A Practical and Scalable Synthesis of a Glucokinase Activator via Diastereomeric Resolution and Palladium-Catalyzed C-N Coupling Reaction

Article abstract of DOI:10.1021/acs.oprd.6b00415

Here we describe the research and development of a process for the practical synthesis of glucokinase activator (R)-1 as a potential drug for treating type-2 diabetes. The key intermediate, chiral α-arylpropionic acid (R)-2, was synthesized in high diastereomeric excess through the diasteromeric resolution of 7 without the need for a chiral resolving agent. The counterpart 2-aminopyrazine derivative 3 was synthesized using a palladium-catalyzed C-N coupling reaction. This efficient process was demonstrated at the pilot scale and yielded 19.0 kg of (R)-1. Moreover, an epimerization process to obtain (R)-7 from the undesired (S)-7 was developed.

Full text of DOI:10.1021/acs.oprd.6b00415

Article
pubs.acs.org/OPRD
A Practical and Scalable Synthesis of a Glucokinase Activator via
Diastereomeric Resolution and Palladium-Catalyzed C−N Coupling
Reaction
Yohei Yamashita,*,†,§ Yasuhiro Morinaga,† Makoto Kasai,† Takao Hashimoto,† Yuji Takahama,†
Atsushi Ohigashi,† Satoshi Yonishi,‡ and Motohiro Akazome§
†Process Chemistry Labs., Astellas Pharma Inc., 160-2 Akahama, Takahagi-shi, Ibaraki 318-0001, Japan
‡Astellas Research Technologies Co., Ltd., 21 Miyukigaoka, Tsukuba-shi, Ibaraki 305-8585, Japan
§Department of Applied Chemistry and Biotechnology, Graduate School of Engineering, Chiba University, 1-33 Yayoicho, Inageku,
Chiba 263-8522, Japan
ABSTRACT: Here we describe the research and development of a process for the practical synthesis of glucokinase activator
(R)-1 as a potential drug for treating type-2 diabetes. The key intermediate, chiral α-arylpropionic acid (R)-2, was synthesized in
high diastereomeric excess through the diasteromeric resolution of 7 without the need for a chiral resolving agent. The
counterpart 2-aminopyrazine derivative 3 was synthesized using a palladium-catalyzed C−N coupling reaction. This efficient
process was demonstrated at the pilot scale and yielded 19.0 kg of (R)-1. Moreover, an epimerization process to obtain (R)-7
from the undesired (S)-7 was developed.
KEYWORDS: glucokinase activator, diastereomeric resolution, epimerization, palladium-catalyzed C−N coupling
selective ring-opening iodination from 4 to 5 was conducted
using iodotrimethylsilane (TMSI) according to the published
method.6 Alkylation at the α-position of carboxylic acid ester 6
with 5 followed by Suzuki coupling and hydrolysis of the ester
yielded 7 as a 1:1 diastereomeric mixture. To obtain the desired
diastereomer, (R)-4-benzyl-2-oxazolidinone7 8 was introduced,
and removal of the ketal group followed by silica gel
chromatography yielded diastereomerically pure (R)-9. The
carbonyl group of (R)-9 was reprotected, and the oxazolidinone
and ketal were removed to yield (R)-2. Subsequent coupling
with 3 and desilylation yielded the final product (R)-1. Pyrazine
3 was prepared from pyrazinecarbonitrile-N-oxide8 10 via
reductive decyanation and concurrent N-oxide reduction to
give 12. Deacetylation followed by O-silylation gave 3.
INTRODUCTION
■
Glucokinase (GK) is one of four members of the hexokinase
family that catalyze the phosphorylation of glucose to generate
glucose-6-phosphate.1 Efforts to activate GK for pharmaceutical
purposes led to the discovery of small-molecule activators that
are potential drugs for treating type-2 diabetes. Therefore,
numerous pharmaceutical companies conducted extensive
investigations and developed various GK activators that were
tested in clinical trials.2 Among these, (2R)-2-(4-cyclo-
propanesulfonyl-3-cyclopropylphenyl)-N-[5-(hydroxymethyl)-
pyrazin-2-yl]-3-[(R)-3-oxocyclopentyl]propanamide3 ((R)-1)
is a GK activator that requires a readily available and abundant
supply of the active pharmaceutical ingredient (API) for use in
nonclinical studies and clinical trials.
Although suitable for gram-scale preparation of (R)-1, this
route posed several synthetic challenges for scale-up, such as
the use of highly corrosive and unstable TMSI, introduction of
8 as an auxiliary group, emission of hydrogen cyanide (HCN)
during hydrogenation, and the requirement of several
chromatographic purifications. To meet the demand for a
readily available supply of the API, we conducted a process
study to develop a new synthetic approach suitable for pilot-
scale synthesis. In this paper, we describe the development of a
practical manufacturing method using a diastereomeric
resolution and a newly developed C−N coupling, which
provided 19.0 kg of (R)-1 as the first good manufacturing
practice (GMP) lot. In addition, the development of an
epimerization process to obtain (R)-7 from the undesired (S)-7
is also discussed.
The retrosynthetic approach to (R)-1 yields chiral α-
arylpropionic acid (R)-2 and 2-aminopyrazine derivative 3 as
key intermediates through the cleavage of amide bond (Scheme
1). This convergent strategy relies on stereocontrol at the α-
position of the carboxylic acid in (R)-2. The medicinal
chemistry synthesis of (R)-1 involved 15 steps and nine
chromatographic purifications (Scheme 2).3 Chiral bicyclic
ketal 4 was prepared from the corresponding ketone reported
in the literature4 and 1,2-diphenylethane-1,2-diol.5 Stereo-
Scheme 1. Structure and Retrosynthesis of (R)-1
Received: December 9, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a b
,
Scheme 2. Medicinal Chemistry Synthesis of (R)-1
a
Reagents and conditions: (a) TMSI, K2CO3, CH2Cl2, below −10 °C, SiO2 column chromatography, 87%; (b) compound 5, n-BuLi, i-Pr2NH, THF,
N,N′-dimethylpropyleneurea, below −60 °C to rt, SiO2 column chromatography, 72%; (c) cyclopropylboronic acid, Pd(PPh3)4, K3PO4, toluene,
H2O, reflux, SiO2 column chromatography, 89%; (d) NaOH(aq), THF, MeOH, reflux, 100%; (e) pivaloyl chloride, Et3N, THF, rt, then compound
8, n-BuLi, THF, −50 °C to rt; (f) HCl(aq), acetone, reflux, SiO2 column chromatography, 33% for two steps; (g) 2,2-dimethyl-1,3-propanediol,
pyridinium p-toluenesulfonate, toluene, reflux; (h) H2O2(aq), LiOH, THF, 0 °C; (i) HCl(aq), acetone, reflux, 100% for three steps; (j) (COCl)2,
DMF, CH2Cl2, 0 °C, then compound 3, pyridine, CH2Cl2, 0 °C, SiO2 column chromatography, 50%; (k) HCl(aq), THF, rt, SiO2 column
chromatography, 82%; (l) AcOK, 18-crown-6, CH3CN, rt, SiO2 column chromatography, 57%; (m) H2 (0.3 MPa), Pt/C, AcOH, MeOH, rt, SiO2
column chromatography, 88%; (n) potassium carbonate, MeOH, reflux; (o) tert-butylchlorodimethylsilane, imidazole, DMF, rt, SiO2 column
b
chromatography, 75% for two steps. The compound in parentheses was not isolated.
use of lithium bromide instead of NaI resulted in no reaction
(Table 1, entry 4). The reaction could be successfully run at
room temperature using 2 equiv of TMSCl and 2 equiv of NaI
in CH3CN (Table 1, entry 5). Optimization of the amount of
TMSCl revealed that 1.8 equiv was optimal (Table 1, entries
5−7). Eventually, 0.1 equiv of water was added to the reaction
mixture to increase the solubility of NaI for industrial use
without detriment (Table 1, entry 8). In the pilot-scale
synthesis, ring-opening iodination was carried out four times
on a 26.5 kg scale, and 5 was obtained with a purity of 88−89
area %. A THF solution of 5 was used for the next reaction
without purification.
