m, 1
H), 3.58 (s, 1 H), 3.18 (s, 2 H), 2.81 (m, 3 H), 1.81 (m, 1 H), 1.00
(m, 2 H), 0.60 (m, 2 H). 13C NMR (100 MHz, DMSO-d6): δ 166.1,
164.4, 163.5, 161.1, 153.9, 148.9, 146.8, 145.3, 135.8, 128.8 (d, J = 8.0
Hz), 127.0, 124.9, 122.6 (m), 119.4, 116.2 (m), 114.4, 113.7 (m), 91.8,
50.9, 49.1, 46.4, 26.0, 12.2, 6.2. HRMS (ESI) calcd for C26H24BF2N2O4
(MH+) 477.1792, found 477.1788.
(m, 1 H), 0.80 (m, 2 H), 0.29 (m, 1 H). 13C NMR (100 MHz,
DMSO-d6): δ 167.0, 164.6 (m), 163.5, 161.9, 153.6, 151.0, 140.7,
140.3, 136.4, 135.9 (d, J = 10.0 Hz), 129.8 (d, J = 8.0 Hz), 127.8, 126.1
(m), 124.8, 116.6 (d, J = 22.0 Hz), 115.9 (d, J = 10.0 Hz), 115.2,
113.9, 112.0, 54.9, 38.5, 26.6, 12.0, 11.2, 9.0. HRMS (ESI) calcd for
C27H26BF2N2O6S (MH)+, 555.1567, found 555.1565. Anal. Calcd for
C27H25BF2N2O6S: C, 58.50; H, 4.55; N, 5.05; S, 5.78. Found: C,
58.29; H, 4.66; N, 5.14; S, 5.69.
B. Amide Route (Scheme 6). To a solution of 2.00 g (4.20 mmol)
of 23 in 4 mL of pyridine was added 2.00 mL (25.2 mmol) of MsCl at
0 °C. The reaction was stirred at 0 °C for 8 h. After being diluted with
8 mL of acetone, the reaction was quenched with 5 mL of water and
acidified to pH < 7.0 with 2 mL of concentrated HCl. About 1 g of
NaCl was added to saturate the aqueous layer. Layers were separated,
and the aqueous layer was extracted with 16 mL of acetone. The
combined organic layers were distilled in vacuo to 18 mL and heated
to 50 °C. The solution was treated with 8 mL of water and seeded with
crystals of 1 previously prepared from another route. After being
stirred for 30 min, the mixture was treated with an additional 8 mL of
water dropwise. The mixture was cooled to room temperature, filtered,
washed with 8 mL of 1:1 acetone/water, and dried in vacuo at 50 °C
to afford 2.07 g (89%) of 1 as a crystalline solid.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.5b01598.
■
S
Crystallographic data (CIF)
NMR spectra for all new compounds, detailed NMR
studies (NOESY and HMBC) for compound 22, and full
crystallographic tables (PDF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail: royston.c.copley@gsk.com.
*E-mail: shiping.x.xie@gsk.com.
■
Notes
The authors declare no competing financial interest.
(4-((N-(5-Cyclopropyl-2-(4-fluorophenyl)-3-(methylcarbamoyl)-
benzofuran-6-yl)methylsulfonamido)methyl)-2-fluorophenyl)-
boronic Acid (1). A. Ester Route (Scheme 3). To a 2 L reactor were
successively added 38.0 g (70.20 mmol) of 21, 5.21 g (77.2 mmol) of
methylamine hydrochloride, and 29.3 g (77.2 mmol) of O-
benzotriazole-N,N,N′,N′-tetramethyluronium hexafluorophosphate
(HBTU), followed by 266 mL of DMF at ambient temperature.
After being cooled to 10 °C, the solution was treated with 30.7 mL
(175 mmol) of N,N-diisopropylethylamine (Hunig’s base) over 5 min
and stirred for 2 h. The reaction was quenched with 340 mL of water
and 16.1 mL (281 mmol) of acetic acid, followed by addition of 380
mL of EtOAc at room temperature. Layers were separated, and the
aqueous layer was back extracted with 160 and 100 mL of EtOAc
successively. The combined orgnaic layers were successively washed
with 160 mL of water, 160 mL of 0.5 N HCl, and 160 mL of brine, and
filtered through 30 g of Celite 545.
The filtrate was returned to the reactor and concentrated in vacuo
to about 100 mL. After addition of 300 mL of MeCN, the solution was
further concentrated to 150 mL. After addition of 100 mL more
MeCN, 50 mL of n-PrOH, and 45 mL of 0.25 N aqueous HCl, the
mixture was heated to 35 °C to form a clear solution. About 10 mg of
seeds of 1, previously prepared in a smaller scale, was added, and this
initiated crystallization immediately. The mixture was cooled to 10 °C
over 30 min, stirred for 45 min, filtered, washed with 100 and 50 mL of
2:1 water/MeCN successively, and dried at 60 °C in vacuo to give 30.9
g (79%) of 1 as a white crystalline solid: mp 176.1 °C. 1H NMR (400
MHz, DMSO-d6): δ 8.41 (m, 1 H), 8.12 (s, 2 H), 7.93 (m, 2 H), 7.81
(s, 1 H), 7.44 (t, J = 8.0 Hz, 1 H), 7.34−7.38 (m, 2 H), 7.03 (dd, J =
18.0, 8.0 Hz, 1 H), 6.94 (s, 2 H), 4.88 (ABq, ΔδAB = 0.12, JAB = 18.0
Hz, 2 H), 3.24 (s, 3 H), 2.80 (d, J = 4.0 Hz, 3 H), 2.27 (m, 1 H), 0.96
ACKNOWLEDGMENTS
■
We thank Professor W. Clegg and Dr. R. W. Harrington for the
GSK funded crystal structure of 22. We also wish to thank Dr.
