Development of an optimized synthetic and purification process of S-2367 (Velneperit), a novel neuropeptide y (NPY) Y5 receptor antagonist

Article abstract of DOI:10.1021/acs.oprd.5b00023

We developed a simple synthetic route to S-2367 (Velneperit) (1) using trans-1-ethoxycarbonyl-4-aminocyclohexane hydrochloride salt (12) as a starting material. The key step was Na₂WO₄/H₂O₂ oxidation, and we found that it was accelerated in weakly basic conditions. The finding was useful to control one of the critical impurities: 14 in 10. The new process was more efficient than the early process from the viewpoint of the number of reactions, yield, throughput, and EHS (environment, health, and safety) but a quality deviation occurred in the pilot manufacturing: 10 content in 1 was over the upper limit. The cause was presumed to be hydrolysis of 1 during the recrystallization step. After finding it, we developed two reliable purification processes. The first was slurry washing including clever polymorphic control using only acetone and water. The second was salt formation of 10 and rational building of the recrystallization procedure based on solubility to improve removal rate.

Full text of DOI:10.1021/acs.oprd.5b00023

Article
pubs.acs.org/OPRD
Development of an Optimized Synthetic and Purification Process of
S‑2367 (Velneperit), a Novel Neuropeptide Y (NPY) Y5 Receptor
Antagonist
Shinichi Oda,* Kumiko Manaka, Kiyoshi Kakiya, Yasuyuki Hozumi, Yuki Fukui, Sohei Omura,
Makoto Kurashita, Masanori Nishiwaki, Yoshiyuki Takeuchi, and Hideyuki Kitamura
Chemical Development, Production Technology Department, Manufacturing Division and Chemical Development Center, CMC
Development Laboratories, Shionogi & Co., Ltd., 1-3, Kuise Terajima 2-Chome, Amagasaki, Hyogo 660-0813, Japan
S
* Supporting Information
ABSTRACT: We developed a simple synthetic route to S-2367 (Velneperit) (1) using trans-1-ethoxycarbonyl-4-
aminocyclohexane hydrochloride salt (12) as a starting material. The key step was Na2WO4/H2O2 oxidation, and we found
that it was accelerated in weakly basic conditions. The finding was useful to control one of the critical impurities: 14 in 10. The
new process was more efficient than the early process from the viewpoint of the number of reactions, yield, throughput, and EHS
(environment, health, and safety) but a quality deviation occurred in the pilot manufacturing: 10 content in 1 was over the upper
limit. The cause was presumed to be hydrolysis of 1 during the recrystallization step. After finding it, we developed two reliable
purification processes. The first was slurry washing including clever polymorphic control using only acetone and water. The
second was salt formation of 10 and rational building of the recrystallization procedure based on solubility to improve removal
rate.
environmental impact. However, the changes caused a quality
deviation. Herein, we discuss the summary of process
improvement, a new finding on Na2WO4/H2O2 oxidation and
purification strategies of 1 to solve the quality issue.
INTRODUCTION
■
Neuropeptide Y (NPY) is a 36-amino acid peptide neuro-
transmitter that is widely distributed in mammalian central and
peripheral nervous systems. Thus far, five distinct subtypes of G
protein-coupled NPY receptors, Y1, Y2, Y4, Y5, and Y6, are
known. Among them, the Y5 receptor subtype is thought to
play a role in meal initiation and the regulation of energy
balance. Therefore, an antagonist of the NPY Y5 receptor is
considered to have potential as an antiobesity drug.1 S-2367 (1)
(Figure 1), which was discovered by Shionogi Research
RESULT AND DISCUSSIONS
■
Summary of process improvement. In the pilot
manufacturing for preclinical and early clinical studies, the
synthesis of 1 was carried out using the starting material 5.5
Esterification of 5 using SOCl2 in methanol gave methyl ester 6.