RESULTS AND DISCUSSION
■
Ring-Opening Iodination of 4. When we tried ring-
opening iodination of 4 to give 5 using the original method as
the first step, the yield was moderate because of the generation
of impurities9 (Table 1, entry 1). In order to improve the
selectivity, we initiated an optimization of the reaction
conditions. Performing the reaction at a lower temperature
using sodium hydrogen carbonate led to an improved yield,
despite the use of 2.2 equiv of TMSI (Table 1, entry 2).
Chlorotrimethylsilane (TMSCl) together with sodium iodide
(NaI), which is widely used as a substitute for TMSI,10
achieved an acceptable result (Table 1, entry 3). In contrast, the
B
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Table 1. Ring-Opening Iodination of 4
a
b
entry
scale (g)
reagent (equiv)
TMSI (1.5)
base (equiv)
solvent
CH2Cl2
temp. (°C)
5
(HPLC area % )
1
2
3
4
5
6
7
8
0.5
0.2
2.0
0.2
3.0
3.0
3.0
K2CO3 (2.5)
NaHCO3 (3.7)
NaHCO3 (7)
NaHCO3 (3.7)
NaHCO3 (6)
NaHCO3 (6)
NaHCO3 (6)
NaHCO3 (6)
−17
−42
−42
−44
23
60
84
88
TMSI (2.2)
CH2Cl2
TMSCl (4), NaI (3.2)
TMSCl (2.2), LiBr (3.2)
TMSCl (2), NaI (2)
TMSCl (1.8), NaI (2)
TMSCl (1.5), NaI (2)
TMSCl (1.8), NaI (2), H2O (0.1)
CH2Cl2/CH3CN
CH2Cl2/CH3CN
CH3CN
c
ND
83
88
84
88
CH3CN
23
CH3CN
23
3.0
CH3CN
24
a
b
c
Reaction mixture. Measured by HPLC condition A (see the Experimental Section). Not detected.
Scheme 3. Synthetic Strategy for (R)-2
a b
,
Synthesis of Chiral α-Arylpropionic Acid (R)-2. Our
synthetic strategy to synthesize (R)-2 from 7 involving
diastereomeric resolution of the α-arylpropionic acid by salt
formation with a chiral amine is described in Scheme 3.
Although several articles had reported the enantioselective
synthesis of α-arylpropionic acid derivatives through asym-
metric hydrogenation11 and ketene desymmetrization12 using
(R)-pantolactone,13 we gave priority to the development of a
scalable process promptly and chose diastereomeric resolution
for the first pilot-scale synthesis. This route is reliable in the
sense that the diastereomeric excess (de) of (R)-7 can be
improved through repeated recrystallization. Salt formation of 7
with (R)-(+)-1-phenylethylamine was initially attempted,
leading to the production of the (R)-7 salt with a little-
improved de of 30% (Table 2, entry 1). When (S)-(−)-1-
phenylethylamine was used instead, the opposite diastereomer
(S)-7 was obtained with a low de (Table 2, entry 2). Although
several chiral amines were tried with an expectation of
improved yield or de, no solids could be obtained (Table 2,
entries 3−7). Unexpectedly, the desired diastereomer (R)-7
was crystallized in the absence of chiral amine (Table 2, entry
8), suggesting that (R)-7 is less soluble than the diastereomer
(S)-7 and that pure (R)-7 might thus be obtained through
repeated crystallizations. This protocol should be more effective
than the general resolution process requiring salt formation and
subsequent neutralization.
Table 2. Optical Resolution of 7 with Chiral Amine
de of 7
(%)
c
entry
amine (0.5 equiv)
yield (%)
1
2
3
4
5
6
7
8
(R)-(+)-1-phenylethylamine
(S)-(−)-1-phenylethylamine
(R)-(+)-1-(1-naphthyl)ethylamine
cinchonine
53
43
30 (R)
5 (S)
not crystallized
not crystallized
not crystallized
not crystallized
not crystallized
59
−
−
−
−
−
29 (R)
brucine (anhydrous)
(+)-dehydroabietylamine
(1S,2S)-(+)-1,2-diaminocyclohexane
none
a
b
c
50 mg scale. At room temperature. Measured by HPLC condition
B (see the Experimental Section).
lization. We assumed that the bulky cyclic ketal moiety, which
was the only structural difference, would significantly affect the
solubility as well as the ability to crystallize. On this basis, we
therefore concluded that crystallization of 7 was the most
feasible way to improve the de. We then proceeded to the
resolution of 7 in the most promising solvents: toluene and
EtOH (Table 3). Whereas toluene did not significantly improve
the diastereomeric ratio (Table 3, entry 1), two recrystalliza-
tions with EtOH increased the de to 99% (Table 3, entry 2). In
the pilot-scale synthesis, 90%−91% de of (R)-7 was obtained
by crystallization from a solution of mixed diastereomers, after
We next investigated the solubilities of (R)-7 and (S)-7 in
various solvents to identify the best solvent for crystallization
(Figure 1). As anticipated, (R)-7 was less soluble than (S)-7 in
most solvents at 25 °C, and there were large differences in
solubilities, particularly in EtOH, i-PrOH, and toluene. On the
other hand, interestingly, unprotected (R)-2 showed an
opposite trend of solubility, which suggested that diastereo-
merically enriched (S)-2 could be obtained through crystal-
C
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
epimerization and recovery of (S)-7. First, we screened bases to
deprotonate the α-position of the carboxylic acid using readily
available (R)-7 (Table 4). The conditions using sodium
methoxide and sodium ethoxide did not lead to sufficient
epimerization even at 60 °C (Table 4, entries 1 and 2).
Although epimerization was observed using potassium tert-
butoxide, concomitant decomposition occurred (Table 4, entry
3). Sufficient epimerization was observed using sodium hydride
at 60 °C, and a diastereomeric mixture was obtained (Table 4,
entry 4). Partial epimerization was observed when a large excess
of lithium amide was used (Table 4, entry 5), and neither
lithium bis(trimethylsilyl)amide nor lithium diisopropylamide
gave good results (Table 4, entries 6 and 7). We next attempted
epimerization through ketene formation. When (R)-7 was
treated with p-toluenesulfonyl chloride (p-TsCl) together with
1-methylimidazole, epimerization proceeded completely in
CH2Cl2 or CH3CN with slight degradation (Table 4, entries
8 and 9). Although the combination of p-TsCl and 1-
methylimidazole is widely used for mild esterification and
amide formation14 without racemization, (R)-7 substituted with
an electron-withdrawing sulfonyl group was easily activated at
25 °C to generate the ketene under basic conditions.
