Process development of C-N cross-coupling and enantioselective biocatalytic reactions for the asymmetric synthesis of niraparib

Article abstract of DOI:10.1021/op400233z

Process development of the synthesis of the orally active poly(ADP-ribose)polymerase inhibitor niraparib is described. Two new asymmetric routes are reported, which converge on a high-yielding, regioselective, copper-catalyzed Narylation of an indazole derivative as the late-stage fragment coupling step. Novel transaminase-mediated dynamic kinetic resolutions of racemic aldehyde surrogates provided enantioselective syntheses of the 3-aryl-piperidine coupling partner. Conversion of the C-N cross-coupling product to the final API was achieved by deprotection and salt metathesis to isolate the desired crystalline salt form.

Full text of DOI:10.1021/op400233z

Article
pubs.acs.org/OPRD
Process Development of C−N Cross-Coupling and Enantioselective
Biocatalytic Reactions for the Asymmetric Synthesis of Niraparib
Cheol K. Chung,*,† Paul G. Bulger,*,† Birgit Kosjek,† Kevin M. Belyk,† Nelo Rivera,† Mark E. Scott,‡
Guy R. Humphrey,† John Limanto,† Donald C. Bachert,§ and Khateeta M. Emerson§
†Department of Process Chemistry, Merck & Co., Inc., Rahway, New Jersey 07065, United States
‡Department of Medicinal Chemistry, Merck & Co., Inc., Boston, Massachusetts 02115, United States
§Department of Chemical Process Development and Commercialization, Merck & Co., Inc., Rahway, New Jersey 07065, United
States
S
* Supporting Information
ABSTRACT: Process development of the synthesis of the orally active poly(ADP-ribose)polymerase inhibitor niraparib is
described. Two new asymmetric routes are reported, which converge on a high-yielding, regioselective, copper-catalyzed N-
arylation of an indazole derivative as the late-stage fragment coupling step. Novel transaminase-mediated dynamic kinetic
resolutions of racemic aldehyde surrogates provided enantioselective syntheses of the 3-aryl-piperidine coupling partner.
Conversion of the C−N cross-coupling product to the final API was achieved by deprotection and salt metathesis to isolate the
desired crystalline salt form.
Evaluation of alternative fragment-coupling strategies that
avoided the use of azide reagents at elevated temperature was
also desirable, as was elimination of the use of chlorinated
solvents, heavy transition metal catalysts (e.g., Pd, Pt),
specialized equipment (pressurized hydrogenation) and silica
gel purification. Furthermore, detailed investigation of
structure−activity relationships had revealed that the primary
amide substituent (CONH2) on the indazole ring of the
tetracyclic core conferred a level of potency that led to low
occupational exposure limits for penultimate 8 and the API 9.
Dedicated potent compound-handling facilities would be
needed for steps involving the generation, use, or handling of
compounds containing the primary amide motif; thus,
minimization of the number of unit operations would be
beneficial. Collectively, opportunities for sizable improvements
in overall yield and productivity, with concomitant reductions
in cycle time, cost and environmental impact, presented
themselves for the synthesis of the target compound 9. A
further demand on any new route would be that it reliably
deliver material meeting exacting purity specifications with no
new impurities (relative to earlier preclinical batches), including
the undesired enantiomer, at levels above ICH qualification
thresholds.
INTRODUCTION
■
The enzyme poly(ADP-ribose)polymerase (PARP) is involved
in a number of cellular processes, one of these being to assist in
the repair of single-strand breaks in DNA. Overexpression of
the PARP enzyme in certain cancer types has been implicated
in cancer cell proliferation and tumor progression.1 Con-
sequently, inhibitors of PARP are of significant interest in the
oncology field, with a number of small-molecule agents having
been entered into clinical trials.2 A recent report3 from these
laboratories described the development of a kilogram-scale
synthesis of niraparib p-toluenesulfonate monohydrate (9,
Scheme 1), an orally active PARP inhibitor shown to be
efficacious in murine xenograft cancer models.4 This con-
venient route was fit-for-purpose in being amenable to the
production of the quantities of active pharmaceutical ingredient
(API) needed to advance the compound through early
preclinical evaluation and into the clinic. However, it was
envisaged that more efficient, practical chemistry would
subsequently be needed for significantly larger API deliveries
as the program would move towards late-stage development.
We describe herein the results of laboratory-scale development
of alternative routes to the target compound 9. These routes
feature novel bio- and chemocatalytic transformations that
build upon the foundations laid by recent advances in these
areas of catalysis.
RESULTS AND DISCUSSION
■
General Synthetic Strategy. Rather than attempt to
optimize the existing route further, we elected to investigate a
new approach which featured a C−N coupling between
indazole 11 and chiral 3-aryl-piperidine 12 as the convergent
The first kilogram-scale synthesis of compound 9, isolated as
the p-toluenesulfonate salt monohydrate crystalline form,
proceeded in 11 total steps and 11% overall yield (Scheme
1).3 The linear sequence from commercial starting materials 1
and 2 was relatively short, but a number of opportunities were
identified for longer-term route development. A primary
objective was to develop an asymmetric synthesis of the
piperidine unit, as the modest throughput of the chromato-
graphic resolution was recognized to be a significant bottleneck.
Special Issue: Transition Metal-Mediated Carbon-Heteroatom Cou-
pling Reactions
Received: August 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
introducing steric bias. Furthermore, a judicious choice of C-7
substituent could also minimize the number of steps requiring
potent compound handling during the manipulation of
coupling product 10 to the final API 9.
Scheme 1. First kilogram-scale synthesis of compound 9
A number of methods have been reported for the asymmetric
synthesis of 3-aryl-piperidines.7 We proposed an approach that
would expand the scope of an emerging biocatalytic technology
for chiral amine synthesis, namely the use of transaminase
enzymes.8 Starting from racemic open-chain aldehyde 13
(Scheme 3), if selection of the desired (S)-aldehyde enantiomer
Scheme 3. Proposed biocatalytic approach to piperidine 12
by the transaminase enzyme could be accompanied by
racemization of unreacted (R)-aldehyde, then a dynamic kinetic
resolution (DKR) process would result. This, in theory, could
lead to all of the racemic starting material 13 being converted to
(S)-primary amine 14 with high enantioselectivity. This
concept is analogous to the DKR hydrogenation of ketones.9
Incorporation of a suitably positioned electrophilic group, e.g.
an ester or an alkyl halide, into the aldehyde side chain would
allow for intramolecular trapping of amine 14 to generate
piperidine 15 in a single operation.
At the outset of our work we were encouraged to find that
proof of principle of such a transaminase DKR approach had
been demonstrated in elegant studies by Kroutil and co-
workers, who prepared the five-membered ring lactam 4-
phenylpyrrolidin-2-one in up to 92% yield and 68% ee.10,11
However, to the best of our knowledge, application of this
methodology to piperidine homologues has not been reported.
Enantioselective Synthesis of Piperidine Fragment
Using Biocatalysis. The synthesis of transaminase aldehyde
substrate 23 is illustrated in Scheme 4. Friedel−Crafts acylation
of bromobenzene with succinic anhydride 16, using stoichio-
metric AlCl3 as a promoter, afforded ketoacid 17.12 Temper-
ature control during the exothermic quench of the reaction
mixture into aqueous HCl was required to minimize formation
of the debrominated byproduct 18; below 35 °C the level of 18
formed was typically 1−2%, and this was easily rejected during
downstream crystallizations. The crude product 17 was isolated
in 85−90% purity, and contained a number of more nonpolar
byproducts. The purity was upgraded to 99% by extraction into
aqueous basic solution followed by crystallization upon
reacidification, affording acid 17 in 93% overall yield. Acid 17
was then converted to the corresponding isopropyl ester 19 in
93% yield under standard Fisher esterification conditions. After
distillative removal of the 2-propanol solvent and aqueous
workup, isopropyl ester 19 was crystallized from heptanes in
99% purity and 93% yield. The corresponding methyl (20) and
ethyl (21) ester analogues were prepared in a similar manner,
but were found to give much lower yields in subsequent steps.
step (Scheme 2). Anticipated to be challenging was controlling
the regiochemistry of the coupling to the N-2 position, as
literature reports of the N-arylation of indazoles with aryl
halides5 or aryl boron species6 suggested a preference for
coupling at the undesired N-1 position. However, we
anticipated that the presence of the C-7 substituent in indazole
11 could offer a handle to modulate the regiochemistry by
Scheme 2. General retrosynthetic strategy
B
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 4. Preparation of aldehyde 23
Figure 1. Selected results of epoxide rearrangement HTE screening.15
Aldehyde 23 was initially screened against a small panel of
isolated transaminase enzymes,16 using pyridoxal-5-phosphate
(PLP) as a cofactor and i-PrNH2 as the amine donor. As shown
in Table 1, ATA-301 was identified as providing good
The homologation of ketone 19 to aldehyde 23 was effected
using a two-step epoxidation/rearrangement protocol. The
epoxidation step was carried out by premixing solid Me3SOI
and KOt-Bu in DMSO, followed by the addition of a solution
of ketone 19 in 1:1 DMSO/THF. This mixed solvent system
represented an appropriate balance between reaction volume
efficiency (significantly less solvent volume required to
solubilize ketone 19 using THF) and yield (cleaner reaction
profile in DMSO). After a 2 h reaction time at room
temperature followed by aqueous workup, epoxide 22 was
obtained in 76% HPLC assay yield. Identifiable components of
the mass balance of this reaction appeared to result from
competitive nucleophilic attack at the ester group and/or
opening of the epoxide ring. These more polar byproducts were
effectively partitioned into the aqueous phase during workup,
such that the final organic layer contained epoxide 22 at
relatively high purity (∼90 HPLC area %). Epoxide 22 was an
oil, and could not be crystallized. For evaluation of the
rearrangement step, small samples of product were purified by
flash chromatography.