The coupling reaction of 6 with sulfinyl chloride 4, which was
synthesized by oxidation of di-tert-butyl sulfide 2 and
chlorination reaction, produced sulfinamide 7. Oxidation of 7
using H2O2/Mo catalyst, isomerization by NaOMe, and
hydrolysis gave the intermediate 10. The coupling reaction of
10 with 11 through acid chloride and recrystallization using
acetone/water produced purified 1, whose polymorph was form
I (Scheme 1). The process could be tolerable for scale-up, but
there was still room for improvement in synthetic route, yield,
and productivity. The starting material was changed from 5 to
trans-1-ethoxycarbonyl-4-aminocyclohexane hydrochloride salt
(12), which was commercially available.6 The change could
remove esterification and isomerization from the manufacturing
process.
Figure 1. Chemical structure of S-2367 (1).
Laboratories, is a novel, potent, and selective orally available
NPY Y5 antagonist. It has been shown to have a significant
effect on weight loss in clinical studies, suggesting its potential
as a promising antiobesity drug.2 In addition, it can ameliorate
the glucose metabolism, reduce LDL cholesterol and uric acid
levels, lower blood pressure, and improve liver function.3,4
Therefore, the development of an optimized manufacturing
process was required to make it available for commercial supply
as well as clinical studies. Required process optimization
included not only simplifying the synthetic route, reaction
telescoping, and increasing yield, but also increasing the
charged amount and selecting low-toxic and recoverable solvent
because all of them affect the manufacturing cost and
Next we tried to develop a telescoped process using toluene
and water. The solvent of the coupling reaction could be
changed from dimethylformamide (DMF) to toluene/water
but following oxidation of 13 did not proceed well, probably
because 13 was distributed in the toluene layer but oxidant was
Received: December 24, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.5b00023
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Scheme 1. Synthetic Route to S-2367 (1) for Preclinical and Early Clinical Studies
a
Reagents and conditions: (a) H2O2, AcOH; (b) Cl2, 87% (2 steps;) (c) SOCl2, MeOH, washing by MeCN, 93%; (d) 4, Et3N, DMF; (e) H2O2,
(NH4)6Mo7O24·4H2O cat., water, recrystallization by AcOEt and cyclohexane, 81% (2 steps); (f) NaOMe, toluene, and MeOH; (g) water, 90%
(containing 1.5 area % of cis-10), (2 steps); (h) SOCl2, DMF cat., toluene → 11, pyridine (i) recrystallization using acetone and water, 87% (2
steps).
a
Scheme 2. Improved Synthetic Route to the Intermediate 10
a
Reagents and conditions: (a) 4, Et3N, toluene, H2O; (b) NaOH; (c) H2O2, Na2WO4·2H2O cat., H2O, 95% (3 steps).
in water. Hence, we exchanged the order of oxidation and
hydrolysis. Hydrolysis of 13 and oxidation of 14 proceeded
quantitatively. Moreover, the workup procedure could be
simplified: aqueous and organic impurities were removed by
the extraction after the coupling reaction and the hydrolysis,
respectively. After oxidation, 10 could be isolated by
neutralization crystallization with aqueous H2SO4 in 95%
yield (Scheme 2).
With regard to the final step, the amidation reaction
conditions were not changed, but the washing liquid of the
filtrated crude 1 cake was changed from 2-propanol to 50 v/v %
acetone/water to unify the solvents used in the overall
synthesis. That would reduce the tasks such as receiving test
Figure 2. Solubility of 1 (form I and form II) in acetone/water.
and storage of 2-propanol in commercial manufacturing. We
also tried to reduce the bottleneck volume of the solvent
dissolving crude 1 during the recrystallization to increase the
Table 1. Early Process vs Improved Process
charged amount. The solubility of 1 (form I and II) in various
ratios of acetone/water is shown in Figure 2. As the highest
solubility was obtained with acetone/water = 95 v/v %, the
dissolving solvent was changed from 14 V of 100 v/v % acetone
to 10.95 V of 95 v/v % acetone/water. Though crude 1 was
form II, obtaining the desired polymorph form I was not
difficult because form II was quickly transformed to form I in
various ratios of acetone/water. Even if the seed of form II was
added during the recrystallization, it was dissolved and form I
was precipitated.