On the basis of these results, we performed an experiment to
epimerize (S)-7 in the filtrate from the pilot-scale synthesis
(Scheme 5). After solvent exchange from EtOH to CH3CN, 1-
methylimidazole and p-TsCl were added to the solution of (S)-
7 (63% de) followed by stirring at 25 °C for 2 h. After complete
epimerization was observed, crystallization in EtOH yielded
(R)-7 with 71% de. An increase in the de to 99% was achieved
through two recrystallizations with EtOH. The yield was 36%
based on the sum of (R)-7 and (S)-7 in the filtrate. We
concluded that this epimerization process was acceptable for
application to the next pilot-scale synthesis.
Figure 1. Solubilities of chiral carboxylic acids in various solvents (g/
L) at 25 °C as measured by HPLC conditions B and C (see the
Experimental Section).
ab
,
Table 3. Solvent Effect on Recrystallization of (R)-7
New Approach to 2-Aminopyrazine Derivative 3. We
required a new synthetic approach for 3 to avoid emitting
HCN. Our synthetic strategy from commercially available 18
involves the key step of a catalytic C−N coupling reaction
(Scheme 6). Successful conversion from 15 to 14 depended on
an appropriate nucleophilic amine as a nitrogen source. In
practice, intermediate 15 was obtained from 18 in 65% yield
through hydrolysis, reduction to the primary alcohol, and O-
silylation (Scheme 7). A subsequent direct amination15 of 15 to
3 catalyzed by palladium was unsuccessful (Table 5, entry 1).
When a combination of aqueous ammonia and a copper
catalyst16 was used, desilylation predominantly occurred (Table
5, entry 2). The coupling reaction of 15 with lithium
bis(trimethylsilyl)amide17 resulted in a complex mixture
(Table 5, entry 3). The palladium-catalyzed C−N coupling
reaction with diallylamine generated 14b in 45% yield, and
subsequent deallylation using published methods with tetrakis-
(triphenylphosphine)palladium(0)18 and chlorotris-
(triphenylphosphine)rhodium(I)19 afforded amine 3 in 76%
and 78% HPLC yield, respectively (Table 5, entry 4). The yield
of the C−N coupling reaction was substantially improved by
employing benzophenone imine20 as a coupling partner (Table
5, entry 5), and following imine cleavage by hydroxylamine
hydrochloride, 3 was generated in 73% yield. Eventually, we
adopted this approach for the pilot-scale synthesis.21 We next
needed to purify 3 from the reaction mixture containing
impurities such as benzophenone and benzophenone oxime
that were generated during deprotection. Because 3 was oily,
we screened various acids for salt formation and concluded that
fumaric acid salts were the best from the aspect of stability.
c
entry
1
time
solvent
de of 7 (%)
1st
toluene
toluene
toluene
EtOH
89 (R)
93 (R)
95 (R)
96 (R)
99 (R)
2nd
3rd
1st
2
2nd
EtOH
a
b
1.0 g scale. Wet crystals of (R)-7 obtained by optical resolution were
used for recrystallization. Measured by HPLC condition B (see the
c
Experimental Section).
which recrystallization with EtOH afforded (R)-7 with 99% de
(Scheme 4). Subsequent ketal deprotection using hydrochloric
acid in acetone proceeded without a decrease in the de, and 1,2-
diphenylethane-1,2-diol was separated from (R)-2 by phase
separation. Eventually, 27.5 kg of (R)-2 was obtained in 95%
yield based on (R)-7 with 99% de.
Recovery of the Undesired (S)-7. Although we developed
a practical and scalable process toward (R)-7, we did not reuse
the undesired (S)-7 in the filtrate in the first pilot-scale
synthesis. Aiming at yield improvement, we initiated a study of
D
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a b
,
Scheme 4. Scale-Up Route to (R)-2
a
Reagents and conditions: (a) TMSCl, NaI, NaHCO3, CH3CN, H2O, 23 °C; (b) compound 5, t-BuOK, THF, N,N′-dimethylpropyleneurea, −9 °C;
(c) cyclopropylboronic acid, Pd(OAc)2, PPh3, K3PO4(aq), toluene, 87 °C; (d) NaOH(aq), THF, MeOH, 47 °C, 30% for three steps; (e)
b
recrystallization with EtOH, 84%; (f) HCl(aq), acetone, 50 °C, 95%. Compounds in parentheses were not isolated.
a
Table 4. Epimerization Experiments Using (R)-7
b
7
c
d
entry
reagent (equiv)
MeONa (5)
solvent
temp. (°C)
HPLC area %
de (%)
1
2
3
4
5
6
7
8
9
MeOH/THF
EtOH/THF
THF
60
60
25
60
60
40
40
25
25
98
94
37
89
97
19
63
91
91
88 (R)
57 (R)
2 (R)
EtONa (5)
t-BuOK (5)
NaH (2.2)
DMF/THF
THF
1 (R)
LiNH2 (10)
39 (R)
89 (R)
89 (R)
1 (R)
LHMDS (5)
THF
LDA (5)
THF
p-TsCl (1.2), 1-methylimidazole (3)
CH2Cl2
CH3CN
p-TsCl (1.2), 1-methylimidazole (3)
2 (R)
a
b
c
d
0.2 g scale. Reaction mixture. Measured by HPLC condition A (see the Experimental Section). Measured by HPLC condition B (see the
Experimental Section).
During the research, 16 could be purchased as a
commercially available compound, so 16 was employed as a
raw material in the pilot-scale synthesis (Scheme 8). Through
O-silylation, C−N coupling reaction, and imine cleavage, a
THF/toluene solution of 3 was obtained. The solution was
added to a THF solution of fumaric acid to generate crude 3
fumarate, and most of the impurities were present in the filtrate.
After suspension of the crude product with AcOi-Pr, 61.1 kg of
3 fumarate was obtained with a purity of 99 area %. The yield
was 71% based on 16, which was significantly improved
compared with that of the medicinal chemistry synthesis (38%).
Preparation of (R)-1. The remaining steps required to
synthesize (R)-1 were implemented mainly according to the
medicinal chemistry synthesis (Scheme 8). 3 fumarate was
E
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a b
,
Scheme 5. Recovery of the Undesired (S)-7
a
b
Reagents and conditions: (a) p-TsCl, 1-methylimidazole, CH3CN, 25 °C, 50%; (b) recrystallization twice with EtOH, 72%. Compounds in
parentheses were not isolated.
Scheme 6. Synthetic Strategy for 3
a
Scheme 7. Synthesis of 15
a
Reagents and conditions: (a) NaOH(aq), rt, 91%; (b) (COCl)2, DMF, 1,4-dioxane, rt, then NaBH4, 1,4-dioxane, H2O, 5 °C, 75%; (c) tert-
butylchlorodimethylsilane, imidazole, DMF, rt, SiO2 column chromatography, 95%.
neutralized with aqueous KHCO3 before acylation with (R)-2.