Initial attempts at the conversion of epoxide 22 to aldehyde
23 using BF3·OEt2 were not promising, with the reactions being
capricious and low-yielding. Screening of conditions was
therefore performed using microscale high-throughput exper-
imentation (HTE) techniques.13 A total of 12 acids and 8
solvents were evaluated in 96-well plate format, and selected
results are illustrated in Figure 1. Of the acids screened, ZnBr2
and InBr3 were found to be uniquely effective, with a trend to
increasing product formation as solvent polarity decreased.
Although the screening results suggested CH2Cl2 was the
solvent of choice, PhMe was preferred for scale up on the basis
of cost, with the same consideration driving the selection of
ZnBr2 over InBr3 as the Lewis acid promoter. In addition, a
higher-boiling solvent was desirable for development of a
through process from ketone 19 (vide infra). It was found that
the use of a 25 mol % loading of ZnBr2 in PhMe resulted in
>99% conversion of epoxide 22 within 1.5 h at room
temperature (Scheme 4).14 After filtration to remove Zn salts,
aldehyde 23 could be isolated in 91% yield by flash
chromatography.
Table 1. Results of first-pass transaminase screening of
aldehyde 23
entry
enzyme
HPLC area % of 25
ee (%) of 25
1
2
3
4
5
6
7
8
ATA-007
ATA-013
ATA-014
ATA-016
ATA-301
ATA-234
CDX-010
CDX-017
19
84
87
87
81
89
88
72
n.d.
75
59
66
99
0
44
45
conversion to the desired lactam 25 with excellent
enantioselectivity (>99% ee). The use of the corresponding
methyl ester 24 generally resulted in worse reaction profiles
(due to competing background ester hydrolysis) and lower
enantioselectivities, although using ATA-301 lactam 25 was
formed in 99% ee and 68 HPLC area%. Based on these initial
findings, the use of isopropyl ester 23 with enzyme ATA-301
was selected for further optimization.
During the initial screening, variable levels of formation of
ketone 19 (Figure 2) were observed; this was found to result
from oxidative degradation of aldehyde starting material 23
under the basic conditions, occurring even in the absence of
enzyme. This side reaction could be obviated by rigorous
inertion of the vessel under nitrogen. Variation of pH revealed
that a pH of 10.5 was the optimum balance between the rate of
the desired transaminase reaction and that of competing ester
hydrolysis to give acid 26. Sufficient enzyme stability was
maintained at 45 °C to fully consume aldehyde starting material
23 and enable a satisfactory rate of cyclization of amine
intermediate 27 (typically less than 2% of 27 remained at the
C
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 6. Preparation of C−N coupling substrate 31
Figure 2. Proposed structures of side products 19 and 26, and
intermediate 27 formed during the transaminase reaction.
end of reaction). An acceptable reaction profile could be
achieved at a reduced enzyme loading of 50 wt %, albeit at the
expense of extended reaction time.
Under these optimized conditions, lactam 25 was formed in
86% assay yield and >99% ee starting from 35 g of isopropyl
ester 23 (Scheme 5). Isolation of the product was accomplished
with good recovery by aqueous workup followed by
crystallization from IPAc.
Scheme 5. Result of 35-g-scale transaminase reaction of
aldehyde 23
While these results were highly encouraging, there were
improvements to be made for development of a practical
process. Aldehyde 23 was obtained as a solution in PhMe
following the ZnBr2-promoted epoxide rearrangement, but the
PhMe would need to be removed prior to the transaminase
step due to the incompatibility of the enzyme with common
organic solvents. Aldehyde 23 was an oil, precluding its
isolation by crystallization, and was also found to have limited
stability both when stored neat and in solution. We sought a
more stable aldehyde surrogate and hypothesized that the
corresponding bisulfite adduct 28 (Scheme 6) could serve this
purpose, liberating the parent aldehyde in situ under the basic
conditions of the transaminase reaction. As shown in Scheme 6,
treatment of the PhMe solution of crude aldehyde 23 with
aqueous NaHSO3 resulted in formation and extraction of
bisulfite adduct 28 into the aqueous layer.17 Relative to
aldehyde 23, bisulfite 28 demonstrated improved stability in
solution, with a degradation rate of <1% per day at ambient
temperature. After a phase cut and wash with heptane, the
aqueous layer containing 28 could be charged directly to the
transaminase reaction. Performance of bisulfite 28 closely
matched that of using the parent aldehyde 23. In addition,
further screening had identified the ATA-302 enzyme as a more
robust variant and, under the conditions illustrated in Scheme
6, lactam 25 (>99% ee) was isolated in 84% yield from bisulfite
28. The 4-step through process from ester 19 to lactam 25
proceeded reliably on 100 g scale. To achieve full conversion of
bisulfite 28, relatively large enzyme loadings (35 wt % relative
to starting material 28) and reaction times (44 h) were
required. To support larger-scale production in a pilot-plant
setting, optimization of the transaminase enzyme structure to
improve activity through rounds of directed evolution8a would
be warranted.
A number of reducing agents were evaluated for the
conversion of lactam 25 to piperidine 29. DIBAL, Red-Al, or
LiBH4 led to product formation but also resulted in varying
amounts of des-bromo impurity 30 (Scheme 6). Borane
reduction, however, was more efficient and gave minimal
debromination. The use of borane generated in situ from
NaBH4 and BF3·THF was preferred over commercially
available borane solutions for cost, reagent handling, and
volume productivity reasons.18,19 Initially, an elevated reaction
temperature of 55 °C was required to obtain full conversion
and a clean profile. Subsequent optimization studies revealed
that the addition of EtOH (3 equiv relative to lactam 25, 1
equiv relative to NaBH4) allowed the reaction to proceed at
ambient temperature. Hazard evaluation of this reduction was
performed due to three potential safety issues, namely: (1)
potential for thermal hazards, (2) corrosion of glass-lined
equipment due to the formation of HF, and (3) generation of
diborane, a highly toxic and explosive gas. No potential thermal
hazards were identified by DSC screening, and a paper
assessment of the chemistry suggested that generation of
more than trace amounts of fluoride ion was not expected.
Monitoring of the vessel headspace of a trial reaction by FTIR
spectroscopy showed that the level of diborane gas generated
was significantly below the lower explosive limit in air (8400
ppm) at all stages during the process (Figure 3).20
The piperidine product 29 from the reduction was only
weakly crystalline and was difficult to isolate as a solid with
D
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 7. Lactol transaminase route to piperidine 31
Figure 3. Time course of diborane levels in vessel headspace
monitored by FTIR spectroscopy.
good reproducibility; therefore it was converted to the
corresponding p-toluenesulfonate salt 29·TsOH (Scheme 6),
affording good physical properties and impurity control
(isolated solids were >99 wt % purity). The subsequent Boc
protection to give C−N coupling substrate 31 was achieved in
quantitative yield using Boc2O in a biphasic mixture of MTBE
and aqueous NaOH. The product 31 was a low-melting, waxy
solid that was difficult to handle; thus, it was more convenient
to carry forward the crude product solution (following a
solvent-switch into DMAc, vide infra) into the C−N coupling
step. The synthesis of compound 31 was thus accomplished in
nine steps (four isolations) and 44% overall yield from
commercial starting materials.
Alternative Biocatalytic Routes to Piperidine Frag-
ment. In addition to approach described above, proof-of-
concept was also established on a second transaminase DKR-
based route in which lactol 35 (Scheme 7) was employed as an
alternative aldehyde surrogate. Lactone 34 was prepared in two
steps with minor adjustments to a procedure reported
previously from these laboratories.21 Dianion formation from
acid 32 using MeLi or LDA (rather than n-HexLi, to avoid
undesired metal−halogen exchange) followed by addition of 1-
bromo-3-chloropropane gave crude monoalkylated product 33,
which could be cyclized by treatment with DBU in IPAc
(cleaner reaction than in THF). On laboratory scale, lactone 34
was isolated by flash chromatography, although the overall yield
for the two-step process was somewhat variable (60−80%).
Reduction of lactone 34 to lactol 35 was effected using 1 equiv
of DIBAL in 1:1 THF/PhMe at −15 °C. The use of THF
cosolvent was required to maintain complete solubility of the
lactone starting material 34 below ambient temperature. An
excess of DIBAL or higher temperatures led to overreduction to
the diol. After aqueous workup, lactol 35 was isolated as a
stable solid in 89% yield by crystallization from PhMe/heptane.
Lactol 35 proved to be an excellent substrate for trans-
aminase DKR. Little modification was needed to the reaction
conditions that had been developed for aldehyde 23, with the
primary amine product 36 being generated in >98% ee and
>99% HPLC area% purity at full conversion of the starting
material 35 (Scheme 7). The main procedural change was that
the optimum pH for this process was 8.5 rather than 10.5;
possibly at lower pH there is a higher concentration of the
open-chain aldehyde available in equilibrium. The amino
alcohol product 36 had high water solubility, and attempted
recovery by extraction with organic solvents resulted in
significant product loss to the aqueous phase. The workup
was therefore changed to precipitate the spent enzyme at the
end of the reaction using aqueous HCl, followed by removal by
filtration. The acidic filtrate was then adjusted to pH 10 and
treated directly with Boc2O to give alcohol 37, which could be
readily extracted. This crude product was, in turn, reacted with
MsCl in the presence of Et3N to give mesylate 38, which could
be crystallized from hexanes/EtOAc. Mesylate 38 was isolated
in >99% purity and 48% overall yield for this three-step
sequence. HTE screening of the intramolecular cyclization of
mesylate 38 revealed that stronger bases were more effective
(Figure 4); confirmation of representative conditions (NaOH,
DMF) gave Boc-piperidine 31 in 94% assay yield (Scheme 7).