The comparison of the early process with the improved
process is shown in Table 1. The overall yield from the starting
material was increased from 59% to 83%. The number of
chemical reactions was reduced from 6 to 4. The types of
solvents and their quantity could also be reduced. The new
items
early process
improved process
Lot scale (yield of 1 for one batch)
11 kg
59%
348 kg
83%
a
Overall yield
Purity of purified 1
99.5−99.9%
99.6−100.6%
(2 lots)
(6 lots)
a
Chemical reactions
6
7
1
1
4
a
Types of sovents
3
a
Relative total quantity of solvent
0.2
0.8
Relative maximum volume in the
final step
a
From the starting material 5 or 12 to purified 1
process was better from the viewpoint of EHS (environment,
health, and safety) because dimethylformamide (DMF), which
is relatively toxic and difficult to recover, could be substituted
B
DOI: 10.1021/acs.oprd.5b00023
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 3. Fate of Impurity 14 in the Amidation Reaction
by toluene and water.7 Furthermore, the relative maximum
volume of the final step was reduced; that is, production
capacity was increased.
Acceleration effect in weakly basic condition and safety
evaluation on Na2WO4/H2O2 oxidation. The key step of the
improved process was oxidation of 14 because residual 14 in
the intermediate 10 was quantitatively converted to the
impurity 15 with low removal rate in the following steps.
Residual 14 had to be controlled by not more than 0.04 w/w %
in oxidation reaction to meet our target quality of 1 (Scheme
3). It could be achieved by extending reaction time and
increasing the amount of H2O2 and/or catalyst, but they might
increase the processing time and the material cost. At first we
changed the catalyst from (NH4)6Mo7O24·4H2O to Na2WO4·
2H2O because the latter was more inexpensive. As a result of
investigation, we found the interesting new knowledge that the
oxidation reaction in the weakly basic condition of pH 9.0−9.5
was faster than that under nearly neutral conditions (Table 2).
Figure 3. Heat flow of oxidation reaction (total 332 kJ/mol).
in detail.9 The formation of acid chloride and the amidation
reaction was controlled by the IPT (In Process Test),10 and
they indicated that the reactions had successfully proceeded.
The impurity 10 content in crude 1 was 0.05−0.10 w/w % (7
lots), which was equivalent to that in the preliminary flask
experiment. These results suggested that the process before
obtaining crude 1 was not the cause of the impurity exceeding
the specification limit. Next, we examined the material balance
of 10 in the recrystallization step and found purified 1 and the
mother liquor included 0.08 w/w % and 0.10 w/w % of 10,
respectively, which suggested that the absolute quantity of 10
increased during the recrystallization step. It was probably due
to the hydrolysis of 1. We therefore compared the increase in
the amount of 10 in the crude 1 in 95 v/v % acetone/water to
100% acetone, and the result showed that the increase in the
former was about five times higher than that in the latter
(Figure 4). This suggested that the increase of 10 by hydrolysis
Table 2. Optimization of pH in Na2WO4/H2O2 Oxidation
residual 13 (w/w %) after
residual 13 (w/w %) after
entry
pH
addition of H2O2
stirring for 2.5 h
1
2
3
4
8.0
9.0
1.80
1.27
0.77
2.14
0.07
0.00
0.00
0.03
9.5
10.0
The result was unexpected for us because it was reported that
oxidation of 2-octanol using Na2WO4/H2O2 did not proceed
under basic conditions, probably due to the decomposition of
H2O2.8 The reason was not clarified, but the interesting finding
made it possible to control residual 14 in 10 without extending
reaction time and increasing the amount of H2O2 and/or
catalyst. Besides, safety evaluation of the reaction was done. To
avoid the combustion, at least either one of three elements of
combustion (oxygen, heat, and fuel) must be removed. As
oxygen gas was continuously released by decomposition of
H2O2 in the reaction, we removed flammable toluene before
the reaction (partial toluene vapor pressure/atmospheric
pressure ≤0.012). Furthermore, reaction heat could be
controlled by addition of H2O2 (Figure 3), and the maximum
temperature attainable by the synthesis reaction (MTSR; 87.6
°C) was lower than the boiling point of water (100 °C) (total
reaction heat; 332 kJ/mol, process temperature; 25 °C
adiabatic temperature; 62.6 °C). They suggested that the
reaction could be performed safely.
Figure 4. Stability of crude 1 dissolved in acetone and acetone/water
at 55 °C.