In the desilylation step, phase separation (n-heptane/CH3CN)
was introduced to remove silanol, siloxane, or both, which
might inhibit the crystallization of (R)-1. Through clarifying
filtration and recrystallization with aqueous EtOH, 19.0 kg of
(R)-1 was finally obtained in 79% yield based on (R)-2 with
99% de.
was obtained through resolution of the ketal derivative 7, and
recovery of the undesired (S)-7 from a filtrate was achieved
using mild conditions. 2-Aminopyrazine derivative 3 was
synthesized using a new synthetic route involving a
palladium-catalyzed C−N coupling reaction with benzophe-
none imine. This approach will likely be applicable for the
synthesis of other 2-aminopyrazine derivatives.
The advantages of the process developed here compared
with the medicinal chemistry synthesis are as follows: the
overall yield to (R)-1 based on 6 increased from 9% to 19%, the
number of chemical transformations has been reduced, the
emission of HCN was eliminated, several chromatographic
purifications were avoided, and commercially available 16 was
CONCLUSION
■
We achieved our goal of developing a practical and scalable
method to produce a glucokinase activator. For this purpose,
we developed an improved process of ring-opening iodination
to give 5. Diastereomerically pure α-arylpropionic acid (R)-2
F
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Table 5. C−N Coupling Reaction
step 1 (C−N coupling reaction)
conditions (equiv)
step 2 (deprotection)
conditions (equiv)
yield
(%)
entry
1
yield (%)
b
0.5 M NH3 in 1,4-dioxane (5), Pd2(dba)3 (0.01), t-BuDavePhos (0.05), t-BuONa 3: 7
(1.4), 1,4-dioxane, 80 °C
−
−
2
3
4
28 wt % NH3(aq) (10), Cu2O (0.05), NMP, 80 °C
LHMDS (1.1), Pd(dba)2 (0.01), t-Bu3P (0.04), toluene, 90 °C
diallylamine (1.1), Pd2(dba)3 (0.01), rac-BINAP (0.03), t-BuONa (1.2), toluene, 14b: 45 Pd(PPh3)4 (0.01), 1,3-dimethylbarbituric acid
70 °C
3: 0
−
−
−
−
3: 76
14a: 0
b
(3.0), CH2Cl2, 35 °C
b
(PPh3)3RhCl (0.05), CH3CN/H2O, 100 °C
3: 78
3: 73
5
benzophenone imine (1.1), Pd2(dba)3 (0.01), rac-BINAP (0.03), t-BuONa (1.2), 14c: 76 NH2OH·HCl (1.8), AcONa (2.4), MeOH, rt,
toluene, 70 °C
a
b
Measured by HPLC condition D (see the Experimental Section). HPLC yield.
a b
,
Scheme 8. Scale-Up Route to (R)-1
a
Reagents and conditions: (a) tert-butylchlorodimethylsilane, imidazole, DMF, 12 °C; (b) benzophenone imine, Pd2(dba)3, rac-BINAP, t-BuONa,
toluene, 61 °C; (c) NH2OH·HCl, AcONa, MeOH, 22 °C; (d) fumaric acid, toluene, THF, 73% for four steps; (e) suspension with AcOi-Pr, 97%;
(f) KHCO3(aq), AcOi-Pr; (g) POCl3, DMF, CH2Cl2, −3 °C, then compound 3, DMAP, pyridine, CH2Cl2, −1 °C; (h) HCl(aq), THF, 2 °C, 83%
b
for two steps; (i) recrystallization with aqueous EtOH, 95%. Compounds in parentheses were not isolated.
used as a starting material. In addition, several process-related
issues were solved. The first pilot-scale synthesis yielded 19.0 kg
of (R)-1 for the first GMP lot.
spectra were acquired with a JEOL JNM-ECS400 spectrometer
at 400 and 101 MHz, respectively. Chemical shifts (δ) of H
1
NMR spectra are reported in parts per million relative to the
residual nondeuterated solvent peak of DMSO-d6 (2.49 ppm)
as an internal standard. Chemical shifts of proton-decoupled
13C NMR spectra are reported in parts per million relative to
the center line of the septet of DMSO-d6 (39.7 ppm). Infrared
spectra were acquired using attenuated total reflectance (ATR)
measured with a Shimadzu IRAffinity-1 spectrometer. High-
resolution mass spectrometry (HRMS) with electrospray
ionization (ESI) was performed using a Shimadzu LCMS-IT-
TOF mass spectrometer. Elemental analyses were performed
EXPERIMENTAL SECTION
■
General. Starting materials, reagents, and solvents were
obtained from commercial suppliers and used without further
purification. All pilot operations were carried out under a
nitrogen atmosphere. Optical rotations were measured with a
JASCO P-2200 digital polarimeter at 20 °C using the sodium D
line, and optical rotation data are reported as follows: [α]D20
1
(concentration c in g/100 mL, solvent). H and 13C NMR
G
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
with a Yanaco JM10 and a Thermo Scientific Dionex ICS-3000,
and HPLC was performed with a Hitachi L-2000 system.
HPLC Conditions. A: Waters XBridge C18 150 mm × 4.6
mm, 3.5 μm; elution A, 10 mM KH2PO4 solution (pH 2);
elution B, CH3CN; isocratic, 40% A; flow rate, 1.0 mL/min; at
45 °C; wavelength, 215 nm; retention times, 7 9.4 min, 4 11.1
min, 5 25.7 min. B: Daicel CHIRALCEL OD-RH 150 mm ×
4.6 mm, 5 μm; elution A, 10 mM KH2PO4 solution (pH 2);
elution B, CH3CN; isocratic, 40% A; flow rate, 1.0 mL/min; at
40 °C; wavelength, 230 nm; retention times, (R)-7 7.7 min,
(S)-7 9.8 min. C: Imtact Unison UK-Phenyl 250 mm × 4.6
mm, 3 μm; elution A, 10 mM KH2PO4 solution (pH 2); elution
B, CH3CN; isocratic, 50% A; flow rate, 1.0 mL/min; at 45 °C;
wavelength, 215 nm; retention times, (S)-2 17.2 min, (R)-2
18.3 min. D: YMC-Pack Pro-C18 RS 150 mm × 4.6 mm, 5 μm;
elution A, 2 mM KH2PO4 solution (pH 7); elution B, CH3CN/
i-PrOH (50/50); gradient, 80% A to 20% A over 50 min; flow
rate, 1.0 mL/min; at 50 °C; wavelength, 275 nm; retention
time, 3 13.3 min. E: Daicel CHIRALPAC AD-H 250 mm × 4.6
mm, 10 μm; elution A, n-hexane; elution B, EtOH; isocratic,
50% A; flow rate, 0.7 mL/min; at 25 °C; wavelength, 254 nm;
retention times, (R)-1 9.7 min, (S)-1 14.9 min.