Although less extensively developed than the approach using
bisulfite adduct 28, this lactol transaminase route offers a viable
alternative. The use of chloroaldehyde 39 or the corresponding
bisulfite adduct 40 as substrates was also briefly explored, as
these were expected to give piperidine 29 directly from the
transaminase DKR process (Scheme 8). In practice, however,
both the synthesis of the substrates and their performance in
the transaminase reaction were plagued by low yields and
instability issues, and this approach was not pursued at length.
E
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
indazoles 41−46 (Table 2) that were either commercially
available or readily prepared in one step from carboxylic acid
41. Key results from the output of several hundred microscale
screening experiments are summarized in Table 2 and Figure 5.
Figure 4. Screening results of the intramolecular cyclization of
mesylate 38.15
Scheme 8. Alkyl chloride transaminase approach
Figure 5. Selected ligand screening results of C−N coupling of
indazole 46.15
C−N Coupling. With bromide 31 in hand, a regioselective
C−N coupling with the indazole heterocycle was required to
efficiently assemble the tetracyclic core of the target API.
Copper ligand catalysis5 was first evaluated using a series of
Table 2. Regioselectivity of C−N coupling with indazoles 41−46
a
b
c
All reactions were run for 24 h. Conversion of bromide starting material 31 was measured by HPLC. Significant competing ester hydrolysis was
d
observed. DMI = 1,3,dimethyl-2-imidazolidinone.
F
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 9. Optimized C−N coupling of Boc-protected
piperidine 31 with indazole 46
Figure 6. Byproducts formed during the C−N coupling.
Scheme 10. C−N coupling of indazole 46 with unprotected
piperidine 29·TsOH
The carboxylic acid (41) and methyl ester (42) indazole
derivatives displayed no regioselectivity (Table 2, entries 1 and
2), while higher selectivity for arylation at the desired N-2
position was observed with indazoles 43−46 (entries 3−6).
The good regioselectivity observed using cyano-substituted
indazole 43 (entry 3) confounds a simplistic analysis that steric
bulk at the C-7 position on the indazole ring is the sole
determining factor. In addition to their excellent regioselectiv-
ity, both amide derivatives 45 and 46 were appealing in that
they offered a more streamlined endgame to the API; no late-
stage amidation step would be required, and their correspond-
ing C−N coupling products 47 (penultimate to the API) were
determined to not require potent compound handling
conditions. Ultimately, the tert-butylamide 45 was selected
over the cumyl derivative 46 on the basis of atom-economy and
raw material cost.
The performance of the Cu-catalyzed C−N coupling was
found to be dependent on a number of parameters. The choice
of ligand was critical. Although diamine 51 was effective under a
broad range of screening conditions (Figure 5), 8-hydroxy-
quinoline 5222 was competitive using the specific conditions of
K2CO3 as base in DMAc. Both ligands gave very high levels of
regioselectivity (>100:1) and chemoselectivity (no coupling at
the amide nitrogen was observed). The ligands were differ-
entiated on cost (8-hydroxyquinoline 52 being cheaper) and
subsequent catalyst loading studies, which showed that the
coupling yield dropped precipitously as the amount of diamine
ligand 51 was lowered from 40 mol % relative to aryl bromide
31, while systems using 8-hydroxyquinoline 52 were more
robust. Reaction temperatures above 100 °C were required to
achieve full conversion of aryl bromide 31. CuBr was the
optimum copper source, with CuI or Cu2O being much less
efficient. The optimum ligand:Cu ratio was 2:1. Mixing CuBr
and 8-hydroxyquinoline 52 in DMAc resulted in formation of a
homogeneous solution which was stable for at least 18 h prior
to addition of the substrates. However, preforming the catalyst
was not a requirement, and it was found to be operationally
easier to simply add all the reaction components sequentially to
the vessel. Indeed, a Cu(II)/bis(8-hydroxyquinoline) complex
is commercially available but was not as effective as using the
species generated in situ from CuBr and ligand 52. The
reaction was tolerant to the presence of at least 10% (by
volume) water but was found to be highly oxygen sensitive;
intrusion of oxygen resulted in stalling of the reaction and
degradation of unreacted indazole 46. To obtain consistent
results, degassing of the reaction mixture prior to heating,
followed by a nitrogen sweep through the vessel headspace
during the reaction, was performed. The reaction is
heterogeneous, as the K2CO3 base does not fully dissolve. On
laboratory scale with overhead mechanical stirring, powdered
and granular K2CO3 were found to be equally effective.
Palladium catalysis of the C−N coupling was also screened, but
only the expensive, sterically hindered phosphine ligand 5723
(Figure 5) gave good conversion, accompanied by an impurity
profile inferior to that of the Cu-catalyzed reactions.
The optimized C−N coupling sequence is illustrated in
Scheme 9. Indazole 46 was prepared from acid 41 and tert-
butylamine through a CDI-mediated amide bond formation,
which proved superior to mixed-anhydride or HATU amidation
protocols. Removal of the CO2 (by sparging or applying
vacuum) generated during the acyl imidazole activation step
facilitated the subsequent displacement by tert-butylamine. A
slight excess of CDI was used (1.2 equiv), although significant
overcharge beyond this was found to lead to formation of
elevated levels of 1,3-di-tert-butylurea, which could not be
efficiently rejected during the crystallization of indazole 46 by
addition of water to the reaction mixture.
Under the optimized conditions shown in Scheme 9,
coupling of bromide 31 in DMAc with a near stoichiometric
amount of indazole 46 at 5 mol % copper loading proceeded to
>99% conversion within 24 h. The observed regioselectivity
was >200:1 for the N-2 isomer 56. Minor byproducts identified
in the mixture were debrominated piperidine 30 and
G
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
compound 58 (Figure 6), the latter resulting from C−O
coupling between ligand 52 and the aryl bromide 31. After
filtration to remove inorganic salts (filtration rate was improved
by the addition of Celite), the desired product 56 could be
directly crystallized by addition of aqueous solution. The use of
10% aqueous citric acid afforded slightly improved copper
removal compared to pure water or aqueous ammonia, with
lower copper levels being retained in the product cake (<100
ppm vs 300−500 ppm). Compound 56 was isolated in 94%
yield and >99% weight % purity.
Proof-of-concept was also demonstrated on the use of
piperidine tosylate salt 29·TsOH in the C−N coupling. When
tosylate salt 29·TsOH was subjected to the conditions that had
been optimized for the corresponding Boc-protected substrate
31, product 59 was formed in a lower assay yield of 78%
(Scheme 10), with acetyl derivative 60 and homocoupling
adduct 61 being major byproducts. A small screening study
identified DMSO as a solvent that gave an improved reaction
profile, generating the desired product 59 in 90% assay yield. In
addition to 8-hydroxyquinoline 52, ligands 51 and 53 were also
effective. Given the potential for energetic decomposition of
DMSO,24 accelerating rate calorimetry was used to confirm that
there were no thermal safety issues with running a process
under these conditions. Although not optimized further, this
coupling approach would allow a more expedient route to the
API by eliminating the Boc protection step.
selection of solvents evaluated, ortho-xylene was the most
effective. In practice, when a slurry of penultimate 56 in o-
xylene was treated with excess MSA at 40 °C, formation of 9·
MSA with <0.1% tert-butylamide 59·MSA remaining was
achieved within 3 h.
Isolation of the API was facilitated by exploiting the large
difference in aqueous solubility between the desired p-
toluenesulfonate salt monohydrate crystalline form 9 (<1
mg/mL) and that of the mesylate salt 9·MSA resulting from the
deprotection reaction (>100 mg/mL). Once the deprotection
was complete, mesylate salt 9·MSA was partitioned into an
aqueous layer by dilution with water, and the organic phase was
cut away. Control of the batch temperature below 25 °C during
the water addition was required to minimize hydrolysis of the
primary amide group to give undesired carboxylic acid 62·MSA.
Addition of an aqueous solution of p-TsOH then resulted in
salt metathesis and solubility-driven crystallization to afford the
desired salt polymorph 9 in 95% yield. This material was
isolated in excellent purity (99.8 HPLC area%, >99.7% ee) and
met all the required specifications.
CONCLUSION
■
In summary, two new synthetic routes to the PARP inhibitor
niraparib (9) have been developed which feature novel bio- and
chemocatalytic transformations (Scheme 12). The scope of
biocatalytic transaminase reactions has been expanded to
encompass highly efficient and stereoselective dynamic kinetic
resolution processes using a bisulfite adduct or lactol as a more
stable aldehyde surrogate. Incorporation of appropriate func-
tional group handles into the transaminase reaction substrates
enabled a concise, enantioselective synthesis of the 3-aryl-
piperidine ring system. Convergent fragment coupling was
achieved through a copper-catalyzed C−N arylation with an
indazole derivative that proceeded with high chemo- and
regioselectivity. Development of these and other steps was
facilitated by extensive use of microscale high-throughput
experimentation techniques, which significantly expanded the
parameter space for reaction screening while concomitantly
reducing the time, resources and material needed for hit-finding
and optimization.