Deviation in the pilot manufacturing. In the pilot
manufacturing the impurity 10 in purified 1 was 0.08 w/w %
∼ 0.09 w/w %, which did not meet the specification of not
more than 0.06 w/w %. It was controlled at 0.03 w/w % ∼ 0.04
w/w % in the previous pilot manufacturing for the clinical
studies, and the flask experiment conducted using the improved
process. Since it seemed that the cause lay in the amidation
reaction and/or the recrystallization step, we investigated them
during the dissolving step was the cause of the deviation. We
did not pay attention to the hold time of the dissolution and
could not find the issue in the flask experiment conducted
before the pilot manufacturing.
Purification by solvent-mediated polymorphic trans-
formation to remove acid component enclosed in crude 1.
It was an option that the dissolving solvent was reverted to
100% acetone but the throughput was poor, and the risk of
C
DOI: 10.1021/acs.oprd.5b00023
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
deviation from the specification limit could not be completely
removed because residual 10 content varied in the flask
experiments. Considering these points, we decided to build the
process that can reliably remove 10 using 95 v/v % acetone/
water. The dissolution of crude 1 in 95 v/v % acetone/water
was highly acidic (about pH 1) because crude 1 included 0.02−
0.14 equiv of HCl and/or H2SO3 derived from SOCl2, and it
was thought to cause acidic hydrolysis.11 We considered that
removing the residual acid would decrease the hydrolysis. At
first, we tried purging the acidic gas from the reaction mixture
by nitrogen and increasing the amount of pyridine, but they
were not effective. Increasing the amount of washing liquid of
the filtrated crude 1 cake could reduce the rate of hydrolysis but
not stop it completely. From these insights, we assumed that
the acid component would be enclosed in the crystal, which
meant that crude 1 had to be dissolved to remove acid. As
additional operation should be minimized, it was unwise to
repeat the recrystallization process. On the other hand, solvent-
mediated polymorphic transformation by slurrying consists of
the dissolution of the unstable form and the nucleation and
growth of stable form. By the method, crude 1 was expected to
be dissolved by a small amount of solvent keeping low
temperature and concentration, which would prevent hydrol-
ysis. Since the crude 1 was form II, we searched the effective
solvent for transformation of form II to form I and found that it
was realized by slurrying in the mixture of acetone and water
for a short time.12 This led to sufficient removal of the residual
acid enclosed in the crystal and improved the stability of the
dissolution (Figure 5). However, the polymorph type of crude
and prevention of hydrolysis of 1. Furthermore, the removal
rate of 10 of the recrystallization step was expected to be
improved because the cause of the fluctuation in the removal
rate from 36% to 73% in the flask experiments was considered
to be the possible weak interaction of acid 10 with the basic
pyridyl group of 1; the salt formation could prevent this (Figure
6). Monitoring of the pilot manufacturing showed that excess
Figure 6. Possible interaction of 1 and 10.
base was needed since the residual acid included in crude 1
varied from 0.02 to 0.14 equiv due to the inconsistent washing
of the cake; 0.2 equiv of base was used to deal with the highest
amount of the residual acid. The pKa of the conjugated acid of
the base used for the neutralization needed to be more than 6
for completion of the salt exchange, since the calculated pKa of
10 was 4.2. However, the strong base sodium hydroxide could
not be used because its excess use caused basic hydrolysis of 1.
Then we examined several weak bases (Table 3). NaHCO3
Table 3. Salt Formation and Increase Amount of 10
a
entry
compd
pKa
Increase of 10 (w/w %)
1
2
3
4
5
compound 10
NaHCO3
4.2
6.35
4.76
3.86
9.25
0.04
0.02
0.00
0.01
sodium acetate
sodium lactate
ammmonia
a
Crude 1 and 0.2 equiv of base in 95% acetone/water were heated at
55 °C for 6 h.
(pKa of carbon dioxide at 25 °C; 6.35) also caused slight basic
hydrolysis, but the removal rate was improved to 80% due to
the completion of salt formation (Entry 2). Sodium acetate
(pKa of acetic acid at 25 °C; 4.76) and sodium lactate (pKa of
lactic acid at 25 °C; 3.86) made the crude 1 dissolution stable,
but the pKa was not sufficient for salt formation and the
generated acid could be a new impurity (Entries 3 and 4).