Richard Matsuoka for data integrity check and Dr. Pek Chong
for helpful discussions during manuscript preparation.
REFERENCES
■
(1) World Health Organization (WHO). Hepatitis C Fact Sheet No.
164, updated April 2014: http://www.who.int/mediacentre/
factsheets/fs164/en/index.html (accessed July 7, 2015).
(2) (a) Beaulieu, P. L. Expert Opin. Ther. Pat. 2009, 19, 145.
(b) Powdrill, M. H.; Bernatchez, J. A.; Gotte, M. Viruses 2010, 2, 2169.
(3) (a) Voitenleitner, C.; Crosby, R.; Walker, J.; Remlinger, K.;
Vamathevan, J.; Wang, A.; You, S.; Johnson, J., III; Woldu, E.; Van
Horn, S.; Horton, J.; Creech, K.; Shotwell, J. B.; Hong, Z.; Hamatake,
R. Antimicrob. Agents Chemother. 2013, 57, 5216. (b) 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. J.
Med. Chem. 2014, 57, 1902.
(4) Kneteman, N. M.; Howe, A. Y. M.; Gao, T.; Lewis, J.; Pevear, D.;
Lund, G.; Douglas, D.; Mercer, D. F.; Tyrrell, D. L. J.; Immermann, F.;
Chaudhary, I.; Speth, J.; Villano, S. A.; O'Connell, J.; Collett, M.
Hepatology 2009, 49, 745.
I
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(5) (a) Nenitzescu, C. D. Bull. Soc. Chim. Romania 1929, 11, 37.
(b) Allen, G.; Pidacks, C.; Weiss, M. J. Am. Chem. Soc. 1966, 88, 2536.
(6) We purchased ethyl 4-fluorobenzoylacetate. It can be prepared
from 4-fluorobenzoic acid and a malonate or Meldrum’s acid. See:
Raillard, S. P.; Chen, W.; Sullivan, E.; Bajjalieh, W.; Bhandari, A.; Baer,
T. A. J. Comb. Chem. 2002, 4, 470.
(7) We purchased 1,4-dibromo-2-fluoro-5-nitrobenzene. It can be
prepared from readily available 1,4-dibromo-2-fluorobenzene with 70%
HNO3 and concentrated H2SO4 at room temperature. See: Muro, F.;
Iimura, S.; Yoneda, Y.; Chiba, J.; Watanabe, T.; Setoguchi, M.; Iigou,
Y.; Takayama, G.; Yokoyama, M.; Takashi, T.; Nakayama, A.;
Machinaga, N. Bioorg. Med. Chem. 2008, 16, 9991.
(8) (a) Chen, C.; Dormer, P. G. J. Org. Chem. 2005, 70, 6964.
(b) Willis, M. C.; Taylor, D.; Gillmore, A. T. Org. Lett. 2004, 6, 4755.
(c) Ackermann, L.; Kaspar, L. T. J. Org. Chem. 2007, 72, 6149.
(d) Eidamshaus, C.; Burch, J. D. Org. Lett. 2008, 10, 4211.
(9) For a recent review on the synthesis of benzofurans, see: Sharma,
U.; Naveen, T.; Maji, A.; Manna, S.; Maiti, D. Angew. Chem. 2013, 125,
12901.
(10) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
(11) We typically observed 1−2% of debromination which was
removed in the crystallization of the product from MeOH.
(12) An analogous sequence starting from β-keto amide 17 as shown
in Figure 2 was only briefly investigated because the reaction was less
clean. The amide 17 was prepared from β-keto ester 3 and
methylamine in the presence of Candida antarctica lipase B (CAL-
B), albeit in only about 50% yield. For lipase catalyzed conversion of β-
keto esters to β-keto amides, see: Garcia, M. J.; Rebolledo, F.; Gotor,
V. Tetrahedron 1994, 50, 6935.
(13) Figueras, F.; Coq, B. J. Mol. Catal. A: Chem. 2001, 173, 223.
(14) Burkhardt, E. R.; Coleridge, B. M. Tetrahedron Lett. 2008, 49,
5152.
(15) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C.
A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849.
(16) Yamada, K.-i.; Nakano, M.; Maekawa, M.; Akindele, T.;
Tomioka, K. Org. Lett. 2008, 10, 3805.
(17) Uhlmann, E.; Peyman, A.; Breipohl, G.; Will, D. W. Angew.
Chem., Int. Ed. 1998, 37, 2796.
(18) Schildknecht, H.; Korning, W.; Siewerdt, R.; Krauss, D. Justus
Liebigs Ann. Chem. 1970, 734, 116.
(19) Barbier, M. Liebigs Ann. Chem. 1987, 1987, 545.
(20) (a) Barbier, M. Liebigs Ann. Chem. 1991, 1991, 393. (b) Barbier,
M. Org. Prep. Proced. Int. 1991, 23, 676.
J
DOI: 10.1021/acs.joc.5b01598
J. Org. Chem. XXXX, XXX, XXX−XXX