Deprotection and API Isolation. Removal of both the
tert-butyl amide and piperidine-Boc protecting groups from
penultimate 56 could be accomplished in a single step under
strongly acidic conditions (Scheme 11). While the Boc group
Scheme 11. Conversion of penultimate 56 to API 9
The route using bisulfite adduct 28 comprises 12 steps with 7
isolations, and proceeds in 40% overall yield for the longest
linear sequence of 11 steps. The approach using lactol 35 was
less extensively developed, but proceeds in 23−30% overall
yield for the longest linear sequence of 9 steps, comparing
favorably with the 11% overall yield of the first kilogram-scale
synthesis. In both of these new routes, several reactions could
be telescoped to reduce cycle time and the number of
isolations, and potent compound handling operations were
minimized by developing a streamlined process from
penultimate 56 in which the final API could be isolated by
using a salt metathesis-driven crystallization. While scope for
further optimization exists, both routes address key issues
hindering further scale-up of the previously described syntheses
of niraparib (9). They would also be expected to have greatly
reduced cost and environmental impact, and further demon-
strate the enabling potential of continuing advances in catalysis
technology on chemical process development.
was cleaved readily, more forcing conditions were required to
fully liberate the primary amide. Treatment of penultimate 56
with neat anhydrous methanesulfonic acid (MSA, 22 equiv) at
40 °C resulted in 98% conversion to the MSA salt of the API
(9·MSA) within 1 h. However, deprotection of the remaining
2% of tert-butylamide intermediate 59·MSA slowed dramati-
cally, and required more than 24 h to reach full conversion. It
was found that use of an aromatic cosolvent to scavenge tert-
butyl cations significantly enhanced the reaction rate. Of a small
EXPERIMENTAL SECTION
■
General. All reactions were carried out under N2
atmosphere. THF was dried by storing over molecular sieves.
H
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 12. Summary of new routes to niraparib 9
All other commercial chemicals were used as received.
Temperatures refer to internal batch temperatures, unless
otherwise stated. Reactions were monitored by reverse-phase
HPLC on a Hewlett-Packard 1100 Series instrument, using the
following conditions as a standard method - Column: Ascentis
Express C18 column (100 mm × 4.6 mm, 2.7 μm fused-core
particle size); Eluent: (A) 0.1% aq H3PO4, (B) MeCN;
Gradient: linear ramp from 10% to 95% B over 6 min, then
hold at 95% B for 2 min, then 2 min post time at 10% B; Flow
rate: 1.8 mL/min; Sample injection volume: 5 μL; Column
temperature: 40 °C; UV detection: 210 nm. HPLC assay yields
and weight % purities were obtained using analytical standards
give 265 g of 17 (99 wt %, 93% corrected yield) as a white
powder. Mp 147 °C; HPLC tR 3.47 min; 1H NMR (400 MHz,
DMSO-d6) δ 12.15 (s, 1H), 7.93−7.89 (m, 2H), 7.76−7.72 (m,
2H), 3.23 (t, J = 6.5 Hz, 2H), 2.58 (t, J = 6.5 Hz, 2H); 13C
NMR (101 MHz, DMSO-d6) δ 197.7, 173.7, 135.4, 131.8,
129.9, 127.2, 33.1, 27.8; IR (neat) 2893, 1694, 1670, 1583,
1408, 1197, 839, 790, 645 cm−1; HRMS calcd for C10H10BrO3
[M + H]+ 256.9813, found 256.9818.
Isopropyl 4-(4-Bromophenyl)-4-oxobutanoate (19).
Concentrated H2SO4 (2.13 mL, 39.9 mmol) was charged to a
slurry of acid 17 (205 g, 797 mmol) in 2-propanol (4 L) at 20
− 25 °C, and the mixture was heated to gentle reflux (82−83
°C). After 40 h the solution was cooled, and then concentrated
to a volume of 350−400 mL at 35−40 °C. The resulting slurry
was dissolved in MTBE (1.2 L), and aq Na2CO3 (prepared by
dissolving 35 g Na2CO3 in 500 mL water) was added in one
portion at 20−25 °C, and the biphasic mixture was stirred for
10 min. The layers were separated, and the organic layer was
washed with water (1 × 400 mL) and filtered. The filtrate was
concentrated and solvent-switched by distillation under vacuum
at 35−40 °C (bath temperature) into heptane to a final volume
of ∼600 mL. The resulting slurry was aged at 20−25 °C for 4 h,
then at 2−5 °C for a further 1 h. Filtration, washing with chilled
(∼5 °C) heptane (1 × 200 mL), and drying in vacuo at 20−25
°C under a N2 sweep for 16 h afforded 224 g of 19 (99 wt %,
93% corrected yield) as a white powder. Mp 62 °C; HPLC tR
5.17 min; 1H NMR (400 MHz, CDCl3) δ 7.85−7.81 (m, 2H),
7.60−7.57 (m, 2H), 5.01 (sept, J = 6.5 Hz, 1H), 3.24 (t, J = 6.5
1
of known purity. H and 13C NMR spectra were recorded on
Bruker Avance NMR Spectrometers. Chemical shifts are
reported in ppm relative to the residual deuterated solvent.
Melting points were obtained on a Buchi B-545 Melting Point
̈
instrument and are uncorrected. Optical rotations were
measured on a Perkin-Elmer 241 polarimeter, and data are
reported as [α]20 (concentration in g/100 mL, solvent).
D
Infrared spectra were recorded as thin films on a Nicolet Nexus
570 FT-IR instrument. High-resolution mass spectra were
recorded on a Micromass QTOF Ultima API US mass
spectrometer by electrospray ionization.
4-(4-Bromophenyl)-4-oxobutanoic Acid (17). Bromo-
benzene (695 mL, 6.6 mol) and succinic anhydride 16 (110 g,
1.1 mol) were combined at 20−25 °C, and slurry was cooled to
2−5 °C. AlCl3 (293 g, 2.2 mol) was added in one portion, and
after 30 min the slurry was allowed to warm to 25 °C over 5 h,
and then aged at that temperature for a further 80 h. The HCl
evolved during the reaction was vented with a gentle N2 sweep
through the vessel headspace into a NaOH scrubber. A separate
vessel was charged with water (1.2 L) and 37% aq HCl (220
mL) and the solution cooled to 0−5 °C. The Friedel−Crafts
reaction mixture was then transferred over a period of 30 min
into the stirred HCl solution (EXOTHERMIC), maintaining
the temperature of the quenched mixture below 35 °C. The
resulting white slurry was aged at 10 °C for 1 h, and was then
filtered and the cake washed with water (2 × 800 mL). The
product wet cake was then slurried in MTBE (500 mL), and aq
NaOH (200 mL 10 M aq NaOH dissolved in 1 L water) was
added over 5 min at 20−25 °C. After a further 15 min the
layers were separated. The aqueous layer was acidified to pH 1
using 37% aq HCl (∼250 mL) at 20−25 °C. The resulting
slurry was aged at 10 °C for a further 1 h and then filtered. The
cake was washed with water (2 × 500 mL), then dried in vacuo
under a N2 sweep at 20−25 °C for 4 h then at 50 °C for 16 h to
Hz, 2H), 2.71 (t, J = 6.5 Hz, 2H), 1.23 (d, J = 6.5 Hz, 6H); 13
C
NMR (101 MHz, CDCl3) δ 197.1, 172.1, 135.3, 131.8, 129.5,
128.2, 68.0, 33.3, 28.5, 21.7; IR (neat) 2980, 2917, 1721, 1679,
1582, 1310, 1170, 1115, 784 cm−1; HRMS calcd for
C13H16BrO3 [M + H]+ 299.0283, found 299.0283.
Isopropyl 3-(2-(4-Bromophenyl)oxiran-2-yl)-
propanoate (22). Trimethylsulfoxonium iodide (230 g, 1.04
mol) was slurried in DMSO (300 mL) at 20−25 °C. KOt-Bu
(113 g, 1.01 mol) was added in one portion, followed by
additional DMSO (300 mL). After 1.5 h a solution of ketone
19 (223 g, 0.75 mmol) in a mixture of THF (250 mL) and
DMSO (150 mL) was added over 45 min at 20−25 °C. After a
further 2 h the reaction was diluted with hexanes (1 L), then
quenched by the addition of chilled (∼5 °C) water (600 mL)
over 5 min, maintaining the temperature below 25 °C. The
layers were separated, and the organic layer was washed with
water (1 × 500 mL) and then sat. aq NaCl (1 × 500 mL).
HPLC assay gave 176 g (76% yield) of epoxide 22 in the final
I
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
organic solution, which was held overnight at 20−25 °C before
being used in the rearrangement step. For characterization, a
sample of epoxide 22 was isolated by flash column
chromatography on silica gel (gradient elution from 19:1 to
9:1 n-hexane/EtOAc), affording a light-yellow oil. HPLC tR
171.7, 140.7, 131.9, 128.7, 120.9, 48.4, 39.0, 31.1, 27.7; IR
(neat) 3305, 2950, 2885, 1658, 1612, 1488, 1463, 819, 752
cm−1; HRMS calcd for C11H13BrNO [M + H]+ 254.0181,
found 254.0191.