Ammonia (pKa of the conjugate acid; 9.25) made the
dissolution stable, probably because any excess was volatilized
though its pKa was high (Entry 5). We hesitated to use it
because of acute aquatic toxicity with a long-lashing effect.
Considering these points, we chose NaHCO3 for further
examination. In the pilot manufacturing the variant which could
critically influence the stability was the acid content in crude 1
because it was difficult to strictly control the cake washing. The
stabilities of crude 1 including various equivalents of residual
acid dissolved in 95% acetone/water with 0.2 equiv of NaHCO3
are shown in Figure 7. The dissolution including more acid did
not show decomposition, but a slight amount of basic
hydrolysis was observed using crude 1 with less acid. After
the salt formation could eliminate the possibility of
intermolecular interaction, the driving force of separation was
the difference of the respective solubility. The solubility of form
I at 58 °C and the sodium salt of 10 and the concentration
Figure 5. Stability of crude and semipurified 1 dissolved in acetone/
water at 55 °C.
1 washed by 50% acetone/water changed from form II to form
I within 2 days at room temperature. Since it was crucial to
maintain the polymorph form before the slurry washing, we
investigated the other washing liquids to extend the lifetime of
form II and found washing it with purified water allowed it to
be maintained for at least 12 days at room temperature. By the
clever polymorph control using only acetone and water, the
purified 1 including 0.00% of 10 could be produced in the pilot
manufacturing.
Dramatic improvement of removal rate of 10 by salt
formation and rational building of the recrystallization
procedure based on its solubility. Toward commercial supply,
we tried to eliminate the solvent-mediated polymorph trans-
formation because it could increase the costs of operation and
facility installation. As the hydrolysis was caused by the residual
acid, we tried neutralizing crude 1 dissolution in 95 v/v %
acetone/water. This was expected to lead to stable dissolution
D
DOI: 10.1021/acs.oprd.5b00023
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
bottleneck volume but caused the increase of residual 10 in
purified 1 in the pilot manufacturing. After finding it, we
developed two reliable purification processes. The first was
slurry washing including clever polymorphic control using only
acetone and water. The second was salt formation of 10 to
improve stability and build a rational recrystallization process
based on solubility by excluding the possibility of ionic
interaction of 1 and 10. The latter process led to reliable
control of the impurity 10 without increasing maximum
volume.
EXPERIMENTAL SECTION
■
All experiments were run under a nitrogen atmosphere.
Solvents and reagents were obtained from commercial sources
and used without further purification. High performance liquid
chromatographic (HPLC) analysis was carried out using a
Shimadzu LC-10ADVP. Reaction heat was measured using
Figure 7. Stability of crude 1 (influence of residual acid).
transition of the modified process are shown in Figure 8. In the
existing process purified 1 was precipitated from the dissolution
1
Mettler Toledo RC1e. H NMR spectra were recorded on a
300 MHz Varian FT spectrometer. XRPD was measured using
Rigaku RINT TTR III. Melting points were measured with
Rigaku Industrial Corporation Thermo plus DSC8230S.
Infrared spectra were recorded with Thermo Fisher Scientific
MAGNA560. The solvent composition after distillation was
simulated using CHEMCAD 5.3.0
Preparation of the intermediate 10 by the improved
process. To the reactor water (660 kg) were charged 12 (330
kg, 1589 mol) and toluene (828 kg) after the nitrogen purge.
The mixture was adjusted to 15 °C (internal temperature), and
triethylamine (354 kg, 3495 mol) was charged. To the mixture
was charged slowly 4 (232 kg 1652 mol) in toluene (285 kg).