(2R)-2-(4-Cyclopropanesulfonyl-3-cyclopropylphen-
yl)-3-[(2R,3R,7R)-2,3-diphenyl-1,4-dioxaspiro[4.4]nonan-
7-yl]propanoic Acid ((R)-7). A 2000 L reactor was charged
with (1S,4′R,5R,5′R)-4′,5′-diphenylspiro[bicyclo[3.1.0]hexane-
2,2′-[1,3]dioxolane] (4) (26.5 kg, 90.6 mol) and CH3CN (420
kg). Water (137.6 g, 7.64 mol), NaI (24.8 kg, 166 mol),
NaHCO3 (42.7 kg, 508 mol), and TMSCl (16.1 kg, 148 mol)
were added to the mixture at 23 °C in six portions at 30 min
intervals. After 2 h of stirring at 23 °C, the mixture was added
to a solution of Na2SO3 (36.7 kg) and K2CO3 (40.1 kg) in
water (790 kg), and the reactor was washed with CH3CN/
water (42 kg/26 kg). The organic layer was washed three times
with NaCl solution (26 kg in 130 kg of water) and
concentrated to 52 L under vacuum. The procedure described
above was repeated one more time, and the combined CH3CN
solution was azeotropically distilled three times with THF (230
kg) to 105 L under vacuum. The residue was filtered to remove
inorganic salts, and the reactor was washed with THF (47 kg)
to give 5 as a THF solution. A 1500 L reactor was charged with
t-BuOK (21.2 kg, 189 mol), N,N′-dimethylpropyleneurea (58
kg), and THF (160 kg). A solution of methyl (3-bromo-4-
cyclopropanesulfonylphenyl)acetate (6) (52.4 kg, 157 mol) in
THF (68 kg) and N,N′-dimethylpropyleneurea (58 kg) were
added dropwise to the t-BuOK solution at −8 °C, and the
reactor was washed with THF (23 kg). After 1 h of stirring at
−11 °C, the solution of 5 was added dropwise to the mixture at
−9 °C, and the reactor was washed with THF (23 kg). After 9
h of stirring at −9 °C, the reaction mixture was added to a
mixture of NH4Cl solution (42 kg in 420 kg of water) and
AcOEt (380 kg), and the reactor was washed with AcOEt (94
kg) and water (52 kg). After phase separation, activated carbon
(5.2 kg) and AcOEt (68 kg) were added to the organic layer.
After 1 h of stirring, the mixture was filtered through powdered
cellulose, and the reactor was washed with AcOEt (94 kg). The
filtrate was washed with NaCl solution (52 kg in 260 kg of
water), and the organic layer was concentrated to 157 L under
vacuum followed by azeotropic distillation three times with
toluene (230 kg) to 157 L under vacuum. K3PO4 solution (133
kg in 210 kg of water), toluene (549 kg), cyclopropylboronic
acid (17.6 kg, 205 mol), triphenylphosphine (1.65 kg, 6.29
mol), and Pd(OAc)2 (707 g, 3.15 mol) were added, and the
mixture was stirred at 87 °C for 20 h and then cooled.
Trithiocyanuric acid (5.2 kg), activated carbon (5.2 kg), and
water (259 kg) were added, and the mixture was stirred at 27
°C for 24 h. The mixture was filtered through powdered
cellulose, and the reactor was washed with toluene (230 kg).
After phase separation, the organic layer was washed with NaCl
solution (52 kg in 260 kg of water) and concentrated to 157 L
under vacuum, followed by azeotropic distillation three times
with THF (230 kg) to 157 L under vacuum. THF (47 kg),
MeOH (130 kg), and NaOH solution (24 wt % sodium
hydroxide, 62.9 kg in 58 kg of water) were added, and the
mixture was stirred at 47 °C for 2 h. Water (420 kg) and n-
heptane (180 kg) were added to the reaction mixture, and the
aqueous layer was washed with n-heptane (180 kg). HCl
solution (36 wt % hydrochloric acid, 35 kg in 26 kg of water)
was added to the aqueous layer until the pH reached 5.5. The
aqueous solution was extracted twice with AcOEt (470 kg, 240
kg), and the combined organic layers were washed with NaCl
solution (52 kg in 260 kg of water). The organic layer was
concentrated to 157 L under vacuum, followed by azeotropic
distillation three times with EtOH (210 kg) under vacuum, and
then EtOH (680 kg) was added to the residue. The resulting
slurry was stirred at 54 °C for 1 h and then at 22 °C for 43 h.
The crystals were filtered, washed with EtOH (210 kg), and
dried at 40 °C for 16 h under vacuum to give crude (R)-7 (26.7
kg, 46.6 mol; 29.7% yield based on 6, 91.0% de). A filtrate was
collected and used for an epimerization experiment on (S)-7
(filtrate A). This procedure was repeated one more time to
obtain crude (R)-7 (26.5 kg, 46.3 mol; 29.5% yield based on 6,
90.2% de). A 5000 L reactor was charged with crude (R)-7
(52.7 kg, 92.0 mol) and EtOH (748 kg). The mixture was
heated at 78 °C to obtain a clear solution. Activated carbon (5.3
kg) and EtOH (21 kg) were added, and the mixture was stirred
at 78 °C for 0.5 h. The mixture was filtered through powdered
cellulose, and the reactor was washed with EtOH (63 kg). The
filtrate was cooled to 24 °C and then stirred for 3 h. The
crystals were filtered, washed with EtOH (130 kg), and dried at
40 °C for 12 h under vacuum to give (R)-7 (44.3 kg, 77.4 mol;
84.1% yield based on crude (R)-7, 98.5% de). [α]2D0 −48.8 (c
1
1.00, MeOH); H NMR (DMSO-d6, 400 MHz) δ 12.57 (s,
1H), 7.78 (d, J = 8.2 Hz, 1H), 7.39−7.28 (m, 7H), 7.23−7.06
(m, 5H), 4.73−4.64 (m, 2H), 3.69 (t, J = 7.8 Hz, 1H), 3.08−
2.94 (m, 1H), 2.79 (tt, J = 8.4, 5.1 Hz, 1H), 2.22 (dd, J = 12.8,
6.9 Hz, 1H), 2.12−2.00 (m, 2H), 1.96−1.73 (m, 4H), 1.67 (dd,
J = 12.8, 9.2 Hz, 1H), 1.50−1.25 (m, 1H), 1.17−1.01 (m, 6H),
0.96−0.83 (m, 2H); 13C NMR (DMSO-d6, 101 MHz) δ 174.2,
145.7, 143.1, 137.8, 136.7, 136.6, 128.7, 128.43, 128.37, 126.91,
126.89, 125.3, 124.9, 118.0, 84.6, 84.2, 49.4, 43.3, 36.6, 34.7,
31.9, 29.2, 12.2, 10.6, 10.5, 5.6; FTIR (ATR, cm−1) 3011, 1725,
1294, 1192, 1137, 1119, 1004, 771, 700, 528; HRMS (ESI) m/z
[M + Na]+ calcd for C34H36O6S 595.2130, found 595.2183.
Anal. Calcd for C34H36O6S: C, 71.30; H, 6.34; S, 5.60. Found:
C, 71.21; H, 6.37; S, 5.63.
Recovery of the Undesired (2S)-2-(4-Cyclopropane-
sulfonyl-3-cyclopropylphenyl)-3-[(2R,3R,7R)-2,3-diphen-
yl-1,4-dioxaspiro[4.4]nonan-7-yl]propanoic Acid ((S)-7).