Sodium 2-(4-Bromophenyl)-1-hydroxy-5-isopropoxy-
5-oxopentyl sulfite (28). To the toluene solution of aldehyde
23 was charged a solution of NaHSO3 (24.7 g, 238 mmol) in
water (140 mL) in one portion at 20−25 °C, and the biphasic
mixture was stirred vigorously for 3 h. The layers were
separated, and the aqueous layer washed with heptanes (1 × 60
mL). The aqueous layer (234.2 g) contained 71.2 g bisulfite
adduct 28 (30.4 wt % solution, 90% yield from epoxide 22) by
1H NMR assay against a water-soluble internal standard (3-
(trimethylsilyl)-1-propanesulfonic acid sodium salt). This
solution was held overnight at 20−25 °C before being used
in the transaminase step. For characterization, a sample of
bisulfite adduct 28 was prepared using an alternative procedure
as described in the Supporting Information, affording a white
powder as a 1:1 mixture of diastereoisomers. Mp 133−135 °C;
HPLC tR 3.51/3.57 min; 1H NMR (400 MHz, D2O) δ 7.51 (d,
J = 8.5 Hz, 2H), 7.49 (d, J = 8.5 Hz, 2H), 7.25 (d, J = 8.5 Hz,
2H), 7.24 (d, J = 8.5 Hz, 2H), 4.84 (sept, J = 6.5 Hz, 1H), 4.83
(sept, J = 6.5 Hz, 1H), 4.66 (d, J = 7.0 Hz, 1H), 4.58 (d, J = 5.5
Hz, 1H), 3.25−3.16 (m, 2H), 2.53−2.46 (m, 1H), 2.44−2.36
(m, 1H), 2.20−1.97 (m, 6H), 1.18 (d, J = 6.5 Hz, 3H), 1.18 (d,
J = 6.5 Hz, 3H), 1.15 (d, J = 6.5 Hz, 3H), 1.14 (d, J = 6.5 Hz,
3H); 13C NMR (101 MHz, D2O) δ 175.5, 175.3, 140.1, 138.7,
131.5, 131.3, 131.1, 130.5, 120.4, 86.9, 86.3, 69.4, 69.4, 46.6,
46.0, 32.3, 27.7, 24.7, 23.8, 21.0; IR (neat) 3290, 2979, 1714,
1374, 1177, 1104, 1037, 834, 642 cm−1; HRMS calcd for
C14H23BrNO6S [M(acid) + NH4]+ 412.0429, found 412.0431.
(S)-3-(4-Bromophenyl)piperidin-1-ium 4-methylben-
zenesulfonate (29·TsOH). To a slurry of piperidinone 25
(10.3 g, 97.5 wt %, 39.4 mmol) in THF (100 mL) was added
NaBH4 (4.47 g, 118 mmol) in one portion at 0−2 °C. EtOH
(6.9 mL, 118 mmol) was then added over 20 min. After a
further 1 h, BF3·THF (13.0 mL, 118 mmol) was added over 1
h. The temperature rose to 7.5 °C during this addition (GAS
EVOLVED). The resulting milky solution was aged at 5 °C for
1 h then at 20−25 °C for 16 h. The reaction was then cooled to
0−5 °C then slowly quenched with 8 mL MeOH (GAS
EVOLVED). The mixture was aged for 10 min until offgassing
subsided, then 37% aq HCl (9.7 mL) was added (EXOTHER-
MIC − batch temperature rose to 17 °C during the HCl
addition). The mixture was then heated to 45 °C and aged for 2
h to complete decomplexation of the product-borane complex.
The mixture was cooled to 15 °C, diluted with IPAc (75 mL)
and water (80 mL), then adjusted to pH 8 using 28% aq
NH4OH (14 mL). The resulting biphasic mixture was aged for
1 h at 20−25 °C, then the layers were separated. To the organic
layer was added water (75 mL), then the pH was adjusted to
10.5 using 50 wt % aq NaOH (∼2.5 mL). The layers were
separated, and the organic layer was washed with 10% aq NaCl
(1 × 50 mL), filtered, and solvent-switched by distillation under
vacuum at 35−40 °C (bath temperature) into IPAc to a final
volume of ∼90 mL. HPLC assay gave 9.4 g (37.4 mmol, 96%
yield) of piperidine 29. The IPAc solution was heated to 40 °C,
and p-TsOH·H2O (8.2 g, 43 mmol) was added in portions over
5 min. The resulting slurry was further heated to 50 °C, aged
for 2 h, then cooled to 20−25 °C and aged for 16 h. Filtration,
washing with IPAc (3 × 15 mL), and drying in vacuo at 20−25
°C under a N2 sweep for 16 h afforded 14.9 g of 29·TsOH
(99.4 wt %, 96% corrected yield) as a white powder. Mp 156
1
5.53 min; H NMR (500 MHz, CDCl3) δ 7.48−7.46 (m, 2H),
7.27−7.24 (m, 2H), 4.97 (sept, J = 6.5 Hz, 1H), 2.99 (d, J = 5.0
Hz, 1H), 2.71 (d, J = 5.0 Hz, 1H), 2.52 (ddd, J = 14.5, 10.0, 6.0
Hz, 1H), 2.36 (ddd, J = 16.0, 10.0, 6.0 Hz, 1H), 2.26 (ddd, J =
16.0, 10.0, 6.0 Hz, 1H), 2.05 (ddd, J = 14.5, 10.0, 6.0 Hz, 1H),
1.21 (d, J = 6.5 Hz, 3H), 1.20 (d, J = 6.5 Hz, 3H); 13C NMR
(126 MHz, CDCl3) δ 172.3, 138.4, 131.6, 127.7, 121.6, 67.9,
59.1, 55.5, 30.3, 29.9, 21.7, 21.7; IR (neat) 2978, 2933, 1724,
1179, 1106, 1077, 1009, 820 cm−1; HRMS calcd for
C14H18BrO3 [M + H]+ 313.0439, found 313.0451.
Isopropyl 4-(4-Bromophenyl)-5-oxopentanoate (23).
The crude solution of epoxide 22 (HPLC assay 59.5 g 22, 190
mmol) was solvent-switched by distillation under vacuum at
35−40 °C (bath temperature) into toluene to a final volume of
∼450 mL. Anhydrous ZnBr2 (10.7 g, 47.5 mmol) was added in
one portion at 22 °C. The batch warmed from 22 to 30 °C over
30 min (no external cooling was applied). After 1.5 h the slurry
(batch temperature 22 °C) was filtered, and the filtrate was
washed with 10% aq NaCl (1 × 130 mL). For characterization,
a sample of aldehyde 23 was isolated by flash column
chromatography on silica gel (gradient elution from 19:1 to
4:1 n-hexane/EtOAc), affording a yellow oil. HPLC tR 5.3 min;
1H NMR (500 MHz, CDCl3) δ 9.67 (d, J = 1.5 Hz, 1H), 7.53−
7.51 (m, 2H), 7.08−7.05 (m, 2H), 4.99 (sept, J = 6.5 Hz, 1H),
3.61 (ddd, J = 8.0, 6.5, 1.5 Hz, 1H), 2.43−2.36 (m, 1H), 2.30−
2.18 (m, 2H), 2.02−1.95 (m, 1H), 1.23 (d, J = 6.5 Hz, 3H),
1.21 (d, J = 6.5 Hz, 3H); 13C NMR (126 MHz, CDCl3) δ
199.3, 172.2, 134.5, 132.4, 130.6, 122.0, 67.9, 57.4, 31.6, 24.8,
21.8, 21.8.
(S)-5-(4-Bromophenyl)piperidin-2-one (25). Pyridoxal-
5-phosphate (1.51 g, 6.1 mmol) was dissolved in 0.2 M aq
borate buffer (487 mL) containing 1 M aq i-PrNH2 was added
at 20−25 °C. To the resulting solution (pH 10.5) was charged
ATA-302 enzyme (56 g) with gentle agitation. Once the
enzyme had fully wetted, DMSO (81 mL) was added, and the
mixture heated to 45 °C. 923.9 g of a 17.2 wt % bisulfite adduct
solution (1H NMR assay 158.9 g 28, 380.1 mmol) was charged
over 3 h. During the addition, and subsequent reaction, the pH
of the mixture was maintained at 10.5 by addition of 8 M aq i-
PrNH2. The reaction was aged for 44 h before being cooled to
30 °C and extracted with a mixture of 3:2 v/v 2-propanol/IPAc
(1 × 500 mL). The aqueous layer was back-extracted with a
mixture of 3:7 2-propanol/IPAc (1 × 500 mL). The combined
organic layers were washed with sat. aq NaCl (2 × 300 mL).
The resulting organic layer, containing 85 g of piperidinone 28
(by HPLC assay) at 99.3% ee (SFC assay), was filtered, and the
filtrate was concentrated and solvent-switched by distillation
under vacuum at 35−40 °C (bath temperature) into IPAc to a
final volume of ∼200 mL. The resulting slurry was aged at 20−
25 °C overnight, then between −5 and 0 °C for 1.5 h.
Filtration, washing with chilled (∼5 °C) IPAc (2 × 40 mL), and
drying in vacuo at 20−25 °C under a N2 sweep for 16 h
afforded 84 g of 25 (97.5 wt %, 84% corrected yield) as a white
powder. Mp 207−207.5 °C; HPLC tR 3.54 min; [α]25D +51.9 (c
1
1.38, CHCl3); H NMR (400 MHz, CDCl3) δ 7.48−7.45 (m,
2H), 7.14−7.12 (m, 2H), 6.48 (br s, 1H), 3.51−3.46 (m, 1H),
3.34 (t, J = 11.0, 11.0 Hz, 1H), 3.07−2.99 (m, 1H), 2.57−2.44
(m, 2H), 2.12−1.97 (m, 2H); 13C NMR (101 MHz, CDCl3) δ
J
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
°C; HPLC tR 1.2 (p-TsOH), 2.4 (piperidine 29) min; [α]25
The batch was then heated to 63−65 °C and aged for 3 h,
before being allowed to cool to 22 °C overnight. The slurry was
then filtered, washing the cake with EtOAc (1 × 100 mL), and
the combined filtrates were concentrated to afford an orange
solid which was purified by flash chromatography on silica gel
(gradient elution from 3:1 to 1:1 hexanes/EtOAc). Product-
containing fractions were concentrated and the residue slurried
in 3:1 v/v hexanes/EtOAc (60 mL) at 22 °C for 2 h. Filtration
and drying in vacuo at 22 °C under a N2 sweep for 5 h afforded
22.95 g of 34 (80% overall yield) as a white powder. Mp 78−79
D
+5.0 (c 0.24, DMSO); 1H NMR (400 MHz, DMSO-d6) δ 8.84
(br d, J = 10.5 Hz, 1H), 8.51 (br d, J = 10.5 Hz, 1H), 7.55−7.49
(m, 4H), 7.24−7.21 (m, 2H), 7.14 (dd, J = 8.5, 1.0 Hz, 2H),
3.28 (br t, J = 10.0, 10.0 Hz, 2H), 3.04−2.85 (m, 3H), 2.28 (s,
3H), 1.88−1.76 (m, 2H), 1.73−1.62 (m, 2H); 13C NMR (101
MHz, DMSO-d6) δ 145.0, 141.2, 138.3, 131.6, 129.5, 128.3,
125.6, 120.2, 47.9, 43.1, 38.7, 29.2, 22.3, 20.9; IR (neat) 2944,
2816, 1609, 1488, 1222, 1159, 1117, 1006, 678 cm−1; HRMS
calcd for C11H15BrN [M + H]+ 240.0388, found 240.0381.