After stirring for 1 h, the reaction mixture was adjusted to 25
°C and the aqueous layer was removed. To the toluene layer
were added water (653 kg) and 25% aq NaOH (636 kg, 3972
mol) at 25 °C. Then it was heated to 45 °C, stirred for 2 h, and
cooled to 25 °C. After the toluene layer was removed, the
aqueous layer was concentrated to remove residual toluene and
ethanol. To the aqueous solution was added 20 w/w % aq
H2SO4 to adjust the pH 9.0. After the addition of catalyst
Na2WO4·2H2O (5.2 kg, 16 mol) was charged slowly 35% H2O2
aqueous solution (309 kg, 3178 mol) at 25 °C, and the mixture
was stirred for 2.5 h. Then 13.2 w/w % aq H2SO3 (1365 kg)
was added to quench the reaction. It was heated to 40 °C, and
20% aq H2SO4 was charged to adjust the pH 2.5. The slurry
was cooled to 25 °C and stirred for 30 min. The precipitated
intermediate 10 was filtered, washed with water, and dried in
vacuum (397 kg, 95%). 10 is already known, and the 1H NMR
data corresponded to that reported.5 1H NMR (300 MHz,
CDCl3) δ 3.99 (d, 1H, J = 9 Hz), 3.67 (s, 3H), 3.27 (m, 1H),
2.00−2.32 (m, 5H), 1.49−1.62 (m, 2H), 1.39 (s, 3H), 1.16−
1.32 (m, 2H).
Preparation of purified 1 by the improved process. 10
(475 kg, 1804 mol) was charged with a catalytic amount of
DMF (2.6 kg, 36 mol) and toluene (1604 kg). The resulting
slurry was heated to about 50 °C, and then thionyl chloride
(236 kg, 1984 mol) was dropped slowly. After the reaction
mixture was stirred about 30 min, it was cooled to 30 °C. Next,
pyridylamine (11) (322 kg, 1984 mol) and pyridine (314 kg,
3968 mol) dissolved in toluene were added, and the resulting
slurry was stirred about 1 h. After the reaction was completed,
purified water (1425 L) was added. The crude 1, whose
polymorph was form II, was filtered and washed with purified
water. Half the amount of crude 1 was used in the following
step. To crude 1 and acetone (3040 kg) was added sodium
Figure 8. Modified concentration transition of the recrystallization
process.
in 95% acetone/water during the concentration but the
solubility of the sodium salt in the ratio was low. The process
could cause the precipitation of 10 salt, and it could be
enclosed in the crystal of 1. To improve the removal effect, the
precipitation point of form I was set at 80 v/v % acetone/water
in which the solubility of sodium salt of 10 was sufficiently
large. The first concentration was performed under the
solubility of form I to certainly prevent the precipitation.
After that water was added to 80% acetone/water of
supersaturation concentration where visible nucleation of
form I occurred. Then concentration, water addition, and
cooling were followed. By the changes, the removal rate of 10
was dramatically and reproducibly increased to more than 83%.
Using the recrystallization procedure, we could produce
purified 1 with no 10 at an average 94.7% yield (348 kg/lot,
6 lots). The rational building of the recrystallization procedure
led to the highly reliable control of impurity 10 without
increasing maximum volume.
CONCLUSION
■
An optimized process to manufacture 1 could be developed.
The telescopic process to synthesize 10 was realized by
changing the starting material, coupling reaction conditions,
and the order of oxidation and hydrolysis reaction from the
early process. Then we found the interesting knowledge that
Na2WO4/H2O2 oxidation could be accelerated in weakly basic
conditions, and it was useful to control the residual 14 in 10.
Changing the dissolving solvent from acetone to 95 v/v %
acetone/water in the recrystallization step reduced the
E
DOI: 10.1021/acs.oprd.5b00023
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
obtained slurry was cooled to 10 °C and stirred for 60 min. The
purified 1 was filtered, washed with 50 v/v % of acetone/water, and
dried in vacuum.
(10) The reaction mixtures were quenched by methanol and analyzed
using HPLC (215 nm).
(11) The quantity of the included acid was determined as follows.
1.00 g of wet crude 1 was measured accurately, which was dissolved in
30 mL of 95% acetone/water. To the solution was added dropwise
0.001 N sodium hydoroxide solution. The end point of neutralization
was determined by pH meter. The number of equivalents was
calculated based on dried crude S-2367.