Filtrate (377 mL) from the pilot-scale synthesis of (R)-7
(filtrate A, 11.3 mmol of (S)-7, 63.3% de) was added to a 500
mL three-neck round-bottom flask. The filtrate was concen-
trated under vacuum, followed by azeotropic distillation three
times with CH3CN (64 mL) under vacuum. CH3CN (96 mL),
1-methylimidazole (2.7 mL, 34 mmol) and p-TsCl (2.60 g, 13.6
mmol) were added, and the mixture was stirred at 25 °C for 2
H
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
h. AcOEt (32 mL), CH3CN (13 mL), and 5 wt % NaHCO3
solution (32 mL) were added to the reaction mixture, and the
organic layer was washed three times with 5 wt % NaHCO3
solution (19 mL) and twice with 25 wt % NaCl solution (19
mL), yielding the solution of epimerized 7 (0% de). The
solution was concentrated under vacuum and charged with
toluene (32 mL), THF (32 mL), and 1 M NaOH solution
(12.8 mL), and the aqueous layer was washed three times with
toluene (32 mL). The aqueous layer was neutralized with 1 M
HCl solution and extracted twice with AcOEt (64 mL, 32 mL).
The organic layer was washed with 25 wt % NaCl solution (16
mL), concentrated under vacuum, and azeotropically distilled
twice with EtOH (64 mL) under vacuum. EtOH (128 mL) and
activated carbon (0.64 g) were added, and the mixture was
stirred at reflux for 1 h. The mixture was filtered through
powdered cellulose, and the filtrate was concentrated under
vacuum. EtOH (38 mL) was added, and the slurry was stirred
at 78 °C until the solid was dissolved. After the mixture was
cooled to 57 °C, seed crystals (6.6 mg) were added, and the
solution was stirred at 23 °C for 3 days. The crystals were
filtered, washed with EtOH (3.2 mL), and dried at 40 °C for 1
day under vacuum to give crude (R)-7 (3.22 g, 5.62 mmol;
49.7% yield based on filtrate A, 70.9% de). Crude (R)-7 (3.15 g,
5.50 mmol) and EtOH (63 mL) were added to a 100 mL three-
neck round-bottom flask. The slurry was stirred at 78 °C until
the solid was dissolved. The solution was slowly cooled to 23
°C, followed by stirring for 1 day. The crystals were filtered and
washed with EtOH (9.5 mL). The wet crystals were dissolved
in EtOH (63 mL) at 78 °C. The solution was slowly cooled to
24 °C, followed by stirring for 2 days. The crystals were filtered,
washed with EtOH (9.5 mL), and dried at 40 °C under vacuum
to give (R)-7 (2.28 g, 3.98 mmol; 72.4% yield based on crude
(R)-7, 98.8% de).
5.63, 5.60; FTIR (ATR, cm−1) 2965, 1724, 1696, 1598, 1313,
1288, 1152, 1125, 881, 534; HRMS (ESI) m/z [M + Na]+ calcd
for C20H24O5S 399.1242, found 399.1245. Anal. Calcd for
C20H24O5S: C, 63.81; H, 6.43; S, 8.52. Found: C, 63.53; H,
6.46; S, 8.61.
5-({[tert-Butyl(dimethyl)silyl]oxy}methyl)pyrazin-2-
amine Mono-(2E)-but-2-enedioate (3 Fumarate). A 1000
L reactor was charged with (5-chloropyrazin-2-yl)methanol
(16) (35.5 kg, 246 mol), imidazole (33.4 kg, 491 mol), and
DMF (202 kg). tert-Butylchlorodimethylsilane (44.4 kg, 295
mol) was added at 11 °C, and the mixture was stirred at 12 °C
for 2 h. Water (71 kg) and n-heptane (121 kg) were added to
the reaction mixture, and the aqueous layer was extracted with
n-heptane (72 kg). The combined organic layer was washed
twice with NaCl solution (18 kg in 178 kg of water) and water
(107 kg). The organic layer was concentrated to 80 L under
vacuum, followed by azeotropic distillation with AcOEt (96 kg)
and toluene (92 kg) to 110 L under vacuum to give a solution
of 15. Toluene (521 kg), benzophenone imine (66.8 kg, 369
mol), tris(dibenzylideneacetone)dipalladium(0) (2.25 kg, 2.46
mol), rac-BINAP (4.60 kg, 7.39 mol), and sodium tert-butoxide
(40.1 kg, 417 mol) were added, and the mixture was stirred at
61 °C for 2 h and then cooled. Trithiocyanuric acid (3.6 kg),
activated carbon (7.1 kg), water (355 kg), and toluene (20 kg)
were added, and the mixture was stirred at 22 °C for 24 h. The
mixture was filtered through powdered cellulose, and the
reactor was washed with toluene (103 kg). After phase
separation, the organic layer was washed with NaCl solution
(18 kg in 178 kg of water) and concentrated to 190 L under
vacuum, followed by azeotropic distillation with methanol (84
kg) to 157 L under vacuum to give a solution of 14c. The
solution was charged with methanol (281 kg), hydroxylamine
hydrochloride (30.8 kg, 443 mol), and sodium acetate (48.4 kg,
590 mol), and the mixture was stirred at 22 °C for 3 h. Toluene
(307 kg) and K2CO3 solution (68 kg in 497 kg of water) were
added to the reaction mixture, and the organic layer was
washed twice with NaCl solution (18 kg in 178 kg of water)
and concentrated to 250 L under vacuum, followed by
azeotropic distillation with toluene (154 kg) to 250 L under
vacuum to give a solution of 3. The solution was charged with
toluene (368 kg) and THF (32 kg) and slowly added to a
fumaric acid solution (28.5 kg, 246 mol in 410 kg of THF) at
52 °C, and then the reactor was washed with toluene (61 kg).
The resulting slurry was stirred at 50 °C for 1 h and then at 25
°C for 17 h. The crystals were filtered, washed with toluene
(154 kg), and dried at 40 °C for 17 h under vacuum to give
crude 3 fumarate (64.1 kg, 180 mol; 73.2% yield based on 16).
A 1000 L reactor was charged with crude 3 fumarate (63.0 kg,
177 mol) and AcOi-Pr (280 kg). The mixture was stirred at 25
°C for 1 h, and the crystals were filtered, washed with AcOi-Pr
(150 kg), and dried at 60 °C for 15 h under vacuum to give 3
fumarate (61.1 kg, 172 mol; 97.2% yield based on crude 3
(2R)-2-(4-Cyclopropanesulfonyl-3-cyclopropylphen-
yl)-3-[(R)-3-oxocyclopentyl]propanoic Acid ((R)-2). A
1000 L reactor was charged with (R)-7 (22.0 kg, 38.4 mol),
acetone (260 kg), and HCl (36 wt % hydrochloric acid, 4.73 kg
in 3.7 kg of water), and the mixture was stirred at 50 °C for 3 h
and then cooled. Toluene (95 kg) and NaHCO3 solution (11
kg in 220 kg of water) were added to the reaction mixture. The
organic layer was washed with NaHCO3 solution (5.5 kg in 110
kg of water), and HCl solution (36 wt % hydrochloric acid 15.8
kg in 12 kg of water) was added to the combined aqueous layer
until the pH reached 2.3. The solution was extracted twice with
AcOEt (200 kg, 99 kg), and the combined organic layer was
washed with NaCl solution (13 kg in 66 kg of water). The
procedure described above was repeated one more time, and
the combined organic layer was concentrated to 66 L under
vacuum, followed by azeotropic distillation three times with
AcOEt (200 kg) to 66 L under vacuum. After n-heptane (210
kg) was added to the residue, the mixture was stirred at 40 °C
for 3 h. The resulting slurry was stirred at 75 °C for 1 h and
then at 23 °C for 2 h. The crystals were filtered, washed with
AcOEt/n-heptane (20 kg/62 kg), and dried at 40 °C for 12 h
under vacuum to give (R)-2 (27.5 kg, 73.0 mol; 95.0% yield,
1
fumarate). H NMR (DMSO-d6, 400 MHz) δ 7.88 (s, 1H),
7.79 (d, J = 1.4 Hz, 1H), 6.61 (s, 2H), 6.34 (s, 2H), 4.55 (s,
2H), 0.86 (s, 9H), 0.04 (s, 6H); 13C NMR (DMSO-d6, 101
MHz) δ 166.0, 155.1, 141.5, 139.9, 134.0, 131.0, 64.0, 25.8,
18.0, −5.1; FTIR (ATR, cm−1) 3453, 2928, 2857, 1707, 1647,
1417, 1251, 1114, 833, 775; HRMS (ESI) m/z [M + Na]+ calcd
for C11H21N3OSi 262.1352, found 262.1310. Anal. Calcd for
C11H21N3OSi·C4H4O4: C, 50.68; H, 7.09; N, 11.82. Found: C,
50.72; H, 7.12; N, 11.75.