(S)-tert-Butyl 3-(4-Bromophenyl)piperidine-1-carbox-
ylate (31). Method A: From Tosylate Salt 29·TsOH. NaOH
(1 N aq 72.7 mL, 72.7 mmol) was charged to a slurry of
tosylate salt 29·TsOH (25 g, 60.6 mmol) in MTBE (375 mL)
at 20−25 °C. After 5 min Boc2O (13.4 mL, 57.6 mmol) was
added over 3 min to the biphasic mixture (temperature rose
from 20 to 24 °C during the addition). After aging for a further
4.5 h, the layers were separated. The organic layer was washed
with water (2 × 100 mL), filtered, diluted with DMAc (100
mL), and concentrated by distillation under vacuum at 30 °C
(bath temperature) to give 120 g of a clear solution (HPLC
assay 21.9 g Boc-piperidinone 31, 18.2 wt % solution) that was
used directly in the C−N coupling step. For characterization, a
sample of product 31 was isolated by flash column
chromatography on silica gel (gradient elution from 20:1 to
10:1 n-hexane/MTBE), affording a colorless oil that crystallized
1
°C; HPLC tR 4.0 min; H NMR (500 MHz, CDCl3) δ 7.50−
7.47 (m, 2H), 7.14−7.11 (m, 2H), 4.50−4.42 (m, 2H), 3.77−
3.72 (m, 1H), 2.32−2.25 (m, 1H), 2.08−1.98 (m, 3H); 13C
NMR (126 MHz, CDCl3) δ 171.9, 137.8, 131.8, 130.0, 121.3,
69.1, 46.5, 28.0, 22.0; IR (neat) 2938, 1723, 1487, 1152, 1071,
1010, 809 cm−1; HRMS calcd for C11H12BrO2 [M + H]+
255.0021, found 255.0019.
3-(4-Bromophenyl)tetrahydro-2H-pyran-2-ol (35). A
solution of lactone 34 (20.3 g, 80 mmol) in 1:1 v/v PhMe/
THF (140 mL) was cooled to −18 °C, and a solution of
DIBAL (55.7 mL of a 1 M solution in PhMe, 84 mmol) was
added over 70 min, maintaining the temperature between −18
and −8 °C. The batch was aged at −15 °C for a further 50 min
before being quenched by the addition of 2 M aq HCl (10 mL)
over 5 min (temperature rose from −13 to −5 °C). Additional
2 M aq HCl (120 mL) was then added over 1 min
(temperature rose from −5 to 15 °C). The mixture was aged
at 22 °C for 1 h, then the layers were separated. The organic
layer was washed with water (1 × 80 mL), filtered and
concentrated to dryness. The residue was dissolved in toluene
(21 mL) at 22 °C, and n-heptane (21 mL) was added. Seed
crystals of lactol product 35 (50 mg) were added, and the slurry
aged overnight. n-Heptane (60 mL) was added over 2 h, and
the slurry was aged at 22 °C for 5 h before being cooled to 5 °C
and aged for a further 1 h. Filtration, washing with n-heptane (2
× 30 mL), and drying in vacuo at 22 °C under a N2 sweep for 5
h afforded 18.68 g of 35 (97 wt % purity, 89% corrected yield)
as a white powder and as a ∼2:1 mixture of anomers. Mp 71−
slowly upon standing at 20−25 °C. Mp 43 °C. HPLC tR 6.39
1
min; [α]25 −62.4 (c 0.37, CHCl3); H NMR (400 MHz,
D
CDCl3) δ 7.44 (d, J = 8.5 Hz, 2H), 7.11 (d, J = 8.5 Hz, 2H),
4.15 (t, J = 12.0, 12.0 Hz, 2H), 2.78−2.61 (m, 3H), 2.02−1.99
(m, 1H), 1.78−1.73 (m, 1H), 1.65−1.54 (m, 2H), 1.47 (s, 9H);
13C NMR (101 MHz, CDCl3) δ 154.8, 142.5, 131.5, 128.8,
120.3, 79.6, 50.3, 44.1, 42.0, 31.7, 28.5, 25.3; IR (neat) 2939,
2857, 1685, 1418, 1362, 1136, 820, 765 cm−1; HRMS calcd for
C16H23BrNO2 [M + H]+ 340.0912, found 340.0920.
Method B: From Mesylate 38. To a solution of mesylate 38
(200 mg, 0.46 mmol) in DMF (2 mL) was added powdered
NaOH (31 mg, 0.78 mmol) in one portion at 22 °C. After
aging for 22 h, HPLC assay indicated 146 mg of piperidine
product 31 in solution (94% assay yield). The mixture was not
processed further.
1
72 °C; HPLC tR 3.82 min; H NMR (400 MHz, CDCl3) δ
7.46−7.41 (m, 2H (major anomer) and 2H (minor anomer)),
7.20 (d, J = 8.5 Hz, 2H (minor)), 7.15−7.12 (m, 2H (major)),
5.18 (d, J = 3.0 Hz, 1H (minor)), 4.74 (d, J = 8.0 Hz, 1H
(major)), 4.09−4.02 (m, 1H (major) and 1H (minor)), 3.64−
3.58 (m, 1H (major) and 1H (minor)), 3.31 (br s, 1H
(minor)), 2.94 (dt, J = 12.5, 3.0 Hz, 1H (minor)), 2.61−2.55
(m, 1H (major)), 2.27−2.16 (m, 1H (minor)), 2.02−1.96 (m,
3-(4-Bromophenyl)tetrahydro-2H-pyran-2-one (34). A
solution of acid 32 (24.2 g, 112 mmol) in THF (400 mL) was
cooled to −50 °C and MeLi (72 mL of a 3 M solution in
diethoxymethane, 216 mmol) was added over 20 min,
maintaining the temperature between −50 and −40 °C. The
slurry was then warmed to −5 °C over 1 h, and 1-bromo-3-
chloropropane (12.1 mL, 123 mmol) added over 5 min at −5
to 0 °C. The batch was then aged at 0−5 °C for 1.5 h before
being quenched by the addition of 1 M aq NaOH (160 mL) in
one portion (temperature rose from 5 to 14 °C). The mixture
was concentrated by distillation under vacuum at 35 °C (bath
temperature) to remove most of the THF, and the residue was
partitioned between 1:1 v/v n-heptane/MTBE (100 mL) and
water (50 mL). The layers were separated, and the aqueous
layer was adjusted to pH 1 using 37% aq HCl. The aqueous
layer was extracted with EtOAc (2 × 150 mL), and the
combined organic layers were washed with water (1 × 100
mL), dried over MgSO4, filtered and concentrated to afford
35.33 g of crude acid 33 (89.6 HPLC area% purity) as a
colorless oil. This oil was dissolved in EtOAc (235 mL) at 25
°C and DBU (16.9 mL, 112 mmol) was added in two portions
over 5 min (temperature rose from 25 to 36 °C over 5 min).
1H (major)), 1.84−1.65 (m, 3H (major) and 3H (minor)); 13
C
NMR (101 MHz, CDCl3) δ 140.6 (major), 140.6 (minor),
131.4 (major), 131.2 (minor), 130.0 (minor), 129.6 (major),
125.5 (minor), 120.4 (major), 98.7 (major), 93.8 (minor), 66.0
(major), 59.3 (minor), 48.5 (major), 46.1 (minor), 30.0
(major), 25.5 (minor), 25.0 (major), 22.9 (minor); IR (neat)
3257, 2929, 2856, 1488, 1051, 1010, 980, 808 cm−1; HRMS
calcd for C11H13BrNaO2 [M + Na]+ 278.9997, found 279.0007.