(12) It was monitored by the concentration transition of mother
liquor and the specific peak strength 1686 cm−1 of form II and 1708
cm−1 of form I in the infrared spectrum. Sugata, Y.; Ide, Y.; Sakata, T.;
Omura, S. WO2009136617A1.
bicarbonate (15 kg, 180 mol) dissolved in purified water (165
L), and the resulting mixture was adjusted to 55 °C (internal
temperature) and stirred for about 1 h. The precipitating salt
was filtered and washed with acetone (290 kg). Then acetone
was distilled until the internal volume was reduced to 4023 L,
and then purified water (735 L) was added. The slurry was
distilled until the internal volume was reduced to 3032 L,
purified water (1011 L) was added, and the reaction mixture
was cooled to 10 °C (internal temperature) and stirred for 60
min. The purified 1 was filtered, washed with 50 v/v % of
acetone/water, and dried in vacuum (348 kg; 94.7%). S-2367
1
(1) is already known, and the H NMR data corresponded to
that reported.2 1H NMR (300 MHz, DMSO-d6) δ 10.84 (1H,
s), 8.69 (1H, s), 8.28 (1H, d, J = 8.7 Hz), 8.10−8.18 (1H, m),
6.79 (1H, d, J = 8.4 Hz), 3.00−3.15 (1H, m), 2.40−2.54 (1H,
m), 1.82−2.00 (4H, m), 1.25−1.75 (4H, m), 1.27 (9H, s),
melting point; 224.6 °C, crude S-2367 (form II) 2θ = 12.5
0.2, 14.3 0.2, 16.8 0.2, 17.8 0.2, 18.4 0.2, 19.0 0.2,
21.6 0.2, purified S-2367 (form I) 2θ = 10.2 0.2, 14.5
0.2, 16.8 0.2, 17.1 0.2, 17.7 0.2, 18.3 0.2, 19.1 0.2,
20.2 0.2, 20.8 0.2, 26.5 0.2, melting point; 219.6 °C.
ASSOCIATED CONTENT
* Supporting Information
XRPD of form I and form II. This material is available free of
charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
* E-mail: shinichi.oda@shionogi.co.jp.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to T. Masui, K. Yoshida, A. Sugimoto, Y.
Kitaura, T. Fukui, and T. Unno for helpful discussions and their
contributions to this research.
REFERENCES
■
(1) Levens, N. R.; Della-Zuana, O. Curr. Opin. Investig. Drugs 2003, 4,
1198−1204.
(2) Kawanishi, Y.; Takenaka, H.; Hanasaki, K.; Okada, T.
WO0137826A1.
(3) Yukioka, H. WO2008047769.
(4) Kobayashi, Y.; Yoshioka, Y. WO2013002313A1.
(5) Kawanishi, Y.; Uenaka, M.; Matsuura, M. WO03076374A1.
(6) Masui, T.; Yoshida, K. WO2006106800A1.
(7) Prat, D.; Pardigon, O.; Flemming, H. W.; Letestu, S.; Ducandas,
V.; Isnard, P.; Guntrum, E.; Senac, T.; Ruisseau, S.; Cruciani, P.;
Hosek, P. Org. Process. Res. Dev. 2013, 17, 1517−1525.
(8) Hida, T.; Nogusa, H. Tetrahedron 2009, 65, 270−274.
(9) The whole procedure of the step was as follows. To the reactor
were charged 10, DMF (0.02 equiv), and toluene (4 V). The resulting
slurry was heated to 50 °C, and then thionyl chloride (1.1 equiv) was
charged slowly. After the reaction mixture had been stirred for 30 min,
it was cooled to 30 °C. To the reaction mixture was charged a solution
of 11 (1.1 equiv) and pyridine (2.2 equiv) in toluene, and then this
was stirred for 30 min. After the reaction was completed, purified
water (3 V) was added. The crystallized crude 1, whose polymorph
was form II, was filtered and washed with 50% acetone/water (2 V)
and purified water (2 V). The obtained crude 1 was dissolved in
acetone (10.5 V) and purified water (0.45V) at 55 °C, filtered to
remove the foreign matter, and flushed with acetone (1 V). Next,
acetone was removed by distillation, and the residue was adjusted to 5
wt %. To this was added purified water (5 V) at 55 °C, and the
F
DOI: 10.1021/acs.oprd.5b00023
Org. Process Res. Dev. XXXX, XXX, XXX−XXX