1
98.9% de). [α]2D0 −121.1 (c 1.00, MeOH); H NMR (DMSO-
d6, 400 MHz) δ 12.59 (s, 1H), 7.77 (d, J = 8.2 Hz, 1H), 7.33
(dd, J = 8.2, 1.8 Hz, 1H), 7.05 (d, J = 1.4 Hz, 1H), 3.78−3.61
(m, 1H), 3.08−2.96 (m, 1H), 2.78 (tt, J = 8.4, 5.1 Hz, 1H),
2.35−2.20 (m, 1H), 2.15−1.78 (m, 7H), 1.56−1.29 (m, 1H),
1.18−1.02 (m, 6H), 0.93−0.84 (m, 2H); 13C NMR (DMSO-d6,
101 MHz) δ 218.3, 174.2, 145.6, 143.1, 137.9, 128.8, 125.3,
125.0, 49.1, 44.1, 38.4, 38.0, 34.5, 31.9, 28.9, 12.2, 10.6, 10.4,
(2R)-2-(4-Cyclopropanesulfonyl-3-cyclopropylphen-
yl)-N-[5-(hydroxymethyl)pyrazin-2-yl]-3-[(R)-3-
I
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
oxocyclopentyl]propanamide ((R)-1). A 1000 L reactor was
charged with 3 fumarate (38.3 kg, 108 mol), AcOi-Pr (460 kg),
and KHCO3 solution (53.7 kg in 410 kg of water). The mixture
was stirred at 43 °C for 1 h, and the organic layer was washed
twice with water (270 kg) and NaCl solution (35 kg in 140 kg
of water). The solution was filtered, and the reactor was washed
with AcOi-Pr (23 kg). After phase separation, the organic layer
was concentrated to 54 L under vacuum, followed by
azeotropic distillation with toluene (230 kg) to 54 L under
vacuum. CH2Cl2 (350 kg) was added, and the mixture was
filtered to remove inorganic salts. The reactor was washed with
CH2Cl2 (73 kg) to give the solution of 3. N,N-Dimethyl-4-
aminopyridine (4.37 kg, 35.8 mol) and pyridine (28.4 kg, 359
mol) were added to the solution. A 1000 L reactor was charged
with (R)-2 (27.0 kg, 71.7 mol), CH2Cl2 (350 kg), and DMF
(2.10 kg, 28.7 mol). POCl3 (10.9 kg, 71.1 mol) was slowly
added at −4 °C, and the mixture was stirred at −3 °C for 1 h.
The resulting acid chloride solution was added dropwise to the
CH2Cl2 solution of 3 at −3 °C, and the reactor was washed
with CH2Cl2 (73 kg). The reaction mixture was stirred at −1
°C for 3 h. Water (270 kg) was added to the reaction mixture,
and the organic layer was washed three times with citric acid
solution (14 kg in 270 kg of water) and once with water (270
kg). The organic layer was filtered, and the reactor was washed
with CH2Cl2 (35 kg). The solution was washed with NaHCO3
solution (14 kg in 270 kg of water) and concentrated to 81 L
under vacuum, followed by azeotropic distillation three times
with THF (120 kg) to 81 L under vacuum. THF (49 kg) and
HCl solution (36 wt % hydrochloric acid, 14.7 kg in 12 kg of
water) were added at 3 °C, and the mixture was stirred at 2 °C
for 3 h. Trithiocyanuric acid (2.7 kg), activated carbon (2.7 kg),
AcOEt (364 kg), and NaHCO3 solution (21 kg in 410 kg of
water) were added, and the mixture was stirred at 19 °C for 8 h.
The mixture was filtered through powdered cellulose, and the
reactor was washed with AcOEt (49 kg). After phase
separation, the organic layer was washed twice with NaHCO3
solution (7.0 kg in 140 kg of water) and once with water (140
kg) and NaCl solution (20 kg in 81 kg of water). The organic
layer was concentrated to 81 L under vacuum, followed by
azeotropic distillation twice with CH3CN (210 kg) to 81 L
under vacuum. CH3CN (250 kg) and n-heptane (92 kg) were
added to the residue, and the lower CH3CN layer was
collected. The CH3CN layer was washed with n-heptane (92
kg, twice) to remove silanol, siloxane, or both. The solution was
concentrated to 81 L under vacuum, followed by azeotropic
distillation three times with AcOEt (240 kg) to 81 L under
vacuum. AcOEt (49 kg) was added, and the mixture was stirred
at 55 °C for 2 h. n-Heptane (92 kg) was added to the slurry at
54 °C over 3 h, and the mixture was stirred at 55 °C for 3 h and
then at 23 °C for 3 h. The crystals were filtered, washed twice
with AcOEt/n-heptane (24 kg/27 kg, 12 kg/84 kg), and dried
at 60 °C for 8 h under vacuum to give crude (R)-1 (28.8 kg,
59.6 mol; 83.1% yield based on (R)-2). A 200 L reactor was
charged with crude (R)-1 (20.0 kg, 41.4 mol), EtOH (31.6 kg),
and water (30 L). The mixture was stirred at 67 °C until the
solid was dissolved. The solution was filtered through a 0.45
μm filter into a 100 L reactor, and the reactor was washed with
hot EtOH/water (15.8 kg/5 L). Water (25 L) was added to the
solution through the 0.45 μm filter, and the solution was cooled
to 12 °C over 6 h and then stirred at 12 °C for 16 h. Water
(180 L) was added through the 0.45 μm filter at 25 °C over 1 h,
and the mixture was stirred at 25 °C for 20 h. The crystals were
filtered, washed with water (100 L), and dried at 60 °C for 19 h
under vacuum to give (R)-1 (19.0 kg, 39.3 mol; 94.9% yield
based on crude (R)-1, 99.0% de). [α]2D0 −128.7 (c 1.00,
1
MeOH); H NMR (DMSO-d6, 400 MHz) δ 11.07 (s, 1H),
9.20 (d, J = 1.4 Hz, 1H), 8.41 (d, J = 1.4 Hz, 1H), 7.79 (d, J =
8.2 Hz, 1H), 7.41 (dd, J = 8.2, 1.8 Hz, 1H), 7.15 (d, J = 1.8 Hz,
1H), 5.52 (t, J = 5.7 Hz, 1H), 4.56 (d, J = 6.0 Hz, 2H), 4.04 (t, J
= 7.6 Hz, 1H), 3.03−2.97 (m, 1H), 2.79 (tt, J = 8.4, 5.1 Hz,
1H), 2.25−1.81 (m, 8H), 1.53−1.47 (m, 1H), 1.17−1.12 (m,
2H), 1.08−1.02 (m, 4H), 0.89−0.84 (m, 2H); 13C NMR
(DMSO-d6, 101 MHz) δ 218.5, 171.8, 152.1, 147.3, 145.7,
143.2, 140.3, 138.2, 134.8, 129.0, 125.3, 125.1, 62.5, 49.9, 44.4,
38.4, 38.2, 34.8, 32.1, 29.1, 12.4, 10.8, 10.7, 5.8; FTIR (ATR,
cm−1) 3544, 3257, 1727, 1692, 1546, 1507, 1363, 1285, 1149,
719; HRMS (ESI) m/z [M + Na]+ calcd for C25H29N3O5S
506.1726, found 506.1747. Anal. Calcd for C25H29N3O5S: C,
62.09; H, 6.04; N, 8.69. Found: C, 61.79; H, 6.19; N, 8.62.