(S)-tert-Butyl (2-(4-Bromophenyl)-5-hydroxypentyl)-
carbamate (37). Pyridoxal-5-phosphate (0.4 g, 1.6 mmol)
was dissolved in 0.2 M aq borate buffer (320 mL) containing 1
M aq i-PrNH2 was added at 22 °C. To the resulting solution
(pH 8.5) was charged ATA-301 enzyme (4 g) with gentle
agitation. Once the enzyme had fully wetted, DMSO (40 mL)
was added, and the mixture was heated to 45 °C. A solution of
lactol 35 (10 g, 38.9 mmol) in DMSO (40 mL) was added over
5 min. The mixture was aged at 45 °C for 69 h, during which
K
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
time the pH was maintained at 8.5 by addition of 4 M aq i-
PrNH2, before being cooled to 25 °C. Solka-Floc (5 g) was
added, and the pH of the mixture adjusted to 2.5 using 6 N aq
HCl. The mixture was aged for 1 h then filtered. The aqueous
filtrate (∼500 mL, containing crude amine 36 at 97.4 HPLC
area% purity) was washed with EtOAc (1 × 400 mL), and then
adjusted to pH 10 using 10 M aq NaOH. Boc2O (13.6 g, 62.2
mmol) was added in one portion at 22 °C, and the mixture
aged for 30 min before being extracted with MTBE (1 × 600
mL). The organic layer was washed with water (1 × 200 mL),
sat aq NaCl (1 × 200 mL), dried over MgSO4, filtered and
concentrated to afford crude alcohol 37 (85 wt % purity, 95
HPLC area% purity) as a yellow oil that was used directly in the
subsequent mesylation step. Data for crude alcohol 37: HPLC
DMF (1 × 250 mL) followed by water (2 × 250 mL), and
drying in vacuo at 20−25 °C under a N2 sweep afforded 55.3 g
of 46 (86% yield) as a beige powder. Mp 158.5−159 °C; HPLC
1
tR 3.47 min; H NMR (400 MHz, CDCl3) δ 13.07 (br s, 1H),
8.17 (s, 1H), 8.03 (br s, 1H), 7.92 (d, J = 7.5 Hz, 1H), 7.91 (d,
J = 7.5 Hz, 1H), 7.15 (t, J = 7.5 Hz, 1H), 1.45 (s, 9H); 13C
NMR (101 MHz, DMSO-d6) δ 165.7, 138.5, 132.8, 125.5,
124.1, 119.7, 118.8, 51.0, 28.7; IR (neat) 3442, 3374, 2960,
1665, 1649, 1513, 1298, 941, 843, 748, 713 cm−1; HRMS calcd
for C12H16N3O [M + H]+ 218.1293, found 218.1292.
(S)-tert-Butyl 3-(4-(7-(tert-Butylcarbamoyl)-2H-inda-
zol-2-yl)phenyl)piperidine-1-carboxylate (56). To 113 g
of crude Boc-piperidinone 31 solution (18.2 wt % in DMAc,
20.6 g 31, 60.6 mmol) was charged indazole 46 (13.8 g, 63.6
mmol) and K2CO3 (25.6 g, 182 mmol) at 20−25 °C.
Additional DMAc (10 mL) was used to wash residual solids
down into the flask. The mixture was degassed by subsurface
sparging with N2 for 1 h. CuBr (0.444 g, 3.03 mmol) and 8-
hydroxyquinoline 52 (0.889 g, 6.06 mmol) were added, and N2
sparging continued for 30 min. The mixture was then heated to
110 °C and aged for 24 h. After cooling to 40 °C, Celite (14.5
g) was added, and the mixture aged for 1 h before being
filtered, washing the cake with DMAc (1 × 100 mL). The
combined filtrates (HPLC assay 28.6 g 56, 99% yield) were
adjusted to 35 °C, then DMAc (50 mL) and 10% aq citric acid
(26 mL) were added, followed by seed crystals (286 mg) of
product 56. After 1 h additional 10% aq citric acid (60 mL) was
added over 100 min. The resulting slurry was aged for 2 h at 35
°C then at 20−25 °C overnight. Filtration, washing with 2:1 v/
v DMAc/water (1 × 150 mL) followed by water (1 × 300 mL),
and drying in vacuo at 20−25 °C under a N2 sweep afforded
27.24 g of 56 (98.3 wt %, 94% corrected yield) as a light-yellow
powder. Mp 192 °C; HPLC tR 6.49 min; [α]25D −62.8 (c 0.25,
DMSO); 1H NMR (400 MHz, MeOH-d4) δ 9.57 (s, 1H), 8.94
(s, 1H), 8.09 (dd, J = 7.0, 1.0 Hz, 1H), 7.96 (d, J = 8.5 Hz, 2H),
7.95 (dd, J = 8.5, 1.0 Hz, 1H), 7.50 (d, J = 8.5 Hz, 2H), 7.23
(dd, J = 8.5, 7.0 Hz, 1H), 4.17−4.09 (br m, 2H), 2.91−2.86 (br
m, 2H), 2.79−2.73 (m, 1H), 2.05 (d, J = 12.0 Hz, 1H), 1.82−
1.72 (m, 2H), 1.64−1.60 (m, 1H), 1.58 (s, 9H), 1.48 (s, 9H);
13C NMR (101 MHz, MeOH-d4) δ 166.6, 156.7, 147.9, 145.7,
139.9, 130.7, 129.8, 126.5, 125.3, 124.0, 123.3, 121.8, 81.3, 52.7,
43.7, 32.9, 29.5, 28.9, 26.6; IR (neat) 3300, 3102, 2933, 2849,
1684, 1649, 1572, 1363, 1172, 833, 749 cm−1; HRMS calcd for
C28H37N4O3 [M + H]+ 477.2866, found 477.2878.
1
tR 4.51 min; H NMR (500 MHz, CDCl3) δ 7.46−7.44 (m,
2H), 7.06 (d, J = 8.5 Hz, 2H), 4.40 (br s, 1H), 3.59−3.58 (br
m, 2H), 3.48−3.46 (br m, 1H), 3.15 (ddd, J = 14.0, 8.5, 5.0 Hz,
1H), 2.76 (br s, 1H), 1.81−1.74 (m, 1H), 1.63−1.56 (m, 1H),
1.53−1.42 (m, 3H), 1.40 (s, 9H); 13C NMR (126 MHz,
CDCl3) δ 155.9, 141.6, 131.8, 129.6, 120.5, 79.4, 62.6, 46.1,
45.5, 30.3, 29.5, 28.3.
(S)-4-(4-Bromophenyl)-5-((tert-butoxycarbonyl)-
amino)pentyl Methanesulfonate (38). The crude alcohol
37, prepared as described above, was dissolved in CH2Cl2 (120
mL) and cooled to 5 °C. Et3N (10.8 mL, 78 mmol) was added
in one portion, followed by the addition of MsCl (3.6 mL, 46.7
mmol) over 3 min. The slurry was aged at 5 °C for 30 min
before being poured into 0.8 M aq HCl (70 mL) at 22 °C. The
layers were separated, and the organic layer was washed with
1:1 sat aq NaCl/sat aq NaHCO3 (1 × 70 mL), dried over
MgSO4, filtered and concentrated. The residue was dissolved in
3:2 hexanes/EtOAc (25 mL) at 22 °C, which was followed by
rapid crystallization of the product. Additional EtOAc (10 mL)
was added, and the slurry aged overnight. Hexanes (25 mL)
was added over 75 min, and the slurry aged for a further 4 h.
Filtration, washing with 9:1 hexanes/EtOAc (1 × 20 mL)
followed by hexanes (1 × 20 mL), and drying in vacuo at 22 °C
under a N2 sweep overnight afforded 8.2 g of 38 (99.4 wt %
purity, 48% overall yield from lactol 35) as a cream powder. Mp
1
98 °C; HPLC tR 5.07 min; [α]25 − 36.6 (c 1.07, CHCl3); H
D
NMR (400 MHz, CDCl3) δ 7.47−7.44 (m, 2H), 7.06−7.03 (m,
2H), 4.44 (br s, 1H), 4.15 (t, J = 6.0 Hz, 2H), 3.44 (br s, 1H),
3.16−3.11 (m, 1H), 2.97 (s, 3H), 2.76 (br s, 1H), 1.83−1.76
(m, 1H), 1.68−1.56 (m, 3H), 1.40 (s, 9H); 13C NMR (101
MHz, CDCl3) δ 155.8, 140.9, 131.9, 129.5, 120.7, 79.4, 69.5,
46.0, 45.2, 37.3, 29.0, 28.3, 26.9; IR (neat) 3387, 2983, 2931,
1677, 1517, 1347, 1168, 918, 775 cm−1; HRMS calcd for
C17H27BrNO5S [M + H]+ 436.0793, found 436.0802.
Niraparib p-Toluenesulfonate Monohydrate (9).3 All
operations in this step were performed in a dedicated potent
compound-handling facility.
To a slurry of amide 56 (20.0 g, 98.3 wt %, 41.2 mmol) in o-
xylene (40 mL) was charged MSA (60 mL, 924 mmol) at 22 °C
(GAS EVOLVED; EXOTHERMICwith no external cooling
the batch temperature rose from 22 to 36 °C over 5 min). After
5 min the biphasic mixture was warmed to 40 °C, aged for 2.5
h, then cooled to below 20 °C. Water (140 mL) was added
over 23 min, maintaining the temperature below 25 °C. The
layers were then separated, and the aqueous layer was washed
with PhMe (1 × 20 mL) and filtered. A solution of p-TsOH·
H2O (11.8 g, 61.9 mmol) in water (23.5 mL) was prepared, and
6 mL added over 20 min to the aqueous filtrate at 20 °C. Seed
crystals of product salt 9 (400 mg) were introduced, and the
resulting slurry aged for 30 min. The remainder of the aq p-
TsOH solution was then added over 75 min, and the mixture
aged overnight. Filtration, washing with water (3 × 100 mL),
and drying in vacuo at 20−25 °C under a N2 sweep overnight
N-(tert-Butyl)-1H-indazole-7-carboxamide (46). CDI
(59.1 g, 97 wt %, 354 mmol) was added to a solution of
indazole 41 (50.3 g, 95 wt %, 295 mmol) in DMF (150 mL) at
20−25 °C (GAS EVOLVED). After 30 min, moderate vacuum
(50−100 mmHg) was briefly applied to the reaction mixture to
ensure complete removal of the CO2 byproduct. After a further
25 min, t-BuNH2 (62.5 mL, 589 mmol) was added over 3 min
(EXOTHERMIC). With no external cooling, the batch
temperature rose from 21 to 38 °C over 40 min. The mixture
was then aged at 40 °C for a further 2.5 h. After cooling to 20
°C, water (100 mL) was added, maintaining the temperature
below 25 °C. The solution was then seeded with crystalline
product 46 (150 mg), and the slurry was aged at 20−25 °C for
30 min. Water (500 mL) was then added over 80 min, and the
slurry aged overnight. Filtration, washing with 9:1 v/v water/
L
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(9) (a) Ohkuma, T.; Li, J.; Noyori, R. Synlett 2004, 1383−1386.
(b) Chung, J. Y. L.; Mancheno, D.; Dormer, P. G.; Variankaval, N.;
Ball, R. G.; Tsou, N. N. Org. Lett. 2008, 10, 3037−3040. (c) Xu, F.;
Chung, J. Y. L.; Moore, J. C.; Liu, Z.; Yoshikawa, N.; Hoerrner, R. S.;
Lee, J.; Royzen, M.; Cleator, E.; Gibson, A. G.; Dunn, R.; Maloney, K.
M.; Alam, M.; Goodyear, A.; Lynch, J.; Yasuda, N.; Devine, P. N. Org.
Lett. 2013, 15, 1342−1345.
afforded 20.6 g of 9 (99.8 HPLC area % purity, 97.3 wt %, 95%
corrected yield) as white powder.
ASSOCIATED CONTENT
■
S
* Supporting Information
Procedure for the preparation of solid bisulfite adduct 28, chiral
SFC analysis of piperidinone 25, chiral HPLC analysis of
aminoalcohol 36, and copies of 1H and 13C NMR spectra for all
new compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(10) Koszelewski, D.; Clay, D.; Faber, K.; Kroutil, W. J. Mol. Catal. B:
Enzym. 2009, 60, 191−194.
(11) The asymmetric reductive amination of aldehydes via DKR
using small-molecule organocatalysis had previously been reported by
List and co-workers: Hoffmann, S.; Nicoletti, M.; List, B. J. Am. Chem.
Soc. 2006, 128, 13074−13075.
(12) Seed, A. J.; Sonpatki, V.; Herbert, M. R. Org. Synth. 2002, 79,
204−206.
(13) (a) Rubin, A. E.; Tummala, S.; Both, D. A.; Wang, C.; Delaney,
E. J. Chem. Rev. 2006, 106, 2794−2810. (b) Shultz, C. S.; Krska, S. W.
Acc. Chem. Res. 2007, 40, 1320−1326. (c) Belyk, K. M.; Xiang, B.;
Bulger, P. G.; Leonard, W. R., Jr.; Balsells, J.; Yin, J.; Chen, C. Org.
Process Res. Dev. 2010, 14, 692−700.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: cheol_chung@merck.com.
*E-mail: paul_bulger@merck.com
Notes
The authors declare no competing financial interest.
(14) The use of ZnBr2 in PhMe to promote epoxide rearrangement
has been reported previously. For an example, see: Ramachandran, P.
V.; Krzeminski, M. P. Tetrahedron Lett. 1999, 40, 7879−7881.
(15) The y-axis values refer to the HPLC area % ratios of desired
product to an internal reference standard, and are a qualitative measure
of relative reaction performance.
ACKNOWLEDGMENTS
■
We thank Dr. Tom J. Novak for assistance with HRMS analysis,
Zainab Pirzada and Tanja Brkovic for analytical support, and
Cheng-yi Chen, Pintipa Grongsaard, Phil Jones and Debra J.
Wallace for helpful discussions.
(16) Available from Codexis, Inc., 200 Penobscot Drive, Redwood
City, CA 94063, United States.
(17) Using an alternative procedure (see the Supporting
Information), solid bisulfite adduct 28 could be isolated by
crystallization; however, the more streamlined through-process was
preferred.
(18) (a) Guercio, G.; Manzo, A. M.; Goodyear, M.; Bacchi, S.; Curti,
S.; Provera, S. Org. Process Res. Dev. 2009, 13, 489−493. (b) Atkins, W.
J., Jr.; Burkhardt, E. R.; Matos, K. Org. Process Res. Dev. 2006, 10,
1292−1295.
(19) For a discussion of potential hazards associated with the use of
BH3·THF on scale, see: am Ende, D. J.; Vogt, P. F. Org. Process Res.
Dev. 2003, 7, 1029−1033.
(20) For experimental details, see the Supporting Information. Note
that different vessel configurations and headspace nitrogen sweep
conditions may give different concentrations of diborane exiting the
reactor.
(21) Rosen, J. D.; Nelson, T. D.; Huffman, M. A.; McNamara, J. M.
Tetrahedron Lett. 2003, 44, 365−368.
(22) (a) Pu, Y.-M.; Ku, Y.-Y.; Grieme, T.; Henry, R.; Bhatia, A. V.
Tetrahedron Lett. 2006, 47, 149−153. (b) Pu, Y.-M.; Ku, Y.-Y.; Grieme,
T.; Black, L. A.; Bhatia, A. V.; Cowart, M. Org. Process Res. Dev. 2007,
11, 1004−1009.
(23) Burgos, C. H.; Barder, T. E.; Huang, X.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2006, 45, 4321−4326.
(24) See: Wang, Z.; Richter, S. M.; Gates, B. D.; Grieme, T. A. Org.
Process Res. Dev. 2012, 16, 1994−2000 and references therein.
REFERENCES
(1) Jagtap, P.; Szabo,
■
́
C. Nat. Rev. Drug Discovery 2005, 4, 421−440.
(2) (a) Kummar, S.; Chen, A.; Parchment, R. E.; Kinders, R. J.; Ji, J.;
Tomaszewski, J. E.; Doroshow, J. H. BMC Medicine 2012, 10, 25−29.
(b) Park, S. R.; Chen, A. Hematol. Oncol. Clin. N. Am. 2012, 26, 649−
670. (c) Smith, K. L.; Isaacs, C. Curr. Breast Cancer Rep. 2011, 3, 165−
171.
(3) Wallace, D. J.; Baxter, C. A.; Brands, K. J. M.; Bremeyer, N.;
Brewer, S. E.; Desmond, R.; Emerson, K. M.; Foley, J.; Fernandez, P.;
Hu, W.; Keen, S. P.; Mullens, P.; Muzzio, D.; Sanjoz, P.; Tan, L.;
Wilson, R. D.; Zhou, G.; Zhou, G. Org. Process Res. Dev. 2011, 15,
831−840.
(4) Jones, P.; Altamura, S.; Boueres, J.; Ferrigno, F.; Fonsi, M.;
Giomini, C.; Lamartina, S.; Monteagudo, E.; Ontoria, J. M.; Orsale, M.
V.; Palumbi, M. C.; Pesci, S.; Roscilli, G.; Scarpelli, R.; Schultz-
Fademrecht, C.; Toniatti, C.; Rowley, M. J. Med. Chem. 2009, 52,
7170−7185.
(5) Antilla, J. C.; Baskin, J. M.; Barder, T. E.; Buchwald, S. L. J. Org.
Chem. 2004, 69, 5578−5587.
(6) (a) Joubert, N.; Basle, E.; Vaultier, M.; Pucheault, M. Tetrahedron
́
Lett. 2010, 51, 2994−2997. (b) Hari, Y.; Shoji, Y.; Aoyama, T.
Tetrahedron Lett. 2005, 46, 3771−3774. (c) Lam, P. Y. S.; Vincent, G.;
Clark, C. G.; Deudon, S.; Jadhav, P. K. Tetrahedron Lett. 2001, 42,
3415−3418. (d) Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.;
Winters, M. P.; Chan, D. M. T.; Combs, A. Tetrahedron Lett. 1998, 39,
2941−2944.
(7) Selected examples: (a) Chiral auxiliary-mediated DKR: Amat,
́
M.; Canto, M.; Llor, N.; Escolano, C.; Molins, E.; Espinosa, E.; Bosch,
J. J. Org. Chem. 2002, 67, 5343−5351. (b) Alkene hydrogenation:
Verendel, J. J.; Zhou, T.; Li, J.-Q.; Paptchikhine, A.; Lebedev, O.;
Andersson, P. G. J. Am. Chem. Soc. 2010, 132, 8880−8881. (c)
́
Asymmetric alkylation: Michon, C.; Bethegnies, A.; Capet, F.; Roussel,
P.; de Filippis, A.; Gomez-Pardo, D.; Cossy, J.; Agbossou-Niedercorn,
F. Eur. J. Org. Chem. 2013, 4979−4985.
(8) For reviews and selected applications, see: (a) Savile, C. K.; Janey,
J. M.; Mundorff, E. C.; Moore, J. C.; Tam, S.; Jarvis, W. R.; Colbeck, J.
C.; Krebber, A.; Fleitz, F. J.; Brands, J.; Devine, P. N.; Huisman, G. W.;
Hughes, G. J. Science 2010, 329, 305−309. (b) Mutti, F. G.; Kroutil,
W. Adv. Synth. Catal. 2012, 354, 3409−3413. (c) Berglund, P.;
Humble, M. S.; Branneby, C. Comprehensive Chirality; Carreria, E. M.,
Yamamoto, H., Eds.; Elsevier; Amsterdam, 2012; Vol. 7, pp 390−401.
M
dx.doi.org/10.1021/op400233z | Org. Process Res. Dev. XXXX, XXX, XXX−XXX