AUTHOR INFORMATION
Corresponding Author
*E-mail: yohei.yamashita@astellas.com.
■
ORCID
Yohei Yamashita: 0000-0001-8586-629X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Dr. M. Hayakawa, Mr. T. Nigawara, Mr. M.
Okumura, and Dr. K. Maki for helpful discussions about the
medicinal chemistry synthesis; Dr. T. Mukuta, Dr. M. Okada,
Mr. T. Takahashi, and Dr. N. Hashimoto for valuable advice;
Mr. K. Takasuka, Ms. N. Taniguchi, and Ms. M. Fujimoto for
development of the HPLC conditions; and all members of the
Astellas Analytical Science Laboratories, Inc. for acquiring
infrared spectra, mass spectra, and optical rotation data.
REFERENCES
■
(1) (a) Iynedjian, P. B. Cell. Mol. Life Sci. 2009, 66, 27−42.
(b) Kawai, S.; Mukai, T.; Mori, S.; Mikami, B.; Murata, K. J. Biosci.
Bioeng. 2005, 99, 320−330.
(2) (a) Matschinsky, F. M. Nat. Rev. Drug Discovery 2009, 8, 399−
416. (b) Pal, M. Drug Discovery Today 2009, 14, 784−792.
(3) Hayakawa, M.; Kido, Y.; Nigawara, T.; Okumura, M.; Kanai, A.;
Maki, K.; Amino, N. PCT Int. Appl. WO/2009/091014 A1 20090723,
2009.
(4) Alorati, A. D.; Bio, M. M.; Brands, K. M. J.; Cleator, E.; Davies, A.
J.; Wilson, R. D.; Wise, C. S. Org. Process Res. Dev. 2007, 11, 637−641.
(5) For a review, see: Okano, K. Tetrahedron 2011, 67, 2483−2512.
(6) Sarabu, R.; Bizzarro, F. T.; Corbett, W. L.; Dvorozniak, M. T.;
Geng, W.; Grippo, J. F.; Haynes, N.-E.; Hutchings, S.; Garofalo, L.;
Guertin, K. R.; Hilliard, D. W.; Kabat, M.; Kester, R. F.; Ka, W.; Liang,
Z.; Mahaney, P. E.; Marcus, L.; Matschinsky, F. M.; Moore, D.; Racha,
J.; Radinov, R.; Ren, Y.; Qi, L.; Pignatello, M.; Spence, C. L.; Steele, T.;
Tengi, J.; Grimsby, J. J. Med. Chem. 2012, 55, 7021−7036.
(7) Evans, D. A.; Weber, A. E. J. Am. Chem. Soc. 1986, 108, 6757−
6761.
(8) Albaneze-Walker, J.; Zhao, M.; Baker, M. D.; Dormer, P. G.;
McNamara, J. Tetrahedron Lett. 2002, 43, 6747−6750.
(9) On the basis of MS analysis of a reaction mixture, a major
impurity was thought to be formed by β-elimination of HI from 5.
(10) (a) Morita, T.; Okamoto, Y.; Sakurai, H. J. Chem. Soc., Chem.
Commun. 1978, 20, 874−875. (b) Olah, G. A.; Narang, S. C.; Gupta,
B. G. B.; Malhotra, R. J. Org. Chem. 1979, 44, 1247−1251.
(11) (a) Magnus, N. A.; Braden, T. M.; Buser, J. Y.; DeBaillie, A. C.;
Heath, P. C.; Ley, C. P.; Remacle, J. R.; Varie, D. L.; Wilson, T. M.
J
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Org. Process Res. Dev. 2012, 16, 830−835. (b) Bachmann, S.; Fettes, A.;
Lautz, C.; Scalone, M. Org. Process Res. Dev. 2013, 17, 1451−1457.
(12) (a) DeBaillie, A. C.; Magnus, N. A.; Laurila, M. E.; Wepsiec, J.
P.; Ruble, J. C.; Petkus, J. J.; Vaid, R. K.; Niemeier, J. K.; Mick, J. F.;
Gunter, T. Z. Org. Process Res. Dev. 2012, 16, 1538−1543.
(b) Yamagami, T.; Moriyama, N.; Kyuhara, M.; Moroda, A.;
Uemura, T.; Matsumae, H.; Moritani, Y.; Inoue, I. Org. Process Res.
Dev. 2014, 18, 437−445.
(13) Grabowski, E. J. J. In Encyclopedia of Reagents for Organic
Synthesis; Paquette, L. A., Ed.; Wiley & Sons: London, 1995; Vol. 6, pp
3893−3896.
(14) Wakasugi, K.; Iida, A.; Misaki, T.; Nishii, Y.; Tanabe, Y. Adv.
Synth. Catal. 2003, 345, 1209−1214.
(15) Surry, D. S.; Buchwald, S. L. J. Am. Chem. Soc. 2007, 129,
10354−10355.
(16) Xu, H.; Wolf, C. Chem. Commun. 2009, 21, 3035−3037.
(17) Lee, S.; Jørgensen, M.; Hartwig, J. F. Org. Lett. 2001, 3, 2729−
2732.
(18) Garro-Helion, F.; Merzouk, A.; Guibe, F. J. Org. Chem. 1993, 58,
6109−6113.
(19) Laguzza, B. C.; Ganem, B. Tetrahedron Lett. 1981, 22, 1483−
1486.
(20) Wolfe, J. P.; Åhman, J.; Sadighi, J. P.; Singer, R. A.; Buchwald, S.
L. Tetrahedron Lett. 1997, 38, 6367−6370.
(21) Besides that, sodium azide was tried as an amine source.
Although azide adduct was obtained, neither the subsequent
hydrogenation nor Staudinger reaction proceeded successfully.
K
DOI: 10.1021/acs.oprd.6b00415
Org. Process Res. Dev. XXXX, XXX, XXX−XXX